content
stringlengths 4
1.04M
| lang
stringclasses 358
values | score
int64 0
5
| repo_name
stringlengths 5
114
| repo_path
stringlengths 4
229
| repo_licenses
listlengths 1
8
|
|---|---|---|---|---|---|
#include <PJONSwitch.h>
#include <PJONSoftwareBitBang.h>
/* Route packets between a collection of buses with the same
or different strategies or media.
PJONSwitch can route between buses of different strategies,
though not shown here. If using only one strategy, some
resources can be saved by using PJONSimpleSwitch instead.
__________ ________ __________
| | Pin 7 | | Pin 12 | |
| DEVICE 1 |_______________| SWITCH |_______________| DEVICE 2 |
|__________| Bus 0.0.0.1 |________| Bus 0.0.0.2 |__________| */
PJONSwitch2<SoftwareBitBang, SoftwareBitBang> router;
void setup() {
router.get_strategy_0().set_pin(7);
router.get_strategy_1().set_pin(12);
router.get_bus(0).set_bus_id((const uint8_t[4]){0, 0, 0, 1});
router.get_bus(1).set_bus_id((const uint8_t[4]){0, 0, 0, 2});
router.begin();
}
void loop() {
router.loop();
}
| Arduino | 4 | solhuebner/PJON | examples/routing/ARDUINO/Network/Switch/Switch/Switch.ino | [
"Apache-2.0"
]
|
theory Leftpad
imports Main
begin
(* What does it mean to be padded? A string can be split into two parts where the first part
is all padding and the second is the original string: *)
definition isPadded where "isPadded padChar unpadded padded
\<equiv> \<exists> n. set (take n padded) \<subseteq> { padChar } \<and> drop n padded = unpadded"
(* The star of the show is the `leftPad` function ... *)
definition leftPad where "leftPad padChar targetLength s
\<equiv> replicate (targetLength - length s) padChar @ s"
(* ... which satisfies the spec: *)
lemma isPadded_leftPad: "isPadded padChar s (leftPad padChar targetLength s)"
unfolding isPadded_def leftPad_def by (intro exI [where x = "targetLength - length s"], auto)
lemma length_leftPad: "length (leftPad padChar targetLength s) = max targetLength (length s)"
unfolding leftPad_def by auto
(* But, hold on, we used `replicate` from the standard library which made this very easy:
there are a lot of facts about `replicate` already proven. Let's assume we don't know about it:
what facts do we need? How about these: *)
locale replication =
fixes abstract_replicate
assumes length_replicate[simp]: "length (abstract_replicate n c) = n"
assumes set_replicate[simp]: "set (abstract_replicate n c) = (if n = 0 then {} else {c})"
(* This works: *)
context replication
begin
definition abstract_leftPad where "abstract_leftPad padChar targetLength s
\<equiv> abstract_replicate (targetLength - length s) padChar @ s"
lemma "isPadded padChar s (abstract_leftPad padChar targetLength s)"
unfolding isPadded_def abstract_leftPad_def by (intro exI [where x = "targetLength - length s"], auto)
lemma "length (abstract_leftPad padChar targetLength s) = max targetLength (length s)"
unfolding abstract_leftPad_def by auto
end
(* Of course it's important that the real `replicate` function satisfies this spec: *)
interpretation stdlib: replication replicate by (unfold_locales, auto)
(* However, if we didn't know about it then we could write our own quite easily: *)
fun myReplicate :: "nat \<Rightarrow> 'a \<Rightarrow> 'a list" where
"myReplicate 0 c = []"
| "myReplicate (Suc n) c = c # myReplicate n c"
(* ... and show it satisfies the spec too: *)
interpretation mine: replication myReplicate proof
fix n and c :: 'a
show "length (myReplicate n c) = n" by (induct n, auto)
show "set (myReplicate n c) = (if n = 0 then {} else {c})" by (induct n, auto)
qed
(* In fact, the `length_replicate` and `set_replicate` functions are a complete description of
`replicate`, in the sense that `replicate` is the only function that satisfies them: *)
context replication
begin
lemma "abstract_replicate = replicate"
proof (intro ext)
fix n :: nat and c :: 'a
{
fix cs
assume p: "length cs = n" "set cs = (if n = 0 then {} else {c})"
hence "set cs = (if length cs = 0 then {} else {c})" by simp
hence "cs = replicate (length cs) c"
proof (induct cs)
case Nil show ?case by simp
next
case (Cons c0 cs)
from Cons.prems have "set cs = (if length cs = 0 then {} else {c})" by (cases cs, auto)
with Cons.hyps have hyp: "cs = replicate (length cs) c" by simp
with Cons.prems show ?case by simp
qed
}
note p = this
have "abstract_replicate n c = replicate (length (abstract_replicate n c)) c" by (intro p, simp_all)
thus "abstract_replicate n c = replicate n c" by simp
qed
end
(* Similarly, the spec for `leftPad` is a complete description of it, in the sense that `leftPad`
is the only function that satisfies it: *)
lemma leftpad_unique:
assumes isPadded_leftPad': "\<And>padChar targetLength s. isPadded padChar s (leftPad' padChar targetLength s)"
assumes length_leftPad': "\<And>padChar targetLength s. length (leftPad' padChar targetLength s) = max targetLength (length s)"
shows "leftPad' = leftPad"
proof (intro ext)
fix padChar :: 'a and targetLength s
from isPadded_leftPad [of padChar] obtain n where n:
"set (take n (leftPad padChar targetLength s)) \<subseteq> { padChar }"
"drop n (leftPad padChar targetLength s) = s"
unfolding isPadded_def by blast
from isPadded_leftPad' obtain n' where n':
"set (take n' (leftPad' padChar targetLength s)) \<subseteq> { padChar }"
"drop n' (leftPad' padChar targetLength s) = s"
unfolding isPadded_def by blast
{
fix xs xs'
assume "set xs \<subseteq> { padChar }" "set xs' \<subseteq> { padChar }" "length xs = length xs'"
hence "xs = xs'"
proof (induct xs arbitrary: xs')
case Nil thus ?case by simp
next
case (Cons x xs xxs')
from Cons obtain x' xs' where xxs': "xxs' = x' # xs'" by (cases xxs', auto)
with Cons show ?case by auto
qed
}
note padding_eq = this
have append_cong: "\<And>ws xs ys zs. ws = ys \<Longrightarrow> xs = zs \<Longrightarrow> ws @ xs = ys @ zs" by simp
have "leftPad' padChar targetLength s = take n' (leftPad' padChar targetLength s) @ drop n' (leftPad' padChar targetLength s)" by simp
also have "... = take n (leftPad padChar targetLength s) @ drop n (leftPad padChar targetLength s)"
proof (intro append_cong)
from n n' show "drop n' (leftPad' padChar targetLength s) = drop n (leftPad padChar targetLength s)" by simp
have "length (take n' (leftPad' padChar targetLength s)) = length (take n (leftPad padChar targetLength s))"
proof (cases s)
case Cons
from n have "length s = length (drop n (leftPad padChar targetLength s))" by simp
also have "... = length (leftPad padChar targetLength s) - n" by simp
also from length_leftPad [of padChar] have "... = max targetLength (length s) - n" by simp
finally have n_eq: "n = max targetLength (length s) - length s" using Cons by auto
from n' have "length s = length (drop n' (leftPad' padChar targetLength s))" by simp
also have "... = length (leftPad' padChar targetLength s) - n'" by simp
also from length_leftPad' have "... = max targetLength (length s) - n'" by simp
finally have n'_eq: "n' = max targetLength (length s) - length s" using Cons by auto
show ?thesis by (simp add: length_leftPad length_leftPad' n_eq n'_eq)
next
case Nil
with n n' length_leftPad [of padChar] length_leftPad'
show ?thesis by auto
qed
thus "take n' (leftPad' padChar targetLength s) = take n (leftPad padChar targetLength s)"
by (intro padding_eq n n')
qed
also have "... = leftPad padChar targetLength s" by simp
finally show "leftPad' padChar targetLength s = leftPad padChar targetLength s" .
qed
end
| Isabelle | 5 | RomainGehrig/lets-prove-leftpad | isabelle/Leftpad.thy | [
"CC0-1.0"
]
|
/*
* Copyright (c) 2016-present, RxJava Contributors.
*
* Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in
* compliance with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software distributed under the License is
* distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See
* the License for the specific language governing permissions and limitations under the License.
*/
/**
* Classes representing so-called hot sources, aka <strong>subjects</strong>, that implement a base reactive class and
* the respective consumer type at once to allow forms of multicasting events to multiple
* consumers as well as consuming another base reactive type of their kind.
* <p>
* Available subject classes with their respective base classes and consumer interfaces:
* <br>
* <table border="1" style="border-collapse: collapse;" summary="The available subject classes with their respective base classes and consumer interfaces.">
* <tr><td><b>Subject type</b></td><td><b>Base class</b></td><td><b>Consumer interface</b></td></tr>
* <tr>
* <td>{@link io.reactivex.rxjava3.subjects.Subject Subject}
* <br> {@link io.reactivex.rxjava3.subjects.AsyncSubject AsyncSubject}
* <br> {@link io.reactivex.rxjava3.subjects.BehaviorSubject BehaviorSubject}
* <br> {@link io.reactivex.rxjava3.subjects.PublishSubject PublishSubject}
* <br> {@link io.reactivex.rxjava3.subjects.ReplaySubject ReplaySubject}
* <br> {@link io.reactivex.rxjava3.subjects.UnicastSubject UnicastSubject}
* </td>
* <td>{@link io.reactivex.rxjava3.core.Observable Observable}</td>
* <td>{@link io.reactivex.rxjava3.core.Observer Observer}</td>
* </tr>
* <tr>
* <td>{@link io.reactivex.rxjava3.subjects.SingleSubject SingleSubject}</td>
* <td>{@link io.reactivex.rxjava3.core.Single Single}</td>
* <td>{@link io.reactivex.rxjava3.core.SingleObserver SingleObserver}</td>
* </tr>
* <tr>
* <td>{@link io.reactivex.rxjava3.subjects.MaybeSubject MaybeSubject}</td>
* <td>{@link io.reactivex.rxjava3.core.Maybe Maybe}</td>
* <td>{@link io.reactivex.rxjava3.core.MaybeObserver MaybeObserver}</td>
* </tr>
* <tr>
* <td>{@link io.reactivex.rxjava3.subjects.CompletableSubject CompletableSubject}</td>
* <td>{@link io.reactivex.rxjava3.core.Completable Completable}</td>
* <td>{@link io.reactivex.rxjava3.core.CompletableObserver CompletableObserver}</td>
* </tr>
* </table>
* <p>
* The backpressure-aware variants of the {@code Subject} class are called
* {@link org.reactivestreams.Processor}s and reside in the {@code io.reactivex.processors} package.
* @see io.reactivex.rxjava3.processors
*/
package io.reactivex.rxjava3.subjects;
| Java | 4 | Jawnnypoo/RxJava | src/main/java/io/reactivex/rxjava3/subjects/package-info.java | [
"Apache-2.0"
]
|
/**
* General purpose sample objects that can be used with tests.
*/
package org.springframework.tests.sample.objects;
| Java | 1 | nicchagil/spring-framework | spring-core/src/test/java/org/springframework/tests/sample/objects/package-info.java | [
"Apache-2.0"
]
|
rule display_help
command = echo 'Error: Execute Ninja in a build directory: `ninja -C build/release`'
description = Help the poor user
build help: display_help
default help
| Ninja | 3 | liubangchen/seastar | build.ninja | [
"Apache-2.0"
]
|
FROM tomcat:jdk16-openjdk
COPY . /usr/local/tomcat/webapps/
RUN ["javac", "-cp", ".:/usr/local/tomcat/lib/servlet-api.jar", "-d", "/usr/local/tomcat/webapps/private/WEB-INF/classes/", "/usr/local/tomcat/webapps/private/HelloWorld.java"]
EXPOSE 8080
CMD ["catalina.sh", "run"]
| Dockerfile | 4 | rveerama1/istio | tests/integration/security/fuzz/backends/tomcat/Dockerfile | [
"Apache-2.0"
]
|
= Induction : Proof by Induction
> module Induction
First, we import all of our definitions from the previous chapter.
> import Basics
Next, we import the following \idr{Prelude} modules, since we'll be dealing with
natural numbers.
> import Prelude.Interfaces
> import Prelude.Nat
For \idr{import Basics} to work, you first need to use `idris` to compile
`Basics.lidr` into `Basics.ibc`. This is like making a .class file from a .java
file, or a .o file from a .c file. There are at least two ways to do it:
- In your editor with an Idris plugin, e.g. [Emacs][idris-mode]:
Open `Basics.lidr`. Evaluate \idr{idris-load-file}.
There exists similar support for [Vim][idris-vim], [Sublime
Text][idris-sublime] and [Visual Studio Code][idris-vscode] as well.
[idris-mode]: https://github.com/idris-hackers/idris-mode
[idris-vim]: https://github.com/idris-hackers/idris-vim
[idris-sublime]: https://github.com/idris-hackers/idris-sublime
[idris-vscode]: https://github.com/zjhmale/vscode-idris
- From the command line:
Run \mintinline[]{sh}{idris --check --total --noprelude src/Basics.lidr}.
Refer to the Idris man page (or \mintinline[]{sh}{idris --help} for
descriptions of the flags.
> %access public export
> %default total
== Proof by Induction
We proved in the last chapter that \idr{0} is a neutral element for \idr{+} on
the left using an easy argument based on simplification. The fact that it is
also a neutral element on the _right_...
```coq
Theorem plus_n_O_firsttry : forall n:nat,
n = n + 0.
```
... cannot be proved in the same simple way in Coq, but as we saw in `Basics`,
Idris's \idr{Refl} just works.
To prove interesting facts about numbers, lists, and other inductively defined
sets, we usually need a more powerful reasoning principle: _induction_.
Recall (from high school, a discrete math course, etc.) the principle of
induction over natural numbers: If \idr{p n} is some proposition involving a
natural number \idr{n} and we want to show that \idr{p} holds for _all_ numbers
\idr{n}, we can reason like this:
- show that \idr{p Z} holds;
- show that, for any \idr{k}, if \idr{p k} holds, then so does \idr{p (S k)};
- conclude that \idr{p n} holds for all \idr{n}.
In Idris, the steps are the same and can often be written as function clauses by
case splitting. Here's how this works for the theorem at hand.
> plus_n_Z : (n : Nat) -> n = n + 0
> plus_n_Z Z = Refl
> plus_n_Z (S k) =
> let inductiveHypothesis = plus_n_Z k in
> rewrite inductiveHypothesis in Refl
In the first clause, \idr{n} is replaced by \idr{Z} and the goal becomes \idr{0
= 0}, which follows by \idr{Refl}exivity. In the second, \idr{n} is replaced by
\idr{S k} and the goal becomes \idr{S k = S (plus k 0)}. Then we define the
inductive hypothesis, \idr{k = k + 0}, which can be written as \idr{plus_n_Z k},
and the goal follows from it.
> minus_diag : (n : Nat) -> minus n n = 0
> minus_diag Z = Refl
> minus_diag (S k) = minus_diag k
==== Exercise: 2 stars, recommended (basic_induction)
Prove the following using induction. You might need previously proven results.
> mult_0_r : (n : Nat) -> n * 0 = 0
> mult_0_r n = ?mult_0_r_rhs
> plus_n_Sm : (n, m : Nat) -> S (n + m) = n + (S m)
> plus_n_Sm n m = ?plus_n_Sm_rhs
> plus_comm : (n, m : Nat) -> n + m = m + n
> plus_comm n m = ?plus_comm_rhs
> plus_assoc : (n, m, p : Nat) -> n + (m + p) = (n + m) + p
> plus_assoc n m p = ?plus_assoc_rhs
$\square$
==== Exercise: 2 stars (double_plus)
Consider the following function, which doubles its argument:
> double : (n : Nat) -> Nat
> double Z = Z
> double (S k) = S (S (double k))
Use induction to prove this simple fact about \idr{double}:
> double_plus : (n : Nat) -> double n = n + n
> double_plus n = ?double_plus_rhs
$\square$
==== Exercise: 2 stars, optional (evenb_S)
One inconvenient aspect of our definition of \idr{evenb n} is that it may need
to perform a recursive call on \idr{n - 2}. This makes proofs about \idr{evenb
n} harder when done by induction on \idr{n}, since we may need an induction
hypothesis about \idr{n - 2}. The following lemma gives a better
characterization of \idr{evenb (S n)}:
> evenb_S : (n : Nat) -> evenb (S n) = not (evenb n)
> evenb_S n = ?evenb_S_rhs
$\square$
== Proofs Within Proofs
\ \todo[inline]{Edit the section}
In Coq, as in informal mathematics, large proofs are often broken into a
sequence of theorems, with later proofs referring to earlier theorems. But
sometimes a proof will require some miscellaneous fact that is too trivial and
of too little general interest to bother giving it its own top-level name. In
such cases, it is convenient to be able to simply state and prove the needed
"sub-theorem" right at the point where it is used. The `assert` tactic allows us
to do this. For example, our earlier proof of the \idr{mult_0_plus} theorem
referred to a previous theorem named \idr{plus_Z_n}. We could instead use
`assert` to state and prove \idr{plus_Z_n} in-line:
> mult_0_plus' : (n, m : Nat) -> (0 + n) * m = n * m
> mult_0_plus' n m = Refl
The `assert` tactic introduces two sub-goals. The first is the assertion itself;
by prefixing it with `H:` we name the assertion `H`. (We can also name the
assertion with `as` just as we did above with `destruct` and `induction`, i.e.,
`assert (0 + n = n) as H`.) Note that we surround the proof of this assertion
with curly braces `{ ... }`, both for readability and so that, when using Coq
interactively, we can see more easily when we have finished this sub-proof. The
second goal is the same as the one at the point where we invoke `assert` except
that, in the context, we now have the assumption `H` that `0 + n = n`. That is,
`assert` generates one subgoal where we must prove the asserted fact and a
second subgoal where we can use the asserted fact to make progress on whatever
we were trying to prove in the first place.
The `assert` tactic is handy in many sorts of situations. For example, suppose
we want to prove that `(n + m) + (p + q) = (m + n) + (p + q)`. The only
difference between the two sides of the `=` is that the arguments `m` and `n` to
the first inner `+` are swapped, so it seems we should be able to use the
commutativity of addition (`plus_comm`) to rewrite one into the other. However,
the `rewrite` tactic is a little stupid about _where_ it applies the rewrite.
There are three uses of `+` here, and it turns out that doing `rewrite ->
plus_comm` will affect only the _outer_ one...
```idris
plus_rearrange_firsttry : (n, m, p, q : Nat) ->
(n + m) + (p + q) = (m + n) + (p + q)
plus_rearrange_firsttry n m p q = rewrite plus_comm in Refl
```
```
When checking right hand side of plus_rearrange_firsttry with expected type
n + m + (p + q) = m + n + (p + q)
_ does not have an equality type ((n1 : Nat) ->
(n1 : Nat) -> plus n1 m1 = plus m1 n1)
```
To get \idr{plus_comm} to apply at the point where we want it to, we can
introduce a local lemma using the \idr{where} keyword stating that \idr{n + m =
m + n} (for the particular \idr{m} and \idr{n} that we are talking about here),
prove this lemma using \idr{plus_comm}, and then use it to do the desired
rewrite.
> plus_rearrange : (n, m, p, q : Nat) ->
> (n + m) + (p + q) = (m + n) + (p + q)
> plus_rearrange n m p q = rewrite plus_rearrange_lemma n m in Refl
> where
> plus_rearrange_lemma : (n, m : Nat) -> n + m = m + n
> plus_rearrange_lemma = plus_comm
== More Exercises
==== Exercise: 3 stars, recommended (mult_comm)
Use \idr{rewrite} to help prove this theorem. You shouldn't need to use
induction on \idr{plus_swap}.
> plus_swap : (n, m, p : Nat) -> n + (m + p) = m + (n + p)
> plus_swap n m p = ?plus_swap_rhs
Now prove commutativity of multiplication. (You will probably need to define and
prove a separate subsidiary theorem to be used in the proof of this one. You may
find that \idr{plus_swap} comes in handy.)
> mult_comm : (m, n : Nat) -> m * n = n * m
> mult_comm m n = ?mult_comm_rhs
$\square$
==== Exercise: 3 stars, optional (more_exercises)
\ \todo[inline]{Edit}
Take a piece of paper. For each of the following theorems, first _think_ about
whether (a) it can be proved using only simplification and rewriting, (b) it
also requires case analysis (`destruct`), or (c) it also requires induction.
Write down your prediction. Then fill in the proof. (There is no need to turn in
your piece of paper; this is just to encourage you to reflect before you hack!)
> lte_refl : (n : Nat) -> True = lte n n
> lte_refl n = ?lte_refl_rhs
> zero_nbeq_S : (n : Nat) -> 0 == (S n) = False
> zero_nbeq_S n = ?zero_nbeq_S_rhs
> andb_false_r : (b : Bool) -> b && False = False
> andb_false_r b = ?andb_false_r_rhs
> plus_ble_compat_l : (n, m, p : Nat) ->
> lte n m = True -> lte (p + n) (p + m) = True
> plus_ble_compat_l n m p prf = ?plus_ble_compat_l_rhs
> S_nbeq_0 : (n : Nat) -> (S n) == 0 = False
> S_nbeq_0 n = ?S_nbeq_0_rhs
> mult_1_l : (n : Nat) -> 1 * n = n
> mult_1_l n = ?mult_1_l_rhs
> all3_spec : (b, c : Bool) ->
> (b && c) || ((not b) || (not c)) = True
> all3_spec b c = ?all3_spec_rhs
> mult_plus_distr_r : (n, m, p : Nat) -> (n + m) * p = (n * p) + (m * p)
> mult_plus_distr_r n m p = ?mult_plus_distr_r_rhs
> mult_assoc : (n, m, p : Nat) -> n * (m * p) = (n * m) * p
> mult_assoc n m p = ?mult_assoc_rhs
$\square$
==== Exercise: 2 stars, optional (beq_nat_refl)
\ \todo[inline]{Edit}
Prove the following theorem. (Putting the \idr{True} on the left-hand side of
the equality may look odd, but this is how the theorem is stated in the Coq
standard library, so we follow suit. Rewriting works equally well in either
direction, so we will have no problem using the theorem no matter which way we
state it.)
> beq_nat_refl : (n : Nat) -> True = n == n
> beq_nat_refl n = ?beq_nat_refl_rhs
$\square$
==== Exercise: 2 stars, optional (plus_swap')
\ \todo[inline]{Edit}
The `replace` tactic allows you to specify a particular subterm to rewrite and
what you want it rewritten to: `replace (t) with (u)` replaces (all copies of)
expression `t` in the goal by expression `u`, and generates `t = u` as an
additional subgoal. This is often useful when a plain \idr{rewrite} acts on the
wrong part of the goal.
Use the `replace` tactic to do a proof of \idr{plus_swap'}, just like
\idr{plus_swap} but without needing `assert (n + m = m + n)`.
> plus_swap' : (n, m, p : Nat) -> n + (m + p) = m + (n + p)
> plus_swap' n m p = ?plus_swap__rhs
$\square$
==== Exercise: 3 stars, recommended (binary_commute)
Recall the \idr{incr} and \idr{bin_to_nat} functions that you wrote for the
\idr{binary} exercise in the `Basics` chapter. Prove that the following diagram
commutes:
bin --------- incr -------> bin
| |
bin_to_nat bin_to_nat
| |
v v
nat ---------- S ---------> nat
That is, incrementing a binary number and then converting it to a (unary)
natural number yields the same result as first converting it to a natural number
and then incrementing. Name your theorem \idr{bin_to_nat_pres_incr} ("pres" for
"preserves").
Before you start working on this exercise, please copy the definitions from your
solution to the \idr{binary} exercise here so that this file can be graded on
its own. If you find yourself wanting to change your original definitions to
make the property easier to prove, feel free to do so!
$\square$
==== Exercise: 5 stars, advanced (binary_inverse)
This exercise is a continuation of the previous exercise about binary numbers.
You will need your definitions and theorems from there to complete this one.
(a) First, write a function to convert natural numbers to binary numbers. Then
prove that starting with any natural number, converting to binary, then
converting back yields the same natural number you started with.
(b) You might naturally think that we should also prove the opposite direction:
that starting with a binary number, converting to a natural, and then back
to binary yields the same number we started with. However, this is not true!
Explain what the problem is.
(c) Define a "direct" normalization function -- i.e., a function \idr{normalize}
from binary numbers to binary numbers such that, for any binary number b,
converting to a natural and then back to binary yields \idr{(normalize b)}.
Prove it. (Warning: This part is tricky!)
Again, feel free to change your earlier definitions if this helps here.
$\square$
| Idris | 5 | diseraluca/software-foundations | src/Induction.lidr | [
"MIT"
]
|
//Read a text (non-ironman) EU4 savefile and scan for matters of interest. Provides info to networked clients.
//TODO: Show a coalition as if it's a war?
//TODO: Raise highlighting priority of building upgrade options by a console command - or hide upgrades
//by default, show them only when explicitly requested (like the building highlighter), and then always
//have it at PRIO_EXPLICIT
//TODO: Allow input on the primary console, stateless infodumps only
constant SAVE_PATH = "../.local/share/Paradox Interactive/Europa Universalis IV/save games";
constant PROGRAM_PATH = "../.steam/steam/steamapps/common/Europa Universalis IV"; //Append /map or /common etc to access useful data files
#ifdef QUIET
//Run "pike -DQUIET eu4_parse.pike ...." to avoid warnings from the LR Parser module. Obviously,
//parsing save files won't work in that form.
object parser;
#else
Parser.LR.Parser parser = Parser.LR.GrammarParser.make_parser_from_file("eu4_parse.grammar");
#endif
int retain_map_indices = 0;
class maparray {
//Hybrid mapping/array. Can have key-value pairs with string keys, and also an array
//of values, indexed numerically.
mapping keyed = ([]);
array indexed = ({ });
object addkey(string key, mixed value) {
if (retain_map_indices && mappingp(value)) value |= (["_index": sizeof(keyed)]);
keyed[key] = value;
return this;
}
object addidx(mixed value) {indexed += ({value}); return this;}
protected int _sizeof() {return sizeof(keyed) + sizeof(indexed);}
protected mixed `[](string|int key) {return intp(key) ? indexed[key] : keyed[key];}
protected mixed `[]=(string key, mixed val) {return keyed[key] = val;}
protected mixed `->(string key) {
switch (key) {
case "keyed": return keyed;
case "indexed": return indexed;
case "addkey": return addkey;
case "addidx": return addidx;
default: return keyed[key];
}
}
protected string _sprintf(int type, mapping p) {return sprintf("<%*O/%*O>", p, keyed, p, indexed);}
//Enable foreach(maparray();int i;mixed val) - but not, unfortunately, foreach(maparray,mixed val)
protected Array.Iterator _get_iterator() {return get_iterator(indexed);}
}
mapping|array|maparray coalesce(mixed ret_or_brace, mixed ret) {
if (ret_or_brace != "{") ret = ret_or_brace;
//Where possible, simplify a maparray down to just a map or an array
if (!sizeof(ret->indexed)) return ret->keyed;
if (!sizeof(ret->keyed)) return ret->indexed;
//Sometimes there's a mapping, but it also has an array of empty mappings after it.
if (Array.all(ret->indexed, mappingp) && !Array.any(ret->indexed, sizeof)) return ret->keyed;
return ret;
}
maparray makemapping(mixed name, mixed _, mixed val) {return maparray()->addkey(name, val);}
maparray addmapping(maparray map, mixed name, mixed _, mixed val) {
//Note that, sometimes, an array is defined by simply assigning multiple times.
//I have no way of distinguishing an array of one element in that form from a
//simple entry; and currently, since this is stateless, I can't properly handle
//an array of arrays.
if (arrayp(map[name])) map[name] += ({val});
else if (map[name]) map[name] = ({map[name], val});
else map->addkey(name, val);
return map;
}
maparray makearray(mixed val) {return maparray()->addidx(val);}
maparray addarray(maparray arr, mixed val) {return arr->addidx(val);}
mapping emptymaparray() {return ([]);}
mapping low_parse_savefile(string|Stdio.Buffer data, int|void verbose) {
if (stringp(data)) data = Stdio.Buffer(data); //NOTE: Restricted to eight-bit data. Since EU4 uses ISO-8859-1, that's not a problem. Be aware for future.
data->read_only();
string ungetch;
string|array next() {
if (string ret = ungetch) {ungetch = 0; return ret;}
data->sscanf("%*[ \t\r\n]");
while (data->sscanf( "#%*s\n%*[ \t\r\n]")); //Strip comments
if (!sizeof(data)) return "";
if (array str = data->sscanf("\"%[^\"]\"")) {
//How are embedded quotes and/or backslashes handled?
return ({"string", str[0]});
}
if (array digits = data->sscanf("%[-0-9.]")) return ({"string", digits[0]});
if (array|string word = data->sscanf("%[0-9a-zA-Z_]")) {
word = word[0];
if ((<"yes", "no">)[word]) return ({"boolean", word == "yes"});
//Hack: this one element seems to omit the equals sign for some reason.
if (word == "map_area_data") ungetch = "=";
return ({"string", word});
}
return data->read(1);
}
string|array shownext() {mixed tok = next(); write("%O\n", tok); return tok;}
//while (shownext() != ""); return 0; //Dump tokens w/o parsing
return parser->parse(verbose ? shownext : next, this);
}
//File-like object that reads from a string. Potentially does a lot of string copying.
class StringFile(string basis) {
int pos = 0;
int seek(int offset, string|void whence) {
switch (whence) {
case Stdio.SEEK_SET: pos = offset; break;
case Stdio.SEEK_CUR: pos += offset; break;
case Stdio.SEEK_END: pos = sizeof(basis) + offset; break;
case 0: pos = offset + sizeof(basis) * (offset < 0); break; //Default is SEEK_END if negative, else SEEK_SET
}
return pos;
}
int tell() {return pos;}
string(8bit) read(int len) {
string ret = basis[pos..pos+len-1];
pos += len;
return ret;
}
void stat() { } //No file system stats available.
}
mapping parse_savefile_string(string data, int|void verbose) {
if (has_prefix(data, "PK\3\4")) {
//Compressed savefile. Consists of three files, one of which ("ai") we don't care
//about. The other two can be concatenated after stripping their "EU4txt" headers,
//and should be able to be parsed just like an uncompressed save. (The ai file is
//also the exact same format, so if it's ever needed, just add a third sscanf.)
object zip = Filesystem.Zip._Zip(StringFile(data));
sscanf(zip->read("meta") || "", "EU4txt%s", string meta);
sscanf(zip->read("gamestate") || "", "EU4txt%s", string state);
if (meta && state) data = meta + state; else return 0;
}
else if (!sscanf(data, "EU4txt%s", data)) return 0;
if (verbose) write("Parsing %d bytes...\n", sizeof(data));
return low_parse_savefile(data, verbose);
}
mapping parse_savefile(string data, int|void verbose) {
sscanf(Crypto.SHA256.hash(data), "%32c", int hash);
string hexhash = sprintf("%64x", hash);
mapping cache = Standards.JSON.decode_utf8(Stdio.read_file("eu4_parse.json") || "{}");
if (cache->hash == hexhash) return cache->data;
mapping ret = parse_savefile_string(data, verbose);
Stdio.write_file("eu4_parse.json", string_to_utf8(Standards.JSON.encode((["hash": hexhash, "data": ret]))));
return ret;
}
mapping(string:string) L10n;
void parse_localisation(string data) {
data = utf8_to_string(data);
sscanf(data, "%*s\n%{ %s:%*d \"%s\"\n%}", array info);
L10n |= (mapping)info;
}
string tabulate(array(string) headings, array(array(mixed)) data, string|void gutter, int|void summary) {
if (!gutter) gutter = " ";
array info = ({headings}) + (array(array(string)))data;
array(int) widths = map(Array.transpose(info)) {return max(@sizeof(__ARGS__[0][*]));};
//Hack: First column isn't size-counted or guttered. It's for colour codes and such.
string fmt = sprintf("%%%ds", widths[1..][*]) * gutter;
//If there's a summary row, insert a ruler before it. (You can actually have multiple summary rows if you like.)
if (summary) info = info[..<summary] + ({({headings[0]}) + "\u2500" * widths[1..][*]}) + info[<summary-1..];
return sprintf("%{%s" + fmt + "\e[0m\n%}", info);
}
int threeplace(string value) {
//EU4 uses three-place fixed-point for a lot of things. Return the number as an integer,
//ie "3.142" is returned as 3142.
if (!value) return 0;
sscanf(value, "%d.%s", int whole, string frac);
return whole * 1000 + (int)sprintf("%.03s", frac);
}
int interest_priority = 0;
array(string) interesting_province = ({ });
enum {PRIO_UNSET, PRIO_SITUATIONAL, PRIO_IMMEDIATE, PRIO_EXPLICIT};
void interesting(string id, int|void prio) {
if (prio < interest_priority) return; //We've already had higher priority markers
if (prio > interest_priority) {interest_priority = prio; interesting_province = ({ });} //Replace with new highest prio
if (!has_value(interesting_province, id)) interesting_province += ({id}); //Retain order but avoid duplicates
}
mapping prov_area = ([]);
mapping province_info;
mapping building_types; array building_id;
mapping building_slots = ([]);
multiset(string) area_has_level3 = (<>);
void analyze_cot(mapping data, string name, string tag, function write) {
mapping country = data->countries[tag];
array maxlvl = ({ }), upgradeable = ({ }), developable = ({ });
foreach (country->owned_provinces, string id) {
mapping prov = data->provinces["-" + id];
if (!prov->center_of_trade) continue;
int dev = (int)prov->base_tax + (int)prov->base_production + (int)prov->base_manpower;
int need = prov->center_of_trade == "1" ? 10 : 25;
array desc = ({
sprintf("%s %04d %s", prov->owner, 9999-dev, prov->name),
prov->center_of_trade,
id,
sprintf("%s\tLvl %s\tDev %d\t%s", id, prov->center_of_trade, dev, string_to_utf8(prov->name)),
});
if (prov->center_of_trade == "3") {maxlvl += ({desc}); area_has_level3[prov_area[id]] = 1;}
else if (dev >= need) upgradeable += ({desc});
else developable += ({desc});
}
sort(maxlvl); sort(upgradeable); sort(developable);
int level3 = sizeof(country->merchants->envoy); //You can have as many lvl 3 CoTs as you have merchants.
if (sizeof(maxlvl)) write("Max level CoTs (%d/%d):\n%{%s\n%}\n", sizeof(maxlvl), level3, maxlvl[*][-1]);
else write("Max level CoTs: 0/%d\n", level3);
level3 -= sizeof(maxlvl);
string colorize(string color, array info, int prio) {
//Colorize if it's interesting. It can't be upgraded if not in a state; also, not all level 2s
//can become level 3s, for various reasons.
array have_states = data->map_area_data[prov_area[info[2]]]->?state->?country_state->?country;
if (!have_states || !has_value(have_states, tag)) return info[-1] + " [is territory]";
if (info[1] == "2") {
if (area_has_level3[prov_area[info[2]]]) return info[-1] + " [other l3 in area]";
if (level3-- <= 0) return info[-1] + " [need merchants]";
}
interesting(info[2], prio);
return color + info[-1];
}
if (sizeof(upgradeable)) write("Upgradeable CoTs:\n%{%s\e[0m\n%}\n", colorize("\e[1;32m", upgradeable[*], PRIO_IMMEDIATE));
if (sizeof(developable)) write("Developable CoTs:\n%{%s\e[0m\n%}\n", colorize("\e[1;36m", developable[*], PRIO_SITUATIONAL));
}
object calendar(string date) {
sscanf(date, "%d.%d.%d", int year, int mon, int day);
return Calendar.Gregorian.Day(year, mon, day);
}
void analyze_leviathans(mapping data, string name, string tag, function write) {
if (!has_value(data->dlc_enabled, "Leviathan")) return;
mapping country = data->countries[tag];
array projects = ({ });
foreach (country->owned_provinces, string id) {
mapping prov = data->provinces["-" + id];
if (!prov->great_projects) continue;
mapping con = prov->great_project_construction || ([]);
foreach (prov->great_projects, string project) {
mapping proj = data->great_projects[project];
projects += ({({
(int)id - (int)proj->development_tier * 10000,
({"", id, "Lvl " + proj->development_tier, prov->name, L10n[project],
con->great_projects != project ? "" : //If you're upgrading a different great project in this province, leave this one blank (you can't upgrade two at once)
sprintf("%s%d%%, due %s",
con->type == "2" ? "Moving: " : "", //Upgrades are con->type "1", moving to capital is type "2"
threeplace(con->progress) / 10, con->date),
}),
})});
//write("Project: %O\n", proj);
}
//if (con) write("Construction: %O\n", con);
}
sort(projects);
if (sizeof(projects)) write("%s\n", string_to_utf8(tabulate(({""}) + "ID Tier Province Project Upgrading" / " ", projects[*][-1], " ", 0)));
write("\nFavor cooldowns:\n");
object today = calendar(data->date);
array cooldowns = ({ });
mapping cd = country->cooldowns || ([]);
foreach ("gold men sailors" / " ", string tradefor) {
string date = cd["trade_favors_for_" + tradefor];
if (!date) {cooldowns += ({({"", "---", "--------", String.capitalize(tradefor)})}); continue;}
int days = today->distance(calendar(date)) / today;
cooldowns += ({({"", days, date, String.capitalize(tradefor)})});
}
write("%s\n", string_to_utf8(tabulate(({"", "Days", "Date", "Trade for"}), cooldowns, " ", 0)));
}
int count_building_slots(mapping data, string id) {
//Count building slots. Not perfect. Depends on the CoTs being provided accurately.
//Doesn't always give the terrain bonus.
int slots = 2 + building_slots[id]; //All cities get 2, plus possibly a bonus from terrain and/or a penalty from climate.
mapping prov = data->provinces["-" + id];
if (prov->buildings->?university) ++slots; //A university effectively doesn't consume a slot.
if (area_has_level3[prov_area[id]]) ++slots; //A level 3 CoT in the state adds a building slot
//TODO: Modifiers, incl event flags
int dev = (int)prov->base_tax + (int)prov->base_production + (int)prov->base_manpower;
return slots + dev / 10;
}
mapping(string:string) manufactories = ([]); //Calculated from building_types
void analyze_furnace(mapping data, string name, string tag, function write) {
mapping country = data->countries[tag];
array maxlvl = ({ }), upgradeable = ({ }), developable = ({ });
int seen = 0;
foreach (country->owned_provinces, string id) {
mapping prov = data->provinces["-" + id];
if (prov->trade_goods != "coal") continue;
if (!seen) {write("Coal-producing provinces:\n"); seen = 1;}
int dev = (int)prov->base_tax + (int)prov->base_production + (int)prov->base_manpower;
mapping bldg = prov->buildings || ([]);
mapping mfg = bldg & manufactories;
if (bldg->furnace) write("%s\tHas Furnace\tDev %d\t%s\n", id, dev, string_to_utf8(prov->name));
else if (building_id[(int)prov->building_construction->?building] == "furnace")
write("%s\t%s\tDev %d\t%s\n", id, prov->building_construction->date, dev, string_to_utf8(prov->name));
else if (sizeof(mfg)) write("\e[1;31m%s\tHas %s\tDev %d\t%s\e[0m\n", id, values(mfg)[0], dev, string_to_utf8(prov->name));
else {
int slots = count_building_slots(data, id);
int buildings = sizeof(bldg);
if (prov->building_construction) {
//There's something being built. That consumes a slot, but if it's an
//upgrade, then that slot doesn't really count. If you have four slots,
//four buildings, and one of them is being upgraded, the game will show
//that there are five occupied slots and none open; for us here, it's
//cleaner to show it as 4/4.
++buildings;
string upg = building_id[(int)prov->building_construction->building];
while (string was = building_types[upg]->make_obsolete) {
if (bldg[was]) {--buildings; break;}
upg = was;
}
}
interesting(id, PRIO_IMMEDIATE); //TODO: Should it always be highlighted at the same prio? Should it always even be highlighted?
write("\e[1;%dm%s\t%d/%d bldg\tDev %d\t%s%s\e[0m\n", buildings < slots ? 32 : 36, id, buildings, slots, dev,
string_to_utf8(prov->name), prov->settlement_growth_construction ? " - SETTLER ACTIVE" : ""); //Can't build while there's a settler promoting growth);
}
}
if (seen) write("\n");
}
void analyze_upgrades(mapping data, string name, string tag, function write) {
mapping country = data->countries[tag];
mapping upgradeables = ([]);
foreach (country->owned_provinces, string id) {
mapping prov = data->provinces["-" + id];
if (!prov->buildings) continue;
string constructing = building_id[(int)prov->building_construction->?building]; //0 if not constructing anything
foreach (prov->buildings; string b;) {
mapping bldg = building_types[b]; if (!bldg) continue; //Unknown building??
if (bldg->influencing_fort) continue; //Ignore forts - it's often not worth upgrading all forts. (TODO: Have a way to request forts too.)
mapping target;
while (mapping upgrade = building_types[bldg->obsoleted_by]) {
[string techtype, int techlevel] = upgrade->tech_required;
if ((int)country->technology[techtype] < techlevel) break;
//Okay. It can be upgraded. But before we report it, see if we can go another level.
//For instance, if you have a Marketplace and Diplo tech 22, you can upgrade to a
//Trade Depot, but could go straight to Stock Exchange.
target = bldg->obsoleted_by;
bldg = upgrade;
}
if (target && target != constructing) {interesting(id, PRIO_SITUATIONAL); upgradeables[target] += ({prov->name});}
}
}
foreach (sort(indices(upgradeables)), string b) {
write("Can upgrade %d buildings to %s\n", sizeof(upgradeables[b]), b);
write("==> %s\n", string_to_utf8(upgradeables[b] * ", "));
}
}
void analyze_findbuildings(mapping data, string name, string tag, function write, string highlight) {
mapping country = data->countries[tag];
foreach (country->owned_provinces, string id) {
mapping prov = data->provinces["-" + id];
//Building shipyards in inland provinces isn't very productive
if (building_types[highlight]->build_trigger->?has_port && !province_info[id]->?has_port) continue;
mapping bldg = prov->buildings || ([]);
int slots = count_building_slots(data, id);
int buildings = sizeof(bldg);
if (prov->building_construction) {
//Duplicate of the above
++buildings;
string upg = building_id[(int)prov->building_construction->building];
while (string was = building_types[upg]->make_obsolete) {
if (bldg[was]) {--buildings; break;}
upg = was;
}
}
if (buildings < slots) continue; //Got room. Not a problem. (Note that the building slots calculation may be wrong but usually too low.)
//Check if a building of the highlight type already exists here.
int gotone = 0;
foreach (prov->buildings; string b;) {
while (string upg = building_types[b]->make_obsolete) b = upg;
if (b == highlight) {gotone = 1; break;}
}
if (gotone) continue;
interesting(id, PRIO_EXPLICIT);
int dev = (int)prov->base_tax + (int)prov->base_production + (int)prov->base_manpower;
write("\e[1;32m%s\t%d/%d bldg\tDev %d\t%s\e[0m\n", id, buildings, slots, dev, string_to_utf8(prov->name));
}
}
mapping(string:array) interesting_provinces = ([]);
void analyze(mapping data, string name, string tag, function|void write, string|void highlight) {
if (!write) write = Stdio.stdin->write;
interesting_province = ({ }); interest_priority = 0; area_has_level3 = (<>);
write("\e[1m== Player: %s (%s) ==\e[0m\n", name, tag);
({analyze_cot, analyze_leviathans, analyze_furnace, analyze_upgrades})(data, name, tag, write);
if (highlight) analyze_findbuildings(data, name, tag, write, highlight);
//write("* %s * %s\n\n", tag, Standards.JSON.encode((array(int))interesting_province)); //If needed in a machine-readable format
interesting_provinces[tag] = interesting_province;
}
//Not currently triggered from anywhere. Doesn't currently have a primary use-case.
void show_tradegoods(mapping data, string tag, function|void write) {
//write("Sevilla: %O\n", data->provinces["-224"]);
//write("Demnate: %O\n", data->provinces["-4568"]);
mapping prod = ([]), count = ([]);
mapping country = data->countries[tag];
float prod_efficiency = 1.0;
foreach (country->owned_provinces, string id) {
mapping prov = data->provinces["-" + id];
//1) Goods produced: base production * 0.2 + flat modifiers (eg Manufactory)
int production = threeplace(prov->base_production) / 5;
//2) Trade value: goods * price
float trade_value = production * (float)data->change_price[prov->trade_goods]->current_price / 1000;
//3) Prod income: trade value * national efficiency * local efficiency * (1 - autonomy)
float local_efficiency = 1.0, autonomy = 0.0; //TODO.
float prod_income = trade_value * prod_efficiency * local_efficiency * (1.0 - autonomy);
//Done. Now gather the stats.
prod[prov->trade_goods] += prod_income;
count[prov->trade_goods]++;
}
float total_value = 0.0;
array goods = indices(prod); sort(-values(prod)[*], goods);
foreach (goods, string tradegood) {
float annual_value = prod[tradegood];
if (annual_value > 0) write("%.2f/year from %d %s provinces\n", annual_value, count[tradegood], tradegood);
total_value += annual_value;
}
write("Total %.2f/year or %.4f/month\n", total_value, total_value / 12);
}
void analyze_flagships(mapping data, function|void write) {
if (!write) write = Stdio.stdin->write; //Wait. How does this even work?? FIXME - shouldn't it fail, and make me use stdout properly?!?
array flagships = ({ });
foreach (data->countries; string tag; mapping country) {
//mapping country = data->countries[tag];
if (!country->navy) continue;
foreach (Array.arrayify(country->navy), mapping fleet) {
foreach (Array.arrayify(fleet->ship), mapping ship) {
if (!ship->flagship) continue;
string cap = "";
if (ship->flagship->is_captured) {
string was = ship->flagship->original_owner;
cap = " CAPTURED from " + (data->countries[was]->name || L10n[was] || was);
}
flagships += ({({
string_to_utf8(sprintf("\e[1m%s\e[0m - %s: \e[36m%s %q\e[31m%s\e[0m",
country->name || L10n[tag] || tag, fleet->name,
String.capitalize(ship->type), ship->name, cap)),
//Measure size without colour codes or UTF-8 encoding
sizeof(sprintf("%s - %s: %s %q%s",
country->name || L10n[tag] || tag, fleet->name,
String.capitalize(ship->type), ship->name, cap)),
ship->flagship->modification * ", ",
})});
//write("%O\n", ship->flagship);
}
}
}
if (!sizeof(flagships)) return;
write("\n\e[1m== Flagships of the World ==\e[0m\n");
sort(flagships);
int width = max(@flagships[*][1]);
foreach (flagships, array f) f[1] = " " * (width - f[1]);
write("%{%s %s %s\n%}", flagships);
}
mapping transform(string ... types) {
mapping ret = ([]);
foreach (types, string type) {
sscanf(type, "%s: %{%s %}", string value, array keys);
foreach (keys, [string key]) ret[key] = value;
}
return ret;
}
mapping ship_types = transform(
"heavy_ship: early_carrack carrack galleon wargalleon twodecker threedecker ",
"light_ship: barque caravel early_frigate frigate heavy_frigate great_frigate ",
"galley: galley war_galley galleass galiot chebeck archipelago_frigate ",
"transport: war_canoe cog flute brig merchantman trabakul eastindiaman ",
);
void analyze_wars(mapping data, multiset(string) tags, function|void write) {
if (!write) write = Stdio.stdin->write;
foreach (values(data->active_war || ({ })), mapping war) {
if (!mappingp(war)) continue; //Dunno what's with these, there seem to be some strings in there.
//To keep displaying the war after all players separate-peace out, use
//war->persistent_attackers and war->persistent_defenders instead.
int is_attacker = war->attackers && sizeof((multiset)war->attackers & tags);
int is_defender = war->defenders && sizeof((multiset)war->defenders & tags);
if (!is_attacker && !is_defender) continue; //Irrelevant bickering somewhere in the world.
//If there are players on both sides of the war, show "attackers" and "defenders".
//But if all players are on one side of a war, show "allies" and "enemies".
string atk = "\U0001f5e1\ufe0f", def = "\U0001f6e1\ufe0f";
int defender = is_defender && !is_attacker;
if (defender) [atk, def] = ({def, atk});
write("\n\e[1;31m== War: %s - %s ==\e[0m\n", war->action, string_to_utf8(war->name));
//war->action is the date it started?? Maybe the last date when a call to arms is valid?
//war->called - it's all just numbers, no country tags. No idea.
//Ticking war score is either war->defender_score or war->attacker_score and is a positive number.
float ticking_ws = (float)(war->attacker_score || "-" + war->defender_score);
if (defender) ticking_ws = -ticking_ws;
//Overall war score?? Can't figure that out. It might be that it isn't stored.
//war->participants[*]->value is the individual contribution. To turn this into a percentage,
//be sure to sum only the values on one side, as participants[] has both sides of the war in it.
array armies = ({ }), navies = ({ });
array(array(int)) army_total = ({allocate(8), allocate(8)});
array(array(int)) navy_total = ({allocate(6), allocate(6)});
foreach (war->participants, mapping p) {
mapping country = data->countries[p->tag];
int a = has_value(war->attackers, p->tag), d = has_value(war->defenders, p->tag);
if (!a && !d) continue; //War participant has subsequently peaced out
string side = sprintf("\e[48;2;%d;%d;%dm%s ",
a && 30, //Red for attacker
tags[p->tag] && 60, //Cyan or olive for player
d && 30,
a ? atk : def, //Sword or shield
);
//I don't know how to recognize that eastern_militia is infantry and muscovite_cossack is cavalry.
//For land units, we can probably assume that you use only your current set. For sea units, there
//aren't too many (and they're shared by all nations), so I just hard-code them.
mapping unit_types = mkmapping(values(country->sub_unit), indices(country->sub_unit));
mapping mil = ([]), mercs = ([]);
if (country->army) foreach (Array.arrayify(country->army), mapping army) {
string merc = army->mercenary_company ? "merc_" : "";
foreach (Array.arrayify(army->regiment), mapping reg) {
//Note that regiment strength is eg "0.807" for 807 men. We want the
//number of men, so there's no need to re-divide.
mil[merc + unit_types[reg->type]] += reg->strength ? threeplace(reg->strength) : 1000;
}
}
if (country->navy) foreach (Array.arrayify(country->navy), mapping navy) {
foreach (Array.arrayify(navy->ship), mapping ship) {
mil[ship_types[ship->type]] += 1; //Currently not concerned about hull strength. You either have or don't have a ship.
}
}
int mp = threeplace(country->manpower);
int total_army = mil->infantry + mil->cavalry + mil->artillery + mil->merc_infantry + mil->merc_cavalry + mil->merc_artillery;
armies += ({({
-total_army * 1000000000 - mp,
({
side,
country->name || L10n[p->tag] || p->tag,
mil->infantry, mil->cavalry, mil->artillery,
mil->merc_infantry, mil->merc_cavalry, mil->merc_artillery,
total_army, mp,
sprintf("%3.0f%%", (float)country->army_professionalism * 100.0),
sprintf("%3.0f%%", (float)country->army_tradition),
}),
})});
army_total[d] = army_total[d][*] + armies[-1][1][2..<2][*];
int sailors = (int)country->sailors; //Might be 0, otherwise is eg "991.795" (we don't care about the fraction, this means 991 sailors)
int total_navy = mil->heavy_ship + mil->light_ship + mil->galley + mil->transport;
navies += ({({
-total_navy * 1000000000 - sailors,
({
side,
country->name || L10n[p->tag] || p->tag,
mil->heavy_ship, mil->light_ship, mil->galley, mil->transport, total_navy, sailors,
sprintf("%3.0f%%", (float)country->navy_tradition),
}),
})});
navy_total[d] = navy_total[d][*] + navies[-1][1][2..<1][*];
}
armies += ({
//The totals get sorted after the individual country entries. Their sort keys are
//guaranteed positive, and are such that the larger army has a smaller sort key.
//Easiest way to do that is to swap them :)
({1 + army_total[1][-2] + army_total[1][-1], ({"\e[48;2;50;0;0m" + atk + " ", ""}) + army_total[0] + ({"", ""})}),
({1 + army_total[0][-2] + army_total[0][-1], ({"\e[48;2;0;0;50m" + def + " ", ""}) + army_total[1] + ({"", ""})}),
});
navies += ({
({1 + navy_total[1][-2] + navy_total[1][-1], ({"\e[48;2;50;0;0m" + atk + " ", ""}) + navy_total[0] + ({""})}),
({1 + navy_total[0][-2] + navy_total[0][-1], ({"\e[48;2;0;0;50m" + def + " ", ""}) + navy_total[1] + ({""})}),
});
sort(armies); sort(navies);
write("%s\n", string_to_utf8(tabulate(({" "}) + "Country Infantry Cavalry Artillery Inf$$ Cav$$ Art$$ Total Manpower Prof Trad" / " ", armies[*][-1], " ", 2)));
write("%s\n", string_to_utf8(tabulate(({" "}) + "Country Heavy Light Galley Transp Total Sailors Trad" / " ", navies[*][-1], " ", 2)));
}
}
multiset(object) connections = (<>);
mapping last_parsed_savefile;
class Connection(Stdio.File sock) {
Stdio.Buffer incoming = Stdio.Buffer(), outgoing = Stdio.Buffer();
string notify, highlight;
protected void create() {
//write("%%%% Connection from %s\n", sock->query_address());
sock->set_buffer_mode(incoming, outgoing);
sock->set_nonblocking(sockread, 0, sockclosed);
}
void sockclosed() {connections[this] = 0; sock->close();}
string find_country(mapping data, string country) {
foreach (data->players_countries / 2, [string name, string tag])
if (lower_case(country) == lower_case(name)) country = tag;
if (data->countries[country]) return country;
outgoing->sprintf("Player or tag %O not found - try%{ %O%} or any country tag\n", country, data->players_countries);
sock->write(""); //Force a write callback (shouldn't be necessary??)
}
void inform(mapping data) {
//A savefile has been parsed. Notify this socket (if applicable).
if (!notify) return;
string tag = find_country(data, notify); if (!tag) return;
analyze(data, notify, tag, outgoing->sprintf, highlight);
analyze_wars(data, (<tag>), outgoing->sprintf);
sock->write(""); //Ditto
}
void cycle_provinces(string country) {
if (!last_parsed_savefile) return;
string tag = find_country(last_parsed_savefile, country); if (!tag) return;
if (!interesting_provinces[tag]) analyze(last_parsed_savefile, "Province finder", tag); //Should this be sent to /dev/null instead of the console?
if (!sizeof(interesting_provinces[tag])) {sock->close("w"); return;}
[string id, array rest] = Array.shift(interesting_provinces[tag]);
interesting_provinces[tag] = rest + ({id});
//Note: Ignores buffered mode and writes directly. I don't think it's possible to
//put a "shutdown write direction when done" marker into the Buffer.
sock->write(id + "\n");
sock->close("w");
}
void sockread() {
while (array ret = incoming->sscanf("%s\n")) {
string cmd = String.trim(ret[0]), arg = "";
sscanf(cmd, "%s %s", cmd, arg);
switch (cmd) {
case "notify":
notify = arg; connections[this] = 1;
if (last_parsed_savefile) inform(last_parsed_savefile);
break;
case "province": cycle_provinces(arg); break;
case "highlight": case "hl": case "build": case "building": case "buildings": {
//Request highlighting of provinces in which a particular building could be built if you had a slot.
//Example: "highlight shipyard" ==> any province with no shipyard and no building slots gets highlighted.
//Typing "highlight" without an arg, or any invalid arg, will give a list of building IDs.
arg = replace(lower_case(arg), " ", "_");
if ((<"none", "off", "-">)[arg]) {
highlight = 0;
outgoing->sprintf("Highlighting disabled.\n");
sock->write("");
break;
}
string tag = last_parsed_savefile && find_country(last_parsed_savefile, notify);
if (!tag) break;
if (!building_types[arg]) {
array available = ({ });
mapping tech = last_parsed_savefile->countries[tag]->technology;
int have_mfg = 0;
foreach (building_types; string id; mapping bldg) {
[string techtype, int techlevel] = bldg->tech_required || ({"", 100}); //Ignore anything that's not a regular building
if ((int)tech[techtype] < techlevel) continue; //Hide IDs you don't have the tech to build
if (bldg->obsoleted_by) continue; //Show only the baseline building for each type
if (bldg->manufactory && !bldg->show_separate) {have_mfg = 1; continue;} //Collect regular manufactories under one name
if (bldg->influencing_fort) continue; //You won't want to check forts this way
available += ({id});
}
if (have_mfg) available += ({"manufactory"}); //Note that building_types->manufactory is technically valid
outgoing->sprintf("Valid IDs: %s\n", sort(available) * ", ");
outgoing->sprintf("Or use 'highlight none' to disable.\n");
sock->write("");
break;
}
//If you say "highlight stock_exchange", act as if you said "highlight marketplace".
while (string older = building_types[arg]->make_obsolete) arg = older;
highlight = arg;
analyze_findbuildings(last_parsed_savefile, notify, tag, outgoing->sprintf, arg);
sock->write("");
break;
}
case "flagship": case "flagships": case "flag": case "fs":
analyze_flagships(last_parsed_savefile, outgoing->sprintf);
sock->write("");
break;
case "war": case "wars": {
if (arg == "") {
foreach (last_parsed_savefile->active_war || ({ }), mapping war) {
outgoing->sprintf("\n\e[1;31m== War: %s - %s ==\e[0m\n", war->action, string_to_utf8(war->name));
if (war->attackers) outgoing->sprintf(string_to_utf8("\U0001f5e1\ufe0f %{ %s%}\n"), war->attackers);
if (war->defenders) outgoing->sprintf(string_to_utf8("\U0001f6e1\ufe0f %{ %s%}\n"), war->defenders);
}
sock->write("");
break;
}
string tag = find_country(last_parsed_savefile, arg); if (!tag) break;
analyze_wars(last_parsed_savefile, (<tag>), outgoing->sprintf);
sock->write("");
break;
}
default: sock->write(sprintf("Unknown command %O\n", cmd)); break;
}
}
}
}
void sock_connected(object mainsock) {while (object sock = mainsock->accept()) Connection(sock);}
Stdio.File parser_pipe = Stdio.File();
void process_savefile(string fn) {parser_pipe->write(fn + "\n");}
void done_processing_savefile() {
parser_pipe->read();
mapping data = Standards.JSON.decode_utf8(Stdio.read_file("eu4_parse.json") || "{}")->data;
if (!data) {werror("Unable to parse save file (see above for errors, hopefully)\n"); return;}
write("\nCurrent date: %s\n", data->date);
foreach (data->players_countries / 2, [string name, string tag]) analyze(data, name, tag);
//analyze_flagships(data);
analyze_wars(data, (multiset)(data->players_countries / 2)[*][1]);
indices(connections)->inform(data);
last_parsed_savefile = data;
}
class ClientConnection {
inherit Connection;
protected void create(Stdio.File sock) {
::create(sock);
Stdio.stdin->set_read_callback(stdinread);
Stdio.stdin->set_close_callback(stdineof);
}
void sockread() {
//Display only complete lines, to avoid disruption of input text
while (array ret = incoming->sscanf("%s\n")) write("%s\n", ret[0]);
}
void sockclosed() {::sockclosed(); exit(0);}
void stdinread(mixed _, string data) {sock->write(data);}
void stdineof() {sock->close("w");}
}
class PipeConnection {
inherit Connection;
void sockread() {
while (array ret = incoming->sscanf("%s\n")) {
[string fn] = ret;
write("Reading save file %s\n", basename(fn));
string raw = Stdio.read_file(fn); //Assumes ISO-8859-1, which I think is correct
parse_savefile(raw);
sock->write("*"); //Signal the parent. It can read it back from the cache.
}
}
}
int main(int argc, array(string) argv) {
if (argc > 1 && argv[1] == "--parse") {
//Parser subprocess, invoked by parent for asynchronous parsing.
PipeConnection(Stdio.File(3)); //We should have been given fd 3 as a pipe
return -1;
}
if (argc > 1 && argv[1] == "--timeparse") {
object start = System.Timer();
#define TIME(x) {float tm = gauge {x;}; write("%.3f\t%.3f\t%s\n", start->get(), tm, #x);}
string raw; TIME(raw = Stdio.read_file(SAVE_PATH + "/mp_autosave.eu4"));
mapping data; TIME(data = parse_savefile_string(raw));
write("Parse successful. Date: %s\n", data->date);
return 0;
}
if (argc > 2) {
//First arg is server name/IP; the rest are joined and sent as a command.
//If the second arg is "province", then the result is fed as keys to EU4.
//Otherwise, this is basically like netcat/telnet.
Stdio.File sock = Stdio.File();
string writeme = sock->connect(argv[1], 1444, argv[2..] * " " + "\n");
if (!writeme) exit(0, "Unable to connect to %s : 1444\n", argv[1]);
sock->write(writeme); //TBH there shouldn't be any residual data, since it should be a single packet.
if (argv[2] != "province") {ClientConnection(sock); return -1;}
string province = "";
while (string data = sock->read(1024, 1)) {
if (data == "") break;
province += data;
}
sock->close();
if (String.trim(province) != "") Process.create_process(({"xdotool",
/*"search", "--name", "Europa Universalis IV",*/ //Doesn't always work. Omitting this assumes that EU4 has focus.
"key", "--delay", "125", //Hurry the typing along a bit
"f", @(String.trim(province) / ""), "Return", //Send "f", then type the province ID, then hit Enter
}))->wait();
return 0;
}
//Load up some info that is presumed to not change. If you're modding the game, this may break.
catch {L10n = Standards.JSON.decode_utf8(Stdio.read_file(".eu4_localisation.json"));};
if (!mappingp(L10n)) {
L10n = ([]);
foreach (glob("*_l_english.yml", get_dir(PROGRAM_PATH + "/localisation")), string fn)
parse_localisation(Stdio.read_file(PROGRAM_PATH + "/localisation/" + fn));
Stdio.write_file(".eu4_localisation.json", Standards.JSON.encode(L10n, 1));
}
mapping areas = low_parse_savefile(Stdio.read_file(PROGRAM_PATH + "/map/area.txt"));
foreach (areas; string areaname; array|maparray provinces)
foreach (provinces;; string id) prov_area[id] = areaname;
mapping terrains = low_parse_savefile(Stdio.read_file(PROGRAM_PATH + "/map/terrain.txt"));
//Terrain info is used below.
mapping climates = low_parse_savefile(Stdio.read_file(PROGRAM_PATH + "/map/climate.txt"));
//For simplicity, I'm not looking up static_modifiers or anything - just arbitrarily flagging Arctic regions.
foreach (climates->arctic, string id) building_slots[id] -= 1;
retain_map_indices = 1;
building_types = low_parse_savefile(Stdio.read_file(PROGRAM_PATH + "/common/buildings/00_buildings.txt"));
retain_map_indices = 0;
building_id = allocate(sizeof(building_types));
foreach (building_types; string id; mapping info) {
if (info->manufactory) manufactories[id] = info->show_separate ? "Special" : "Basic";
//Map the index to the ID, counting from 1, but skipping the "manufactory" pseudo-entry
//(not counting it and collapsing the gap).
if (id != "manufactory") building_id[info->_index + (info->_index < building_types->manufactory->_index)] = id;
}
foreach (({"adm", "dip", "mil"}), string cat) {
mapping tech = low_parse_savefile(Stdio.read_file(PROGRAM_PATH + "/common/technologies/" + cat + ".txt"));
foreach (tech->technology; int level; mapping effects) {
//The effects include names of buildings, eg "university = yes".
foreach (effects; string id;) if (mapping bld = building_types[id]) {
bld->tech_required = ({cat + "_tech", level});
if (bld->make_obsolete) building_types[bld->make_obsolete]->obsoleted_by = id;
}
}
}
/* It is REALLY REALLY hard to replicate the game's full algorithm for figuring out which terrain each province
has. So, instead, let's ask for a little help - from the game, and from the human. And then save the results.
Unfortunately, it's not possible (as of v1.31) to do an every_province scope that reports the province ID in a
log message. It also doesn't seem to be possible to iterate over all provinces and increment a counter, as the
every_province scope skips sea provinces (which still consume province IDs).
I would REALLY like to do something like this:
every_province = {
limit = {
has_terrain = steppe
is_wasteland = no
}
log = "PROV-TERRAIN: steppe [This.ID] [This.GetName]"
}
and repeat for each terrain type. A technique others have done is to cede the provinces to different countries,
save, and parse the savefile; this is slow, messy, and mutates the save, so it won't be very useful in Random
New World. (Not that I'm going to try to support RNW, but it should be easier this way if I do in the future.)
Since we can't do it the easy way, let's do it the hard way instead. For each province ID, for each terrain, if
the province has that terrain, log a message. If it's stupid, but it works........ no, it's still stupid.
*/
province_info = Standards.JSON.decode(Stdio.read_file(".eu4_provinces.json") || "0");
if (!mappingp(province_info)) {
//Build up a script file to get the info we need.
//We assume that every province that could be of interest to us will be in an area.
Stdio.File script = Stdio.File(SAVE_PATH + "/../prov.txt", "wct");
script->write("log = \"PROV-TERRAIN-BEGIN\"\n");
foreach (sort(indices(prov_area)), string provid) {
script->write(
#"%s = {
set_variable = { which = terrain_reported value = -1 }
if = {
limit = { has_port = yes is_wasteland = no }
log = \"PROV-TERRAIN: %<s has_port=1\"
}
", provid);
foreach (terrains->categories; string type; mapping info) {
script->write(
#" if = {
limit = { has_terrain = %s is_wasteland = no }
log = \"PROV-TERRAIN: %s terrain=%[0]s\"
}
", type, provid);
}
script->write("}\n");
}
//For reasons of paranoia, iterate over all provinces and make sure we reported their
//terrain types.
script->write(#"
every_province = {
limit = { check_variable = { which = terrain_reported value = 0 } is_wasteland = no }
log = \"PROV-TERRAIN-ERROR: Terrain not reported for province [This.GetName]\"
}
log = \"PROV-TERRAIN-END\"
");
script->close();
//See if the script's already been run (yes, we rebuild the script every time - means you
//can rerun it in case there've been changes), and if so, parse and save the data.
string log = Stdio.read_file(SAVE_PATH + "/../logs/game.log") || "";
if (!has_value(log, "PROV-TERRAIN-BEGIN") || !has_value(log, "PROV-TERRAIN-END"))
exit(0, "Please open up EU4 and, in the console, type: run prov.txt\n");
string terrain = ((log / "PROV-TERRAIN-BEGIN")[-1] / "PROV-TERRAIN-END")[0];
province_info = ([]);
foreach (terrain / "\n", string line) {
//Lines look like this:
//[effectimplementation.cpp:21960]: EVENT [1444.11.11]:PROV-TERRAIN: drylands 224 - Sevilla
sscanf(line, "%*sPROV-TERRAIN: %d %s=%s", int provid, string key, string val);
if (!provid) continue;
mapping pt = province_info[(string)provid] || ([]); province_info[(string)provid] = pt;
pt[key] = String.trim(val);
}
Stdio.write_file(".eu4_provinces.json", Standards.JSON.encode(province_info));
}
foreach (province_info; string id; mapping provinfo) {
mapping terraininfo = terrains->categories[provinfo->terrain];
if (!terraininfo) continue; //TODO: What happens if we have a broken terrain name??
int slots = (int)terraininfo->allowed_num_of_buildings;
if (slots) building_slots[id] += slots;
}
object proc = Process.spawn_pike(({argv[0], "--parse"}), (["fds": ({parser_pipe->pipe(Stdio.PROP_NONBLOCK|Stdio.PROP_BIDIRECTIONAL|Stdio.PROP_IPC)})]));
parser_pipe->set_nonblocking(done_processing_savefile, 0, parser_pipe->close);
//Find the newest .eu4 file in the directory and (re)parse it, then watch for new files.
array(string) files = SAVE_PATH + "/" + get_dir(SAVE_PATH)[*];
sort(file_stat(files[*])->mtime, files);
if (sizeof(files)) process_savefile(files[-1]);
object inot = System.Inotify.Instance();
string new_file; int nomnomcookie;
inot->add_watch(SAVE_PATH, System.Inotify.IN_CLOSE_WRITE | System.Inotify.IN_MOVED_TO | System.Inotify.IN_MOVED_FROM) {
[int event, int cookie, string path] = __ARGS__;
//EU4 seems to always save into a temporary file, then rename it over the target. This
//sometimes includes renaming the target out of the way first (eg old_autosave.eu4).
//There are a few ways to detect new save files.
//1) Watch for a CLOSE_WRITE event, which will be the temporary file (eg autosave.tmp).
// When you see that, watch for the next MOVED_FROM event for that same name, and then
// the corresponding MOVED_TO event is the target name. Assumes that the file is created
// in the savegames directory and only renamed, never moved in.
//2) Watch for all MOVED_TO events, and arbitrarily ignore any that we don't think are
// interesting (eg if starts with "old_" or "older_").
//3) Watch for any CLOSE_WRITE or MOVED_TO. Wait a little bit. See what the newest file in
// the directory is. Assumes that the directory is quiet apart from what we care about.
//Currently using option 1. Change if this causes problems.
switch (event) {
case System.Inotify.IN_CLOSE_WRITE: new_file = path; break;
case System.Inotify.IN_MOVED_FROM: if (path == new_file) {new_file = 0; nomnomcookie = cookie;} break;
case System.Inotify.IN_MOVED_TO: if (cookie == nomnomcookie) {nomnomcookie = 0; process_savefile(path);} break;
}
};
inot->set_nonblocking();
Stdio.Port mainsock = Stdio.Port();
mainsock->bind(1444, sock_connected, "::", 1);
return -1;
}
| Pike | 5 | Rosuav/shed | eu4_parse.pike | [
"MIT"
]
|
# Copyright 2018 The gRPC Authors
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Reference implementation for status mapping in gRPC Python."""
import collections
import sys
from google.rpc import status_pb2
import grpc
from ._common import GRPC_DETAILS_METADATA_KEY
from ._common import code_to_grpc_status_code
class _Status(
collections.namedtuple('_Status',
('code', 'details', 'trailing_metadata')),
grpc.Status):
pass
def from_call(call):
"""Returns a google.rpc.status.Status message corresponding to a given grpc.Call.
This is an EXPERIMENTAL API.
Args:
call: A grpc.Call instance.
Returns:
A google.rpc.status.Status message representing the status of the RPC.
Raises:
ValueError: If the gRPC call's code or details are inconsistent with the
status code and message inside of the google.rpc.status.Status.
"""
if call.trailing_metadata() is None:
return None
for key, value in call.trailing_metadata():
if key == GRPC_DETAILS_METADATA_KEY:
rich_status = status_pb2.Status.FromString(value)
if call.code().value[0] != rich_status.code:
raise ValueError(
'Code in Status proto (%s) doesn\'t match status code (%s)'
% (code_to_grpc_status_code(rich_status.code), call.code()))
if call.details() != rich_status.message:
raise ValueError(
'Message in Status proto (%s) doesn\'t match status details (%s)'
% (rich_status.message, call.details()))
return rich_status
return None
def to_status(status):
"""Convert a google.rpc.status.Status message to grpc.Status.
This is an EXPERIMENTAL API.
Args:
status: a google.rpc.status.Status message representing the non-OK status
to terminate the RPC with and communicate it to the client.
Returns:
A grpc.Status instance representing the input google.rpc.status.Status message.
"""
return _Status(code=code_to_grpc_status_code(status.code),
details=status.message,
trailing_metadata=((GRPC_DETAILS_METADATA_KEY,
status.SerializeToString()),))
__all__ = [
'from_call',
'to_status',
]
if sys.version_info[0] >= 3 and sys.version_info[1] >= 6:
from . import _async as aio # pylint: disable=unused-import
__all__.append('aio')
| Python | 5 | warlock135/grpc | src/python/grpcio_status/grpc_status/rpc_status.py | [
"Apache-2.0"
]
|
-- @shouldFailWith UnknownName
module Main where
import Prelude (unit, pure)
main = do
pure unit
pure unit
| PureScript | 2 | andys8/purescript | tests/purs/failing/2109-discard.purs | [
"BSD-3-Clause"
]
|
frequency,raw,error,smoothed,error_smoothed,equalization,parametric_eq,fixed_band_eq,equalized_raw,equalized_smoothed,target
20.00,6.04,4.22,6.04,4.22,-4.22,-3.70,-1.82,1.82,1.82,1.82
20.20,6.04,4.17,6.04,4.17,-4.17,-3.72,-1.87,1.87,1.87,1.86
20.40,6.04,4.12,6.04,4.13,-4.13,-3.75,-1.93,1.91,1.91,1.91
20.61,6.04,4.09,6.04,4.08,-4.09,-3.77,-1.98,1.95,1.95,1.94
20.81,6.04,4.04,6.04,4.04,-4.05,-3.80,-2.04,1.98,1.98,1.99
21.02,6.04,4.00,6.04,4.01,-4.02,-3.82,-2.10,2.02,2.02,2.03
21.23,6.04,3.98,6.04,3.98,-3.99,-3.85,-2.16,2.05,2.05,2.05
21.44,6.04,3.96,6.04,3.96,-3.96,-3.87,-2.22,2.08,2.08,2.07
21.66,6.04,3.94,6.04,3.95,-3.94,-3.89,-2.28,2.10,2.10,2.09
21.87,6.04,3.93,6.03,3.93,-3.91,-3.91,-2.34,2.13,2.12,2.10
22.09,6.03,3.90,6.03,3.90,-3.89,-3.93,-2.41,2.14,2.14,2.12
22.31,6.01,3.88,6.01,3.88,-3.88,-3.95,-2.48,2.13,2.14,2.13
22.54,6.00,3.87,5.99,3.86,-3.87,-3.97,-2.55,2.13,2.12,2.13
22.76,5.98,3.85,5.97,3.85,-3.87,-3.99,-2.62,2.11,2.10,2.13
22.99,5.96,3.84,5.98,3.85,-3.88,-4.00,-2.70,2.08,2.10,2.12
23.22,6.00,3.88,5.99,3.87,-3.89,-4.02,-2.77,2.10,2.10,2.12
23.45,6.04,3.92,6.04,3.91,-3.91,-4.04,-2.85,2.13,2.12,2.12
23.69,6.08,3.95,6.08,3.96,-3.93,-4.05,-2.92,2.14,2.15,2.13
23.92,6.13,4.00,6.12,3.98,-3.96,-4.06,-3.00,2.17,2.16,2.13
24.16,6.14,4.00,6.14,4.00,-3.98,-4.08,-3.09,2.16,2.16,2.14
24.40,6.14,3.99,6.14,4.00,-3.99,-4.09,-3.17,2.15,2.15,2.15
24.65,6.14,4.00,6.14,4.00,-4.00,-4.10,-3.25,2.14,2.14,2.14
24.89,6.14,4.00,6.14,4.00,-4.00,-4.11,-3.33,2.13,2.13,2.14
25.14,6.14,4.00,6.14,4.00,-4.00,-4.12,-3.42,2.13,2.13,2.14
25.39,6.14,3.99,6.14,3.99,-4.00,-4.13,-3.50,2.14,2.14,2.15
25.65,6.14,3.99,6.14,3.99,-4.00,-4.13,-3.59,2.14,2.14,2.15
25.91,6.14,4.00,6.14,4.00,-4.00,-4.14,-3.67,2.14,2.14,2.14
26.16,6.14,4.00,6.14,4.00,-4.00,-4.15,-3.76,2.14,2.14,2.14
26.43,6.14,4.00,6.14,4.00,-4.00,-4.15,-3.84,2.14,2.14,2.13
26.69,6.14,4.00,6.14,4.00,-4.00,-4.15,-3.93,2.14,2.14,2.13
26.96,6.14,3.99,6.14,4.00,-4.00,-4.16,-4.01,2.14,2.14,2.14
27.23,6.14,3.99,6.14,4.00,-4.00,-4.16,-4.09,2.14,2.14,2.14
27.50,6.14,4.00,6.14,4.00,-4.00,-4.16,-4.16,2.13,2.13,2.14
27.77,6.14,4.01,6.14,4.00,-4.01,-4.16,-4.24,2.13,2.13,2.13
28.05,6.14,4.01,6.14,4.01,-4.01,-4.16,-4.31,2.13,2.13,2.13
28.33,6.14,4.02,6.14,4.01,-4.01,-4.16,-4.38,2.13,2.13,2.12
28.62,6.14,4.02,6.14,4.02,-4.01,-4.16,-4.44,2.13,2.13,2.12
28.90,6.14,4.01,6.14,4.02,-4.01,-4.16,-4.50,2.13,2.13,2.13
29.19,6.14,4.02,6.14,4.01,-4.01,-4.15,-4.56,2.13,2.13,2.12
29.48,6.14,4.00,6.14,4.01,-4.01,-4.15,-4.61,2.13,2.13,2.14
29.78,6.14,4.01,6.14,4.01,-4.01,-4.14,-4.65,2.13,2.13,2.13
30.08,6.14,4.00,6.14,4.01,-4.01,-4.14,-4.69,2.13,2.13,2.13
30.38,6.14,4.01,6.14,4.01,-4.01,-4.13,-4.72,2.13,2.13,2.12
30.68,6.14,4.02,6.14,4.01,-4.01,-4.12,-4.75,2.12,2.12,2.12
30.99,6.14,4.02,6.14,4.02,-4.02,-4.12,-4.77,2.12,2.12,2.12
31.30,6.14,4.02,6.14,4.02,-4.02,-4.11,-4.78,2.12,2.12,2.12
31.61,6.14,4.02,6.14,4.02,-4.02,-4.10,-4.78,2.12,2.12,2.12
31.93,6.14,4.02,6.14,4.02,-4.02,-4.09,-4.78,2.12,2.12,2.12
32.24,6.14,4.03,6.14,4.02,-4.02,-4.08,-4.78,2.11,2.11,2.11
32.57,6.14,4.02,6.14,4.03,-4.03,-4.07,-4.77,2.11,2.11,2.11
32.89,6.14,4.02,6.14,4.03,-4.03,-4.06,-4.75,2.11,2.11,2.11
33.22,6.14,4.04,6.14,4.03,-4.03,-4.05,-4.72,2.10,2.10,2.10
33.55,6.14,4.04,6.14,4.04,-4.04,-4.04,-4.70,2.10,2.10,2.10
33.89,6.14,4.05,6.14,4.04,-4.05,-4.03,-4.66,2.09,2.09,2.09
34.23,6.14,4.04,6.14,4.05,-4.05,-4.01,-4.63,2.08,2.08,2.10
34.57,6.14,4.05,6.14,4.05,-4.06,-4.00,-4.58,2.08,2.08,2.08
34.92,6.14,4.05,6.14,4.05,-4.06,-3.99,-4.54,2.08,2.08,2.08
35.27,6.14,4.06,6.14,4.06,-4.06,-3.98,-4.49,2.08,2.09,2.08
35.62,6.14,4.06,6.14,4.06,-4.04,-3.96,-4.44,2.09,2.10,2.08
35.97,6.14,4.06,6.13,4.05,-4.03,-3.95,-4.39,2.11,2.10,2.08
36.33,6.10,4.01,6.10,4.02,-4.00,-3.94,-4.34,2.10,2.10,2.08
36.70,6.06,3.99,6.06,3.98,-3.97,-3.92,-4.28,2.08,2.08,2.07
37.06,6.01,3.94,6.01,3.94,-3.95,-3.91,-4.23,2.06,2.07,2.07
37.43,5.98,3.89,5.98,3.90,-3.92,-3.89,-4.17,2.06,2.06,2.09
37.81,5.96,3.87,5.95,3.87,-3.89,-3.88,-4.12,2.07,2.07,2.08
38.19,5.94,3.86,5.94,3.85,-3.87,-3.87,-4.06,2.07,2.07,2.08
38.57,5.94,3.85,5.94,3.85,-3.86,-3.85,-4.01,2.08,2.08,2.09
38.95,5.94,3.84,5.94,3.84,-3.85,-3.84,-3.95,2.09,2.09,2.10
39.34,5.94,3.84,5.94,3.84,-3.84,-3.82,-3.90,2.09,2.09,2.09
39.74,5.94,3.85,5.94,3.84,-3.84,-3.81,-3.85,2.10,2.10,2.09
40.14,5.94,3.84,5.94,3.85,-3.83,-3.80,-3.80,2.11,2.11,2.10
40.54,5.94,3.84,5.94,3.85,-3.82,-3.78,-3.76,2.12,2.12,2.09
40.94,5.94,3.85,5.92,3.83,-3.81,-3.77,-3.71,2.13,2.11,2.09
41.35,5.89,3.80,5.89,3.80,-3.79,-3.76,-3.67,2.09,2.09,2.09
41.76,5.84,3.75,5.85,3.76,-3.78,-3.74,-3.63,2.06,2.07,2.08
42.18,5.82,3.73,5.83,3.74,-3.77,-3.73,-3.59,2.05,2.06,2.09
42.60,5.83,3.75,5.82,3.74,-3.76,-3.72,-3.55,2.06,2.06,2.08
43.03,5.84,3.77,5.83,3.76,-3.77,-3.71,-3.52,2.07,2.06,2.07
43.46,5.84,3.77,5.84,3.78,-3.79,-3.70,-3.49,2.05,2.05,2.07
43.90,5.84,3.80,5.84,3.79,-3.81,-3.69,-3.46,2.03,2.03,2.04
44.33,5.84,3.81,5.83,3.80,-3.83,-3.67,-3.44,2.01,2.01,2.03
44.78,5.84,3.83,5.85,3.83,-3.83,-3.66,-3.41,2.01,2.02,2.01
45.23,5.86,3.85,5.88,3.88,-3.82,-3.65,-3.39,2.04,2.07,2.01
45.68,5.91,3.92,5.88,3.89,-3.80,-3.64,-3.37,2.11,2.08,1.99
46.13,5.88,3.90,5.81,3.83,-3.78,-3.64,-3.36,2.10,2.03,1.98
46.60,5.69,3.71,5.70,3.73,-3.76,-3.63,-3.35,1.93,1.95,1.98
47.06,5.54,3.57,5.61,3.64,-3.73,-3.62,-3.34,1.81,1.88,1.97
47.53,5.58,3.62,5.57,3.61,-3.69,-3.61,-3.33,1.89,1.88,1.96
48.01,5.63,3.66,5.60,3.65,-3.66,-3.60,-3.32,1.97,1.94,1.96
48.49,5.63,3.68,5.63,3.68,-3.63,-3.60,-3.32,2.00,2.00,1.95
48.97,5.62,3.69,5.60,3.67,-3.62,-3.59,-3.32,2.00,1.98,1.93
49.46,5.55,3.64,5.54,3.63,-3.62,-3.59,-3.32,1.93,1.92,1.91
49.96,5.47,3.57,5.51,3.60,-3.63,-3.58,-3.32,1.84,1.88,1.89
50.46,5.49,3.59,5.49,3.60,-3.64,-3.58,-3.33,1.85,1.85,1.90
50.96,5.53,3.64,5.50,3.62,-3.64,-3.57,-3.34,1.88,1.86,1.89
51.47,5.53,3.66,5.53,3.65,-3.65,-3.57,-3.34,1.88,1.88,1.87
51.99,5.53,3.67,5.54,3.68,-3.65,-3.57,-3.36,1.88,1.89,1.86
52.51,5.55,3.68,5.55,3.69,-3.66,-3.57,-3.37,1.89,1.89,1.86
53.03,5.57,3.70,5.54,3.68,-3.67,-3.57,-3.38,1.90,1.87,1.86
53.56,5.53,3.67,5.53,3.68,-3.68,-3.57,-3.39,1.85,1.85,1.86
54.10,5.50,3.65,5.52,3.67,-3.69,-3.57,-3.41,1.81,1.83,1.85
54.64,5.52,3.68,5.51,3.67,-3.70,-3.57,-3.43,1.82,1.81,1.84
55.18,5.53,3.69,5.52,3.69,-3.71,-3.57,-3.44,1.82,1.81,1.83
55.74,5.53,3.70,5.54,3.71,-3.71,-3.57,-3.46,1.82,1.83,1.82
56.29,5.54,3.71,5.55,3.73,-3.71,-3.58,-3.48,1.83,1.84,1.82
56.86,5.57,3.76,5.55,3.74,-3.71,-3.58,-3.49,1.86,1.84,1.81
57.42,5.54,3.75,5.53,3.73,-3.71,-3.59,-3.51,1.83,1.82,1.79
58.00,5.49,3.70,5.50,3.71,-3.71,-3.59,-3.53,1.78,1.79,1.79
58.58,5.46,3.67,5.46,3.68,-3.70,-3.60,-3.55,1.75,1.75,1.78
59.16,5.43,3.66,5.43,3.66,-3.70,-3.61,-3.56,1.73,1.74,1.77
59.76,5.43,3.67,5.43,3.68,-3.69,-3.62,-3.58,1.74,1.74,1.76
60.35,5.43,3.70,5.43,3.70,-3.69,-3.63,-3.59,1.74,1.74,1.72
60.96,5.43,3.73,5.40,3.70,-3.69,-3.64,-3.61,1.74,1.71,1.70
61.57,5.36,3.69,5.37,3.70,-3.69,-3.65,-3.62,1.66,1.67,1.66
62.18,5.31,3.68,5.33,3.70,-3.70,-3.66,-3.63,1.60,1.63,1.63
62.80,5.31,3.72,5.29,3.70,-3.72,-3.67,-3.64,1.59,1.57,1.59
63.43,5.28,3.72,5.27,3.71,-3.73,-3.69,-3.65,1.55,1.54,1.56
64.07,5.23,3.71,5.25,3.73,-3.73,-3.70,-3.66,1.50,1.52,1.52
64.71,5.23,3.75,5.23,3.74,-3.74,-3.72,-3.66,1.49,1.49,1.48
65.35,5.21,3.75,5.20,3.75,-3.74,-3.73,-3.67,1.47,1.46,1.45
66.01,5.18,3.76,5.17,3.75,-3.74,-3.75,-3.68,1.44,1.43,1.42
66.67,5.11,3.73,5.12,3.74,-3.74,-3.77,-3.68,1.37,1.38,1.38
67.33,5.06,3.72,5.07,3.73,-3.73,-3.79,-3.69,1.33,1.33,1.34
68.01,5.03,3.72,5.01,3.71,-3.71,-3.81,-3.69,1.31,1.29,1.31
68.69,4.95,3.68,4.96,3.70,-3.69,-3.83,-3.70,1.26,1.27,1.27
69.37,4.90,3.67,4.90,3.67,-3.66,-3.85,-3.70,1.24,1.24,1.22
70.07,4.85,3.67,4.83,3.64,-3.63,-3.88,-3.70,1.21,1.19,1.18
70.77,4.74,3.59,4.76,3.62,-3.61,-3.90,-3.71,1.13,1.15,1.15
71.48,4.68,3.57,4.67,3.57,-3.58,-3.92,-3.71,1.09,1.09,1.11
72.19,4.61,3.55,4.58,3.52,-3.55,-3.95,-3.72,1.05,1.02,1.05
72.91,4.47,3.46,4.51,3.50,-3.52,-3.98,-3.73,0.95,0.99,1.01
73.64,4.45,3.48,4.47,3.50,-3.48,-4.00,-3.74,0.97,0.99,0.96
74.38,4.43,3.51,4.41,3.49,-3.44,-4.03,-3.75,0.98,0.97,0.92
75.12,4.38,3.50,4.32,3.45,-3.41,-4.06,-3.76,0.97,0.91,0.87
75.87,4.17,3.34,4.19,3.36,-3.37,-4.09,-3.77,0.79,0.82,0.83
76.63,4.03,3.24,4.08,3.30,-3.34,-4.12,-3.79,0.69,0.75,0.79
77.40,4.01,3.26,4.01,3.27,-3.30,-4.15,-3.80,0.71,0.71,0.75
78.17,4.04,3.33,3.98,3.27,-3.26,-4.19,-3.82,0.78,0.72,0.70
78.95,3.93,3.24,3.93,3.26,-3.22,-4.22,-3.85,0.71,0.72,0.68
79.74,3.85,3.20,3.86,3.22,-3.18,-4.25,-3.87,0.67,0.69,0.64
80.54,3.79,3.17,3.79,3.18,-3.16,-4.29,-3.90,0.63,0.63,0.61
81.35,3.75,3.17,3.74,3.17,-3.15,-4.32,-3.93,0.60,0.59,0.57
82.16,3.71,3.17,3.69,3.15,-3.17,-4.36,-3.96,0.54,0.53,0.54
82.98,3.68,3.16,3.67,3.16,-3.21,-4.40,-4.00,0.47,0.46,0.52
83.81,3.65,3.17,3.70,3.22,-3.28,-4.44,-4.03,0.37,0.42,0.48
84.65,3.79,3.34,3.78,3.33,-3.38,-4.47,-4.08,0.41,0.40,0.45
85.50,3.94,3.52,3.91,3.49,-3.51,-4.51,-4.12,0.43,0.40,0.42
86.35,4.07,3.68,4.08,3.69,-3.67,-4.55,-4.17,0.40,0.41,0.39
87.22,4.22,3.86,4.24,3.88,-3.85,-4.59,-4.22,0.37,0.39,0.35
88.09,4.42,4.10,4.39,4.07,-4.05,-4.64,-4.27,0.37,0.34,0.32
88.97,4.56,4.27,4.55,4.27,-4.26,-4.68,-4.33,0.30,0.29,0.28
89.86,4.69,4.44,4.70,4.45,-4.46,-4.72,-4.39,0.22,0.24,0.25
90.76,4.85,4.64,4.85,4.64,-4.66,-4.76,-4.46,0.19,0.19,0.21
91.66,5.03,4.84,5.02,4.84,-4.85,-4.80,-4.52,0.18,0.17,0.19
92.58,5.19,5.05,5.19,5.04,-5.03,-4.85,-4.59,0.16,0.16,0.14
93.51,5.34,5.23,5.33,5.22,-5.19,-4.89,-4.67,0.14,0.14,0.11
94.44,5.46,5.37,5.46,5.38,-5.35,-4.93,-4.74,0.11,0.11,0.08
95.39,5.56,5.50,5.56,5.50,-5.50,-4.98,-4.82,0.06,0.06,0.06
96.34,5.64,5.62,5.63,5.61,-5.63,-5.02,-4.91,0.01,-0.00,0.02
97.30,5.70,5.71,5.71,5.72,-5.75,-5.07,-4.99,-0.05,-0.04,-0.01
98.28,5.77,5.81,5.78,5.83,-5.85,-5.11,-5.08,-0.08,-0.07,-0.04
99.26,5.89,5.96,5.86,5.93,-5.93,-5.16,-5.17,-0.05,-0.07,-0.07
100.25,5.92,6.02,5.93,6.03,-6.00,-5.20,-5.26,-0.08,-0.07,-0.10
101.25,5.98,6.10,5.96,6.09,-6.04,-5.24,-5.35,-0.07,-0.08,-0.13
102.27,5.96,6.11,5.95,6.11,-6.08,-5.29,-5.44,-0.12,-0.12,-0.16
103.29,5.92,6.11,5.92,6.11,-6.10,-5.33,-5.54,-0.18,-0.18,-0.20
104.32,5.87,6.08,5.86,6.08,-6.10,-5.37,-5.63,-0.24,-0.24,-0.22
105.37,5.81,6.05,5.82,6.06,-6.10,-5.42,-5.73,-0.29,-0.28,-0.24
106.42,5.78,6.05,5.78,6.06,-6.09,-5.46,-5.83,-0.31,-0.31,-0.27
107.48,5.76,6.07,5.75,6.07,-6.08,-5.50,-5.92,-0.32,-0.32,-0.31
108.56,5.73,6.08,5.74,6.09,-6.07,-5.54,-6.01,-0.34,-0.33,-0.35
109.64,5.71,6.09,5.69,6.08,-6.06,-5.58,-6.10,-0.36,-0.37,-0.38
110.74,5.66,6.08,5.64,6.06,-6.06,-5.62,-6.19,-0.40,-0.42,-0.42
111.85,5.57,6.01,5.60,6.05,-6.06,-5.66,-6.28,-0.49,-0.46,-0.44
112.97,5.55,6.04,5.55,6.03,-6.05,-5.69,-6.36,-0.50,-0.50,-0.49
114.10,5.55,6.06,5.51,6.03,-6.04,-5.73,-6.43,-0.49,-0.53,-0.52
115.24,5.46,6.00,5.49,6.04,-6.03,-5.76,-6.50,-0.57,-0.53,-0.55
116.39,5.45,6.03,5.43,6.01,-6.00,-5.80,-6.56,-0.55,-0.56,-0.58
117.55,5.40,6.00,5.37,5.97,-5.96,-5.83,-6.62,-0.56,-0.59,-0.60
118.73,5.28,5.91,5.32,5.95,-5.92,-5.86,-6.67,-0.64,-0.59,-0.63
119.92,5.27,5.90,5.26,5.90,-5.88,-5.89,-6.71,-0.61,-0.62,-0.64
121.12,5.22,5.88,5.18,5.84,-5.84,-5.91,-6.74,-0.62,-0.66,-0.66
122.33,5.10,5.78,5.09,5.77,-5.80,-5.94,-6.76,-0.70,-0.71,-0.68
123.55,4.98,5.66,5.01,5.70,-5.76,-5.96,-6.78,-0.78,-0.75,-0.68
124.79,4.95,5.65,4.98,5.67,-5.72,-5.98,-6.78,-0.77,-0.75,-0.70
126.03,5.00,5.71,5.00,5.71,-5.69,-6.00,-6.77,-0.69,-0.68,-0.71
127.29,5.05,5.77,5.00,5.73,-5.66,-6.02,-6.76,-0.61,-0.66,-0.72
128.57,5.00,5.74,4.95,5.69,-5.64,-6.03,-6.73,-0.65,-0.69,-0.74
129.85,4.81,5.56,4.88,5.63,-5.64,-6.04,-6.70,-0.83,-0.76,-0.76
131.15,4.80,5.57,4.82,5.59,-5.63,-6.05,-6.65,-0.84,-0.82,-0.77
132.46,4.83,5.61,4.80,5.59,-5.63,-6.06,-6.60,-0.81,-0.83,-0.78
133.79,4.84,5.63,4.83,5.63,-5.64,-6.06,-6.54,-0.80,-0.80,-0.79
135.12,4.84,5.65,4.84,5.65,-5.64,-6.07,-6.48,-0.80,-0.80,-0.82
136.48,4.85,5.66,4.85,5.67,-5.66,-6.07,-6.40,-0.81,-0.81,-0.82
137.84,4.85,5.68,4.85,5.69,-5.68,-6.06,-6.32,-0.83,-0.83,-0.84
139.22,4.85,5.70,4.85,5.70,-5.70,-6.06,-6.24,-0.85,-0.85,-0.86
140.61,4.85,5.72,4.85,5.72,-5.71,-6.05,-6.15,-0.86,-0.86,-0.88
142.02,4.85,5.72,4.86,5.74,-5.72,-6.04,-6.06,-0.87,-0.86,-0.87
143.44,4.85,5.74,4.85,5.74,-5.72,-6.03,-5.96,-0.87,-0.87,-0.89
144.87,4.85,5.77,4.80,5.72,-5.72,-6.02,-5.86,-0.88,-0.92,-0.92
146.32,4.73,5.67,4.76,5.70,-5.72,-6.00,-5.76,-0.99,-0.96,-0.95
147.78,4.69,5.65,4.72,5.69,-5.72,-5.98,-5.66,-1.03,-1.00,-0.96
149.26,4.72,5.72,4.70,5.70,-5.72,-5.96,-5.56,-1.00,-1.02,-1.00
150.75,4.69,5.74,4.67,5.72,-5.71,-5.93,-5.46,-1.02,-1.04,-1.05
152.26,4.63,5.74,4.60,5.71,-5.70,-5.90,-5.35,-1.07,-1.10,-1.11
153.78,4.48,5.64,4.53,5.70,-5.68,-5.87,-5.25,-1.21,-1.15,-1.16
155.32,4.44,5.66,4.44,5.66,-5.65,-5.84,-5.15,-1.22,-1.21,-1.22
156.88,4.41,5.69,4.36,5.65,-5.61,-5.81,-5.05,-1.20,-1.25,-1.29
158.44,4.21,5.57,4.25,5.61,-5.55,-5.77,-4.96,-1.35,-1.30,-1.36
160.03,4.15,5.58,4.10,5.53,-5.50,-5.74,-4.86,-1.35,-1.40,-1.43
161.63,3.94,5.43,3.96,5.45,-5.46,-5.70,-4.77,-1.52,-1.50,-1.49
163.24,3.82,5.36,3.84,5.38,-5.44,-5.65,-4.68,-1.62,-1.60,-1.54
164.88,3.77,5.35,3.73,5.32,-5.44,-5.61,-4.59,-1.67,-1.70,-1.58
166.53,3.70,5.32,3.74,5.36,-5.45,-5.56,-4.50,-1.75,-1.70,-1.62
168.19,3.77,5.42,3.80,5.45,-5.47,-5.52,-4.42,-1.70,-1.67,-1.65
169.87,3.93,5.61,3.86,5.54,-5.50,-5.47,-4.34,-1.57,-1.64,-1.68
171.57,3.93,5.61,3.92,5.61,-5.53,-5.42,-4.27,-1.60,-1.61,-1.69
173.29,3.91,5.60,3.95,5.64,-5.56,-5.37,-4.19,-1.65,-1.61,-1.69
175.02,3.91,5.60,3.89,5.58,-5.57,-5.31,-4.12,-1.66,-1.68,-1.69
176.77,3.90,5.59,3.85,5.54,-5.54,-5.26,-4.05,-1.64,-1.69,-1.69
178.54,3.72,5.41,3.82,5.51,-5.49,-5.20,-3.99,-1.77,-1.66,-1.69
180.32,3.79,5.46,3.74,5.42,-5.42,-5.15,-3.93,-1.63,-1.67,-1.68
182.13,3.75,5.41,3.66,5.33,-5.34,-5.09,-3.87,-1.59,-1.67,-1.66
183.95,3.49,5.15,3.60,5.27,-5.26,-5.03,-3.82,-1.77,-1.65,-1.66
185.79,3.51,5.17,3.47,5.13,-5.18,-4.97,-3.76,-1.67,-1.71,-1.66
187.65,3.44,5.11,3.37,5.04,-5.11,-4.91,-3.71,-1.67,-1.73,-1.67
189.52,3.25,4.92,3.37,5.05,-5.04,-4.85,-3.67,-1.79,-1.67,-1.67
191.42,3.35,5.03,3.33,5.02,-4.99,-4.79,-3.62,-1.64,-1.66,-1.69
193.33,3.38,5.08,3.26,4.96,-4.95,-4.72,-3.58,-1.58,-1.69,-1.70
195.27,3.17,4.88,3.19,4.90,-4.92,-4.66,-3.54,-1.75,-1.73,-1.71
197.22,3.03,4.75,3.13,4.86,-4.89,-4.60,-3.51,-1.86,-1.76,-1.72
199.19,3.09,4.83,3.11,4.85,-4.87,-4.53,-3.47,-1.78,-1.76,-1.74
201.18,3.22,4.97,3.09,4.84,-4.83,-4.46,-3.44,-1.61,-1.74,-1.75
203.19,3.05,4.81,3.06,4.83,-4.77,-4.40,-3.41,-1.72,-1.70,-1.76
205.23,2.87,4.65,2.98,4.75,-4.70,-4.33,-3.38,-1.83,-1.72,-1.78
207.28,2.93,4.71,2.85,4.64,-4.63,-4.27,-3.36,-1.71,-1.79,-1.78
209.35,2.78,4.58,2.75,4.56,-4.56,-4.20,-3.33,-1.78,-1.80,-1.80
211.44,2.60,4.43,2.67,4.49,-4.49,-4.13,-3.31,-1.89,-1.82,-1.83
213.56,2.55,4.38,2.50,4.34,-4.41,-4.06,-3.29,-1.86,-1.91,-1.83
215.69,2.46,4.30,2.41,4.25,-4.32,-3.99,-3.27,-1.87,-1.91,-1.84
217.85,2.24,4.08,2.39,4.24,-4.24,-3.92,-3.25,-2.00,-1.85,-1.85
220.03,2.40,4.25,2.33,4.18,-4.15,-3.86,-3.23,-1.76,-1.83,-1.85
222.23,2.37,4.24,2.26,4.12,-4.07,-3.79,-3.21,-1.70,-1.81,-1.87
224.45,2.09,3.95,2.20,4.06,-3.98,-3.71,-3.19,-1.89,-1.78,-1.86
226.70,2.05,3.90,2.04,3.89,-3.89,-3.64,-3.17,-1.85,-1.85,-1.86
228.96,1.98,3.82,1.90,3.74,-3.81,-3.57,-3.15,-1.83,-1.91,-1.84
231.25,1.77,3.60,1.84,3.67,-3.72,-3.50,-3.14,-1.95,-1.88,-1.83
233.57,1.75,3.56,1.77,3.59,-3.63,-3.43,-3.12,-1.88,-1.86,-1.81
235.90,1.79,3.59,1.72,3.52,-3.54,-3.35,-3.10,-1.75,-1.82,-1.80
238.26,1.68,3.46,1.68,3.47,-3.44,-3.28,-3.08,-1.76,-1.76,-1.79
240.64,1.59,3.35,1.62,3.38,-3.34,-3.20,-3.05,-1.75,-1.72,-1.76
243.05,1.55,3.28,1.54,3.27,-3.23,-3.12,-3.03,-1.68,-1.69,-1.73
245.48,1.49,3.19,1.49,3.19,-3.13,-3.04,-3.01,-1.64,-1.64,-1.70
247.93,1.40,3.06,1.37,3.04,-3.02,-2.96,-2.98,-1.63,-1.65,-1.67
250.41,1.25,2.90,1.19,2.84,-2.91,-2.88,-2.95,-1.66,-1.72,-1.65
252.92,0.99,2.60,1.07,2.68,-2.78,-2.80,-2.93,-1.80,-1.72,-1.62
255.45,0.93,2.52,1.02,2.61,-2.65,-2.72,-2.90,-1.72,-1.63,-1.59
258.00,1.09,2.65,1.00,2.57,-2.52,-2.63,-2.86,-1.43,-1.52,-1.57
260.58,1.04,2.59,0.97,2.52,-2.40,-2.54,-2.83,-1.36,-1.42,-1.55
263.19,0.81,2.32,0.84,2.36,-2.28,-2.45,-2.80,-1.48,-1.45,-1.51
265.82,0.64,2.14,0.62,2.12,-2.19,-2.36,-2.76,-1.55,-1.57,-1.50
268.48,0.46,1.93,0.52,1.99,-2.09,-2.26,-2.73,-1.64,-1.58,-1.47
271.16,0.44,1.87,0.50,1.94,-1.99,-2.16,-2.69,-1.55,-1.49,-1.43
273.87,0.60,2.02,0.47,1.88,-1.88,-2.06,-2.65,-1.28,-1.41,-1.42
276.61,0.46,1.84,0.45,1.83,-1.77,-1.96,-2.61,-1.31,-1.32,-1.38
279.38,0.29,1.64,0.40,1.76,-1.69,-1.85,-2.57,-1.40,-1.28,-1.35
282.17,0.34,1.66,0.30,1.63,-1.62,-1.75,-2.53,-1.29,-1.32,-1.33
284.99,0.29,1.58,0.16,1.46,-1.55,-1.64,-2.49,-1.26,-1.38,-1.29
287.84,0.05,1.32,0.11,1.38,-1.43,-1.53,-2.45,-1.38,-1.32,-1.27
290.72,-0.12,1.11,0.09,1.33,-1.29,-1.41,-2.41,-1.41,-1.19,-1.24
293.63,0.26,1.47,0.09,1.30,-1.15,-1.30,-2.37,-0.89,-1.06,-1.21
296.57,0.13,1.30,0.01,1.19,-1.03,-1.19,-2.33,-0.90,-1.02,-1.17
299.53,-0.20,0.94,-0.19,0.96,-0.95,-1.09,-2.29,-1.15,-1.13,-1.14
302.53,-0.54,0.58,-0.46,0.65,-0.89,-0.98,-2.25,-1.43,-1.35,-1.13
305.55,-0.58,0.50,-0.49,0.60,-0.83,-0.89,-2.21,-1.42,-1.32,-1.08
308.61,-0.37,0.69,-0.37,0.69,-0.79,-0.80,-2.17,-1.16,-1.16,-1.06
311.69,-0.03,1.00,-0.18,0.85,-0.76,-0.72,-2.13,-0.79,-0.94,-1.03
314.81,-0.13,0.87,-0.06,0.95,-0.76,-0.65,-2.10,-0.89,-0.82,-1.00
317.96,-0.06,0.92,-0.08,0.91,-0.79,-0.59,-2.06,-0.85,-0.87,-0.99
321.14,-0.14,0.84,-0.21,0.76,-0.81,-0.55,-2.03,-0.96,-1.02,-0.98
324.35,-0.29,0.67,-0.23,0.73,-0.81,-0.52,-1.99,-1.10,-1.04,-0.96
327.59,-0.32,0.62,-0.17,0.78,-0.77,-0.50,-1.96,-1.09,-0.94,-0.94
330.87,-0.00,0.93,-0.12,0.81,-0.72,-0.50,-1.93,-0.72,-0.83,-0.93
334.18,0.03,0.94,-0.16,0.76,-0.68,-0.51,-1.90,-0.65,-0.83,-0.91
337.52,-0.37,0.53,-0.28,0.63,-0.65,-0.53,-1.87,-1.03,-0.93,-0.91
340.90,-0.53,0.35,-0.40,0.49,-0.65,-0.56,-1.84,-1.18,-1.05,-0.89
344.30,-0.38,0.50,-0.39,0.49,-0.64,-0.60,-1.82,-1.03,-1.04,-0.88
347.75,-0.18,0.70,-0.25,0.63,-0.62,-0.65,-1.79,-0.80,-0.87,-0.88
351.23,-0.12,0.75,-0.12,0.75,-0.59,-0.70,-1.77,-0.71,-0.71,-0.87
354.74,-0.12,0.75,-0.12,0.75,-0.57,-0.75,-1.74,-0.69,-0.69,-0.87
358.28,-0.16,0.70,-0.25,0.62,-0.58,-0.81,-1.72,-0.75,-0.83,-0.86
361.87,-0.36,0.50,-0.29,0.57,-0.62,-0.86,-1.70,-0.98,-0.91,-0.86
365.49,-0.43,0.42,-0.30,0.56,-0.68,-0.91,-1.68,-1.11,-0.98,-0.86
369.14,-0.09,0.77,-0.32,0.55,-0.73,-0.97,-1.67,-0.82,-1.05,-0.87
372.83,-0.21,0.66,-0.16,0.71,-0.78,-1.02,-1.65,-0.99,-0.93,-0.87
376.56,-0.19,0.68,0.04,0.91,-0.85,-1.07,-1.64,-1.04,-0.80,-0.87
380.33,0.41,1.27,0.17,1.03,-0.94,-1.12,-1.62,-0.54,-0.77,-0.86
384.13,0.36,1.22,0.32,1.18,-1.06,-1.16,-1.61,-0.70,-0.75,-0.86
387.97,0.28,1.15,0.37,1.24,-1.18,-1.20,-1.60,-0.90,-0.80,-0.87
391.85,0.32,1.20,0.31,1.18,-1.28,-1.24,-1.59,-0.97,-0.97,-0.88
395.77,0.40,1.27,0.40,1.28,-1.38,-1.28,-1.58,-0.98,-0.97,-0.87
399.73,0.49,1.37,0.55,1.44,-1.44,-1.31,-1.57,-0.96,-0.89,-0.88
403.72,0.75,1.63,0.64,1.53,-1.50,-1.34,-1.56,-0.76,-0.86,-0.88
407.76,0.74,1.63,0.70,1.59,-1.55,-1.37,-1.55,-0.82,-0.85,-0.89
411.84,0.64,1.53,0.75,1.64,-1.60,-1.40,-1.55,-0.96,-0.85,-0.90
415.96,0.76,1.65,0.75,1.64,-1.64,-1.43,-1.54,-0.88,-0.90,-0.90
420.12,0.84,1.73,0.77,1.66,-1.68,-1.45,-1.54,-0.84,-0.91,-0.90
424.32,0.76,1.65,0.79,1.68,-1.69,-1.47,-1.53,-0.93,-0.90,-0.90
428.56,0.75,1.64,0.77,1.66,-1.68,-1.49,-1.53,-0.93,-0.91,-0.89
432.85,0.77,1.65,0.76,1.64,-1.65,-1.51,-1.52,-0.88,-0.89,-0.88
437.18,0.79,1.66,0.80,1.66,-1.62,-1.53,-1.52,-0.83,-0.81,-0.87
441.55,0.83,1.66,0.84,1.68,-1.59,-1.55,-1.52,-0.76,-0.75,-0.83
445.96,0.87,1.67,0.79,1.59,-1.57,-1.56,-1.52,-0.70,-0.79,-0.80
450.42,0.76,1.53,0.74,1.50,-1.56,-1.58,-1.51,-0.80,-0.82,-0.77
454.93,0.59,1.31,0.74,1.46,-1.55,-1.59,-1.51,-0.96,-0.81,-0.72
459.48,0.84,1.51,0.80,1.47,-1.55,-1.61,-1.51,-0.72,-0.75,-0.67
464.07,0.99,1.61,0.93,1.55,-1.56,-1.62,-1.50,-0.57,-0.63,-0.62
468.71,1.06,1.62,1.07,1.64,-1.57,-1.63,-1.50,-0.51,-0.50,-0.56
473.40,1.12,1.64,1.13,1.64,-1.59,-1.64,-1.49,-0.47,-0.46,-0.52
478.13,1.18,1.64,1.17,1.64,-1.61,-1.65,-1.49,-0.44,-0.44,-0.46
482.91,1.22,1.64,1.22,1.64,-1.64,-1.66,-1.49,-0.42,-0.42,-0.42
487.74,1.26,1.64,1.26,1.64,-1.65,-1.67,-1.48,-0.39,-0.39,-0.38
492.62,1.30,1.63,1.30,1.64,-1.65,-1.68,-1.47,-0.35,-0.36,-0.33
497.55,1.33,1.64,1.33,1.64,-1.64,-1.69,-1.47,-0.32,-0.32,-0.31
502.52,1.35,1.63,1.35,1.64,-1.64,-1.70,-1.46,-0.29,-0.29,-0.28
507.55,1.37,1.64,1.37,1.64,-1.64,-1.70,-1.45,-0.28,-0.27,-0.27
512.62,1.39,1.65,1.39,1.65,-1.64,-1.71,-1.44,-0.26,-0.26,-0.26
517.75,1.40,1.65,1.39,1.65,-1.64,-1.72,-1.43,-0.25,-0.25,-0.26
522.93,1.39,1.65,1.39,1.66,-1.65,-1.72,-1.42,-0.27,-0.26,-0.27
528.16,1.38,1.67,1.38,1.67,-1.67,-1.73,-1.41,-0.30,-0.29,-0.30
533.44,1.37,1.69,1.36,1.69,-1.71,-1.73,-1.40,-0.34,-0.34,-0.33
538.77,1.34,1.69,1.32,1.68,-1.73,-1.74,-1.39,-0.40,-0.41,-0.36
544.16,1.30,1.70,1.29,1.69,-1.76,-1.74,-1.38,-0.46,-0.47,-0.41
549.60,1.25,1.70,1.33,1.79,-1.78,-1.75,-1.36,-0.53,-0.45,-0.46
555.10,1.34,1.85,1.35,1.86,-1.79,-1.75,-1.35,-0.45,-0.44,-0.52
560.65,1.45,2.01,1.29,1.86,-1.78,-1.75,-1.34,-0.33,-0.49,-0.57
566.25,1.14,1.75,1.19,1.81,-1.75,-1.75,-1.33,-0.61,-0.56,-0.62
571.92,0.96,1.63,1.03,1.71,-1.70,-1.75,-1.31,-0.74,-0.67,-0.68
577.64,0.93,1.66,0.93,1.66,-1.66,-1.75,-1.30,-0.73,-0.73,-0.74
583.41,0.90,1.67,0.88,1.65,-1.63,-1.75,-1.29,-0.74,-0.76,-0.77
589.25,0.88,1.68,0.75,1.55,-1.63,-1.75,-1.28,-0.75,-0.87,-0.80
595.14,0.61,1.43,0.68,1.50,-1.64,-1.75,-1.27,-1.03,-0.96,-0.82
601.09,0.55,1.38,0.68,1.51,-1.67,-1.74,-1.25,-1.12,-0.99,-0.83
607.10,0.92,1.75,0.82,1.64,-1.70,-1.74,-1.24,-0.79,-0.88,-0.83
613.17,1.02,1.83,1.06,1.86,-1.73,-1.73,-1.23,-0.71,-0.68,-0.81
619.30,1.25,2.03,1.18,1.96,-1.75,-1.73,-1.22,-0.50,-0.56,-0.78
625.50,1.22,1.96,1.16,1.89,-1.76,-1.72,-1.21,-0.54,-0.60,-0.74
631.75,1.02,1.71,1.10,1.79,-1.77,-1.72,-1.20,-0.75,-0.66,-0.69
638.07,1.01,1.65,1.02,1.66,-1.76,-1.71,-1.20,-0.75,-0.73,-0.64
644.45,1.04,1.63,1.02,1.61,-1.73,-1.70,-1.19,-0.69,-0.71,-0.59
650.89,1.08,1.61,1.06,1.59,-1.67,-1.69,-1.18,-0.60,-0.61,-0.53
657.40,1.13,1.61,1.14,1.61,-1.62,-1.68,-1.18,-0.49,-0.48,-0.48
663.98,1.18,1.60,1.25,1.68,-1.59,-1.67,-1.17,-0.42,-0.34,-0.42
670.62,1.37,1.75,1.28,1.66,-1.60,-1.65,-1.17,-0.23,-0.32,-0.38
677.32,1.39,1.73,1.27,1.61,-1.62,-1.64,-1.17,-0.23,-0.35,-0.34
684.10,1.10,1.41,1.26,1.56,-1.64,-1.63,-1.17,-0.54,-0.38,-0.31
690.94,1.26,1.53,1.31,1.58,-1.65,-1.61,-1.17,-0.40,-0.34,-0.27
697.85,1.46,1.71,1.42,1.67,-1.67,-1.60,-1.17,-0.21,-0.25,-0.25
704.83,1.63,1.87,1.51,1.75,-1.67,-1.58,-1.17,-0.04,-0.16,-0.24
711.87,1.46,1.70,1.47,1.71,-1.67,-1.56,-1.17,-0.21,-0.19,-0.24
718.99,1.34,1.59,1.42,1.67,-1.66,-1.55,-1.17,-0.32,-0.24,-0.25
726.18,1.34,1.61,1.38,1.65,-1.65,-1.53,-1.17,-0.31,-0.27,-0.27
733.44,1.46,1.75,1.36,1.65,-1.64,-1.51,-1.18,-0.18,-0.28,-0.29
740.78,1.33,1.64,1.30,1.62,-1.61,-1.49,-1.19,-0.28,-0.31,-0.31
748.19,1.15,1.49,1.19,1.54,-1.58,-1.47,-1.19,-0.43,-0.38,-0.34
755.67,1.09,1.47,1.10,1.48,-1.55,-1.45,-1.20,-0.46,-0.44,-0.38
763.23,1.07,1.48,1.12,1.52,-1.54,-1.42,-1.21,-0.47,-0.42,-0.41
770.86,1.19,1.62,1.13,1.56,-1.53,-1.40,-1.22,-0.34,-0.40,-0.43
778.57,1.17,1.61,1.09,1.53,-1.51,-1.38,-1.22,-0.34,-0.42,-0.44
786.35,0.98,1.43,1.04,1.49,-1.49,-1.36,-1.23,-0.51,-0.45,-0.45
794.22,0.93,1.40,1.01,1.47,-1.46,-1.33,-1.24,-0.53,-0.45,-0.47
802.16,1.04,1.51,1.00,1.47,-1.44,-1.31,-1.25,-0.40,-0.44,-0.47
810.18,1.07,1.54,0.98,1.45,-1.41,-1.28,-1.26,-0.35,-0.44,-0.47
818.28,0.89,1.35,0.89,1.35,-1.38,-1.26,-1.28,-0.50,-0.49,-0.46
826.46,0.76,1.21,0.84,1.29,-1.36,-1.24,-1.29,-0.60,-0.52,-0.45
834.73,0.78,1.22,0.83,1.27,-1.33,-1.21,-1.30,-0.55,-0.50,-0.44
843.08,1.01,1.42,0.88,1.29,-1.30,-1.18,-1.31,-0.30,-0.43,-0.41
851.51,0.91,1.30,0.94,1.33,-1.27,-1.16,-1.32,-0.37,-0.34,-0.39
860.02,0.89,1.25,0.94,1.30,-1.24,-1.13,-1.33,-0.35,-0.31,-0.36
868.62,0.93,1.28,0.86,1.21,-1.20,-1.11,-1.33,-0.28,-0.34,-0.35
877.31,0.82,1.14,0.84,1.16,-1.15,-1.08,-1.34,-0.34,-0.32,-0.32
886.08,0.75,1.04,0.80,1.09,-1.09,-1.05,-1.35,-0.34,-0.29,-0.29
894.94,0.79,1.06,0.74,1.02,-1.01,-1.03,-1.35,-0.22,-0.27,-0.27
903.89,0.70,0.96,0.69,0.94,-0.93,-1.00,-1.35,-0.23,-0.24,-0.26
912.93,0.60,0.84,0.61,0.85,-0.85,-0.97,-1.36,-0.25,-0.23,-0.24
922.06,0.51,0.72,0.51,0.72,-0.77,-0.95,-1.35,-0.27,-0.27,-0.21
931.28,0.45,0.65,0.44,0.64,-0.70,-0.92,-1.35,-0.26,-0.26,-0.20
940.59,0.39,0.57,0.43,0.61,-0.62,-0.89,-1.35,-0.24,-0.19,-0.18
950.00,0.44,0.60,0.42,0.58,-0.54,-0.87,-1.34,-0.11,-0.13,-0.16
959.50,0.44,0.59,0.38,0.52,-0.46,-0.84,-1.33,-0.02,-0.08,-0.15
969.09,0.28,0.41,0.32,0.45,-0.38,-0.81,-1.31,-0.10,-0.06,-0.13
978.78,0.20,0.33,0.21,0.34,-0.31,-0.79,-1.30,-0.11,-0.10,-0.13
988.57,0.13,0.27,0.09,0.22,-0.25,-0.76,-1.28,-0.12,-0.15,-0.14
998.46,0.01,0.14,0.02,0.16,-0.20,-0.73,-1.26,-0.19,-0.18,-0.13
1008.44,-0.04,0.09,-0.01,0.13,-0.18,-0.70,-1.23,-0.22,-0.19,-0.13
1018.53,0.02,0.15,0.00,0.13,-0.20,-0.68,-1.20,-0.18,-0.20,-0.13
1028.71,0.09,0.22,0.06,0.18,-0.25,-0.65,-1.17,-0.16,-0.19,-0.13
1039.00,0.13,0.25,0.15,0.28,-0.33,-0.62,-1.13,-0.20,-0.18,-0.12
1049.39,0.27,0.39,0.30,0.42,-0.43,-0.59,-1.10,-0.17,-0.14,-0.12
1059.88,0.48,0.59,0.48,0.58,-0.54,-0.57,-1.06,-0.06,-0.07,-0.11
1070.48,0.68,0.77,0.64,0.73,-0.64,-0.54,-1.01,0.03,-0.01,-0.09
1081.19,0.76,0.83,0.74,0.81,-0.73,-0.51,-0.97,0.02,0.00,-0.07
1092.00,0.76,0.80,0.79,0.83,-0.80,-0.48,-0.92,-0.04,-0.01,-0.04
1102.92,0.80,0.81,0.81,0.83,-0.84,-0.46,-0.86,-0.04,-0.02,-0.01
1113.95,0.87,0.85,0.85,0.83,-0.84,-0.43,-0.81,0.02,0.01,0.02
1125.09,0.89,0.82,0.89,0.82,-0.82,-0.40,-0.75,0.07,0.07,0.07
1136.34,0.91,0.79,0.91,0.79,-0.78,-0.37,-0.70,0.13,0.14,0.12
1147.70,0.91,0.72,0.92,0.74,-0.72,-0.34,-0.64,0.19,0.20,0.19
1159.18,0.93,0.69,0.90,0.66,-0.66,-0.31,-0.57,0.27,0.24,0.24
1170.77,0.88,0.58,0.88,0.58,-0.59,-0.28,-0.51,0.29,0.29,0.30
1182.48,0.84,0.47,0.87,0.51,-0.52,-0.26,-0.44,0.32,0.35,0.37
1194.30,0.88,0.45,0.86,0.43,-0.45,-0.23,-0.38,0.43,0.41,0.43
1206.25,0.87,0.39,0.86,0.37,-0.37,-0.20,-0.31,0.49,0.48,0.48
1218.31,0.85,0.30,0.86,0.31,-0.30,-0.17,-0.24,0.55,0.56,0.55
1230.49,0.84,0.23,0.85,0.24,-0.23,-0.14,-0.17,0.60,0.61,0.61
1242.80,0.86,0.19,0.84,0.17,-0.17,-0.10,-0.10,0.69,0.67,0.67
1255.22,0.85,0.11,0.85,0.11,-0.12,-0.07,-0.03,0.73,0.74,0.74
1267.78,0.85,0.04,0.87,0.06,-0.06,-0.04,0.05,0.79,0.80,0.81
1280.45,0.90,0.02,0.88,0.00,-0.01,-0.01,0.12,0.88,0.87,0.88
1293.26,0.91,-0.03,0.91,-0.03,0.03,0.02,0.20,0.94,0.94,0.94
1306.19,0.92,-0.08,0.94,-0.06,0.07,0.06,0.27,0.99,1.01,1.00
1319.25,0.97,-0.09,0.95,-0.11,0.10,0.09,0.35,1.07,1.05,1.06
1332.45,0.99,-0.13,0.98,-0.13,0.13,0.12,0.42,1.12,1.11,1.12
1345.77,0.99,-0.18,1.01,-0.16,0.16,0.16,0.50,1.14,1.17,1.17
1359.23,1.05,-0.17,1.03,-0.19,0.19,0.20,0.58,1.24,1.22,1.22
1372.82,1.06,-0.22,1.05,-0.22,0.23,0.23,0.66,1.29,1.28,1.28
1386.55,1.06,-0.27,1.08,-0.25,0.28,0.27,0.74,1.33,1.36,1.33
1400.41,1.10,-0.28,1.09,-0.29,0.32,0.31,0.82,1.42,1.41,1.38
1414.42,1.11,-0.32,1.08,-0.35,0.37,0.35,0.90,1.48,1.45,1.43
1428.56,1.06,-0.43,1.06,-0.43,0.41,0.39,0.98,1.47,1.47,1.49
1442.85,1.03,-0.52,1.04,-0.51,0.45,0.43,1.07,1.48,1.49,1.55
1457.28,1.03,-0.57,1.06,-0.54,0.50,0.47,1.15,1.52,1.56,1.60
1471.85,1.12,-0.54,1.10,-0.55,0.54,0.52,1.23,1.66,1.65,1.66
1486.57,1.19,-0.52,1.15,-0.56,0.59,0.56,1.32,1.78,1.74,1.71
1501.43,1.17,-0.61,1.20,-0.58,0.63,0.61,1.41,1.80,1.83,1.78
1516.45,1.20,-0.66,1.21,-0.64,0.68,0.66,1.49,1.87,1.89,1.86
1531.61,1.26,-0.68,1.22,-0.73,0.72,0.71,1.58,1.98,1.94,1.94
1546.93,1.22,-0.81,1.23,-0.80,0.77,0.76,1.66,1.99,2.00,2.03
1562.40,1.23,-0.89,1.25,-0.87,0.84,0.82,1.75,2.07,2.09,2.12
1578.02,1.28,-0.92,1.29,-0.91,0.91,0.87,1.84,2.19,2.20,2.20
1593.80,1.35,-0.93,1.33,-0.95,0.99,0.93,1.93,2.33,2.31,2.28
1609.74,1.38,-0.99,1.34,-1.03,1.06,0.99,2.02,2.43,2.40,2.37
1625.84,1.32,-1.15,1.34,-1.12,1.12,1.06,2.10,2.44,2.46,2.47
1642.10,1.33,-1.23,1.35,-1.21,1.18,1.12,2.19,2.51,2.53,2.56
1658.52,1.39,-1.27,1.37,-1.28,1.24,1.19,2.28,2.63,2.61,2.66
1675.10,1.45,-1.30,1.42,-1.33,1.30,1.26,2.37,2.75,2.72,2.75
1691.85,1.44,-1.40,1.48,-1.37,1.37,1.34,2.45,2.81,2.85,2.84
1708.77,1.52,-1.42,1.53,-1.41,1.44,1.42,2.54,2.96,2.97,2.94
1725.86,1.61,-1.43,1.56,-1.48,1.51,1.50,2.62,3.11,3.07,3.04
1743.12,1.58,-1.56,1.59,-1.55,1.58,1.59,2.70,3.16,3.16,3.14
1760.55,1.56,-1.68,1.58,-1.66,1.66,1.68,2.78,3.21,3.24,3.24
1778.15,1.58,-1.76,1.56,-1.77,1.75,1.77,2.86,3.32,3.31,3.34
1795.94,1.57,-1.85,1.57,-1.85,1.85,1.86,2.93,3.41,3.42,3.42
1813.90,1.56,-1.96,1.58,-1.94,1.95,1.96,3.00,3.51,3.53,3.52
1832.03,1.60,-2.01,1.57,-2.04,2.06,2.06,3.07,3.66,3.63,3.61
1850.36,1.55,-2.16,1.55,-2.16,2.17,2.17,3.13,3.71,3.71,3.71
1868.86,1.51,-2.29,1.52,-2.27,2.27,2.27,3.19,3.78,3.79,3.80
1887.55,1.50,-2.39,1.51,-2.38,2.37,2.38,3.25,3.87,3.87,3.89
1906.42,1.52,-2.46,1.50,-2.48,2.47,2.48,3.30,3.99,3.97,3.98
1925.49,1.52,-2.57,1.51,-2.58,2.57,2.58,3.35,4.09,4.08,4.09
1944.74,1.51,-2.68,1.54,-2.65,2.67,2.68,3.39,4.18,4.21,4.19
1964.19,1.57,-2.74,1.57,-2.74,2.75,2.77,3.43,4.32,4.32,4.31
1983.83,1.64,-2.77,1.59,-2.82,2.83,2.86,3.46,4.46,4.42,4.41
2003.67,1.61,-2.93,1.62,-2.91,2.89,2.94,3.48,4.50,4.51,4.54
2023.71,1.62,-3.02,1.65,-2.99,2.95,3.02,3.50,4.57,4.60,4.64
2043.94,1.72,-3.04,1.72,-3.04,3.00,3.08,3.52,4.72,4.72,4.76
2064.38,1.83,-3.04,1.81,-3.06,3.06,3.14,3.53,4.89,4.87,4.87
2085.03,1.91,-3.07,1.89,-3.09,3.13,3.19,3.53,5.04,5.02,4.98
2105.88,1.92,-3.17,1.95,-3.13,3.20,3.23,3.53,5.12,5.15,5.09
2126.94,2.00,-3.20,1.99,-3.21,3.27,3.27,3.53,5.27,5.25,5.20
2148.20,2.03,-3.28,1.99,-3.32,3.33,3.30,3.52,5.36,5.32,5.31
2169.69,1.97,-3.45,1.99,-3.43,3.40,3.33,3.51,5.37,5.39,5.42
2191.38,1.97,-3.55,1.99,-3.53,3.46,3.36,3.49,5.43,5.46,5.52
2213.30,2.04,-3.57,2.03,-3.58,3.53,3.38,3.47,5.56,5.56,5.61
2235.43,2.12,-3.59,2.12,-3.59,3.58,3.41,3.45,5.70,5.70,5.71
2257.78,2.21,-3.59,2.20,-3.59,3.62,3.44,3.43,5.82,5.82,5.80
2280.36,2.28,-3.59,2.27,-3.61,3.63,3.47,3.40,5.91,5.90,5.87
2303.17,2.31,-3.64,2.32,-3.63,3.64,3.51,3.37,5.95,5.96,5.95
2326.20,2.36,-3.66,2.36,-3.66,3.64,3.55,3.34,6.00,6.00,6.02
2349.46,2.40,-3.69,2.40,-3.68,3.65,3.60,3.31,6.04,6.05,6.09
2372.95,2.46,-3.67,2.46,-3.67,3.66,3.64,3.28,6.12,6.12,6.13
2396.68,2.51,-3.67,2.51,-3.67,3.70,3.70,3.25,6.21,6.21,6.18
2420.65,2.56,-3.67,2.55,-3.68,3.74,3.76,3.22,6.30,6.29,6.23
2444.86,2.58,-3.71,2.55,-3.74,3.79,3.82,3.19,6.36,6.34,6.29
2469.31,2.53,-3.83,2.55,-3.82,3.84,3.89,3.16,6.37,6.38,6.36
2494.00,2.53,-3.93,2.54,-3.91,3.89,3.96,3.13,6.42,6.44,6.46
2518.94,2.58,-3.98,2.56,-4.00,3.94,4.04,3.11,6.52,6.50,6.56
2544.13,2.62,-4.04,2.62,-4.05,4.00,4.12,3.08,6.61,6.61,6.66
2569.57,2.68,-4.10,2.72,-4.07,4.04,4.20,3.06,6.72,6.76,6.78
2595.27,2.85,-4.07,2.83,-4.08,4.08,4.29,3.03,6.92,6.91,6.92
2621.22,3.00,-4.06,2.96,-4.10,4.10,4.38,3.01,7.10,7.06,7.06
2647.43,3.06,-4.14,3.09,-4.11,4.13,4.47,2.99,7.18,7.22,7.20
2673.90,3.21,-4.14,3.21,-4.14,4.15,4.56,2.98,7.36,7.36,7.35
2700.64,3.33,-4.16,3.31,-4.18,4.18,4.65,2.96,7.51,7.49,7.49
2727.65,3.41,-4.23,3.41,-4.23,4.22,4.75,2.95,7.63,7.63,7.64
2754.93,3.48,-4.30,3.48,-4.30,4.28,4.84,2.94,7.76,7.76,7.78
2782.48,3.54,-4.39,3.54,-4.39,4.36,4.93,2.93,7.90,7.91,7.93
2810.30,3.59,-4.49,3.59,-4.49,4.50,5.02,2.93,8.09,8.09,8.08
2838.40,3.63,-4.60,3.62,-4.60,4.68,5.10,2.93,8.31,8.30,8.23
2866.79,3.63,-4.75,3.64,-4.74,4.88,5.19,2.93,8.51,8.53,8.38
2895.46,3.59,-4.95,3.56,-4.98,5.12,5.26,2.93,8.71,8.68,8.54
2924.41,3.50,-5.21,3.42,-5.29,5.36,5.34,2.94,8.85,8.78,8.71
2953.65,3.18,-5.70,3.25,-5.63,5.57,5.40,2.94,8.75,8.82,8.88
2983.19,3.13,-5.93,3.14,-5.92,5.72,5.46,2.95,8.85,8.86,9.06
3013.02,3.13,-6.11,3.13,-6.11,5.81,5.51,2.97,8.94,8.94,9.24
3043.15,3.32,-6.11,3.30,-6.13,5.84,5.55,2.98,9.16,9.13,9.43
3073.58,3.55,-6.06,3.61,-6.01,5.80,5.59,3.00,9.35,9.41,9.61
3104.32,4.00,-5.80,4.03,-5.77,5.73,5.61,3.02,9.73,9.77,9.80
3135.36,4.53,-5.45,4.46,-5.51,5.66,5.62,3.04,10.19,10.12,9.98
3166.72,4.84,-5.31,4.81,-5.33,5.60,5.63,3.06,10.44,10.40,10.15
3198.38,4.98,-5.30,4.97,-5.32,5.57,5.62,3.09,10.54,10.53,10.28
3230.37,4.96,-5.45,4.97,-5.44,5.58,5.60,3.11,10.54,10.55,10.41
3262.67,4.87,-5.65,4.89,-5.63,5.63,5.58,3.14,10.50,10.52,10.52
3295.30,4.76,-5.86,4.76,-5.85,5.71,5.54,3.17,10.46,10.47,10.62
3328.25,4.60,-6.10,4.56,-6.14,5.81,5.50,3.20,10.41,10.37,10.70
3361.53,4.32,-6.44,4.29,-6.47,5.91,5.44,3.24,10.23,10.21,10.76
3395.15,3.97,-6.84,4.01,-6.80,5.99,5.38,3.27,9.96,10.00,10.81
3429.10,3.75,-7.09,3.76,-7.08,6.05,5.31,3.30,9.80,9.81,10.84
3463.39,3.62,-7.23,3.60,-7.25,6.08,5.24,3.33,9.70,9.68,10.85
3498.03,3.54,-7.31,3.54,-7.31,6.08,5.16,3.37,9.62,9.62,10.85
3533.01,3.62,-7.22,3.66,-7.17,6.05,5.07,3.40,9.66,9.70,10.84
3568.34,3.74,-7.05,3.84,-6.94,5.99,4.98,3.43,9.73,9.83,10.79
3604.02,4.17,-6.53,3.98,-6.72,5.91,4.89,3.46,10.08,9.90,10.70
3640.06,4.00,-6.60,3.95,-6.65,5.80,4.79,3.49,9.79,9.75,10.60
3676.46,3.80,-6.69,3.79,-6.69,5.64,4.69,3.51,9.44,9.43,10.49
3713.22,3.54,-6.81,3.63,-6.72,5.45,4.59,3.54,8.99,9.08,10.35
3750.36,3.66,-6.55,3.74,-6.46,5.22,4.48,3.56,8.88,8.97,10.21
3787.86,4.08,-5.96,4.07,-5.97,4.97,4.38,3.58,9.04,9.04,10.04
3825.74,4.56,-5.29,4.47,-5.38,4.69,4.27,3.59,9.25,9.16,9.85
3864.00,4.87,-4.78,4.83,-4.82,4.40,4.17,3.60,9.27,9.23,9.65
3902.64,4.99,-4.45,5.09,-4.36,4.12,4.06,3.61,9.11,9.21,9.44
3941.66,5.32,-3.92,5.24,-4.00,3.86,3.95,3.62,9.18,9.10,9.24
3981.08,5.36,-3.67,5.36,-3.67,3.62,3.84,3.62,8.98,8.98,9.03
4020.89,5.39,-3.43,5.45,-3.37,3.41,3.73,3.61,8.80,8.85,8.82
4061.10,5.49,-3.12,5.45,-3.16,3.22,3.62,3.61,8.71,8.68,8.61
4101.71,5.46,-2.95,5.45,-2.97,3.06,3.50,3.59,8.52,8.51,8.41
4142.73,5.38,-2.86,5.38,-2.85,2.92,3.39,3.58,8.29,8.30,8.24
4184.15,5.27,-2.79,5.26,-2.80,2.78,3.27,3.56,8.05,8.04,8.06
4226.00,5.14,-2.75,5.15,-2.74,2.66,3.15,3.54,7.79,7.81,7.89
4268.26,5.06,-2.66,5.07,-2.64,2.54,3.02,3.51,7.59,7.61,7.72
4310.94,5.02,-2.50,5.03,-2.49,2.42,2.89,3.48,7.44,7.45,7.52
4354.05,5.03,-2.29,5.01,-2.31,2.30,2.75,3.45,7.33,7.31,7.32
4397.59,4.98,-2.14,4.98,-2.14,2.19,2.61,3.42,7.17,7.17,7.12
4441.56,4.91,-2.00,4.90,-2.01,2.10,2.45,3.38,7.01,7.01,6.91
4485.98,4.80,-1.90,4.78,-1.92,2.04,2.29,3.34,6.84,6.82,6.70
4530.84,4.62,-1.87,4.63,-1.85,1.99,2.11,3.30,6.61,6.63,6.49
4576.15,4.46,-1.81,4.45,-1.83,1.92,1.92,3.26,6.38,6.37,6.27
4621.91,4.28,-1.79,4.26,-1.80,1.78,1.70,3.22,6.06,6.04,6.07
4668.13,4.08,-1.76,4.09,-1.75,1.57,1.47,3.18,5.65,5.66,5.84
4714.81,3.99,-1.60,3.94,-1.65,1.29,1.21,3.14,5.27,5.22,5.59
4761.96,3.96,-1.36,4.01,-1.30,0.95,0.92,3.09,4.90,4.96,5.32
4809.58,4.14,-0.88,4.34,-0.68,0.57,0.60,3.05,4.70,4.91,5.02
4857.67,4.88,0.16,4.79,0.08,0.14,0.25,3.01,5.02,4.93,4.72
4906.25,5.26,0.86,5.07,0.68,-0.32,-0.14,2.97,4.94,4.75,4.40
4955.31,5.21,1.14,5.08,1.00,-0.77,-0.56,2.93,4.43,4.30,4.07
5004.87,4.70,0.94,4.93,1.17,-1.18,-0.99,2.90,3.52,3.74,3.76
5054.91,4.69,1.24,4.82,1.37,-1.52,-1.43,2.86,3.17,3.31,3.45
5105.46,4.98,1.83,4.82,1.67,-1.76,-1.84,2.83,3.22,3.06,3.15
5156.52,4.89,2.03,4.85,2.00,-1.91,-2.18,2.79,2.98,2.94,2.86
5208.08,4.67,2.10,4.69,2.12,-2.01,-2.39,2.76,2.65,2.67,2.57
5260.16,4.40,2.11,4.36,2.07,-2.10,-2.46,2.74,2.30,2.27,2.29
5312.77,4.02,2.00,4.07,2.04,-2.15,-2.35,2.71,1.87,1.92,2.02
5365.89,3.77,1.99,3.82,2.04,-2.15,-2.10,2.69,1.62,1.67,1.78
5419.55,3.62,2.07,3.58,2.02,-2.07,-1.74,2.67,1.55,1.52,1.55
5473.75,3.39,2.02,3.33,1.96,-1.90,-1.31,2.65,1.49,1.43,1.37
5528.49,2.97,1.75,3.01,1.79,-1.65,-0.85,2.63,1.32,1.36,1.22
5583.77,2.56,1.47,2.56,1.46,-1.33,-0.40,2.62,1.23,1.23,1.09
5639.61,2.09,1.08,2.02,1.01,-0.93,0.03,2.61,1.15,1.08,1.01
5696.00,1.41,0.48,1.44,0.50,-0.45,0.42,2.60,0.96,0.99,0.93
5752.96,0.78,-0.10,0.81,-0.07,0.11,0.78,2.60,0.89,0.93,0.88
5810.49,0.19,-0.64,0.16,-0.69,0.72,1.11,2.60,0.91,0.88,0.83
5868.60,-0.44,-1.24,-0.46,-1.28,1.32,1.40,2.60,0.88,0.86,0.80
5927.28,-1.14,-1.91,-1.10,-1.90,1.87,1.67,2.60,0.73,0.76,0.77
5986.56,-1.68,-2.43,-1.70,-2.46,2.31,1.91,2.61,0.63,0.61,0.75
6046.42,-2.26,-2.97,-2.15,-2.89,2.64,2.13,2.62,0.38,0.49,0.71
6106.89,-2.51,-3.17,-2.45,-3.15,2.88,2.33,2.63,0.36,0.43,0.66
6167.96,-2.63,-3.25,-2.58,-3.25,3.03,2.51,2.64,0.40,0.45,0.62
6229.64,-2.67,-3.24,-2.50,-3.14,3.14,2.68,2.66,0.47,0.64,0.57
6291.93,-2.54,-3.06,-2.37,-2.98,3.24,2.83,2.68,0.70,0.87,0.52
6354.85,-2.34,-2.81,-2.33,-2.92,3.32,2.96,2.70,0.98,0.99,0.47
6418.40,-2.60,-3.01,-2.54,-3.11,3.40,3.09,2.73,0.79,0.86,0.41
6482.58,-3.21,-3.58,-2.91,-3.46,3.46,3.20,2.75,0.25,0.55,0.37
6547.41,-3.88,-4.18,-3.28,-3.83,3.51,3.29,2.78,-0.37,0.23,0.30
6612.88,-4.05,-4.29,-3.46,-4.02,3.55,3.37,2.81,-0.50,0.09,0.24
6679.01,-4.08,-4.27,-3.36,-3.93,3.57,3.44,2.85,-0.51,0.21,0.19
6745.80,-3.93,-4.06,-3.09,-3.68,3.55,3.49,2.88,-0.38,0.47,0.13
6813.26,-3.46,-3.52,-2.80,-3.42,3.47,3.53,2.91,0.01,0.67,0.06
6881.39,-3.15,-3.12,-2.52,-3.18,3.33,3.56,2.95,0.18,0.81,-0.03
6950.21,-2.91,-2.78,-2.33,-3.02,3.18,3.58,2.99,0.27,0.85,-0.13
7019.71,-2.57,-2.33,-2.25,-2.97,3.04,3.58,3.02,0.47,0.79,-0.24
7089.91,-2.34,-1.99,-2.17,-2.92,2.95,3.58,3.06,0.61,0.78,-0.35
7160.81,-2.08,-1.59,-2.12,-2.89,2.93,3.56,3.10,0.85,0.81,-0.49
7232.41,-1.32,-0.69,-2.15,-2.93,2.99,3.54,3.13,1.66,0.84,-0.63
7304.74,-1.08,-0.30,-2.25,-3.03,3.08,3.52,3.16,2.00,0.83,-0.78
7377.79,-1.26,-0.33,-2.41,-3.18,3.20,3.50,3.19,1.94,0.79,-0.93
7451.56,-1.58,-0.51,-2.61,-3.36,3.33,3.47,3.22,1.75,0.73,-1.07
7526.08,-1.97,-0.77,-2.79,-3.52,3.46,3.44,3.25,1.49,0.67,-1.20
7601.34,-2.46,-1.17,-2.94,-3.64,3.58,3.41,3.27,1.12,0.65,-1.29
7677.35,-2.92,-1.59,-3.06,-3.73,3.69,3.38,3.29,0.77,0.62,-1.33
7754.13,-3.23,-1.91,-3.16,-3.78,3.76,3.36,3.30,0.53,0.60,-1.32
7831.67,-3.55,-2.29,-3.24,-3.81,3.80,3.33,3.30,0.25,0.56,-1.26
7909.98,-3.87,-2.72,-3.30,-3.82,3.81,3.31,3.31,-0.06,0.51,-1.15
7989.08,-4.03,-3.04,-3.34,-3.81,3.79,3.28,3.30,-0.24,0.45,-0.99
8068.98,-4.53,-3.75,-3.36,-3.77,3.77,3.26,3.29,-0.77,0.40,-0.78
8149.67,-4.88,-4.36,-3.37,-3.72,3.73,3.25,3.28,-1.15,0.36,-0.52
8231.16,-5.22,-5.00,-3.39,-3.67,3.69,3.23,3.26,-1.53,0.31,-0.22
8313.47,-5.49,-5.63,-3.41,-3.62,3.66,3.22,3.23,-1.84,0.25,0.14
8396.61,-5.72,-6.26,-3.45,-3.59,3.62,3.20,3.20,-2.11,0.16,0.54
8480.57,-5.86,-6.84,-3.51,-3.58,3.58,3.19,3.16,-2.28,0.07,0.98
8565.38,-5.80,-7.24,-3.57,-3.55,3.54,3.18,3.12,-2.26,-0.03,1.44
8651.03,-5.78,-7.73,-3.62,-3.52,3.50,3.17,3.07,-2.28,-0.12,1.95
8737.54,-5.90,-8.41,-3.67,-3.47,3.46,3.17,3.02,-2.44,-0.21,2.51
8824.92,-6.22,-9.31,-3.71,-3.42,3.41,3.16,2.97,-2.81,-0.30,3.09
8913.17,-6.50,-10.17,-3.75,-3.35,3.35,3.16,2.92,-3.15,-0.40,3.67
9002.30,-6.70,-10.95,-3.79,-3.29,3.28,3.15,2.86,-3.42,-0.51,4.25
9092.32,-7.53,-12.27,-3.83,-3.21,3.20,3.15,2.80,-4.33,-0.63,4.74
9183.25,-7.85,-12.97,-3.87,-3.13,3.11,3.15,2.74,-4.74,-0.75,5.12
9275.08,-7.99,-13.37,-3.89,-3.04,3.03,3.15,2.68,-4.96,-0.86,5.38
9367.83,-8.57,-14.07,-3.91,-2.95,2.96,3.15,2.62,-5.61,-0.95,5.50
9461.51,-8.58,-14.05,-3.94,-2.87,2.90,3.15,2.56,-5.69,-1.05,5.47
9556.12,-7.14,-12.43,-4.00,-2.81,2.84,3.15,2.50,-4.30,-1.16,5.29
9651.68,-6.82,-11.75,-4.09,-2.77,2.81,3.16,2.44,-4.02,-1.29,4.93
9748.20,-6.90,-11.33,-4.21,-2.77,2.78,3.16,2.38,-4.12,-1.43,4.43
9845.68,-6.89,-10.67,-4.35,-2.79,2.77,3.16,2.33,-4.12,-1.58,3.78
9944.14,-6.58,-9.61,-4.47,-2.79,2.77,3.16,2.27,-3.81,-1.70,3.03
10043.58,-5.41,-7.61,-4.59,-2.79,2.79,3.17,2.22,-2.63,-1.80,2.20
10144.02,-4.23,-5.53,-4.74,-2.80,2.80,3.17,2.18,-1.43,-1.93,1.30
10245.46,-3.32,-3.70,-4.89,-2.82,2.82,3.18,2.13,-0.50,-2.07,0.38
10347.91,-2.16,-1.64,-5.05,-2.84,2.84,3.18,2.09,0.68,-2.21,-0.52
10451.39,-1.66,-0.30,-5.21,-2.86,2.86,3.19,2.05,1.20,-2.35,-1.36
10555.91,-1.41,0.70,-5.37,-2.88,2.88,3.19,2.01,1.47,-2.49,-2.11
10661.46,-1.74,1.01,-5.53,-2.90,2.90,3.20,1.98,1.16,-2.63,-2.75
10768.08,-1.25,2.00,-5.70,-2.92,2.92,3.21,1.95,1.67,-2.78,-3.25
10875.76,-0.73,2.87,-5.86,-2.94,2.94,3.21,1.92,2.21,-2.93,-3.60
10984.52,-0.55,3.31,-6.04,-2.96,2.96,3.22,1.90,2.40,-3.08,-3.86
11094.36,-0.76,3.25,-6.21,-2.98,2.98,3.23,1.88,2.21,-3.23,-4.01
11205.31,-1.06,3.03,-6.39,-2.99,2.99,3.23,1.87,1.93,-3.39,-4.09
11317.36,-1.62,2.55,-6.56,-3.01,3.01,3.24,1.86,1.39,-3.55,-4.17
11430.53,-2.00,2.27,-6.75,-3.03,3.03,3.25,1.85,1.03,-3.71,-4.27
11544.84,-2.47,1.92,-6.93,-3.05,3.05,3.25,1.85,0.58,-3.88,-4.39
11660.29,-2.82,1.73,-7.11,-3.07,3.07,3.26,1.85,0.25,-4.05,-4.55
11776.89,-3.39,1.36,-7.30,-3.09,3.09,3.27,1.85,-0.31,-4.22,-4.75
11894.66,-4.24,0.74,-7.49,-3.10,3.10,3.28,1.87,-1.14,-4.39,-4.98
12013.60,-4.95,0.34,-7.69,-3.12,3.12,3.28,1.88,-1.83,-4.57,-5.29
12133.74,-5.58,0.10,-7.88,-3.14,3.14,3.29,1.90,-2.44,-4.74,-5.68
12255.08,-6.38,-0.22,-8.08,-3.16,3.16,3.30,1.93,-3.22,-4.92,-6.16
12377.63,-6.76,-0.07,-8.28,-3.18,3.18,3.31,1.96,-3.59,-5.11,-6.69
12501.41,-7.00,0.27,-8.49,-3.20,3.20,3.31,1.99,-3.81,-5.29,-7.27
12626.42,-7.17,0.71,-8.69,-3.21,3.21,3.32,2.04,-3.96,-5.48,-7.88
12752.68,-7.18,1.29,-8.90,-3.23,3.23,3.33,2.09,-3.95,-5.67,-8.47
12880.21,-7.15,1.88,-9.11,-3.25,3.25,3.34,2.14,-3.90,-5.86,-9.03
13009.01,-7.23,2.32,-9.32,-3.27,3.27,3.35,2.21,-3.97,-6.06,-9.55
13139.10,-7.09,2.91,-9.54,-3.28,3.28,3.35,2.28,-3.81,-6.26,-10.00
13270.49,-7.21,3.13,-9.76,-3.30,3.30,3.36,2.36,-3.91,-6.46,-10.34
13403.20,-7.27,3.28,-9.98,-3.32,3.32,3.37,2.45,-3.95,-6.66,-10.55
13537.23,-7.62,3.06,-10.20,-3.34,3.34,3.38,2.55,-4.29,-6.87,-10.68
13672.60,-7.73,2.98,-10.43,-3.35,3.35,3.39,2.66,-4.38,-7.07,-10.71
13809.33,-7.76,2.91,-10.66,-3.37,3.37,3.39,2.79,-4.39,-7.28,-10.67
13947.42,-7.67,2.88,-10.89,-3.39,3.39,3.40,2.92,-4.28,-7.50,-10.55
14086.90,-7.33,3.04,-11.12,-3.41,3.41,3.41,3.07,-3.93,-7.71,-10.37
14227.77,-7.08,3.08,-11.36,-3.42,3.42,3.42,3.24,-3.66,-7.93,-10.16
14370.04,-6.88,3.10,-11.59,-3.44,3.44,3.43,3.42,-3.44,-8.15,-9.98
14513.74,-6.85,3.01,-11.83,-3.46,3.46,3.43,3.61,-3.40,-8.38,-9.86
14658.88,-7.24,2.55,-12.08,-3.47,3.47,3.44,3.81,-3.77,-8.60,-9.79
14805.47,-7.69,2.12,-12.32,-3.49,3.49,3.45,4.02,-4.20,-8.83,-9.81
14953.52,-8.27,1.62,-12.57,-3.51,3.51,3.46,4.25,-4.76,-9.06,-9.89
15103.06,-9.08,1.00,-12.82,-3.52,3.52,3.47,4.47,-5.56,-9.30,-10.08
15254.09,-9.70,0.64,-13.07,-3.54,3.54,3.48,4.69,-6.16,-9.53,-10.34
15406.63,-10.81,-0.18,-13.33,-3.56,3.56,3.48,4.90,-7.25,-9.77,-10.63
15560.70,-12.25,-1.28,-13.59,-3.57,3.57,3.49,5.08,-8.68,-10.01,-10.97
15716.30,-13.35,-2.03,-13.85,-3.59,3.59,3.50,5.22,-9.76,-10.26,-11.32
15873.47,-15.03,-3.36,-14.11,-3.61,3.61,3.51,5.30,-11.42,-10.50,-11.67
16032.20,-17.25,-5.21,-14.38,-3.62,3.62,3.51,5.31,-13.63,-10.75,-12.04
16192.52,-18.72,-6.28,-14.64,-3.64,3.64,3.52,5.24,-15.08,-11.00,-12.44
16354.45,-18.93,-6.18,-14.91,-3.66,3.66,3.53,5.10,-15.28,-11.26,-12.75
16517.99,-18.43,-5.47,-15.19,-3.67,3.67,3.54,4.88,-14.76,-11.51,-12.96
16683.17,-17.21,-4.15,-15.46,-3.69,3.69,3.54,4.60,-13.52,-11.77,-13.06
16850.01,-16.02,-2.97,-15.74,-3.71,3.71,3.55,4.27,-12.32,-12.03,-13.05
17018.51,-16.66,-3.72,-16.02,-3.72,3.72,3.55,3.92,-12.94,-12.30,-12.94
17188.69,-18.06,-5.18,-16.30,-3.74,3.74,3.56,3.56,-14.32,-12.57,-12.88
17360.58,-18.44,-5.72,-16.59,-3.75,3.75,3.57,3.20,-14.69,-12.83,-12.72
17534.18,-19.53,-7.02,-16.88,-3.77,3.77,3.57,2.85,-15.76,-13.11,-12.51
17709.53,-20.50,-8.21,-17.17,-3.78,3.78,3.58,2.52,-16.72,-13.38,-12.29
17886.62,-20.81,-8.72,-17.46,-3.80,3.80,3.58,2.22,-17.01,-13.66,-12.09
18065.49,-21.81,-9.71,-17.75,-3.82,3.82,3.58,1.94,-18.00,-13.94,-12.10
18246.14,-23.16,-10.89,-18.05,-3.83,3.83,3.58,1.68,-19.33,-14.22,-12.27
18428.60,-24.94,-12.50,-18.35,-3.85,3.85,3.59,1.46,-21.09,-14.50,-12.44
18612.89,-26.04,-13.44,-18.65,-3.86,3.86,3.58,1.25,-22.18,-14.79,-12.60
18799.02,-25.93,-13.19,-18.96,-3.88,3.88,3.58,1.07,-22.05,-15.08,-12.74
18987.01,-24.41,-11.53,-19.27,-3.89,3.89,3.58,0.91,-20.52,-15.37,-12.88
19176.88,-22.08,-8.78,-19.58,-3.91,3.91,3.57,0.77,-18.17,-15.67,-13.30
19368.65,-18.69,-4.95,-19.89,-3.92,3.92,3.56,0.64,-14.77,-15.96,-13.74
19562.33,-16.73,-2.56,-20.20,-3.94,3.94,3.55,0.53,-12.79,-16.26,-14.17
19757.96,-18.01,-3.42,-20.52,-3.95,3.95,3.53,0.44,-14.06,-16.57,-14.59
19955.54,-19.26,-4.27,-20.84,-3.97,3.97,3.50,0.35,-15.29,-16.87,-14.99
| CSV | 3 | vinzmc/AutoEq | results/innerfidelity/innerfidelity_harman_over-ear_2018/Focal Spirit One/Focal Spirit One.csv | [
"MIT"
]
|
# loads all unit tests in the test directory
(set tests ((NSString stringWithShellCommand:"ls test/test_*.nu") lines))
(tests each:
(do (test)
(load test)))
| Nu | 3 | mattbeshara/nu | tests.nu | [
"Apache-2.0"
]
|
import random
from torch.utils.data import (
DFIterDataPipe,
IterDataPipe,
functional_datapipe,
)
from torch.utils.data.datapipes.dataframe import dataframe_wrapper as df_wrapper
@functional_datapipe('_dataframes_as_tuples')
class DataFramesAsTuplesPipe(IterDataPipe):
def __init__(self, source_datapipe):
self.source_datapipe = source_datapipe
def __iter__(self):
for df in self.source_datapipe:
for record in df.to_records(index=False):
yield record
@functional_datapipe('_dataframes_per_row', enable_df_api_tracing=True)
class PerRowDataFramesPipe(DFIterDataPipe):
def __init__(self, source_datapipe):
self.source_datapipe = source_datapipe
def __iter__(self):
for df in self.source_datapipe:
for i in range(len(df.index)):
yield df[i:i + 1]
@functional_datapipe('_dataframes_concat', enable_df_api_tracing=True)
class ConcatDataFramesPipe(DFIterDataPipe):
def __init__(self, source_datapipe, batch=3):
self.source_datapipe = source_datapipe
self.batch = batch
def __iter__(self):
buffer = []
for df in self.source_datapipe:
buffer.append(df)
if len(buffer) == self.batch:
yield df_wrapper.concat(buffer)
buffer = []
if len(buffer):
yield df_wrapper.concat(buffer)
@functional_datapipe('_dataframes_shuffle', enable_df_api_tracing=True)
class ShuffleDataFramesPipe(DFIterDataPipe):
def __init__(self, source_datapipe):
self.source_datapipe = source_datapipe
def __iter__(self):
size = None
all_buffer = []
for df in self.source_datapipe:
if size is None:
size = df_wrapper.get_len(df)
for i in range(df_wrapper.get_len(df)):
all_buffer.append(df_wrapper.get_item(df, i))
random.shuffle(all_buffer)
buffer = []
for df in all_buffer:
buffer.append(df)
if len(buffer) == size:
yield df_wrapper.concat(buffer)
buffer = []
if len(buffer):
yield df_wrapper.concat(buffer)
@functional_datapipe('_dataframes_filter', enable_df_api_tracing=True)
class FilterDataFramesPipe(DFIterDataPipe):
def __init__(self, source_datapipe, filter_fn):
self.source_datapipe = source_datapipe
self.filter_fn = filter_fn
def __iter__(self):
size = None
all_buffer = []
filter_res = []
for df in self.source_datapipe:
if size is None:
size = len(df.index)
for i in range(len(df.index)):
all_buffer.append(df[i:i + 1])
filter_res.append(self.filter_fn(df.iloc[i]))
buffer = []
for df, res in zip(all_buffer, filter_res):
if res:
buffer.append(df)
if len(buffer) == size:
yield df_wrapper.concat(buffer)
buffer = []
if len(buffer):
yield df_wrapper.concat(buffer)
@functional_datapipe('_to_dataframes_pipe', enable_df_api_tracing=True)
class ExampleAggregateAsDataFrames(DFIterDataPipe):
def __init__(self, source_datapipe, dataframe_size=10, columns=None):
self.source_datapipe = source_datapipe
self.columns = columns
self.dataframe_size = dataframe_size
def _as_list(self, item):
try:
return list(item)
except Exception: # TODO(VitalyFedyunin): Replace with better iterable exception
return [item]
def __iter__(self):
aggregate = []
for item in self.source_datapipe:
aggregate.append(self._as_list(item))
if len(aggregate) == self.dataframe_size:
yield df_wrapper.create_dataframe(aggregate, columns=self.columns)
aggregate = []
if len(aggregate) > 0:
yield df_wrapper.create_dataframe(aggregate, columns=self.columns)
| Python | 4 | sanchitintel/pytorch | torch/utils/data/datapipes/dataframe/datapipes.py | [
"Intel"
]
|
SUMMARY = "A network address manipulation library for Python."
LICENSE = "BSD-3-Clause"
LIC_FILES_CHKSUM = "file://LICENSE;md5=e6345d695ffe3776f68a56fe7962db44"
SRC_URI[md5sum] = "34cad578473b66ad77bc3b2a7613ed4a"
SRC_URI[sha256sum] = "d6cc57c7a07b1d9d2e917aa8b36ae8ce61c35ba3fcd1b83ca31c5a0ee2b5a243"
inherit pypi setuptools3
RDEPENDS:${PN} += " \
${PYTHON_PN}-pprint \
${PYTHON_PN}-xml \
"
| BitBake | 3 | shipinglinux/meta-openembedded | meta-python/recipes-devtools/python/python3-netaddr_0.8.0.bb | [
"MIT"
]
|
package com.baeldung.privateconstructors;
public class PrivateConstructorClass {
private PrivateConstructorClass() {
// in the private constructor
}
}
| Java | 3 | DBatOWL/tutorials | core-java-modules/core-java-lang-oop-constructors/src/main/java/com/baeldung/privateconstructors/PrivateConstructorClass.java | [
"MIT"
]
|
/*
*
* Copyright 2015 gRPC authors.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
*/
#ifndef GRPC_CORE_LIB_SLICE_B64_H
#define GRPC_CORE_LIB_SLICE_B64_H
#include <grpc/support/port_platform.h>
#include <grpc/slice.h>
/* Encodes data using base64. It is the caller's responsibility to free
the returned char * using gpr_free. Returns NULL on NULL input.
TODO(makdharma) : change the flags to bool from int */
char* grpc_base64_encode(const void* data, size_t data_size, int url_safe,
int multiline);
/* estimate the upper bound on size of base64 encoded data. The actual size
* is guaranteed to be less than or equal to the size returned here. */
size_t grpc_base64_estimate_encoded_size(size_t data_size, int multiline);
/* Encodes data using base64 and write it to memory pointed to by result. It is
* the caller's responsibility to allocate enough memory in |result| to fit the
* encoded data. */
void grpc_base64_encode_core(char* result, const void* vdata, size_t data_size,
int url_safe, int multiline);
/* Decodes data according to the base64 specification. Returns an empty
slice in case of failure. */
grpc_slice grpc_base64_decode(const char* b64, int url_safe);
/* Same as above except that the length is provided by the caller. */
grpc_slice grpc_base64_decode_with_len(const char* b64, size_t b64_len,
int url_safe);
#endif /* GRPC_CORE_LIB_SLICE_B64_H */
| C | 5 | mpminardi/grpc | src/core/lib/slice/b64.h | [
"Apache-2.0"
]
|
OUTPUT_FORMAT("elf32-littleunicore32", "elf32-bigunicore32", "elf32-littleunicore32")
OUTPUT_ARCH(unicore32)
ENTRY(_start)
SEARCH_DIR("/usr/unicore/gnu-toolchain-unicore/uc4-1.0-beta-hard-RHELAS5/unicore32-linux/lib");
SECTIONS
{
. = 0x00000000;
. = ALIGN(4);
.text :
{
*(.init)
*(.text)
*(.gnu.linkonce.t*)
/* section information for finsh shell */
. = ALIGN(4);
__fsymtab_start = .;
KEEP(*(FSymTab))
__fsymtab_end = .;
. = ALIGN(4);
__vsymtab_start = .;
KEEP(*(VSymTab))
__vsymtab_end = .;
. = ALIGN(4);
/* section information for modules */
. = ALIGN(4);
__rtmsymtab_start = .;
KEEP(*(RTMSymTab))
__rtmsymtab_end = .;
}
. = ALIGN(4);
.rodata : { *(.rodata) *(.rodata.*) *(.gnu.linkonce.r*) *(.eh_frame) }
. = ALIGN(4);
.ctors :
{
PROVIDE(__ctors_start__ = .);
KEEP(*(SORT(.ctors.*)))
KEEP(*(.ctors))
PROVIDE(__ctors_end__ = .);
}
.dtors :
{
PROVIDE(__dtors_start__ = .);
KEEP(*(SORT(.dtors.*)))
KEEP(*(.dtors))
PROVIDE(__dtors_end__ = .);
}
. = ALIGN(4);
.data :
{
*(.data)
*(.data.*)
*(.gnu.linkonce.d*)
}
. = ALIGN(4);
.nobss : { *(.nobss) }
/*. = 0x00300000*/
. = ALIGN(4);
__bss_start = .;
.bss : { *(.bss) }
__bss_end = .;
/* stabs debugging sections. */
.stab 0 : { *(.stab) }
.stabstr 0 : { *(.stabstr) }
.stab.excl 0 : { *(.stab.excl) }
.stab.exclstr 0 : { *(.stab.exclstr) }
.stab.index 0 : { *(.stab.index) }
.stab.indexstr 0 : { *(.stab.indexstr) }
.comment 0 : { *(.comment) }
.debug_abbrev 0 : { *(.debug_abbrev) }
.debug_info 0 : { *(.debug_info) }
.debug_line 0 : { *(.debug_line) }
.debug_pubnames 0 : { *(.debug_pubnames) }
.debug_aranges 0 : { *(.debug_aranges) }
_end = .;
}
| Linker Script | 4 | Davidfind/rt-thread | bsp/sep6200/sep6200.ld | [
"Apache-2.0"
]
|
diff --git a/node_modules/tsconfig-paths/lib/register.js b/node_modules/tsconfig-paths/lib/register.js
index 311c7fd..ac9feec 100644
--- a/node_modules/tsconfig-paths/lib/register.js
+++ b/node_modules/tsconfig-paths/lib/register.js
@@ -51,7 +51,7 @@ function register(explicitParams) {
explicitParams: explicitParams,
});
if (configLoaderResult.resultType === "failed") {
- console.warn(configLoaderResult.message + ". tsconfig-paths will be skipped");
+ // console.warn(configLoaderResult.message + ". tsconfig-paths will be skipped");
return noOp;
}
var matchPath = match_path_sync_1.createMatchPath(configLoaderResult.absoluteBaseUrl, configLoaderResult.paths, configLoaderResult.mainFields, configLoaderResult.addMatchAll);
| Diff | 2 | amoshydra/cypress | packages/server/patches/tsconfig-paths+3.10.1.patch | [
"MIT"
]
|
((fn (q) (prn (list q (list (quote quote) q)))) (quote (fn (q) (prn (list q (list (quote quote) q))))))
| Arc | 3 | MakeNowJust/quine | quine.arc | [
"Beerware"
]
|
{% include "./include_content.html" %}
| HTML | 0 | jpmallarino/django | tests/template_tests/relative_templates/dir1/dir2/inc2.html | [
"BSD-3-Clause",
"0BSD"
]
|
#![deny(clippy::needless_lifetimes)]
#![allow(dead_code)]
trait Foo {}
struct Bar {}
struct Baz<'a> {
bar: &'a Bar,
}
impl<'a> Foo for Baz<'a> {}
impl Bar {
fn baz<'a>(&'a self) -> impl Foo + 'a {
Baz { bar: self }
}
}
fn main() {}
| Rust | 5 | Eric-Arellano/rust | src/tools/clippy/tests/ui/crashes/needless_lifetimes_impl_trait.rs | [
"ECL-2.0",
"Apache-2.0",
"MIT-0",
"MIT"
]
|
{#
OPNsense® is Copyright © 2021 Frank Wall
OPNsense® is Copyright © 2014 – 2015 by Deciso B.V.
All rights reserved.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED “AS IS” AND ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY,
OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
#}
<script>
$( document ).ready(function() {
// load initial data
var data_get_map = {'frm_GeneralSettings':"/api/qemuguestagent/settings/get"};
mapDataToFormUI(data_get_map).done(function(data){
formatTokenizersUI();
$('.selectpicker').selectpicker('refresh');
});
// link apply button to API set action
$("#saveAct").click(function(){
saveFormToEndpoint(url="/api/qemuguestagent/settings/set",formid='frm_GeneralSettings',callback_ok=function(){
// reconfigure service
ajaxCall(url="/api/qemuguestagent/service/reconfigure", sendData={}, callback=function(data,status) {
});
});
});
});
</script>
<div class="alert alert-info hidden" role="alert" id="responseMsg">
</div>
<div class="col-md-12">
{{ partial("layout_partials/base_form",['fields':generalForm,'id':'frm_GeneralSettings'])}}
</div>
<div class="col-md-12">
<hr/>
<button class="btn btn-primary" id="saveAct" type="button"><b>{{ lang._('Apply') }}</b></button>
<br/>
<br/>
</div>
| Volt | 4 | marcquark/plugins | emulators/qemu-guest-agent/src/opnsense/mvc/app/views/OPNsense/QemuGuestAgent/index.volt | [
"BSD-2-Clause"
]
|
FROM openjdk:8u92-jdk-alpine
COPY /src /src/
RUN mkdir /app && ls /src && javac /src/main/java/com/baeldung/docker/heapsizing/PrintXmxXms.java -d /app
CMD java -version && java $JAVA_OPTS -cp /app com.baeldung.docker.heapsizing.PrintXmxXms | Dockerfile | 2 | DBatOWL/tutorials | docker/heap-sizing/Dockerfile | [
"MIT"
]
|
#if macro
import haxe.macro.Context;
import haxe.macro.Expr;
import haxe.macro.Type;
using haxe.macro.Tools;
#end
macro function f(a, b) {
var t = Context.typeof(a);
var ct = generateDerivedType(t);
var r = macro @:pos(b.pos) ($b : $ct);
return r;
}
#if macro
function generateDerivedType(t:Type):ComplexType {
var cl = switch (t) {
case TInst(_.get() => cl, _): cl;
case _: throw "nope";
};
var derivedName = cl.name + "__Derived";
try {
Context.getType(derivedName);
} catch (e) {
var derivedFields:Array<Field> = [for (f in cl.fields.get()) {pos: f.pos, name: f.name, kind: FVar(f.type.toComplexType()), meta: [{name: ":optional", pos: f.pos}]}];
Context.defineType({
pos: cl.pos,
pack: [],
name: derivedName,
kind: TDStructure,
fields: derivedFields,
}, cl.module);
}
return TPath({pack: [], name: derivedName});
}
#end | Haxe | 4 | Alan-love/haxe | tests/server/test/templates/issues/Issue9449/Macro.hx | [
"MIT"
]
|
#
# Issues mass recenter, gathers mass positions before and after
#
# 11/8/2006 Original J. Eakins
# 12/28/2006 Modified F. Vernon
# 08/13/2007 more modifications by J. Eakins
# 02/26/2009 modified by J.Eakins to add lead based checking
#
#
# use diagnostics ;
use strict ;
use Datascope ;
use orb ;
use archive ;
use Getopt::Std ;
our ( $opt_a, $opt_d, $opt_D, $opt_f, $opt_m, $opt_n, $opt_N, $opt_p, $opt_s, $opt_S, $opt_t, $opt_V, $opt_v, $opt_x );
our (%recenter);
{
our (%dl_mv, %sensor_mv);
our (%dl_snname);
my (@dl_mv);
my ($pfsource,$orbname,$orb,$pf,$mv,$out_of_range,$pfmass,$pfobj,$nomrcd);
my ($chan, $chansub);
our ($lead, %lead_map) ;
my ($target,$cmdorb,$statorb,$dl,$dlname,$targetsrc);
my ($when,$subject,$cmd,$prob,$n,$nmax,$complete);
my ($dltype,$delay_interval,$mrc_delay,$ref);
my ($prog_name,$mailtmp,$host,$Problems,$Success,$Neutral);
my (@dataloggers,@sources,@nomrcd,@mrcd,@xclude,@done,@prior);
my (@db,@dbcalibration,@dbj) ;
my ($nrecs,$row,$snname,$mvdlsta);
my (%nomrcd);
my ($now, $t) ;
if (! getopts('a:d:D:m:N:p:S:s:t:x:fnvV') || (@ARGV < 2 || @ARGV > 3 )) {
print STDERR "getopts or number of arguments failure.\n";
&usage;
}
$now = time();
$t = strtime($now);
# $opt_n = 1; # for testing only
%recenter = ();
@done = ();
$complete = 0 ;
$pfmass = $opt_S . ".pf" ;
$pfobj = "pfobj" ;
$cmdorb = $ARGV[0] ;
$statorb = $ARGV[1] ;
if ($#ARGV == 2) {
$target = $ARGV[2] ;
$targetsrc = $target . "/pf/st";
}
$pfsource = $targetsrc || ".*/pf/st" ;
$nmax = $opt_t || 3 ;
elog_notify("\n$0 start time");
#
# Set up mail
#
$prog_name = $0 ;
$prog_name =~ s".*/"" ;
$mailtmp = "/tmp/#$prog_name.$$";
&savemail($mailtmp) if $opt_m ;
chop ($host = `uname -n` ) ;
$Problems = 0;
$Success = 0;
$Neutral = 0;
elog_notify("\n$0 start time");
&cmdline() ;
#
# read pf file
#
$pf = $opt_p || $prog_name ;
$dltype = pfget($pf, "dltype");
$chan = pfget($pf, "chan");
$mrc_delay = pfget($pf, "mrc_delay");
$delay_interval = pfget($pf, "delay_interval");
$out_of_range = pfget($pf, "out_of_range");
$mv = $out_of_range ;
$ref = pfget($pf, "sensor_lead") ;
%lead_map = %$ref ;
$ref = pfget($pf, "sensor_mv") ;
%sensor_mv = %$ref ;
$dl = $opt_d || $dltype ; # This should default to q330
$mv = $opt_a || $mv ; # this gets modified later if opt_D
$chansub = "chan=~/$chan/";
#
#
#
if ($opt_a && $opt_D) {
print STDERR "-a value of $opt_a will be overwritten by values in database since -D was used\n\n";
}
if ($opt_x) {
@xclude = split(/,/,$opt_x);
}
if ($opt_s) {
@dataloggers = split(/,/,$opt_s);
}
#
# open database for sensor specific mv
#
if ($opt_D) {
@db = dbopen($opt_D,"r") ;
@dbcalibration = dblookup(@db,"","calibration","","");
@dbj = dbsubset(@dbcalibration, "endtime >= $now || endtime == 9999999999.99900" ) ;
print STDERR "Subsetting: $chansub&&dlsta!='-' \n" if $opt_V;
@dbj = dbsubset(@dbj, $chansub ) ;
@dbj = dbsubset(@dbj, "dlsta != '-'" ) ;
print STDERR "Subsetting based on -N: $opt_N\n" if ($opt_N && $opt_v) ;
@dbj = dbsubset(@dbj, "$opt_N" ) if $opt_N;
$nrecs = dbquery(@dbj, dbRECORD_COUNT);
foreach $row (0..$nrecs-1) {
$dbj[3] = $row ;
($mvdlsta,$snname,$lead) = dbgetv(@dbj, qw (dlsta snname lead) ) ;
$mvdlsta = &trim($mvdlsta);
$snname = &trim($snname);
$dl_snname{"$mvdlsta"} = $snname;
if (!$snname || !defined($sensor_mv{$snname} ) ) {
print STDERR "Database error! Can't get sensor/mv definition for: $mvdlsta, using default.\n";
$snname = "default"; #
$mv = $out_of_range; #
$dl_snname{"$mvdlsta"} = $snname;
print STDERR " Using default sensor type with mv value of $mv for $mvdlsta\n\n";
}
#
# How do I check to make sure there are no duplicates? For instance, if the dlsta is incorrect
# in the db (i.e. sta = F07A but dlsta = TA_F06A), two values of TA_F06A will be pushed to dl_mv
#
if (grep (/$mvdlsta/, @dl_mv) ) {
print STDERR "\nDatabase error. Duplicate dlsta value: $mvdlsta\n";
print STDERR "Skipping duplicate. Please fix database!\n";
next;
} else {
push(@dl_mv,$mvdlsta);
$dl_mv{"$mvdlsta"} = $sensor_mv{$snname};
}
}
}
#
# check inputs
#
$Problems = &check_inputs($Problems,$mv) ;
if ($Problems) {
$subject = "Problems - $prog_name $host $target $orbname ";
&sendmail($subject, $opt_m, $mailtmp) if $opt_m ;
print "\n$subject \n\n";
exit (1);
}
#
# open input orb
#
$orb = orbopen($statorb,"r");
if ( $orb < 0 ) {
$Problems++;
printf "\nProblem #$Problems\n" ;
print STDERR "Failed to open orb '$statorb' for reading\n" ;
}
orbselect( $orb, $pfsource);
($when, @sources) = orbsources ( $orb );
if ($#sources < 0 ) {
$Problems++;
printf "\nProblem #$Problems\n" ;
print STDERR "Number of sources: ". @sources . ".\n";
print STDERR "Check the target name: $target\n";
print STDERR "No data available. Exiting.\n";
}
if ($Problems) {
$subject = "Problems - $prog_name $host $target $orbname ";
&sendmail($subject, $opt_m, $mailtmp) if $opt_m ;
print "\n$subject \n\n";
exit (1);
}
#
# make $nmax attempts to mass recenter all stations
#
for ($n = 0; $n < $nmax; $n++) {
if ($n > 0) {
print STDERR "Going into sleep mode... for $delay_interval secs.\n" if $opt_v;
sleep $delay_interval unless $opt_n;
}
#
# Make list of stations to recenter
#
%recenter = ();
&get_masspos($mv,$orb,@sources);
@mrcd = sort keys %recenter;
if ($#mrcd == -1 ) {
print STDERR "\n\nNo more mass recenters needed\n";
$Neutral++ if ($n == 0);
$complete = 1 ;
last;
}
printf STDERR "\n %d Stations with mass positions greater than maximum mass voltage:\n", $#mrcd+1 if ($opt_v) ;
print STDERR "\n@mrcd \n\n" if $opt_v;
if ($opt_x) {
@mrcd = remain(\@mrcd,\@xclude);
print STDERR "Stations after -opt_x $opt_x :\n" if $opt_v;
print STDERR "@mrcd \n\n" if $opt_v;
}
if ($opt_s) {
@mrcd = &intersect(\@mrcd,\@dataloggers);
print STDERR "Stations after -opt_s $opt_s :\n" if $opt_v;
print STDERR "@mrcd \n\n" if $opt_v;
}
push (@done,@mrcd);
print STDERR "\nSending mass recenters to @mrcd\n";
print STDERR "\n\*\*\* TEST MODE ONLY \*\*\*\n" if $opt_n;
foreach $dlname (@mrcd) {
$target = $recenter{$dlname};
printf STDERR "\n%s massrecenter at %s \n", $dlname, strtime(now()) ;
$cmd = "dlcmd $cmdorb $target $dl $dlname massrecenter -duration 8 > /dev/null " ;
$cmd = "dlcmd $cmdorb $target $dl $dlname massrecenter -duration 8 " if $opt_v;
print STDERR "$cmd\n";
$prob = $Problems;
$Problems = &run($cmd,$Problems) unless $opt_n;
print STDERR "\*\*\* No MRC done - TEST MODE \*\*\*\n\n" if $opt_n;
print STDERR "Sleep $mrc_delay between massrecenters...\n" if $opt_v;
sleep $mrc_delay unless $opt_n;
$Success++ if ($prob == $Problems);
}
last if ($opt_f);
}
@done = get_unique( @done ) ;
if (! $complete) {
&get_masspos($mv,$orb,@sources);
@nomrcd = sort keys %recenter;
@done = remain(\@done,\@nomrcd) ;
printf STDERR "\nMass recenters successful at: @done \n", $#done ;
if ($#nomrcd > -1) {
$Problems++;
printf STDERR "\nProblem #$Problems\n" ;
print STDERR "Mass recenters unsuccessful at: @nomrcd \n" ;
# if ($opt_S) {
# $ref = pfget($opt_S, "");
# prettyprint (\$ref);
# exit;
# @prior = sort keys %{$ref};
# print STDERR "prior @prior \n";
# pfnew($pfobj);
# foreach $nomrcd (@nomrcd) {
# $nomrcd{$nomrcd} = now();
# }
# pfput("nomrcd",\%nomrcd,$pfobj);
# pfwrite($pfmass,$pfobj);
# }
}
} elsif ($done[0] =~ /\w/ ) {
printf STDERR "\nMass recenters successful at: @done \n", $#done ;
}
if ($Problems) {
$subject = "Problems - $prog_name $host $orbname @nomrcd";
elog_notify ("$subject");
&sendmail($subject, $opt_m, $mailtmp) if $opt_m ;
elog_notify ("$subject");
exit (1);
} elsif ($Success) {
$subject = "Success - $prog_name $host $orbname @done";
elog_notify ("$subject");
&sendmail($subject, $opt_m, $mailtmp) if $opt_m ;
elog_notify ("$subject");
exit (0);
} elsif ($Neutral) {
$subject = "No MRCs - $prog_name $host $orbname ";
elog_notify ("$subject");
&sendmail($subject, $opt_m, $mailtmp) if $opt_m ;
elog_notify ("$subject");
exit (0);
}
exit;
}
# start subs here
sub usage {
print STDERR <<END;
\nUSAGE: $0 [-m "mail_list"] [-n] [-f] [-p pf] [-d dltype] [-v] [-a voltage_trigger | -D database] [-N subset] [-t max_retries] [-s "dl_sta1,dl_sta2,..."] [-x exclude] cmd_orb status_orb [target]
END
exit(1);
}
sub cmdline { # &cmdline() ;
printf STDERR "\ncommand line: $0 " ;
printf STDERR " -m $opt_m" if $opt_m;
printf STDERR " -v" if $opt_v;
printf STDERR " -V" if $opt_V;
printf STDERR " -n" if $opt_n;
printf STDERR " -f" if $opt_f;
printf STDERR " -d $opt_d" if $opt_d;
printf STDERR " -p $opt_p" if $opt_p;
printf STDERR " -x $opt_x" if $opt_x;
printf STDERR " -s $opt_s" if $opt_s;
printf STDERR " -a $opt_a" if $opt_a;
printf STDERR " -D $opt_D" if $opt_D;
printf STDERR " -N $opt_N" if $opt_N;
printf STDERR " @ARGV\n\n" ;
return ;
}
sub check_inputs { # $Problems = &check_inputs($Problems,$mv) ;
my ($Problems,$mv) ;
if ($mv < 0) {
$Problems++;
printf "\nProblem #$Problems\n" ;
print STDERR "Voltage trigger value must be positive.\n";
}
if ($opt_f && !$opt_s) { # Don't allow recenter for all stations if they aren't out of range
$Problems++;
printf "\nProblem #$Problems\n" ;
print STDERR "Cannot force mass recenter for all stations\n";
}
if ($opt_s =~/\|/) {
$Problems++;
printf "\nProblem #$Problems\n" ;
print STDERR "Use commas to separate dl_sta values.\n";
}
return ($Problems);
}
sub check_masspos {# &check_masspos($pf,$mv,$srcname);
#
# get mass positions for one source name
#
my ($pf,$mv,$srcname) = @_ ;
my ($ref,$dlsta,$sta) ;
my ($m0,$m1,$m2,$m3,$m4,$m5,$mc,$con,$masspo);
my (@dlsta,@mc,@recenter,@xclude,@dataloggers) ;
our ($lead, %lead_map) ;
our ($snname) ;
our (%dl_mv,%sensor_mv);
our (%dl_snname);
@recenter = ();
$srcname =~ s/\/.*// ;
if ($opt_x) {
@xclude = split(/,/,$opt_x);
}
if ($opt_s) {
@dataloggers = split(/,/,$opt_s);
}
$ref = pfget($pf, "");
@dlsta = sort keys %{$ref->{dls}};
if ($opt_x) {
@dlsta = remain(\@dlsta,\@xclude);
print STDERR "Stations after -opt_x $opt_x :\n" if $opt_v;
print STDERR "@dlsta \n\n" if $opt_v;
}
if ($opt_s) {
@dlsta = &intersect(\@dlsta,\@dataloggers);
print STDERR "Stations after -opt_s $opt_s :\n" if $opt_v;
print STDERR "@dlsta \n\n" if $opt_v;
}
foreach $dlsta (@dlsta) {
$sta = $dlsta ;
$sta =~ s/.*_// ;
$m0 = $ref->{dls}{$dlsta}{"m0"};
$m1 = $ref->{dls}{$dlsta}{"m1"};
$m2 = $ref->{dls}{$dlsta}{"m2"};
$m3 = $ref->{dls}{$dlsta}{"m3"};
$m4 = $ref->{dls}{$dlsta}{"m4"};
$m5 = $ref->{dls}{$dlsta}{"m5"};
$con = $ref->{dls}{$dlsta}{"con"};
# printf "%s %s %s %s %s %s %s \n", $dlsta, $m0, $m1, $m2, $m3, $m4, $m5 if $opt_v;
# printf "\n%s %s %s %s %s %s %s %s \n", $dlsta, $m0, $m1, $m2, $m3, $m4, $m5, $con ;
next if ($con !~ /yes/);
# assume that calibration.lead is filled in and then only check the appropriate masspos
# This works around 6ch Q330 case where default/dead sensor reports mass of 2.0 V
# and the 2.0V reading triggers recenter of all Trillium sensors
#
if ( $lead = $lead_map{A} ) {
@mc = qw(m0 m1 m2);
} elsif ( $lead = $lead_map{B} ) {
@mc = qw(m3 m4 m5);
} else { # lead is undefined or NULL
@mc = qw(m0 m1 m2 m3 m4 m5) ; # check all mass pos.
}
foreach $mc (@mc) {
$masspo = $ref->{dls}{$dlsta}{$mc};
next unless ($masspo =~ /\d/);
if (!$dl_mv{$dlsta}) { # this allows mrc to happen for dlsta's which are incorrect
$dl_mv{$dlsta} = $mv ;
}
if ( ($opt_D && ( abs($masspo) >= $dl_mv{$dlsta}) ) || $opt_f ) {
printf STDERR "%7s %10s %-20s %4s %4s %4s %4s %4s %4s \n", $dlsta, $srcname, $dl_snname{"$dlsta"}, $m0, $m1, $m2, $m3, $m4, $m5 ;
push(@recenter,$dlsta);
$recenter{$dlsta} = $srcname;
last;
} elsif ( (!$opt_D && abs($masspo) >= $mv) || $opt_f ) {
printf STDERR "%7s %10s %4s %4s %4s %4s %4s %4s \n", $dlsta, $srcname, $m0, $m1, $m2, $m3, $m4, $m5 ;
push(@recenter,$dlsta);
$recenter{$dlsta} = $srcname;
last;
}
}
}
return;
}
sub get_masspos {# &get_masspos($mv,$orb,@sources);
#
# get mass positions for all source names
#
my ($mv,$orb,@sources) = @_ ;
my ($pktid,$srcname,$pkttime,$pkt,$nbytes,$result,$src);
my ($net,$sta,$chan,$loc,$suffix,$subcode,$type,$desc,$pf);
my ($value, $stype) ;
#my (%sensor_mv);
our (%sensor_mv);
if ($opt_D) {
print STDERR "\n\n";
printf STDERR "Checking for mass positions greater than or equal to:\n";
foreach $stype (sort keys %sensor_mv) {
$mv = $sensor_mv{$stype} ;
printf STDERR "$mv for $stype\n";
}
} else {
printf STDERR "\n\nMass positions greater than or equal to $mv\n";
}
if ($opt_D) {
printf STDERR "\ndl_sta sourcename sensor m0 m1 m2 m3 m4 m5\n";
printf STDERR "======= ========== ====== ==== ==== ==== ==== ==== ====\n";
} else {
printf STDERR "\ndl_sta sourcename m0 m1 m2 m3 m4 m5\n";
printf STDERR "======= ========== ==== ==== ==== ==== ==== ====\n";
}
foreach $src (@sources) {
$srcname = $src->srcname() ;
orbselect ( $orb, $srcname ) ;
($pktid, $srcname, $pkttime, $pkt, $nbytes) = orbget ( $orb, "ORBNEWEST" ) ;
if (!defined $pktid) {
next ;
}
if ( $nbytes == 0 ) {
next ;
}
($result, $pkt) = unstuffPkt ( $srcname, $pkttime, $pkt, $nbytes ) ;
if ( $result ne "Pkt_pf" ) {
if( $opt_V ) {
print "Received a $result, skipping\n" ;
}
next;
}
($net, $sta, $chan, $loc, $suffix, $subcode) = $pkt->parts() ;
($type, $desc) = $pkt->PacketType() ;
$pf = $pkt->pf ;
if ( defined $pf ) {
&check_masspos($pf,$mv,$srcname);
}
}
return;
}
sub prettyprint {
my $val = shift;
my $prefix = "";
if (@_) { $prefix = shift ; }
if (ref($val) eq "HASH") {
my @keys = sort ( keys %$val );
my %hash = %$val;
foreach my $key (@keys) {
my $newprefix = $prefix . "{". $key . "}" ;
prettyprint ($hash{$key}, $newprefix) ;
}
} elsif (ref($val) eq "ARRAY") {
my $i = 0;
my @arr = @$val;
foreach my $entry ( @$val ) {
my $newprefix = $prefix . "[". $i . "]" ;
prettyprint ($arr[$i], $newprefix) ;
$i++;
}
} else {
print $prefix, " = ", $val, "\n";
}
}
sub trim {
my @out = @_;
for (@out) {
s/^\s+//;
s/\s+$//;
}
return wantarray ? @out : $out[0];
}
| XProc | 5 | jreyes1108/antelope_contrib | bin/rt/dlautomrc/dlautomrc.xpl | [
"BSD-2-Clause",
"MIT"
]
|
module parent {
private module child {
proc secretFunction(a: int) {
return a*3;
}
}
module sibling {
use parent only child; // use instead of import, and utilize only list
proc main() {
writeln(child.secretFunction(11));
}
}
}
| Chapel | 3 | jhh67/chapel | test/visibility/private/moduleSymbols/accessPrivateSiblingSubmodule2.chpl | [
"ECL-2.0",
"Apache-2.0"
]
|
// Checks that unions use type based qualification. Regression test for issue #90268.
#![feature(untagged_unions)]
use std::cell::Cell;
union U { i: u32, c: Cell<u32> }
const C1: Cell<u32> = {
unsafe { U { c: Cell::new(0) }.c }
};
const C2: Cell<u32> = {
unsafe { U { i : 0 }.c }
};
const C3: Cell<u32> = {
let mut u = U { i: 0 };
u.i = 1;
unsafe { u.c }
};
const C4: U = U { i: 0 };
const C5: [U; 1] = [U {i : 0}; 1];
fn main() {
// Interior mutability should prevent promotion.
let _: &'static _ = &C1; //~ ERROR temporary value dropped while borrowed
let _: &'static _ = &C2; //~ ERROR temporary value dropped while borrowed
let _: &'static _ = &C3; //~ ERROR temporary value dropped while borrowed
let _: &'static _ = &C4; //~ ERROR temporary value dropped while borrowed
let _: &'static _ = &C5; //~ ERROR temporary value dropped while borrowed
}
| Rust | 4 | ohno418/rust | src/test/ui/consts/qualif-union.rs | [
"ECL-2.0",
"Apache-2.0",
"MIT-0",
"MIT"
]
|
; Compile me with Jasmin 2.1+ (https://sourceforge.net/projects/jasmin/)
.class public synthetic ExampleSynthetic
.super java/lang/Object
| Jasmin | 1 | mauguignard/cbmc | jbmc/regression/jbmc/class-literals/ExampleSynthetic.j | [
"BSD-4-Clause"
]
|
ObjectWrapper = Lazy({}).__proto__
ArrayWrapper = Lazy([]).__proto__
ConcatenatedSequence = Lazy().concat().__proto__
largs = (args)-> Lazy.generate ((i)->args[i]), args.length
prepVarargs = (args)->
Lazy()
.concat(largs args)
.toArray()
addExtras = (proto)->
ext =
__proto__: proto
pick: -> wrap (if arguments.length > 0
proto.pick.call @, prepVarargs arguments
else proto.pick.call @, arguments[0])
keys: -> wrap proto.keys.call @
values: -> wrap proto.values.call @
concat: -> wrap proto.concat.apply @, arguments
take: -> wrap proto.take.apply @, arguments
drop: -> wrap proto.drop.apply @, arguments
L_ArrayWrapper = addExtras ArrayWrapper
L_ObjectWrapper = addExtras ObjectWrapper
L_ConcatenatedSequence = addExtras ConcatenatedSequence
patch = (obj, proto)->
obj.__proto__ = proto
obj
wrap = (result)->
patch result, switch result.__proto__
when ArrayWrapper then L_ArrayWrapper
when ObjectWrapper then L_ObjectWrapper
when ConcatenatedSequence then L_ConcatenatedSequence
else result.__proto__
L = ->
wrap (if arguments.length > 1
Lazy().concat largs arguments
else Lazy(arguments[0]))
patch L, L_ArrayWrapper
(window ? global).L = L
| Literate CoffeeScript | 3 | zot/Leisure | METEOR-OLD/private/build/l.litcoffee | [
"Zlib"
]
|
trait TestList
"""
Source of unit tests for a PonyTest object.
See package doc string for further information and example use.
"""
fun tag tests(test: PonyTest)
"""
Add all the tests in this suite to the given test object.
Typically the implementation of this function will be of the form:
```pony
fun tests(test: PonyTest) =>
test(_TestClass1)
test(_TestClass2)
test(_TestClass3)
```
"""
| Pony | 4 | presidentbeef/ponyc | packages/ponytest/test_list.pony | [
"BSD-2-Clause"
]
|
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
*/
/* PySpark style CSS overwrite */
/* Lato font (from its parent) does not make any difference when it's bold. Defaults to 'Source Sans Pro' */
body {
font-family:"Source Sans Pro",sans-serif!important;
}
h1,h2 {
color:#1B5162!important;
}
h3 {
color: #555555
}
/* Top menu */
#navbar-main {
background: #1B5162!important;
box-shadow: 0px 2px 4px rgba(0, 0, 0, 0.11);
}
#navbar-main-elements li.nav-item a {
color: rgba(255, 255, 255, 0.8);
}
#navbar-main-elements li.active a {
font-weight: 600;
color: #FFFFFF!important;
}
.col-9 {
flex: 0 0 80%;
max-width: 80%;
}
/* Left panel size */
@media (min-width: 768px) {
.col-md-3 {
flex: 0 0 20%;
max-width: 20%;
}
}
/* Top menu right button */
.navbar-toggler {
color:rgba(255,255,255,.5)!important;
border-color:rgba(255,255,255,.5)!important;
}
.navbar-toggler-icon {
background-image:url("data:image/svg+xml;charset=utf-8,%3Csvg xmlns='http://www.w3.org/2000/svg' width='30' height='30'%3E%3Cpath stroke='rgba(255,255,255,.5)' stroke-linecap='round' stroke-miterlimit='10' stroke-width='2' d='M4 7h22M4 15h22M4 23h22'/%3E%3C/svg%3E")!important;
}
/* Left bar list colors */
.bd-sidebar .nav>.active>a {
color: #1B5162!important;
}
.bd-sidebar .nav>li>a:hover {
color: #1B5162!important;
}
.bd-sidebar .nav>.active:hover>a,.bd-sidebar .nav>.active>a {
color: #1B5162!important;
}
u.bd-sidebar .nav>li>ul>.active:hover>a,.bd-sidebar .nav>li>ul>.active>a {
color: #1B5162!important;
}
/* Right bar list colors */
.toc-entry>.nav-link.active {
color: #1B5162!important;
border-left: 2px solid #1B5162!important;
}
| CSS | 3 | kesavanvt/spark | python/docs/source/_static/css/pyspark.css | [
"BSD-2-Clause",
"Apache-2.0",
"CC0-1.0",
"MIT",
"MIT-0",
"ECL-2.0",
"BSD-3-Clause-No-Nuclear-License-2014",
"BSD-3-Clause"
]
|
if ! (( $+commands[fig] )); then
return
fi
# If the completion file doesn't exist yet, we need to autoload it and
# bind it to `fig`. Otherwise, compinit will have already done that
if [[ ! -f "$ZSH_CACHE_DIR/completions/_fig" ]]; then
autoload -Uz _fig
typeset -g -A _comps
_comps[fig]=_fig
fi
fig completion zsh >| "$ZSH_CACHE_DIR/completions/_fig" &|
| Shell | 4 | residwi/ohmyzsh | plugins/fig/fig.plugin.zsh | [
"MIT"
]
|
.class public LHelloWorld;
.super Ljava/lang/Object;
.method public static main([Ljava/lang/String;)V
.registers 2
sget-object v0, Ljava/lang/System;->out:Ljava/io/PrintStream;
const/high16 v1, 0x7f020000
invoke-virtual {v0, v1}, Ljava/io/PrintStream;->println(Ljava/lang/String;)V
return-void
.end method | Smali | 3 | CrazyLeoJay/Apktool | brut.apktool/apktool-lib/src/test/resources/aapt2/testapp/smali/HelloWorld.smali | [
"Apache-2.0"
]
|
/**
Copyright 2015 Acacia Team
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/
package org.acacia.partitioner.hadoop;
import x10.io.FileWriter;
import x10.io.File;
import x10.io.FileReader;
import x10.io.IOException;
import x10.util.StringBuilder;
import x10.regionarray.Array;
import org.acacia.util.Utils;
public class HadoopOrchestrator {
/**
* Default constructor
*/
public def this() {
}
/**
* This method automatically setsup the hadoop cluster on the intendend machines using the configuration information
* provided by the user
*
* Note on 9th Feb 2013 : COnstruction of this method was postponed for getig ready for IEEE Cloud 2013.
*
*/
public static def setupHadoop(){
val hadoopHome:String = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.home");
//First we setup the masters
var hadoopHosts:String = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.masters");
var hadoopHostsArray:Rail[String] = hadoopHosts.split(",");
var outFile:File = new File(hadoopHome + "/conf/masters");
var writer:FileWriter = new FileWriter(outFile);
for (item in hadoopHostsArray.range()){
writer.write((hadoopHostsArray(item) + "\n").bytes());
}
writer.close();
//Second we setup the slaves
hadoopHosts = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.slaves");
hadoopHostsArray = hadoopHosts.split(",");
outFile = new File(hadoopHome + "/conf/slaves");
writer = new FileWriter(outFile);
for (item in hadoopHostsArray.range()){
writer.write((hadoopHostsArray(item) + "\n").bytes());
}
writer.close();
//Next we setup the hadoop-env.conf
//conf/hadoop-env.sh
val data:String = readAlltext(hadoopHome + "/conf/hadoop-env.sh");
val dataArr:Rail[String] = data.split("\n");
val ind:Int = data.indexOf("JAVA_HOME");
//outFile = new File(hadoopHome + "conf/hadoop-env.sh");
//writer = new FileWriter(outFile);
Console.OUT.println("ind : " + ind);
}
private static def readAlltext(val fileName:String):String{
var builder:StringBuilder = new StringBuilder();
var reader:FileReader = new FileReader(new File(fileName));
var line:String = null;
while (true){
try{
line = reader.readLine();
builder.add(line);
}catch(IOException){
//We assume that at this point we completed reading all the file contents.
break;
}
}
return builder.toString();
}
public static def isHadoopRunning():Boolean{
var result:Boolean = false;
return result;
}
public static def isJobDone():Boolean{
val fileTmp:File = new File("/tmp/jobdone");
return fileTmp.exists();
}
public static def startHadoop():void{
Console.OUT.println("Starting Hadoop");
val hadoopLoc:String = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.home");
val namenode:String = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.namenode");
var flag:Boolean = false;
for(line in x10.xrx.Runtime.execForRead("ssh " + namenode + " " + hadoopLoc + "/bin/start-all.sh").lines()){
Console.OUT.println(line);
if(line.indexOf("Stop it first.") != -1n){
flag = true;
}
}
if(flag){
Console.OUT.println("Hadoop is already running.");
}else{
Console.OUT.println("Done starting Hadoop");
}
}
public static def submitJob(val args:String):void{
val hadoopLoc:String = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.home");
Console.OUT.println("Submitting the job : " + args);
for(line in x10.xrx.Runtime.execForRead(hadoopLoc + "/bin/hadoop " + args).lines()){
Console.OUT.println(line);
}
Console.OUT.println("Submitted job to Hadoop");
//
// while(!HadoopOrchestrator.isJobDone()){
// System.sleep(5000);
// }
//
// Console.OUT.println("Done executing job on Hadoop");
}
public static def stopHadoop():void{
Console.OUT.println("Stopping Hadoop");
val hadoopLoc:String = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.home");
val namenode:String = Utils.call_getAcaciaProperty("org.acacia.partitioner.hadoop.namenode");
for(line in x10.xrx.Runtime.execForRead("ssh " + namenode + " " + hadoopLoc + "/bin/stop-all.sh").lines()){
Console.OUT.println(line);
}
Console.OUT.println("Done stopping Hadoop");
}
public static def reStartHadoop():void{
stopHadoop();
startHadoop();
}
} | X10 | 4 | mdherath/Acacia | src/org/acacia/partitioner/hadoop/HadoopOrchestrator.x10 | [
"Apache-2.0"
]
|
(module
(memory 1 2)
;; import a "yield" function that receives the current value,
;; then pauses execution until it is resumed later.
(import "env" "yield" (func $yield (param i32)))
(export "memory" (memory 0))
;; simple linear progression in a loop
(func "linear" (result i32)
(local $x i32)
(loop $l
(call $yield (local.get $x))
(local.set $x
(i32.add (local.get $x) (i32.const 10))
)
(br $l)
)
)
;; exponential in a loop
(func "exponential" (result i32)
(local $x i32)
(local.set $x
(i32.const 1)
)
(loop $l
(call $yield (local.get $x))
(local.set $x
(i32.mul (local.get $x) (i32.const 2))
)
(br $l)
)
)
;; just some weird numbers, no loop
(func "weird" (result i32)
(call $yield (i32.const 42))
(call $yield (i32.const 1337))
(call $yield (i32.const 0))
(call $yield (i32.const -1000))
(call $yield (i32.const 42))
(call $yield (i32.const 314159))
(call $yield (i32.const 21828))
(unreachable)
)
)
| WebAssembly | 4 | phated/binaryen | test/unit/input/asyncify-coroutine.wat | [
"Apache-2.0"
]
|
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Copyright (C) 2016 - 2019 Pinard Liu([email protected])\n",
"\n",
"https://www.cnblogs.com/pinard\n",
"\n",
"Permission given to modify the code as long as you keep this declaration at the top\n",
"\n",
"用scikit-learn进行LDA降维 https://www.cnblogs.com/pinard/p/6249328.html"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<mpl_toolkits.mplot3d.art3d.Path3DCollection at 0x9e17f60>"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import numpy as np\n",
"import matplotlib.pyplot as plt\n",
"from mpl_toolkits.mplot3d import Axes3D\n",
"%matplotlib inline\n",
"from sklearn.datasets.samples_generator import make_classification\n",
"X, y = make_classification(n_samples=1000, n_features=3, n_redundant=0, n_classes=3, n_informative=2,\n",
" n_clusters_per_class=1,class_sep =0.5, random_state =10)\n",
"fig = plt.figure()\n",
"ax = Axes3D(fig, rect=[0, 0, 1, 1], elev=30, azim=20)\n",
"ax.scatter(X[:, 0], X[:, 1], X[:, 2],marker='o',c=y)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[0.43377069 0.3716351 ]\n",
"[1.21083449 1.0373882 ]\n"
]
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAD8CAYAAABjAo9vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzsnXd4VEX3xz9zt++mJ/TemxSlqyiC2AvYfyp2sffX3vVVsffXrig2xA4iIF16kd6RUEJJb9v33ju/P25IstkNBA2Ecj/P4yN7y9zZze6ZmTPnfI+QUmJiYmJicuSg1HUHTExMTExqF9Owm5iYmBxhmIbdxMTE5AjDNOwmJiYmRximYTcxMTE5wjANu4mJickRhmnYTUxMTI4wTMNuYmJicoRhGnYTExOTIwxrXTw0IyNDtmzZsi4ebWJiYnLYsmTJkjwpZb19XVcnhr1ly5YsXry4Lh5tYmJictgihNhak+tMV4yJiYnJEYZp2E1MTEyOMEzDbmJiYnKEYRp2ExMTkyMM07CbmJiYHGHUSVSMiYlJ3SKlRAZ+AP/noJeAYxAi4TaEJaOuu2ZSC5iG3cTkKESW/hcC34MMGAcCY5ChyZAxAaEk123nTP41pivGxOQoQ2o54B9TYdQBUEEvQfq/qbN+mdQepmE3MTnaUNeAsMc5EYLw/IPeHZPaxzTsJiZHG0oDkFqcExawNDvo3TGpfUzDbmJylCFsncDamtgtNhvCc3VddMmkljENu4nJUYhI+wTsfQE74ASlHiL1bYS1bV13zaQWMKNiTEwqIaWE0B9I/w+AjnCdD86zEOLImgMJJQ2R9hlSLwDdB5YmR9x7PJoxDbuJSSVkycMQnAjSb7yOLDRep7yNEKKOe1f7CCUNlLS67oZJLWMO0SYmZcjIGghMKDfqxsEAhP+EyJK665iJyX5iGnYTkz2E5wFxokVkEBmafdC7Y2LyT/nXhl0I4RRCLBRCLBdCrBZCPF0bHTMxOeiIJOJ7J21mNqbJYUVt+NhDwCAppVcIYQNmCyF+l1KamQ6HOZkrtzL6mbFsXJJJ0w6NuPLxi+lyfIe67taBw3k6lD4HsuoJBZzn1EWP9ovMokI+WLyQVTnZdMyox009+9AuPb2uu2VSB/xrwy6llIC37KWt7L+Yn4bJ4cX6xX9z38AnCQfCSCnZvSWHlbPW8vh399L37J513b0DglCSIPVDZOFtVHbJiJQ3EJZ9lpmsU1blZHPZD2MIqSqalKzPz+P3TRv5YtiF9GzUpK67Z3KQqRUfuxDCIoRYBuQAf0gpF9RGuyZ1xwf3fU7IHzLC/8oIBcK8c8enUceONIS9D6L+XETq+4iUdxH15yMcJ9V1t/bJMzOn449E0Mr+NpqUBNQIT06fWsc9M6kLasWwSyk1KWUPoCnQRwhxTNVrhBAjhBCLhRCLc3Nza+OxJgeQDUs2xz2esz2PUCB8kHtzcBHCZhh4R39EXE2VQ49l2bviHl+bl4t+BA/EJvGp1agYKWURMAM4I865D6WUvaSUverVO7SXtSaQXC8p7nG704bNcfSkPwS8AX54Yzz3DXySZy95lRWz1tR1l+KSaHfEPe6y2Tjyou9N9kVtRMXUE0KklP3bBZwKrPu37ZrULZc+cD4Od7SxcLjtnHfr6Vgsljrq1cEl4AtyW5+H+eyxb1gxaw1//jCfR856nh/f+q2uuxbDNd2PxWmNHnCdVitXdO1+RCZWmeyd2pixNwKmCyFWAIswfOzja6Fdk1ogHIowadR0Hjv3BV68+m1Wz11fo/vOvfk0Lv7PuTjcDlyJTuxOG0OGn8x1z11+gHt86PD7x1PJ2ZpLyG+4nqSEkD/EJw9/ja/Ev4+7Dy639O7Lee074rBYSLTbcVgsnN6mHff1P7Guu2ZSB4i62Ajr1auXXLx48UF/7tFGJBzh3pOfZMuqbQR9IYQQ2F12rv3vZVx4d83C9wK+IDnb8shonIon2XOAe3xocd8pT7JiZqzrxZPs5omx93Hcqd3qoFd7J9/vZ0txIc2TU6jnPrr+XkcDQoglUspe+7rOzDw9gpn+zZxyow6GwFXIH+KTR76mtNBLKBBi/aJN7N6SU20bLo+TFp2aHnVGHSC1fjLxvBiappOYlnDwO1QD0t1uejZqYhr1o5yjZxfsKGT2jwvKjXplbHYrnz72DVNGz0JRBGpYpV3P1jz14/2k1Ks+w1JTNSZ8PJWJn05DSsnp1wzkrBtPxWa3Hci38Y+RMgShOUAI7P0RSsp+3X/+7Wcy/7e/CPkrPkOhCNIbpdL22Fa13FsTk9rDNOxHMAlpHoQQMXHnakRj0qfTiYQi5cfWL9zEk0Nf4s05z8VtS0rJk8NeYtn01eWGbtvaHcz5aSEv/vHEIbdBJ8MLkYU373kFUkUmPoriuazGbXQd0IkbRl7Bxw9+icVmQdd00hun8cLERw+592tiUhnTsB/BnHPTacwaO698828PUteJhNSoY2pEY9OyLezanE2j1g1i2lozbwPLZ6yOmr2G/CHWLtzEsumrOHZQ1wPzJv4BUvcjC28C6Ys+Ufo80tFrv4pJDL39TE67eiDrFm4iKS2BNj1amkbd5JDH9LEfwXTu154bXrgCu9OGO8mFK9FJeuNUGrWKNdwAFqtCYXZR3HMrZ62JmuHvIegNsnLW2lrt978mPLOaExFk4Kf9bs6d6OK4wV1pe2wr06ibHBaYM/YjnKF3nMWpw09m5Z9r8SS76XJCBz5+8Cu2rdsRc23AG6R5p/i6IikNUrA5bWjeaJ+9w20nteH++a4POHoA0OOc0EAvPdi9MdkLMrIGWfoyRFaAkgGemxGuoeYA+i8xZ+xHECX5pSyfuZpdm7OjjiekeOh/bi+6ndQZi8VCeuPUuPdbbVaWz4ifWXnSRf3iJiYpFoWBlx7/7ztfmziOBxnHsAs3wjnk4PfHJC4ysgFZ8H8QngOyFLRMKH0K6fuorrt22GMa9iMAKSUfPjCay5rexJNDX+KGrvdy/+Cn8RX74l5fsDu+uwVgd2b80Ed3oouXpjxB/eYZOD0OnAlOMpqkMXLS4ySmHlqhf8LSEBJuBZywJ6FeuMF+gvGfySGB9L4NMljlYAB8/zMimkz+MaYr5ghg8uczGPfeJCKhSLkffPXcdbx87f946sf7Y65v37M1rgQnAW/0j8pmt9KmR8tqn9O+Zxu+zPwfW1ZvBylpeUzzOl0ySxmB8EKQXrD3Nup3lqEk3IK090cGfgAZQDjPAsdAs2DzoURkBdUqfGs7wWqGlP5TTMO+H2iaxuo56/GXBOhyQodDZqY69pVfY+LVIyGVhb//ha/YF5NcdMKwPox6YgzZW3NRw0Z0jM1ho3nnpnQf2GWvzxJC0OqY5rX7Bv4BMrIGWXAtEMEIZ4wgE+9F8VxXfo2w90DYe9RZH032gbUFhOOoUkoVFFMo8N9gGvYasmX1dh46/Vn8pQGEMJJ6rh95BRfceXad9mvRxKVsWxu7EQqgKAq+kkCMYbfZbbw17zk+f2IMM76bi8WicOrwkxn+5MWHxaaVlCqy4DqQhdEnSt9A2o4zjflhgvDcigwvAyqvHJ3gOg+hHBqTpsMVUyumBmiaxhUtbiF/Z7QhcbgdvPTH43TuXzfl4nKz8rm2411RseWVSWuUwjfbP0BRjiz3gwzNRRbdFhunjgDXBSjJL9RJv0z2Hz0wEUr/C3ohYAH3JYjEBzGqbJpUpaZaMeaMvQasmbuBQGkw5ng4EGbc+5NrzbD7in2MfXUcs76fj8vj4LzbzmDIVSdXa5j/GD0TXdPinrPardz9/k1HnFEH4hj08hMQ+Ak9shqReB/CcfJB7ZbJ/qO4zkA6TwdZbEQtHSaFTQ51TMNeA3zFfuJVK5BSUlrgjT3xDwgFQtze92Fyt+fS77RchlxciBDT+O2tAZxz1ytxN/2Kc0tiMkjB0DO54O6z6X/uPgf2wxN7b8MPGxcd1HXIwjsg5XWEc3Dcq2RoBtL3GegF4BiE8FyDUOKHgZocWIQQIA6xXIjDHNOw14BjTuxYvslYGafHwYAL+9XKM6Z+NZu8Hfnc/+Ymep1SistjxGEHfBPw7wjjafpOzD3HndqNCR9PJVglusXusHHaVUfubFUoKcjEB6D0ZSBM/GSkILJ0ZFzDrns/BO+7QMA4oGYaGakZ4xBK9SJoJiaHC0fgOr32SUjxcMPIK3G47eWbi06Pgxadm3LK/9VOIYO/pqygTZeCKKMO4PLo2MUMZGR1zD29z+hBpz5tcXoqKh05PQ5OuewEWnRuViv9OlRRPMMR6V+B66LqL9K2xxySeil436bcqAMQBr0Q6f+y1vtpYlIXmDP2GjLszrPo0LsN496fTGmBlwEX9uOU/zsRu6N2NnnqN8+geXMfDlfs7FMIDUJzwRYdiqgoCs///ih/fDGTP0bPxOawcdYNp3LSRbWzijjUEbauiOSu6KHpoMcpkK5kxB6LrAFhh5gEmBCEZkLCbQekryYmBxPTsO8Hnft3OGARMOfcNIRfX/2cSEjgcEVHKimKHapxEVhtVs68fjBnXh/fl3xU4LkdvCONrMVyXMbxqljSQcaKmYEApeGB6qGJyUHFdMUcIjRu05B+Fz6BjLdLKwQ4zzj4nTpMEO7LIOFeEMmA1fh/4n0I96Wx11rbgrUNEKcgt7Ah9ZID3l8TkwONGcd+iKEHZqEX3mVMIBUAKyL1XYS9T1137ZBHSh2kvyxsrvo5i9RykUW3QmQl0RuvNrA0RmSMQwjnAe+vicn+YsaxH6YorpMQzgUQWQYIsPUwkzVqiBAKiH1nLApLPUgeicw7HyOqZg8R0HIgMB7ce9mUNTE5xPnXrhghRDMhxHQhxFohxGohxF210bGjGSHsCHsfhL33ATHqUkr+Xr6FVXPWEQ6G933Dv32eugkZGI8ML48p03ew2FFawsRNG1mevdvoQ2QFiHjzmgAyPO+g9Elq+ejez9BLX0GGZhsrjprcJ4PopS+jZ/dDz+6JXvQAUouzeXyAkVoeetF96Nk90LOPQy9+3Ig6MqlzamPGrgL3SSn/EkIkAkuEEH9IKeMLe5tUi5QahOeBtg2sncpm6/uv3bJm3no+f3IMW9dk0aJzU6566lK6HG9s+mZt2MmjZ79Awe5CFIuClJJ7P7yZgZfWvpytlGFk0R0QmgfCYmikW1tD2qi48eJSSub8vJCJn05DjWgMGX4yAy87Pq4OfE3RpeSRqZP5Zf1abBYLmi5pkZLC52e1ID3efgY2sBx4kTMZWoAsGlGmGx8yQi1t3SH146jBfGdpCe8uWsC8rG009CRyU89eDEh8smxFVxbZExxvDEYZkxCK+4D3HYxC4TL/ItBzMEwAEPgJGVkO6T+bKpp1zL827FLKXcCusn+XCiHWAk0A07DvB1LLNYoO6PkgNWPD1NoF0j7dL3/vX1NX8sT5I8vrnObvLGT1nPU888uDdD+lC/ef+jT5OwqRUpJWP8I5V+fhlndRuPkcUlrcbGiZ19Z78r5nGHWCFeqs6gZk8WOI1Ldjrn99xAdM/3Z2uVLl6jnrmP7tbJ799aF/LE72zaoVjNuwjpCmESqTX9hUkM/d05yMHpAKWpVqS8KKcF9S/XvS/UjvWxD4GVDBebohX1BJMnhfSKkhi+6MjuKRfogsQ/p/QJQV3N5ZWsLZX4/GFw6hSsmWoiLU8DL6nbwMm1I5XFMFvRQZGIfwxG4YHxCCE0EWUW7UAQgbk5LwPHCYuvd1Sa0Oq0KIlsCxwII450YIIRYLIRbn5h78ZeOhjix+GLQdZTooQeNHH1lpFCPYD967Z1RM8epQIMx794xi5ay1+IoDSClp1jbIRzPXcfEtufQaWEKC41tk7uno4fW196YCY4hW7gOIQGgaUkb3MXPlVqZ9/WeU/HDQF2L5jNUsm77qH3fh82V/EVCrFO7WdZbs2kmR+xOwdQXsgBOURojUjxGWRnHbklIiC68B/5cgC0CWGLPU/Iv2rzCEuoZo3/6eBwQgWFGT9X+LFuKLhFErua/aJOYQ0ePpA/lBXV7zPuwHUveiez9Fzx+OXnQvMrwUGVlrDEYxF6ugbjgg/TCpObW2eSqESAB+AO6WUsbEjEkpPwQ+BCMqpraeeyQgZRDCc4GqP9gQBH6ExNhiGdWxdU1stqVxPIui3JJy58Ntz+3AnaizRyPMYtGBABRciMz4BWFtU6WPkomfTmPMS79QlFtM5/4duHHkFbTspCADY0HbjXCcCM4zKoScqlbHKUc3YskrCT4tnbYKXY/9WgR9IRZPWsaxg7rW+DOojC8SL2YdFCEIaOmkpY9FajlGXy3N9r4yiCwuM1qVjbJq6M0EJ4Lr/Br2ykK1BSYqhWHOy9qGqkf73beWJiFlvD46wdImzvFoFu7I4oXZM1mfn0cDTwJ39unHsE7Va/BL3YvMvwC03UAQIgIZnFIWfusGqhh3YQNLy332w+TAUiszdmE4BX8AvpJS/lgbbR5VSI1qf+jVil3FJzkjMe7xpIxEQ/MmYrTXtb+X+MKPYWTBNTEbeaOfGcv/7vqMHRt34Svys3jiUu48/mG2LxgKvo8h+DOy+Alj9qqX/dgdJxPvKxYMtzRK1VUiMTUBqy3Wl25z2EhKd6P7RqPnDUPPuwDd95VRPWkfaLpOsqNCbsEidFLtARShk+p00jjR+KyEpT7CWoNqUJG18f8e0o+M7MeqwtqxmugdF8J9cfmrBgmx1yzIbczuQAIyak4mQNgQ7gv2+tglu3ZwzS8/sDx7N0FVZWtxEY9Nn8KoZX9Ve4/0fwXaLipWXtL4d3AiKE6i/75WUNLBcdJe+2Fy4KmNqBgBfAKslVK+9u+7dPQhFA9YOxMrIWmF/Sy+fMn95+NwO6KOOdwOLrn/PDIap3HB3Wfj9DgIBfbyp5deY3ZaRsAX5LuXfiFYSfddSsPF89XrKVT4Wf2GoFaZ5opIfACUVIzaoxAOCvxehYcudnF9l3vI2Z5X3t4Jw/rEVdBULIJBZ/8Mpa+AuhrUVeB9CVl48z4jbP63aAFbiosAyS2d/mLJ0FHMPvcrFp3/OV+c7t9/v72lmTEjrYpwgaXmZdyEUBCp74FILBvgbIATHKeAs2LWf3PPPris0Ytqm8XKu5tuQzhOwlhwW8B6DCLtm32qU74ydzbBKm6pgKryxoK5MSuDckJTKN+kjXoTVkh8HOz9MMyI1Sg9mDYGIf75ZrdJ7VAbrpgTgOHASiHEsrJjj0gpJ9RC23WCEZ2y0NCIth2HsNSv9Wcs+WM53786jvydhfQ6owcX3f04KZYbQIaBoPGDFymIhPv2q92L7j0Xb5GP718dh5SGhO+Fd5/NxfedB8B1z11O574pLJu9nn6nFRM/4ESUFT4w2PV3NhabJVo3C5C6YO7EZKTcToWNDEHwN0gYYfiqMybx9/xX2LV+EplrbPz2ZQYF2RYUyy6eHPoS7y15CTCKZb/w+2M8MfTFKCXNRz4/nfT0l4l6uAwYA09ksSHhWwld18s16D9ZuoSgqnJDh+Xc0mkpbqvRrsOikcwodH8zlP2JV3ecZAxUWpAKt5kA7AjXeTVvB0PnhnqzDMOpF4C9D6KKFtBJLVryyIknM3LOnwCousbxzZrz7OCzURxXGfsUUq1xJMz6vLy4x0OqRmEgQD2PJ/akqGZTWGoIa3tE2iikVAFhGvRDiNqIiplN3LnW4YlUNyMLrjZmrQiQYaTnBpTEu2vtGT+/8zsfP/RVeeWj7Ru2M2X0BN5f8Dwp6btA3YywdQfXWTWKiJHh5cYsWc/DFzqR5dP/BiGwWAWaqrN9/Q40VcNqsyKEoO+gTeDbi468DIOtIrkto0kakWB810c4pLBsdgLHDqhoL2+3k59e/ILVc9YR9IfYsnI7UkaHEOqazvZ1O9ixaRdN2hqblV2O78B3Oz9i7fwNaKpOp/7tsUU+QHqrjCh7+hheAvbeSCn57uVf+O7lXyjJ99KkXSNuevVqSsIhQHJTx2XlRr2CAHjf2a9EJCEsyNSvoOhOY+UAYOuKSB75j0q5CcWzT7/8Fd16cFHnY8gsLCDduo4MRy4omUBHYy9jPwpTNEtOpigndt/DogiSnfG/Z8JzFTI8n+hRXQFLU7C2Na6Jmw9gUpeYwaaVMKIebjRic6WvzLiHwf8ZMjSzVp4R9IeijDqAGpaUFob5/qUXQduFkvw0wn1BjYy67v8eWTAcguMgPIe3bvuODYs3EA6ECXpDRIIRFk5YyjcjK6ItjGiG6pJhbJBwM8KSXn4kKT2RNj1axn++Bgv+SCp/vSMziREDLPz89gTWzt9I5opt1bpMLDYL3iLDH+8vDbB+0SYKs4s45sROdB/YBbvDhhTpxNV1wVau3vjF098x+pnvKck3BpcdG3fx3KWv0Wa3xCp0ku3VRKzo0TNYqeUh/V8jfZ8j1W0xl8vwQsgfBtomEA5Q0hBJjyGsreO3X0vYRSHtuZL08F3IkmeQ+ZegF94YE1m0L+7pdwLOKq4dl9XKdT16Yq8mV0A4ToCEOwBH2b6AGywtEWkfHhb1cY9WDivDLtUsZHAqMrLxwDxAXWvEkVfdyJQBpO+rWnnEllXbsFhjP3Y1rLB4hgv8XyMj65jz80Ju7f0gFze8gcfPf5HMlVtj7pEyAKXPYmxs6UTCgtm/JRAJR//gQoEw49+bXP5aOE9lj987GgWSn0dJiFVFPOfm0wx3TBWsNvAkK2W+YgefjOyKv1RFDccv2VeV1t2a89VzP3BJwxt4YMgzXNXuDh4953n8pWUzROEmOla6/F0hHacTDkX4/rXxMXVfQ4EwzabmYLU42OmvZjZdySDrgfHI3FOQJSORpS8j885G9/6v/LzUC5CFI8rCHH3G4KjnGhvNenWl+moHWXy/oS0vfRgz5yCEFiB9n+xXOwNbtuLFU0+ngScBixAk2Ozc1LM3d/c7fq/3KQk3IOrPQaS8jUj/CpHxO8LS5J+/IZMDzmGxhpJSNb7cwSllWtoRpK0rIvWD2q1mLn1UO9bJ2kmVTqmfHLcaE0B6wwgQYdy7X/DhY9vKjdWC8YtZNm0Vb817jlbHVHJpRFZReTarqQKpx59FVY4Px9bLCFcLTSxLklEAOyTchFKNa2DAhf14965PCUSiDbZidTDk+nsQSRGw9WTZn08g44QtVsXhsnPZw8N47JwXWDZ9Fbomy1f7S6eu5JXr/scTY++D4C8A7Pa7mbyjFaquMLjJVlokhBH6NkryGiK1+KuPki35fH3BPfy6oYjr2ozBaan8uTsRiQ8ChtGm+GFiNgm97yMdAxG2zoZ+jIwzWEkdQpPBNWyf7/mfIHUvhBcRGwobBP8YSLhlv9o7t31HzmnXgaCq4rBaUWo46xZKkpl0dBhxeBh230cQnAqEKgokRJYjS55ApNRiII6tK/FdFE5wnlUrj2jYsj7te7Vh3YL1qJEKA+hwaVx0Uy5qROHTpzMJ+SuMkJQQ8ocY9fi3PP3TAxWNiaQoY+N06zTvECRzjSvqmUIR9Dy9e8VrISB5JITPQwYnYGz+DUXYu1Md7kQX/x33ME+c/2K5a0VTNe77+FYad6z4wSekePAVxUlcqURimodLHhjKl8+MjUqmSq0X4bxr8+jSx8eOzVvxFZyJQ/fzc2Y7Hl9yEkJIdCl4dWUfbu60nDsHBEmul4RiiT8Yt+jclB4NG9Gj4ZPI0EBk6Rtlcg1tEAn3IRx9jQuD08skD6q2EEYGf0PYOiP1XOJGhxAxNj8PGHtzt+xHUlQlhBC4bKaw3JHMYWHY8X9FbAZjGIKTkDJca5XNhXAiE5+Bkscor6VZFspWOb743/LUj/fz1AUvsGHxBqxWia7DDY/t5NgBXrKzPKiR2FmUlJJ1C6q4oKztwdIYtEz2DEj3vLKdBy9uSyRiRQ3r2J02nB4nI14aXuW9CnCcYPhQK7ErM5tvXviJlbPW0LBVfS57cBjdBxrRGt1O6szY7I9ZPmMNalil+8DOuBKiB5Fhd57FZ499G+Ma2YPDZefRb+/h2YtfizLqDZqFeGfiRpwuHbtT0qW3DyV0GSu9Q3h8SX1CevRX9YN13TmmlYfBrW1c9tBQvn7+p6hnOlx2rn32sor36zgZ4aiuDqxO/DwCWT5wCns/pH90nGxLK9j7VtPuv0coaUhLC8OvX/W5jtMO2HNNDm8OD8MeL3UZiJfB+G9R3OchbR2Q/m9Bz0U4BoHrnFobPACSM5J4fdYL7Fr/LUWZr9Kyg4rDLUE6SG7xEFL/Je599ZtHl3oTQkDqh8jC68pKwyl06BHhw4UnMW5UOltXb6fz8e05e8QQkjOS4rZZmZ1/7+bWXg8S9AXRVJ2sDbtY+ec67v5gBKdeYSSd2Ow2ep3WndVz13P/4Gf4e1kmSemJXPyf87jg7rMZdudZZK3fyaRRM7A7bYQCYVwJTqw2K626Nefqpy/FYrXEuGuuf3QXnkQNS9k30moDCFLPMgcYGtPXoGbhm1VrGdy6I//38AUkpHj4+vkfKcoppnnnptz86jUcc2Kn8usD3gA2hw2rLc5X3jEQSp6J84k4EM4zjX/ajwdbDwgvpSJCxAWOkxG2Y/b52f4bRMpLxga5VDFm6S5QUhGJppCqSXwOi0IbeuEdEPqDGDeJtR1Kxm+127mDjNQLITQDkEaCh5LGGzd/wJTRswgFKma0Dredx769l37n9IxtY48MrSwCW3eEkvKP+vLClW8y49s5Man9SekJfLf743KVxU3LMrn7xMeiZtyKVaFNtxbc/s4NdO7XnsLsIrauyaJBi3o0at0gqr2sDTu5+bj7o+7/fs0qElMq3Eo5WTZ0XZDeRKXvr1dREolOugI4u1173j7z3L2+p1Vz1vHGTR+QtWEnisXCwEuP5453ro9Zaei+b6H0OQxftg7YwX05StJD5ddIGUH6f4Tgj4DFWMU5zzvg8dsyvBTpfbOsXmsiuC5EeK7ZZ/z6zC2ZfLBkEbt9pfRv2pzbevelcWLsAC+lzsIdO/hk6RJ2e0sZ2LIV1/boSarLFadVk7rkiCq0IRIfMGJpZRBjxmIzUqiTnqvrrv1rhJIas/F221vXIYRg8uczEYqRVj/ipeFxjTqUzdz34h+vKctnrImr1xIKRMjdnk/Dlkai1pfPjCUciI5r11UrZuLsAAAgAElEQVSdjX9l8sCpT3PerWcw4qXhpDaIP8A0bd+YJm0bsWXVtvLn+UsVElM0tm5w8N8RLdm9zY7A2FDWLg5Ck1jDfuNxRnKS1AuMYt/CBY4TEcK4NmvjLh4+47/lG8eaqjNjzFzydxXy4qTHo9pSPJchHccbew4yjHCeamyaVkIIm6GeeJAUFHUp+XPTL0zfOJ4Ue5ALWqk09ewA/4fgPBWU9tXeO3r5UkbOmVUugJZVXMyEjRv47fLhNHIHkd7XIDAZCCKlSkfdTidXV6ZtOY4NBfmMXbOK3y6/ijTXwZEBNqldDosZOxhFCaT/a4gsBWtbhHs4wtrsAPWwbti1OZvMldto3LYhLbs0I+gPUVrgJa1hChZr7c4KpQwbIXRKPSPiAbil1wNs+isz5lqbw8bY7I/xJBk/8v9rdj15O6qvDepwWXhr/ou07toi5lzO9jxGP/UdCyb8RWmhFyEEVruVs67I4tLbd3HDgE6UFlmihK4sHjubHu2G7jbmIQK4pGMXui7xowS/ZNj1axCKvdzNIlI/RNh78dZtHzHhoyloavRKz+4UvD/3GJoecznC2pbCQICX5v7JxE0bsSiCYR07c0+/E3DX4QajpuvcMO4nFu3YhF+1YlNULAJe6TONM5ptAftJKGkfxb03pKr0+uh/MQJoViEY3rUdj3YeWRbWG/25+FUr32d24JmlJ2K3WLj+2J7cf/yAA/QOTf4JR9SMHUBY0hGJd9R1Nw4IakRl5PC3mPfrYqx2K5qq0b5nG/47/mHqNU3fdwNxkFKr1kWg+0aB902MzUGV1StOYcI37VAsCla7NSoc0+60cfzQPniS3GiaxivXvU6z1jvI25FAdQnHalhl3tjnaN31w6jjhdlF3NLzAXxFvnJja3faOHZQV/6cYMFfKoiERYx6oVXCY44OZPUwinMMbduRN894mQWuv3lp7Fpsdqis+S4LroMGi9iyenuMUQew2VR2rRtHk4bfE/Q8ybAfC9lZWlqulzJ6xTKW7NzBD5dcXp6EIyNrDZ1xkQzO0xBKtNja2rxcVmTvpkliEsc3a16jMMIpmzfx+vy5ZJUU0z49g/uPH0CfJk0BGL9xPQt3ZBFQjZ9oRLcSAR5YeAoDG32OU1Qv3JVZVEi8v40qJalMAL2UeNFfbqvKJa3X8drKPnhVO9MzN9eZYQ+pKluKi6jndpurhn/AYWPYj2S+GfkT88ctIRyMEC5L3V+3cCNv3vIRD3955361pft/Ae+roO9GKunguQPh/r8KAxWcCKWvs2cD8McPMxj14k7Cwd1IaWSDCkXg9DhRwyp9zjqOez+6GYAJH01FhKdw2Z3ZrFnsIRQQCCFjDLFikdgta9GDs1CcFUp/P775G4GSQJSxDQcjzPt1MVLq/P5VOvGiU0L+MHq+n8dOMjZRJ38+g92ZObz+y+ayTdaqBJHFT9K5/zGsW7CRSCg6byAcFrRo7wciTFrzBXn+QVEiWGFNY0NBPgt3ZNGnSVNkycMQmABo6LoNxfIspH6EsPcmomncMuFX5m7fhsCQA053u/n2wktpmBBfaRPg53VreGTaH+WiXEt27eSaX35g1PkX0qdJU35ZvzZGRx5AETqL8xpyYpPqf7rpLnc1mu3QK6NMfrcaVF2hkdvLxpI0MuJpxxwEPl26hNfmz0EgiOgag1q25pXTzqzTFdThxmGVeXqkMv69yVEbpQCRkMqs7+eVy+zWBD0wAUoeB3132YF8KH0R6f+m/BrpfZ89Rr2kwMKnLzQiFFDY45HTIhp2p43LHx7GV1vf48nv/4PLY2Spjn9/Mun1/XTt6+OpzzJp3DIYV/pXKHDSuQVQdDtS21l+fPmMNUTiJGcZ7sA9g0PsTNOV4KBL/w7lrxdPWobL7aVFhxDVToyD4xh6+yAcLgdCqbjI4dIYeF4R9RobA+jKwnr443zGekmI2bNX4ssdjwxM5Lcv3FzWvR1nNm3H8N4tmPH5/Uip8vFfi5m7fRtBVSWgqvgiEbKKC7nz16cp3X0lMhJbuERKyQuzZ8UoLQZVlZGzDemK6lL8QWBTbOC5sZrzUM/j4YRmzWPacFmtNEzpgVFYJD5WobPTn4DLauWGY/e54q91Jv+9kVfnzcYfieCLhAlrGtO3bObBKZMOel8OZ0zDfggQ8MWP+dY1Pa4hrBbv68TOxgLge7tCr0XPKT+zcoEHmy3+DHndok2k1o+uSxoKhNm00kUoqHDcSV4+m7ue+97cht2p4/JouDwadqfOXS9tJ71hhNWLLEz/4nkCPqNPjds0iDKyNcFqkzTv1IxeZ/QoP1avWTptuoQJB/fWlo30+kW8s/AF+p/bC1eCIL1hmCvuzuae1yqKkbROLMFVWeJB06n39d80fnwR00Z8zaUtv+ChSxrwwdONKcw1Zow5WXZevTudOT98x7erV8YYaB2F5QX1CYeWoeVfglS3RJ33RSIUBuMImwHr8/MBuLRL1xjJXgCbotOzxVkI197zKt44/WxObNYCh8WCx2bHY7PzyICBtGh0UzVFvA0f+5jMbui4ubf/iZzUouVen3EgeG/xwpiVSkjT+GPzJkpC1a80TKIxXTGHAMed2pV5vyyKiUhp0aVZ+Wy5RlSaHUehFwARwA6248pCRyXuhPip+EIRJKYmGBoowd+R2laE7RgGXtqHsa/ksiNzNy3aBbE7JYMvKKL3wFIWTktCSug7uISkNI1AUOHpR9sSytzJ87ddwwmvX8hld53F7J8WxJTui0Vis0uSMzTOvqEzFz/4ZFRB69OuOYWnf/6+mkIh5R8GKBk0aVufp396wCgeXTiCqtrD57TYzqurTiKohpBA+q/bSFiaj1AlYa8x4C6bnUjVlUQooPDp4zMI/+e4+J8hEk1akDKI9H2ASH6h/JzbZsNpteINx34ODcuKawxs0YpLu3Tjm1UrEAIswujBh+dehj1535WSEh0OPj5vGLl+H/l+P61SUnHsGShSPy8rxViR2IZIJeS4gq5thrKwbwM89trL2wCQkZVI/w8gfQjnGeA4JW7B6xxffN0dq1AoDARJcuzH7+EoxjTs/xCpF0J4GShJYDv2X1Vlv+nlq1gxcw0hf4hwMILVZsXqsHLvhzftX0OW5qD9HXtcqVeeYCUS70aG54AM0K2/F5tDB2/0kt3utHHW9V2QuYPKJBz8SOHmpCGNmDW2KY9e0YmLbt7O4AsLEcDqRW5OOLMYi00SxkppWOGuOYNYelVzEpbkUm9MJrPv/YG/P9N47Nt7eX3E+/iK/YRDkbi6MjaHZMzyTXgy+kPizSi2ilBHKSUf/ucLdm9zs2aJh2P6eMs2Tw10Hca8U59lc9LpeMIUrnz8QsMd4+iL9FyH9H2EFjGus1itJNZ7l+8vacldv49nTU4OSXNzUCJVB7z4K4PsLfmc3iqNr9cEYgpVNPWUkuE0BpHi0kW4PBGcZRsCihDceFwv3q8yO3VZrdxTJsglhOCJk09hePcezN2+jSSHg0EtW+OW89ALXwbdD86zEe5he02eq+f2UM8d7SsX9u6IehOQejFgK4+HTy/7r7bRfZ9C6RvsyeaWwT/A0Q9S/hfzu+nTpCnj1q9Dr7LXYrUo5RWvTPbNYRPueCihez8E79tl1XR0oyBG2mcIa82r6FSlMKeYce9NYs28DbQ8phlDbz+zPG68psjgNGTR3US7Y5yQ9CSK+8KK69TNSO+7EP6Lv9c04pHLbIQDEoQR0XLji1dy3uWfGUlPSPJ2WXl8eGt2ZDqwWK2EQwqtuzVn45LNSAkpGRE+mLaO5zcPIMfvYWFeI8JlEgD1lVIuL1lAzxa51O8KTeqnYnedSV7BUHZsKuXJoS9FJ2K5LAy+MI+7XsrG2EgVkHAnSoLhU167YCMPnPo0QV8Id4LGXS9v58SzirFYIRgQPH9TSxZPT0QvE0NLa5TKp2vfwJPkZvmM1bxx41N07VdMSYFk5cL6tD22I//54W7O+OYLSkp8tHhoEaI6ReMqtOgQ4MXJQYZOuZSdpcVIBA5FxarojB44nm5puWg6TNvVhqeXD+WnS64oL3enS8nbC+bx8dLFRHQdt83Gf/qfyOVdq89H0EtegcDoMuE2ABfYOiDSvkLEq+x0CCC1PGTuKcTo2gg3Ivk1hHNQ1OGtRUWc9+1o/JEIWpltclmtPD1wMBd1PrAZvocDNQ13NA37fiJD85FFN1X6cQEIsDRBZEytc41qGZyKLH3ZELuyNALP3QjXORBZbqhBWpuCfUBUKKSmaiyfMQ9//nS69vORWK9HWYq9MZu85dT2bFnvQNcqZlcOt4PUBknszszlukd30u1SlUtmDiOgVRiYfvV28OGAiVjQcVgrW0sbWOoj0sexePJG3rnjE3Zn5mB32Tj36p2cPTyHhGS9UiaqE5HxI8Lalh/f+I2PH/oyau/B4dJwJ+oU51nQ9diV0xWPX8TwJy7i0kY3UpwXrdLpcDto+9KZTJJ5hDSVZv9dij0vhG4TBDqmoNsUXBtKsPnDUcqZDpfOI+9vod9pGqGkr7l6/AoUdTH96mVxedu15bN1VRc8uHAg47d34LQ27Xj3rOhMWVXXKQ2FSHI4sOzFtyS13cjcU4kVBXMjkp9DuM6u9t5/y4b8PLYVF9EhvR7Nko19l3V5uWwpKqJDRgatUqovyScD45DFjxNT9BrANQwl+cWYw9uLi3ln0XwW7siiSWIiN/fqy4nNY3MijkaOuDj22kRG1iO9bxgzUktjRMJtCMfAmt3r/7KKUQeQhh9bXVWmEFl3COdghHNw+WspQ8iCq0BdYUjMCisoKZD2DcLSEABFX0mPrreXCV6FoOR39hj1rRsc7Mi0Rxl1MNQm6zXNoDC7mO7He9EsbhRRMUlQhM6b/afEqVwEEAEtH+kfQ+8zrufzje8QDoZZOukd3rjVxy+fpiF16H6Cl3tf28bmNQ62bP6IbZtbsG31dlCq+rsthALVJ3BN//pP+p/Tk3Aoti+RUJANuevR0xMAC3kXtSJl6k52X1eW1Vnm4B7i30rpR9vJ22WjccsQNzy2i35DSgE3TqvClxdczPOz3KTbPsFjDaPqgqBmIcGm8tRxs5m6syXTMmPdZFZFqVnqfnihsUKMKa7hR4am75dhl1I1Sh8qyXt145SGQtw47idW5GRjUxTCmsYpLVuT6/exJjcHi6IQ0XRObtGSt848J34kj3AaYVIx80cFiC+53Sw5mRdPPb3G78cklqPOsMvIOmTBZWXGWRrFEgrvQiY9EeWuqBa9uJoTSlnix372R8s1BovIsrKM2qsR1ub7vrGm7XvfM9resxSWIdACyOIHEGlfGFWjiu+rIrQWYI9fuaTAWi7MVRU1ovH+Xy+zc8PVdEvYhk2pmJV3Ss7HYdlbsY0ghP8Ermf++CW8d89n7Pw7G6Ows8HSPxMY3qszFhuEg5uBzdW2JhRRrQ58KBA2ooKqrE7PuzaXq+7Pxu5ZjYrCZxu68ga9CLRPjhk8ZqW14puJa+iallXlwQlgbY9N28oTxy0j5N1CacRGmiNIgs0YSBQhObPZZsZvr9mgL0MzkL7RxnfNOQThvgKU5GqutoBSc8+47vsCvG8Z3wOhIN3XIBLuirtH9MT0KSzL3k1Y08qde5M3b0JAuZsEYNa2LbyzcD739o+j1+6oLsHJjqjJ761q/6VkY0E+DouFlntZKRztHH2GvfS1CqNeTsCI93YN3begk/M0w61RNaxQamDbP70WqW5D5l9QpoEThvAiZOAHSB2FsPfY5/01IvADsbrdOoQXs2PjJqZ+NYtgnk7/09107uWvFBcuAYW23QRaHBlhMDTam7ZvTJNWr6LmXcGrfadx+9whqLogIhVEXCncSijpzBu3mOf+7/W4kTKaahgbrZrxQctwEjg2g2TdwklNWjB99Ky46rvFeSX88s5E3EkuAl7j7zboggKue3QXLrdxgwON6zqsYENxKhOy2sa0YZMBfAF9j9cfY8bpgOSXkYU3lBXDkDgsYRxVvkJOi0p9V4Cz28XXdsnz+/lu9Uo2FeTTI20DQxuNJsFaNknwrkcGfoS074znUTVqxFZjSWnd/zOUvkp5ZJAEfKOQwoaoUjUrrGlM2LSBSJVNYT2O6zaoqnyzanlcwy6Es0yBdATlD5URSPxPjBbPvpizfSv3TJqAPxJBl5JmScm8f875e3UFHa3UShy7EOJTIUSOEGJVbbR3QCnbEIxBBmNqYMZDuC8Gawtgz/JZwdigfMQoTrwfyNKRFXVVAVBB+pElj+1XO3snfhy8rkvuPflRvn5hMmPfS+fhy1rz+n+aRk9qlRa4G7/AgGFN47axYtZqCnOKEbZuWFJe5uTGXn4/42dGdFxB17QwQknZu2mXfj568MsahD/GRxSFyB/UkI1D6vNZex+yXnyXhhrWmDl2Hne9dyOeJDeuBCdX3pdTbtT34LaqDGi0narfD4Hkq8G/cVyDXZXiYyQoHgj8DOH5GINn/PcR1K2sK27BTb36xJxbm5vDoC8+4e2F8/h5/VpeXBjktAnnkxPYk0YfAm03BMch0j4HpQEIj7FSEG5Ifh5hjR2I9rCpLIPWFw6D7x2qhnsaeQ6fIGW0AY9oWlwjXh2BvSTSCXsvRP15iOSXIeF+SHoKYW2PjFeRqhp2lpYwYtzP5Pn9+CMRgqrKpoJ8Lvt+DJHqRv6jmNqasY8C3gG+qKX2DhyWBqDGq3gj97LcrUAIJ6SPRfp/MeLBLekI9xUIW7f970t4LnErNql/I3X/PmVZa4TjDAh8hxHHDvlBJ19u6sKcrY35ux8k/5mNLV8jFLAw85cUBl1QSPfjfaycn0pW1vG0PK4FEbUFEBsjb7OFWT7pdU6+8mGEvScSJ009RdzTdVGZ8fGATKy+rGB4Hrv+7ljpQOUM1H0jNEnC7N0UDWmKpitkDW9Ds1fjzy3sThsWq5Vvst5n9o8LadDs1rjXDWq0DYeiRRX26Ft/J80TSrBbKxs6CbqvrHRf9WE0ftXK4txG/LkrnfnffsWXF1xM9wYNy88/MGVSVDx7QLMR0RVeXtmHl/vM2HMUQtMRniuh3kyIrDRWnfYe1RY83+0t5fpff2JLUSFWRSGi6ywbthtrvI9XBo3/RMX3zWO30yY1jQ0FRsJUfaePa9qvoGdGNn+XpPDphu5sKjFmyooQ+9zcFMKBHlkBvk9A2JFIQ4I4bVSNioF/t3pVTEipBPyRCH9u28qgVge2oPjhRq0YdinlLCFEy9po60AjEm5FFj1A9MzFCa4Lqv2RxLQhnLUj3yo81RQRsZSFUu4/Ui81Kv0Ep4CSBq6LIPwneiSHnQErQ/+4kIBqNQzXiTol/RvQ+L21OLd4CQUUpo7N4P0nmrJ7mwtdbkAoz+Fw2VEUEUfSV+K2TUQWbDSeJXMpXyFIn+HHtQ+G8GTiVyiCes0y2LU5m7QGYdSwgq9UKXfB7MvQCwme5YUUDWkKiiDS0I2aZMNaEom5Vg2rNGpdH1eCiyFXnYye37nMpRZNujPIA93n89zC/ugWBRRBO3c+1piaoxA7+630yUhYU5TOt5s7MXZzR1QpCeth/jP5d/4Yfi0A3nCY9fmxq0RVWpi6o2WlIwpYjNBXIZS4Es1SSuZnZTJ902SSxDp+3ZJBZqkLrdLHvqYglW7p2bGdVTIMyeMydP8Y8L7Jc8cqXD3zXBq6fIwd/CNOi4rDotM9LYdzmv/NiNlnsDS/OS6bjUcGVFedqqx/oRngHwWEKzaApd9w0WT8sc9osl3e0hi3EIAudXL9B7aY+OHIUedjF87TkYl54H0Nw/Whg2soIumRg98Z9xXgfZ9of70dnGchhM1YqkaWgO4Fe69yed3qkLoXmX8+aJXqc4YXsXTBCUwfY2F2945oK0pI3ObD0cCJt1c9dLeVnEtb0/zFFSAUVv/VlOxtAdSwxh7XghqKGNEhVYxzOCTo3KuwrKh2mNiCyyqEp4N9AIRnEz2zNUq7XfPsWbx2w7s8+0UmSakq37zdgAV/JBH0KwT9Cpq69x+87qnyFY5jIGwOK91Obk3TlluQ4TywdTM0/guuJ54g1tXtVpM5VmP8rnZodiuFi33onai8rwtAMGDB4dSq1asJa5Yyo17hdN9cWMC0zM0MatUa617kFRxRhbftCPeV1V6rS8ntE35l1tb1BFSBRTRGlbFe1ueX9+WLk3/Hbqk88Dkh8aFyw6r7x0DJ80CAY9Pht9O+ozRiJ8EawaIYf3+rIrEqKq/1m8vYnIu4/JgepLlcLNu9i2mZm3HbrJzTrg2NrZMg8JNRT1b3VxNNlgfqWtiHv/2EZs35beN6/FWkiCXQq1FjpJT8tG4Nny77i5JQkMGt2nBb735kuI9OZchai2Mvm7GPl1LGzSIQQowARgA0b96859atW2vluf8UKSOgZYOSVjsuj3/UBxVZ/AAEJ4NwGJtK9h6IlPdA226UvJNlESoyAokPonj28gP3fmREPFTZLA0FBZd160Kx1YXFr6KEdXSbgrQKdtzVhUg9J60eWYJTUdAiOpoaOzuNF3VitesMvT6PGx/PQdclG5Y7CHgtdOrpx+neY8TtiHp/IPMvNvYTpB9wGJ97xk9I70fMHvsTvQcV4nBJfKUKP3+SweiXG8aoRsa8X7tC9tXt8HdJBV1iyw3SYuQy+g7xM/hiP8W5YaZ8n8Hpwxtz5qVzEcJKRULZx6CXIIsfAS022iY/28otgzuUrSAE709dT5PWYWz2PYW8oaTQiqpK6jWK7+P1RSzct2AQU3ZGuwmcVivPDhzMhZ2P4cZxPzNza2aUm8Fh0bihwyru6boKkJD4JIo7tjzgHiZuXMR//piJfx+DIMAl7SI832cVqOvB0hyReGdULVg95/ga7TUZ2BD154BI5pFpf/DbhtV0T91GmiPMle1W0j29EKvY811UiOuyEgmI1A8Q9t57fVJY0xg25isyCwsIlvnUXVYrZ7Rtz6unncmzs6bz7aoV5Zm8NkUhzeVi4hXXkOw8cmQIDnqC0r4Me2UO5wSlPUi9BOkfC5GFYGmBcF/5j8MUpbYDIhvA2hxhbWMY/NyT4vzAnIi00YhqqiXp+ZdDJPZz9ZUo3HByB/Jz7VFZlRIINfew484utH5wEanpiRTnlVYbNhgPT6LG6+OyeOzK5pQW6ggFNFVw+/NZnHZpKTjPQEl5DT20hHXTbmHreoXNa5243TDsju4kuaexZyAK+BRuO60dOTvsRMLV7+s7PVZUNcLpI0pIuCCRz35tQ2S+l4QVBTz1RiZ9h3ix2SNIFMCKQCN6NSEMmYV6MwHdqCcaWUtFmKcdRCIF2UHG/i+dJTNdNOvQiNtGqiQnzAU0lsxM5J2Hm9K5t5d7X9uOPbbAEwA3zT6NqTtjM5I9NjuLb7wFbzjM5T9+R1ZJCXv08Xtl7OL9E8fjsDUysoadA6v9LHTfZ9w26S8mZbWs/o9UhkPRuK3PAG7v0z/ueSl1ZHbHuOfiXo8NpcFfzM3azcszP+WDE37GaVWxCB2XpfqVTBTCg6g/v7zq1d7wRyKMWvYXv25Yh8tq5Yqu3bmgUxfy/X4GjPqIcJVNVKfFyp19+3NznE3rwxUzQakapO41fL9K2j/OEpVaLjJ/aFncehCwIf1jIO0jhH3/v0TC0gQsTSoOhBfFWbYChJGBb6s17FjqQyTWZWKxQlGeLSZVXgCOLB8JS/IQmqQoJ35VJKFIjKCJ2M8r6Fd46NKmFOYIZCWXwzuPNKVNtxLanvQoUqq8d8ej/P5V0/JEIptdZ+oPmbw3NYKnTAJk0rep5O227dWoA+hqmJe/28QPH9bn98ECt74ZEBx3cik9BxZjs+tl708nfqSKBD0f6X0dkXCnEc8f+B7848CSTshxJRZHbxLcGzjhkvUMuakTbXq0RwjB8Da3sDszt/yzyM5Ko+Oxfs6/Lp+qoeCarrCyIL4shAA2FORTGgrhC4dRdQ1davRIy+HVfpNxWHTQd0DRnci0UQh7rNiYHl6FLH0FmziJ+PsRe74HhtRBhivEVd2PrfZzFUJBKg0rZJ9jelzxvQpqFqbsaMmcVTNAarx7/K+kOwNVw/+jeyMpWzWpGDN4OyQ+XSOjDoZ42q29+3Jr775Rx1flZmO3WGIMe1BTmbN96xFl2GtKrRh2IcQ3wEAgQwiRBTwppfykNtqOh1S3IH2fgrrB8Jd6rkVYGu39Hr0AWfRgWSSKMNLtk0ci7PHriO61Le9bRuZeeShhBIgYinkZU6odMNSIihrRcLodyMhapP97kCUI5xBwDK6IoZde4m8a6ntJkALhvhoZnEZlv7GU4PcKEpNVigvib8jW+2FLxfUWQaiZBzSJI8uH06nhTtDJaBRh4wpXVL+O6VvKxbcWIKVg+o+J/PlbCrpmnI+ELfz23YUMre/nv5c+ytY10RWXImGFwjwrE75M4+JbjJXJwqlJe80gLX+fiuTxq1rjL7WUa8IAnHBmMQ5X9REqK+Z5GDcqHW+xhQHnFHPqRZ9jD84G+3EQGMum0nQeXticFQWzQc7BvaaIpr9moZR+T1qjVO798CZys2KrE33wdBMGDi0kOV2PmqVaFDuNkxpTkJVNwooC0CW+rmlEGrhQpV6e2VkhBCZYXpDBtbPO4pchP5YdC7Iz+z2+2HolK7J30ymjPtcdexxN3PnoBZcjZISLWq1j6s4WUXIOAC6LSv/6WRSE3AxqvJUrO9pJtFeMPlJKQ3qCCFhaGxuzCfcZmv5Rew8OsPUmHJxHULNgUzSW5DXkkcUDkKxjcHMLibbwXo06gF+1sTMygHapVmPF5P4/hK36uq01paEnAS3OKtMiBM2S9h3pdiRSW1Ex/1cb7dToWeGlyMJry1QHNYisMGZb6f/P3nmHx1Gdbf93pmzTqhdbkmW5d+NeMAZsmh0MAYNNJ4QSeodAaCmE0EILgdBJCL1jwIAb7hUb9yJX2bLVu7R9Zs73x6zKalem5ntfyHtfl6/LmqpdJWwAACAASURBVN05U3bmOc95yn2/02k9r5QSWXsxGLtoNcbmfmTdpZA12/aYvwtCX5KwPtysAKuqtYKhBb5GP09e/SJL3luJZVr0GJTMTQ9vpN8wW6JMhuaBPgLSX7SNu2N0gtZxADfC1XmrtXCMQKbcA42/pyX0IASkZphMObeWj17MJhxq92IDwgIRto2hb2AalRf2wc67CZSQSe4LO1AaGsktDFGy24kRERgRhQtuqWTmVRU4XXb4ZfiEBk46p5Z7LuiFZQksU1JVvJybjl5Ac709XkeEgyrrFqW0GnYjkliRKX4/BdOQMUYd7NWDtEjYnfH2U9m8/nhXQgEBCLZ+lcTnr4d49KOdOMxd1IUUzpp/Ek0RHRk916Z+Key+qBcFD22ibE8Fd558P5aZoDLDFNx+Vh8eeGsfGTkSO5YsIOlKxm130PjQJqQlEUjS5xyk4aRuFJyTwpytzxIxY7tGDamytzGNrXWZDE6vYWdDOmct6E3IWk/EMvm6rJR3tm3mjcmLGJwaRAg4qsshzum1nTf2DgIJqrBAwHMT53BkTvsyVRey9gLIeCP6/F8L5iG75V94Ie1xFM9pWEKzuf3NUlALEMm3sss3lMu+eJ5uSfWU+rwc8LUYTIN9DVrrPTscTCk4f15fll5yPe4fUQ1pYHYOvdLTKaqpjslV6KrKRcMT0yr/3PGTE9qQjb+PJuBall0GSB+y8YHOdzI2g7mfOGMsI7ZA9neF6CzZKmPKxlpw17QHWPreSoywgWVa7N3cwG0zCqg82OKh+22R7tBce3glHZJvBly0GUQ36APA9YvDn5pjFBDr9aoanH9TBX2H+XF5TBwuE90jYzpDjVQHFRf3xfJoSJeGdKmYqQ5Krx6EL6iz/LNUklJMTvlVNUef2sC5N1Tg8lit4Qe312LwaD9jT7DDOU6PSWrqoWh1TbTmuG8KFRf0pvxXffANSUeoFpldw5gGHNjlZOtXSQmMerwnpqgSIxL/6M57J51Igi7Z+hqN1x7tSiigtN7PUEDlwC4niz9OBiJ8sK8/YUuJxuWj0BSMdCfBPilIKQkHE9MMA/ibVDS9hWzWAixk87OE1r+JiFgopkSYoEQk6XMPck/ObIobfRgS8j1NXDdoLX8YsZTjcovRhUGZ3+ZRuW/9BHyG1ip1F7Es/JEIv1/bp3V1IATcNWIln570HrcPW8XdI1aw/JwuHJnTsbQxCMZOZPAL28Cbe+1t0g9WJbLuMqRZieKehpI9H6XrNpTsOQjXiQQbnqIupLOqMr+dUW/5Qby4Eiy0pISQqeKLaJT6krhw8SmEpYOdCUo8fyj+edqZjM7Nx6GquDWdTLeHv//iFPpnZsV8b199Hdd9/gljX3yGX7z+CrOKtvM/QYT4n8ZPKsYuZTjqdcd9Em3p7gTmIRLPYREwOucf6RSeC6DpMWLrmDVwjI8ROZZGMXu/Xsbu9XvilJAiYcHH/8rksruj8UzpRwY+R0QNt5J0CVIfbsvaWQ0I9y9s/u3DkDYB9v0RjjiP3+WRPPrhHrat9bB3q5twhpP6bRaLP0ij4qCTpjFZyKilSHUEmdjlIKZUWFqci39wOt4NtfgaVYaO9zFxWnVcpyLYxn3C1Aa+XppM9z4hUjMNgn77ja/5ZXcaj+qC1O3a8HCuB70mwJfvCxZ+mEGXghCW0RbHDeV6qD8uj3BXN+4DTaQuKEevDaG7Lbr3DnJon5OgL9aaFO9w8+Kf8/jN70sxI6I16nzHuT0RqkrHSSLoV1k2O5UTZ9axqzGNoJnAixQQyXLi3h29vck6lltFrw7G5Cwu/G05SclmTGeqEEGuvm8/y2YPipmwVMtk2wIn46eUkqRF+OvYhVgS3to7iL1NaTg1i6Xl3RiXU8pX1bkJveHNtTmYlmgtQZx/qJAnt46i1J+MRzMY1M3FYIcD0bHWXvrB/y7IIE1hHV0xcWlRJ0mayMCHCG+sDoC0akkyFgC/jNnuUEwu7b+JSweUoGspWFYdSAtFSIKmgx31qfx5/VEETY2ihgxA4FStby1OLaUBoaVglYI2xA67dhLmzPJ4eOPMs6jy+2gOhylMTYsTFC9paOC0t15rpSSo9vu5c8FcDjTUc10nCeWfKn5Sht32RHXiuU8AJTFTHADaYLtcMA4u0L+7rqPwXICMbIXg51HGPQu0QkTaX4FoGWP9zRBaSOnWVFQ1h45etBFRKN7RvgxL2C3q7bc4RiZMmnWGvZv28+b9a9i/JZ9+w3yce0MFud3brlsIGDzGz6AxAQxLQTvV5KJby9m+1sNNH3WhVleYXljEn0ctba2DVsZK7t4/lK0boM+QAGNPsKtmEumKmCZYDpWL7y7nlHOr+WphMu4kk4akJBondkE67HsgQiZ5T29H9UWwpAALyoqdpKQbvLJ6O2tq8rhq1RjClgKKQiTfQ/OYbIa8th5nt2SOvaqZV8500z5hKAE0wUvNo3j34RFMcu1F2eln3cLkqKce75UJRZKaaU+4R2RUMbukT1ycWjoU/L2Tce1qpPqsXgR7JYMlEYZF1jv78G6sJZzrZuRxvoRkaUnJJll5EaoOtZuQhf1b/GbARiKWglO1uHzZFFZU5LdOLu/uG8Cqqp64VZU43Q/sksgWNs339vbjT+sntp57fRhmzCrlX8emMj6nYxJeozIg+PX8aextSkEIOLbrAR4cs5g0ZwjMsphvG5bFTZ9/zvx9p0eT0bI1jfqvYz5lWGaVTfZmgYLDZg7VR1PiH8HFS+ppirTdd01RGJqT00r9ezhIswxZcy7IBpDRZKtjuC0i/h2FRVrwj7WrCUSNegsChsEza9dwyfBRP7pq1P8kflKGXQgV6Z5uNz3EGHeX7UV3tp/WHek60e7GbE0KaaAkf2sCpY7nIdIeRhrXgbEN1DzQhrR6E9L3EoQWASF6DKjHCHeJG8Phshg4qn3XqQvhPus7n0sLNi7ayl2nPEA4GEZaLvYXOVnycRqvrd1GUorFzo1uait0Bo7ykZZloiu2l6ZpMGScn+dylvObDancO2pp1INrqzC4/8aNXPDqQK667xAOZ+fLVlUFdayHabkHcaiS8Sc20qUgTGXXXNpHWLwbalDCJu1YfpFSEA4pbFyezJ+NiYTRWhdZUlEwNUHRpP5YXo01RQNRbwvT5bXduIqbEdhedek1gzBTdJpUhTdCXXEazeR9sSMacmqrEGk9pgX+JlvI+9TCXTy1bRRhS8Vs39wjBL7hWfiHZ4EqkVHOdOlUqbqgDzW/DEOqk1p9J1nEU1UIBXyN8aGxidMaUAQ4FIutdZkxRh0gbGmU+R0cmVPM8orcmM+cisGMHjsQAkxL8NCm8XETkiHh6W2jGJc9OyahK9G4aEFPdjV6WzYwv7QH0+dn8OW0j1GcsZ7r8+u+YsH+qhiKBYHJyd32MTKrAk1p/zyEgQAiaSb9M47h1gnreXD5UlRFEDEthubk8My00+LuUSLI+luj1TntZrXwemTzc4jk677VGB2xtvRQDCtlCzRFYV99HUNy4t/Tnyp+cjF2kXInOCcATptrAge4piKSLj/8fql/Be8NoHazaU7d0xGZH35jN+dhx9QKEK4pCH1o7BLR/yYtE0h+zzBjT2jE6Wp7QBVF4HRLpv0qCCTZ1+C9BuHovBStM0gpWfXpOu4+9QFC/lBrDNgyBQGfwh8v6cHlk/tz24zePHx9d1Z8kdpaudJ6HQIKC3xcXvg1qoh/8KUFR05toNfAxGLCUtqCEuurs3ln70DWVuUSNhVUDR6btZvR4+pR2r1QekUAJRzvhkbCgj17kyjxJfhNFEGwZzLhHA9oCmaWi9LrBnHgnuEYLoWDtw3DTHeC2mZ4QwVemodnALYodvwqQ7BmQQpLPk3FrRr8e9InDE1PwCOkK0hdIJVYAy11gZnpwtQEz+8YRsiMf51qyjWMsEB3guawJ/QLbymnoI/tmAgBmzopifQbFpnOZibnHsCpGCTrIZyqwYQuh/jdsFUA1Ied+IxEiUjBtrqsWKMuYVbpTPY1dswDCUp8KTy9/WhwHh/zyWubNhDs0LBmodI9uTHhs4L0I8PrAbhw2AjW/uYqXj19BnMv+DXvzDyXzG/RCSqtxijVdMdnJAiB9whEInywfStPrl7Jgn17MBNQDSRC905WChHTbFW2+rngJ+WxQ5SnJf05pHHQTohqvVsFIw6/n4rwXgreS//zJ9mhBv2Of+znzb91YfarmQQDXkZPGc5lD51Det4uu7TRcSRCjU3ySKvODh8p2Yett3/xd68x6+kvOmFIFGxZ5UUotBrz7LwIipqgNEyD0e5DaAl04RRF4vJYNDWopGbEd1maEh7fPIp39g2kLuzm65pcXj7mCwan15CU4uK+q3aw8NORBKJphnB+EpZTQQnFHkt3SPoMaEartmK95s6gKphJGo3H5oIZXw4jnSrNwzNJ/rqGSafVseyztNaYfwtCAZXHbyngkRu748gQ7D5DwzNAj2tdT1x+2rZtbXUXBDJaq932jdzCCM8v3sHK+eOwjFomnLSbvB6xv1VXty+hkXQqBoXeRq4YuJGDPi97GtPpkdxAobet3+BwZYa5nuaYvy1U5hYHMGQi4yr4585+XDcp9v74jUQhTCj3J4FwEU/J4EaobZ6vW9cZ1vXwpcjx6Jwp0rDCHP2vFwgaBv5IBI+uU5iaxtszzsH7DaGUK0ePZdXBkhidWaeqMqlHz07DNz9V/OQ89hYIrRvCedS3MuqJUNbUxB8Xf8nU11/hso8/ZM2hg9+807eFazLtY+qaDhfeWsFbW0J83Pgav3/3VvJ6dUO4JiPcp8YYdWmWY9Wch6yciKw6Hlk9BRmOJ6sCqCyp5qO/f35Y2lspifHQNyzzEvTHWwJTCt4JDY9ZcrfHgZ0OGmsUOjpHUoKmwDWDN/Du8bPw6mE0NQ0j7Q307FmItKdI7baA50+dgdfhwOvQYaQXy6vRrp8JzWGRWxhmzDFNnF64E6fS4eU2rYQ8MAiB6e6k9t2SKEF7nBPPru20EzLQrBIOKDQfEiTPLSMQSJDDSUhi1rbtN/03IkTiU8zqanHmrdcy47ru5PWIN5RHdy3Bq0dQOkyqmmJxZs+dAHRLaubY3JIYoy6lSqMcz8zBI3Frsb+bS41w3eCvY7apwmRUZnEn1wKNoXBck88x3XugJJjUdvnGREnzOnwmNHD9MJk+oWRAQsZHnTmH+lAXCLROvP5IhD11tTy5euU3jjsmrxt/PXEqmW4PLk3DoapM7dOPx046+Qed7/9G/OQ89h8Dh5oaOeWNV/FFwhiWxc6aalYcPMD9x53I6QO+G/l/QrhmQuBT2uLUCggXIvUwJZlEW7prz7frh1v2NYuRdRdB1lxEh/r4TYu3oWoqLZS8HaHpJqYZre2O4rPXMzn9smpUNYIebfgL+hVKS4/ityc8i+77PYQ+a+WoiYRV1i708uBbxa1Jv5aoSntD5tEMurp9XNR3Cy8UjaVXegaK7kGG65ANd3Okp5J15x3LylKFoG8FA+ce4MKrx6NuaEQqguEnB7nnvt0oCtwzYgW1YTdLygqwQhIDFdfeRgI9kolTsQBSN9bhG50dz78YsbCcKsX3juTSA0eQNKUK16xyhNl5nsCzswG1PoSRFRuu0ISFKixClh32UIUZs6oYnlmJriQIY0kIK+fj0AeCawqEFhKbH1LQFA9vnbSP65f1oahOQxGQ7fLx+PgvW7VTY6HyVXUBv1tzFGWBFKTcQk6Sl+ooy6FhGdx+xCqmdNtHcVMKr+0eTKnfy8Qu5ZzSI8BDGyVWAmOd6nTEydvdPvEYlpccwB+JEDINdEVBV1T+NHkaInMysu56MEuip5ULqU8wb185721fABLOGDiYk3r3QRECKSVz9+7mrS2bCJsmpw8YxOn9B6InkNQTqQ/b74I0AJtS2BJZ3Lt2cNy0FDZNZhVt+0aGSYCT+/Znap9+VDQ3k+J0/qwSpu3xX2nYn1y9kuZwKCaREjQM7l28kFP6DUA7jKjwN0GaVVB/BbGt7AL0Ud+sihReaWundjRR0kAG3kN4YznEkzO8iATrcFWzkBZIK9aog11vfc1J/Tj7ugomTG3E16jw4YvZLPnEz4V/+JS1cxwMGn0MJ18YJqd7V/T0XzJ+ykUx8en2xr09XJrJ1Py9vFg0FtOysHxvQtMDtCzX1cgWJqYpkGbfm2PvifDW3lFELJWktEpcyUWAXfHxjwlz2VSbRXWNgxd+k4W/2sOuq/thpDuRmn0ywpC49jfh2NdIwUvbKLlsEA6HiWqYBF0OLAObIExTaMCB7+h8nPmpFL60DSMkospMHfINEnq+sIWKO4cSVJJQhSDV6eLh8Vuo8x/g5aL+1IXcDMus4vOS3hjR+7CnMZ0BqbWt5YdtA+okpbiQFePtTComtkhLCKJVJtBMgXMlHx6/lsqASsRSyPM0J/D+neA6iZLAAC5Z0kTAaGmGsp0VXVF4+PgpjMjLo7J0JXMPFnLz6uMxLAVDqiwtL+Cl3V5+O2QZj2wdFTMxuVWTq8ccGRf2y09OYd6Fv+bNzZtYV1bK4CwX5w/OpktKCkLJQ2R/hjRL7etQ8rh13hfM2b2rNYSzouQAJ/TqzRNTp3H3wvnMKtre6m1vKC/n46LtvHL6jLjSRKEPhOwFyMAsMIoRjpH4xSQaIy8RH3uP/x0PB0UIcpOTv/mLP2H8bA273S69h4jhZ9EhB+XNAUbl5jE4pwsrSvYnzI6HLZOSxoYfJLUl/a9jS921hwnhNbahCy8GaSLcp4HrF7FSfGYZcZbYPrNo63csRp14BA6Xjr+xzatTNYvf/WM/D19bSCQc/7ArikJDrcbzf8rn+T+1ddwKYfLK79/BiBhsXAQfPuPklCt7c/kDmahqfKins7BGU8TBqBxJticClQ8SG4O1wyLVQSe3rD6eNVV5GJZAEwYnF+xFbddwA9AvtZ4vS4dxxzsZNO/5FUXb97PcG2JZoBKlMoClSAI9kym5ZSiOUj/5D21Cb4qgahJHGhSf2pfmI9rCXIbQsAqS6XZKDkcP6svnL6+m4kD8/dYbInx6wju898nF7FntY+SgbAb84o9k9tnBtP4rQMnAdF7FkpffpyFkX99zO4ZzQn4xnpgQkhOh9Qf/q0CgXQREBZECsp62jRYQIqdTXWsdshehqJm8sXkxhvU1HUMqEcvi9i/n8taZZzO83xuMf+kpgu18hICpU9Ycptl0cO+oJTy6eRwNYSdJuuTq0Udx6YjE9BoZbg9XjxmNbPyTXZEWciArI0jPuYjk3yHUPAA2VZTzxe6dMTFsvxFh3t7dzN5VxIc7thFs91nAiLC+vIwl+4uZ1COeKE0o6YikX7f+nQIMyenCxorymJJFh6oyfcDAzm7cfyV+sjH2w0Ea+5DVUzGrzyBUdS5j9Jks3/1vZr73Fld8+hGZnTRIGJZF+g+l+IxsIjHxlAVNf7HpCMKLkY13Ieuvje1604eSMP4pPAg9ntZU0zUenv8HsgsycXtdeFLcDBpjEQk70fTE4Ya0LqkoarxVllJitJM3C/pCzHr6C6oP2THdcFCw/PMU5ryVTuVBPbpP7BhhUzAss5KXjnwJKo8mnp/dRkVDEisq8olYKjIawb2gz9a4ycKtGZzXZzvXrB3EUdPHcumdMzm0u4yCe9bR7dHNdH9oM/l/34bp1fFurkX1GyDBjAgCVYLsV/fiKIlNIEqgypnPovcOUlcZT5jmdFtM+1U1XdMCXDHzH5w59UPe+euHnF94NY9evZE632XsCZ3OY6t3MqxLV3RFQQGKGjK5buU0SnzpSFTACe7p0ZBWx3BKKGrUO0PsaxmxND4+MJjp737Oh9u3sbeuNqHohP0bmDy1ZhX76psTNlyFLcnnB7tzdq8iVv/y33x9+musnVnE5aOPOmySXjY/CYFZ2EIZzfY1+N+2S3ujWHZgf1yMHiBkmny0Y1vCDk9/JMKS/fs6PW5HPHbSyaS73Hh0HQVBkq7TNyPzZ9dg9EPxs/PYpTTsdmmrGgVJUvTZfmzcHKbNmcGyA3BqvwHsqq2J8SwcqsqxhT1Ic3XqMsUgbJqUNNST7nbHdtJp/SC8mvi4d4e/ZcAOvYTXgNNmqxN6f6TzKAgtp83T1aPlmYkTUj2HdOf14mfYuW4vGNvo3Xcuqz89iJFIgFrAEccMZMXHawkHOk+4RrKcVP+ykGC/VKZ8vIgz9bGsvcOHaQqkBaYpOO3iKs67qRIhwBICl8uuS3donTEqtqFps0HWG/uoOrcXCEEXtw+PlrgSIt0RpLY5xMbyMppK6jD/tg61XceOs8RH/t+3oTaEUDrYFGFYpC0opfLXbURTHo9J1+py9u6sidIdtNwniaLA9Muq+NVt5TbzpRuOOrmBh3J28eA1hcz910I+PrCbqtO6Y0gLU0pcmkauN5le6elM6nEcWQV/QdEiIJwIoWFVfkeDI7w2bURgFhIHYTPAkvIC7vhqFCGzgrsXzmNgVg5OVSXUidbn7roaPA49ITEWQJJmP4tCQJIegMgapFneaSGClDK66ui4Eg2A/5/g/Q0AKU4nqqJgdjgvh6qS7nKjKUrcOeuK0qmjlQiFaWksvfgy5uzZzcHGRgZn53BMYY+4UM5/O352hp3wyiiXTAfqWmFxVq8dPLJ5HJsqK7hu7JE8uWalrQdpmozrVsAjJx6eh6UFb23ZxAPLFmNJScSyOLawB4+edDJehwORdCEy8FaHTleVtnhqO0g/MrQE4WyjIRVpTyJ9r0DgLZvozDUV4b3msLJ9Qgj6DloLjX9BhkPk9XQkDMMgYfSJfejZbxuv/zVoR32EimmAFU0oGik6B28eiuWyuz7rjTAL7jJRm2MflU9eySK3MITugPXLvVz151JS09teWtOA0mInyekGaZlt24M+wUcvZOHdUIN/cBrpE/R2TIbx2FKXhRk2aY6EWfDSQkQHEi4hQW2KIFUlWvIY+5mjKghS4t7ZSPL6atyOCAfX12CEOybsbO/9vBsrUJS2cJCqwuCxfp77soirZwxkxdRcpNXuegyD6oCfP0w6jsk9Wio52iXk9NG2Nm6HuPD+5ixe292XkuZkxmaXMbNnkc286D4fJeUWpPcm5uz4gr+uKaa4qc3ZCBgG68tLSXW6Ehp2RQiGZHchPzmFfpmZbKuqjAk7utUIv+rbQRdW6FHyus4qzCwSSzgSwzbq1rSEHrslJTeMO5I5e3bHfaYqCqcP/G4FCy5N57T+/xd6ORx+fobdqiZROMOhWnR128vyiGly5eixXHjEcPbU1ZKTlERXb+JkijT2IJuesEm6lByW11/IvUsOxMQKFxcXc/Pcz3j+lNNtpsiM15ANv7e7UtFAPwIiW4lfkkdbsNtBCB3hvQy8l33rS5ZWMzTeB4QQwi5p1HSZ0Gsv2fgMl9xRzNFTFFbMSSXQrPP+c1mtJZANR3fB0m2jDuAqbkIk6GkPBRQ2LE/htr8fYORxzSSntb3QC95P4x9352MYAtMQDJvQzPUPlZCaafKvh7uyfmkyChbJqyq5/jdlJGmRuDCMlBAwVe7fMAHVaTEqN595Vf44TnnALjpKcI5CkTj7aXR5ew+er2tRwhZCSPyd1MiLDjH+9tvdXotRN8GKSgv02P39kQizdxa1M+zt9k2+CRleFg3JWEgJyw/lcemKadipT4VlFd14sWgYLx9fw8Au19v7qVm8vycpxqi33hvAHwlTkJJCSWMsh74CXDfOXiU8M+2XnPfBO9T4fSD9GJbC9B47Ob2wA9+SNEHtnfCe2NevItXeYMYbZvTB9hBS8vjqFQn3756SSn5KKv+ePoPLP/kQfyTSGvZ5fMrJ5Cd//ybB/0Ni/PwMuz7KflA7wBfRWFLeHZeqcnp0tk9yODiiS+d18NLYi6yZ0fpSYlXy7NrlBI3YhouwZbJkfzFVfh/ZniSEPgSR9YFNWoYKMoCsOjrBfKMg3Kd+50tsDIV4ZMVSPtlpV5HcOCLMBd11RLSMzumSUcMeu5+qgdMdAAzye8HMq6p48b6uGEYb50qwR3KM4dJL/QmNJgiCfgWHU5KSaVIXcpHpCrJldRJ/u60gys9iY+3CZK46oX+0G7bNUxYmHJlT2tqWLiVs+8rD8s9SUR2STVtT2HxsGuMPbaN4+bOMOXEYaz5fixnqcCNNSdO4bJLX1cR0tGq65LZzN/LIDYWEouIdbWRcscIUimIxeKyfJbXdeXzzGA76U+iVXMdvj1jDUV0OoSjQu78PKuPvhCIELi0+nm1z7r8D+hiQfpqq91C0weCKsimY7SqvgqZO0NQ4Z146G69oeyXzU1JQhegk0W9R7ffHfa4qCocaG+mfmUVecgoLf3UpX+15goq65QzPKCU/qbnDSE6767kTeUhpVoF5ELw3QcMt2NU8EnsKcSKS77KvwTCoaO44to2DTfbkM6xLV1ZeeiUby8uIWBbDu+bGlVd+V0RMk1c2ruetrZuImBan9OvPlaPGkuz8duIdP1f87Ay70Loj3WdA4CNaPOSAobK/OZVF5f3pnZHJxR2y/6Zh4mv0401LQmn3wsmmJ9uMehTl/sQxeF1RqfH7YzrYWsmKhBfSX0DWXUVbQlEiUh+Li2vWlNXx/uOfsmXZDrr1z2XmzafSc2hh6+fNjX7OPOuPmMtLSI8KN8wRLs7IC9Oi6zzhFw384554jnlVE0w+vSpm2+7NHox2ikWO8gDBnsmgKXi21ZE1a3/C6jKXx2TS6XYC0KlZlAV0PJrBu89kRznP20Pga2x71KQqMLwajWOzqA05yfX4kBJee7QL7z2bTSigRLtloWDZBioMi9teW8S1f3OiIFAcZqvKkuVQCI5JJ3hON6xCN975FTjqQ0gD8nqE2LHOQyiYON+g6xaRiMDtsXB5LIb/VnDTquNbk45b6nK4YtkUnjlqLkd3PcgIdxlCiLj52aGqzBg0OGab5X8X2fhnIBwlz/JQskvlzmv7Ero7zaQKQwAAIABJREFUMQVAczhEU6geXfXy0Y5t7KurS/C9NrTPEbUgZJrc9eVczhs6jDMGDCY/JYXRGdvAuzPBCAp4foXijafjkDKMbLgdgvOw9XjD4JxkO03mbtAG2iHCqFCGU9NwazrNkfj8Sla7d0IRghG5eYe9ru+Cq2Z/zIqDbSvol9avY/7ePXxy7oU/eNL4KeNnZ9gBRMofwTEG6X8Dw2xms28kC6rG88AJvTixV5/WOnXLsnj1T+/y/uOfYoQNPCluLn3gfH5xaZQvI7KejlZtXE4pJb6UOBV4iTxsmaRwjIacFTYHhjTAMSqOpa68uJKrR99OoDmIETbYvmYXC95azs1vXMeU6RM4UF/POXf+jZqRyejdepG+oJTkdVWU7nbQeKyOJyWMEJCSbnLns8Xcf2UPIiGBZQkUVeHax0aQ16Mo5ph9j/CzZXVSq6FMXVxO09hspCrJfmcfihHvLSqqxYCRfiad1mZ4ipvS+O2a43Dt201nNcWmJqg5owfNY7KRCjh0i0O+FAal1SIlnHVNJRldIjx5ezdktFtWjUgkglBA8MQ1c0lKN5gyo5YVX6Ti8VpMvagW9zE1KGI3fabWMnn8+Xi1ELeUfcFrf+lCdr5d/mh2CEs5XRZTz6klNcsgtzDMxJMbOG7eeXGVJEFT56GN4xiTWsqCNzI525XJrGQfAoGoD6GUNnPe5PEMb9c2H4k0YNX/EV1pv2Ty02uw4Kb3q7luQyfPCKDWX8kpc0+ivLmZgGF8r7K1Cp+Pv69eyTNr1/C3KdM4PmeQnXuKS2o7EJ7pCceQTQ9DcIG9TwsFdGgJeK9F8f4j7vuKEPx6+EheWr82ZsJxaxrXdJCy+7GwpbKClQdjw6Jh0+RQUyNz9uzi1H7fXr/154afp2EXAtynINyn4ADG58D4/vHfe/VP7/Luo58Q8tshjIbqJp6+4WW8aUkcfeZ4u5POiqUxvXrgej4r6Y3PcLUugd2axq1HTsSpHf52CqHDYdTYX77rDZrrfW1iDpbEChk89Jt/oA7L4YbPZhPomwSqIJLjJjAglS6v7MK9q5FLHz+K9+5ZiUez47ijJwc4+4Ygc94sZMQJQ7n0/vNIzVSpLvmM/U2pFHrryXQFOe2Sej79dzaRsL3A1quDdH95F/XTe6I0J+5odbol97+5lxaHKBBSeXrLCNbXdyUzP0jK3gqUuO5OSdX5ffANzbA5CICwpXDz6uN4fdInNv2rW3LcGfWsnpfK6vnxcVczAo2VGv2H+9t47AGoAWBtlc1RYiHQT0yl7zqDef5+pMvEGp4X3FpBSjThGzYVKgKJ+UL2NqUxb08P3ut3Blek9KXx2Xns3FuKLG3G5XGw+OVd1Dy7hj9+8FvcXjevr/8n03NFx1A8LrckKTmqrhSnUSoZllGBam0jU+vFPiMrei2JoUZj1InCNACGlBiGwc1zP+OrS87G4X+VWI5+Ybf/h9cjW2TxWs5EWuB/h8RVMP9qrYLpiBvGHUnIMHh18wYEtrG/dsx4zhr0jfr23wsbK8oTNsr5IxG+OnTw/wz7fyNMw+T9Jz5tNeotCPnD/PuP73D0meMR3quRddfRPumZlxThk1MbeWrHKFYePEDXJC9XjB7L8T07Tz61hwytQPpetGX0nEchki5rpQr4et6mhAo9ojHMHR98SlA3ae3iUQTSoVI1syeFf1xPw26NBU3/4pc9G8CqQ4vs5PwbP+X8GzeCswuGx+C381fw6a7pOJUIIRNOzd/FncMFZ/9hMi/fPcsua7MkojGCoXQudpaWKVFVHYkBUmIEFRqadJCS+sl5JK+pRgaN1kSnBBxdFXzDMunIWBUyVZ7ZPoJnJ9rqUe4kixPPrk1o2FsM4SM3dmfEsVvxJrfn1dZ4Yqs9aYZNlUbTyYQ7I8xZ1R1HbYikonrcTtuzs0y467n9MQlfXbFIdYSoD8dXH5lS4aZtJ5I+v4i/zf0Swm37BZps47d56XZ+f8rNWEJl655amqdlc9615TEVQZZF6/geNYK/3eogRQ/xyNgvMS2LQm81a6tjSeEUIdAUBV1RkEC35BTOGjyUh5YvbVVXSgQhBF9XGhzZ9e1oQn9d9BMJshnZeB+EVyPSHmm3l0FnJaum2dDpKkJVFO44+lhuOnICNf4A2UlJ/7FwiLSaGZKymWNyD7CotAthq+04TlWl23+p1mkL/msNu78pQCSUuHa6ssSW7hLOY5Apd0PTQ1FvxwLXNLql/omHun1zckZKK8YTslvsH6R1ovAX2y3TWZ8g1By86Uk0VDfFjSMk1MowUsS/JGaSjpHuQMnzMqXPAISmI+t+A6HVtHpcgbd5bE05y/YXku0MUur3IhF8sKcvix8/QMqqDxFGmyGX5T4ynt+Bb0AqSVvq4wz85JndsaydKIoEAcmpEbIzwuytEpipDkpuG0r6nEN4ihowkzVMp0rYbyLCFrKDhppEobg59iVUE7BPOt0mR09rIDsvQtEGN7NezGTGNbXousUhXzJ/2XAkqyrtvIKuWIzNKWNXQxbH5hag3NOTRXv20PPAIaYV7uHUk8sIaE6aIzpePdLKeXPVgK95YuuYGG5zgcSUCiJokDbnIAlVL4BIyGDDkipaJp/ZL2ew5MNUnl1Q1Grcg5bGW3vtxP3ZvbajKSb/3DkMVVhELJWT557FLUO/Yk9jfEhPSsn5Q4cxJKcL3VNTGdk1DyEEi4r3saxkf8Jzit5gNEVB6H0h6Xxkw3agfeliAIJzkcbuVs1gIRxYak+EuSdmKEvCV1XZdHXV0is9o9NDujSd/JQfT9O0I6zAJ9BwF0NcGg+PCSMlXLl8Cqur7Ni9pqicMXDwN4zy88Z/rWFPSvWQlOqhoaox7rOeQ7q3/l/xzLTFPawKEKmIwyk1AeFQhCVvvsLuNR/TrVclk6ZbeHJ+Dc7J0PwQsSWPEZBNSN8LiJS7OOPGU/jbTS/HVHZYmsA/JB3h0hJ27glAQ/DCA9fh0nRkZBuEVgEhjFaxZ4Pjc7dww6CFWFKhKeLgzrXHsKisEAzL5t1tP6YErSlM8xFpJApOjJ+8CEWJDdOMzi5jfU0OYUvDTHNSfXa09M+U5D++Gb0qlLDPWRUWwzPatDkDPoUF76XTXhijoE+QRz/aje6QON0W4YCCoklUVXLZkil8VZ3faozdaoRjcw8wJL2G/Nx7OXvC8Qx8+gksj8aeAYU8SSFPLmw5mkQRkixngDuHreCS/psxETyzfSQBQ8OrhwmbKn7TgaM0gFRFZ3xr7X6N6B2PKDTVw9vPduH826rQFItnto9gTdT4vFfcn7CpYUgVox3N5WNbxmLXwcd64RK7f2Jqn348epLdbxExTQrTUll5MHHlDIBDUxkZTVbK8EpijXq78w5/De3E4Df7r6Svege6YqIpkoglCJsa92+YwISGTdx59KTD3YhvhDRKkP5/Q6QIHMMQngti6H473+8ANNwJhBASvNH544WjP2fSZxfjdabzxJRpZH0L3vf/n5CRrUj/+yB9CNdJ4Jwc4/T92PhRDLsQYirwN+xOnBellA/+GOP+J6EoCpc9eD5PXfdSDO2t0+3gsgfPj/muEBqo8VUmHVFf1cB1426hvqqWoM+Jy5PHPx+UPPHxP8jv9XcSt9gbEFoGwKlXnsS781dR9slWLE1BmBbBHsk0nt+PSYU9WNmBS1qETfIPGdz42W/5y9plHFzQwOicEDcP0PDqBi41KiQtYWRmRbQ+28KtGTx55HxmLDidxsYICfUS7CMkDMcU9I2ntf1Vny28vnsw4aDaGm4RIZOUjdX0L2jkpMsqmdWUwqrkPq0SeUiJSzO4rPcGLMuujf/qy2SWf54KCDK6hPE3qdzxjB02aSlYcnvbJr7njp7He/v680FxP1QhmdlzB6cX7kLVB5HtNrAa/owl3XTGp25JQWUwiRtXncDSRV9z+y82cP6Jm7GcGqYpuWrFL/iqOo9QjyTqTswjfc5B1HDn7JDtYYQVPpmdy9bjBrK2Kp+aYFtFVVPERfsIuiIsjswpJccVoFEOZnGJhdHBWAcMg1lF2+mTkYFD1Xhy9QrCptmpUffqOi+eOh0tWgIZbHRRqOsoHWcnoYCSHbOpONCPP62bwYW919IvtY6NtTm8WDSM/c2p9PH5vtX1J8KCfXv4aNM73D/yNRyqia5YEPnaFpXPfBeRkK63DTIwi0TvkUtzMHdGAenpZx+WGuF/ApbvFWh6FDu8ZSFDc8AxDtKe+Y8Z9x9s2IXNYvU0cCJwEPhKCPGxlHLbDx37cJDSguBHSN/rQMgWevZc1Gk9biJMvfg4vGlJvPKHt6k8UE3Pod257MELGHLU90u6vPi716k62IBp2IYr6FcJBSWP3VrAox/s6XxH1X6phBC8+PZdXPLy62zauJuIVyfc1UOyFAzMyKJHWgZvbNmIpiiETZNj+/XmpJP7ctui+a2VAQ2BIH8YGrR1KKNI9Jw7FJOL+27m4fxeuHc1onQIMShhi7TFZXEpPoDyAw56DYpNrGW6grx55IfMfOwE/APSUAImqUvKSF1aQW0anPleFaeGFnP9U0EWu/thJem49jXRc/levtycxJpUF2sXJbN1TVLrEQv7Bxk4ykdB7xCJCDdNE3TV4tze2zmn13YsS7SFccztyPrrWF6RT67naA40J3NYaiQheF8fwdqJkGSE+PNre3ksciwbo+pGGR8fIHV5BUokkRGNl91r2dqcnEo9x3HF2J48uXoVDaGWOnDRej6F3gZen/SxHRZCoqtLWNlzNJcuPiJuTEtKHlu5HEvalVjtoSsKqqLg0jSuGj2WC4YOR1dVbvhiNnP37KJbksaHJ0g8MW+9AOEB59ExY43Jz+eOBRncuiZWUcmj6RzX8/DGtzPM37ub67+YzZuTPiFJbz+5hEFGkI33IzJePPwgsolEQhwKFuku63+dUZdWLTT9lZichfTbodLQQnAd3+m+PwQ/hsc+FtgtpdwLIIR4CzgN+M8a9sY7IPAFraGN5v3I4GeQ+f5hxW47YuL0cUyc/uOUYy37YDVmh2dOWoLtXyURCthyePFwI5LaqgzCTUFC9ywhNUNQdnE/MC2aHIJn1qxBcagM79KVyT16ccbAwWR5PBz98lPcPHgJM3vtwKUa7KjPiAokd55QA9AUSXdPA+5t9ZheDZoiraWNLT08HblXWvDIjQX4m1RqKnQK+gS57O4yBo7yseRFD11ejZ/AfA0qkZBA0yVnFhRReoePcNA2agHg/W05cROI020x88oqRk1qTlj5AFBVqvPMPfnc9vcDOBySQ8VOuvU20DQD05JcsXwqayrz8Jstj3miaaodVEHJ7UdQ8NdN/P0fvVk3vSthS0OrDZG6rCJh6SeAwyXp1jvIgSIXhtGuJ8AlOOKSicwtL2NjeRm+VlWm2GqYZ4/6gmyXv0XVD4BxWes4rdDLrP3xSfnOPHRFCN464yyGdunaauD+vnol8/buJmSa7Gl0cPnSqTw+/ktSHAZOVQG1AJH+tL0qbYf85BQuPGIEr2/eSCDa6ebSNPpkZjK1T7+4Y38bPLhsCaYVYlB6dYJPZZRj6fAQzsnIwNt0VCmTWAjHRDZXVvDUmpXsqqlhUHYO144dz4Cs7E5G+/+A0AqbskF2TEb7kcEvEP+LDXs+UNLu74PAf6ZwNQpp7IXAZ8QKFoTAKIHgHPge3Zw/BhS1E49QkEBv022/395bEO28pS9e/pJgc5Damb2RDqU1rGFpAss0WVtWyraqKrZUVXDnxGO5f9QnjM0+GBWghsFpNZ1S6rZHMKyw9X0V9yE/DZO6YuoK3s11WJrAURZIUK7Yhj1bPDH//8OvezLuhAaWfZa4EiEty2DDci+P3FhAMKDYPDYtnO52XRzClAghcHocSCm56HeHGHGs3cnYwv/e/rpCAcG8tzNYtyiZ+68s5PcvFhP0wYcvpFNdqhPsm8xqVy4BvkMSTwisJI364/MoPlTXKhPo2mOLgSRk3gTMNJ3uv9UpXZhKZEEzrrogAUOj+vTufCQqDhuX75ncQH5SMx0fHZca4frBa7ioz0ZUYfLP93qxanYyhqZRPaNnW3VUO0ggPyU1xmt9dfOGmDrvVVX5TPjkAgakNTLr3EvR9EKklCwq3seHO7ZiSZg+YBCTe/TkjonHMDY/n9c2b8QfDnNqvwGcNXjo9650OdDYgGkpRCwVNZHjIb7Fatsxnno5FoexHI9mYFl2UvqD/cPIbg5yy9wPCRkGMnq8hcV7ef2Ms2L6DA4HaTXbVWvBTwEHeM5BeM6Lm/i+NURnYUAFxH9Oju/HMOyJzjruDRBCXA5cDtC9e/e4Hb4TwutBqAmO4keGl3+vNv0fAydccAyfPjeHSKjtoVVVychjm3A425+sE9JfQDiGIoQddzVNk6I1u1k9+2uCgTDBPilxpYEtls1vRFhcXMxFQ9JjjDrYFC+mZbfOt1eQb28YIxEINCjMfj4dI1knLSOZ/cekUTetO55tdeT8ezd0whyYCOGgwtJP09EcFkJYREJtVsrptjj9skruu7xHDM1AJFXH8mhgSbTqoC0kKMCIGLi7pfNBZg796sL0T61FCIkqJJGgQHNIgn6FAztdvPtMDpGwwsYVXu44pwfb1iajOyThkIJwCTJyiii9drA9QX5bKALf4HSG1e6jLFpCZ3m0TgnoG0dlUn1OL4obJHKUQBklURtCBFxuZMdC9gRwqwaWTDx296QmSGrkvssL2bUwSJI/ghTQPCKTYN+UuHNKdjhJd7s51NjIs+vWsK70EHWBeAUmiaCoIQ1D5KEBdy+cx0c7drR65gv37eXkvv14+MSpnNCrDyf06hM3xvdB1yQvB5sa+ai4L6cX7op5bsEFnvO+cQwhBOfNn0B3l4tTu+8iZKl8sK8/a2sKcGtzYyYxS0oChsGflyzk/bO+eWwpw8jas8E4QKvT2PQIMrwakf70d7zaKJwTSWwiHQjPjO835rfAj2HYDwIF7f7uBpR2/JKU8nngeYDRo0d/u+xTZ1Cz6OxmoXxX4dzvD2mWgXkItD4IJY2L7zuHrSuKOLD9AGYkjKZbpGQY3PRIez1VN3ivQHGObd1StHYP9/zyQYI+u+NUAEmbavENz+z02AEjQlndBgZkqHRMJqkKVPjdpDmDCKAy6GbBoUImKsUkeyKsmZ/C6493oblKRdFN6isakCIDJWwS7upBGN9O9T3mXgCyq5Pf3bub1x7M4dA+Jzl5ES66rYztXye18tZIoHpGD5rG5tizjZQIQ5L39HacZX6MsElTcTWhpzTO23o8qXoQX2EKugumuXbQY0sJRevcrNmQQfX0QrtiKGQSenAjirQIh+znQgYljjI/KUvLaTj+u7Wwq2GTa67cj6b2ZMmhPPwDUpG6QIZinzozSaP6nN5IXSEsaXM0klW+raJPUYMdOou7nxKEkKxb7GXtwuRWEW4hIevjAxy6flBbEho7THLn0cdSXF/H9LdfJxCJxCVf26N/ZhYuTWdbVSUf7dgeJ44xe1cRFxwx/LBcStV+P4+uWMa8vbtxaCpnDxrKVWPGUVRTzee7dqIIwan9B9A/067Jv3H8BO5ZOJ/7Nkwgz9PMmOwyDEvFo0tU1+Q4hbBEqAsE2Fdfz06rB/NLe7T7xKIpnEir1u5Q/VYIzrHf55hIQBBCS5GR7baq03eEEE5Ifw5Zdzl27wCAAck3IfT/TOMW/DiG/SugrxCiJ3AIOAf45unxh8BxlL2MiaPnVRGemf/RQwNIy4+svwHCq0A4QIaRnvNwJf+Op1Y/wOIv1rFrwz4GD+nJ2JNyUYLP2pUvaia4z0HIIFbT3xCOMYTMUdx+0r346mPL0Lq8upuSPDeRnHbL03Zut0NVKfWnoGXGe9YhU6HGGMAfvs6hPqwzPKOCLcXpvHa9Iy5OrEQsvGuraTirNxOza9ncoBHonYynqPE7iI2B1AR7Lh2COKKMp+bvitH/XPxJOqah4E2LkH6yhwMjs1rl7ez7KSn/TX+637ueaOEOgX2SLnt3ghC4LUnV2b14c/gw0rZkkryqkpI7j7BzA6qC3hhOGCVRIpLktVUdDHs0/tMxttPuHl/Qayt9J41j8qYTWLRzAyRpOG7LI/2ZPdRVqq3i4IHhaXZx9/eAS9MIGgamVDht3hlcMWAj0wp2k+YMEzIVdMVWJV3+eSpBf6zhdx700f3ZIrw3jafCadAtOZXrxx3JpB49uWr2LHyRSIzCUHtoioJDVfnLcScCsGR/cafiGEv2FzM0xy5B7JiU9EcinPbWa1T5fRhR0Y+nvlrFq5s24I/YothCCF7esI4bxh7JFaPHcsbAwURMi0dXLuOyZacwNCPANSNzOa7viShaQdw5JEIifdQWdBYs+7YaCzL8Vef0xJGN8D0MO7TQiayE0FKQQXAciVA7d9p+DPxgwy6lNIQQ1wJzsMsdX5ZSbv3BZ3YYCKHZ1Lh119hCukIB4UakPtIq0/WfhGz8vW3UCYGMzu6Bt6gKd+eGxRrry8tQUxWSKqp5qGwKk3v82d4vvA5Zd6ld0UMQ6fewcvZgLDPeYxOWJG1+GdVn9rDtkK7EGCIhBMFwHWuruzI6q7x1WWtJiFgqIcPPsxO/aP3+udtPAkWQWxhk+MRm/E0Kq+al2oRblsRsCkPBSNbelctzI5fw7syvEQmiMRJQVdHK3w7gcFrc9tR+Bk7awcu7jyDH7aNfaotCkEqXHvm4PA1EQiol70boumgb5Rf3w8iOdnkqAtOjEe7mwXnQbx8jGHvw7Lf2crBgKA3jciBoYblUWgPTqugs/I3QBE7FQFNsqoVzBhWweG45e7Is0OONhKYo/Cs0hpffBtiC9OqA5LlzFpJ/STO+BpU9W13sL3KxKUPjdaNnxzYADuetuxSV8Sk51O+pYkOGPbmU+lP449cT+eumcdw8dDVHpFcyLNNOMHq8JqoaHxlLropwc9YgJp19VMz21YcOJjTqCjAyN5/BOTlcPHwk3VNtuugkhwNdUTE7ZP01ReGzXUU8uWYlACf07M2fJh1PdpIdF/5wxzbqg4FWow52Urc22C70IyWmYfDE6hVM69efbimpnD1kKGcNHkLYNHGo6neuYvE6HEwo6M7ykgMxx3ZpGkfkdGFTZUVMOMataVw2MrHkXxzUPMBJrMeOHfZVvrnG/nAQwgmuE37QGN8FP0oRpZTyMyllPyllbynlX36MMb8JQuuBkj0bkfUJIuMdRPZyhPOob97xB0LKAAS/oOOPL60Av/5sJ+tKDxE2TQJGhGq/n2s++4RdNTVIaSHrr496BNFSQemnqfoAZkd+XQAJ540cwUcXXsg1Rx1FksOBN/ovSXfw1GSVbo6F3LTyON7ZNwC/oWFJWFvVlfMXnkJIpgEuTBMaalR+OaKYK/5wiOcWFnHlnw5x418P8sb6rfQf7bdpBFwqA7NzEM6jOPC8heXpPOnYLnJPRk6EM6+qZNFH6TxwQT4Fy4tJbW7mnaez+eeDXXnjiQw+eq7JLv0MKBABR5mf/L9vjRXGkDLGi+8IYVp411QhVRV9RC7S2WaUI1kujHRHAtsuOWNGKbcPW8W9o5ay4pevcsfAe5l9/RIKRHPr8YXfwLO1lpSaBhQimFIgsf+pwqKbp4k8jw9VsQnWRkz0cfqlNVz9i+1xaZDYu9Sh8QsodCdTfd3n7Gio7fBtQbPh4IUdw+mXVtc6h594Vl1imUMpGXdKvMFK78Q71VSVf08/kz8ce1yrUQc4uU+/hPNQxDTZVVONYVkYlsX8vbs54503iERnmK/LShOyS3aG+XvbqqWEEDg17VsZdWk1YjU/h1Xz/9g77wApqqyL/15VdZycIwwMOecMKqKoiKCismbXrKuucV1XXeOu7rpm1pwTmEVXEEkSJI7kzMAAwyQmp45V9b4/euiZnu4h6brqx/kHprvCq+6uW+/de+45l2LW3I30b+GJU8+gQ3wCURYLTosFu6Yxql0Ob06ewrnde2JTVaItVmyqysV9+nHVgMFHNEbhmBII4iFQAuqsreigv3T86jtPhZZz+I1+SpiRl2qba5LZW+8Iy2sG9KLX8MiYLJDhjR39R1WDTKP13WWPtjP8zIH0Sc+gT3oGNw0dzg/FRRhSMjjVwFozmfpok4fXDufhtaN5eO1oDqYZ7KpOl6zTmfnCg7zzRCJet8LQcXXc9e8KbPaD4wv8+8hbu5lwzQlYrVYu7duf/LUFbFi8FVevOKLXVYXw24N7Nk1RbQ6T8b+r5LOXU/B5FKQU7FzvZO77CYw6s4Y1i6LJ3+gMCxxCBpqXnNtqcfVKCL5m2xf4fCK2EpmgNfrp63MyZupgXsxbje+gRooQlF7ZlazntqC49ZCGq/ToRs7vEtpyr4ltzL9wG6/l9WLalz1Ie2cXhlOl8L4ByFaqnYYUaIqJGqGLK9nu5oZuP/Dc1qGEBnGBgonZat4kgfzaatIT1IDufYTAVul14GxRVMzp6uX6h4t48f4sVItECCsSK4/MvBtHVLiuzbUDB/Pw4oUhQdemqkzo0i2iZnyS08kLEyZx8+yvgoHWZxgICHFo0qWkxuNmzq58UqOi0E0Ti6K06b3aEooQh0yhtAVpViErzgazGvCCX0F6ZpMY90/mXHw5eSVFFNXV0TMlla5NefxHTz6VP40aQ1F9PdkxsUelyy7UFEh4A1lzB5iVBPQYOiPinz12Vsz/CL+u0f4SoCQGPEhbqT6WuaNRFSWsKc6QksK6WkLry83I7uRj/IWCeR/Z8DQGVgH2KBs9hnZmyBkDgttZVZUR7QJsItn4BhKTGIufN06YxfVLxiN3erDubUQmWPj7zWmsmQmv/z0Trytwg489twaHPfwmVKIVcm6MYkjHQWTGxDLrh9UIAVq1F1e3OJzba5CqQPGaIEODrtet8OG0NKTZ/KrXo1BeYiU61uDJz3dx7yUd2bQi3J1KmBK11ge6iTAlqe/lI5ry1S3Z5vHJfkafWYs1SvKyaxh6AAAgAElEQVRtQholHQQHXC4sqtIc2AnM2ouv7072U6G2b9UHIq88FAEX5G7j8/c0TD/4HbaAvEIYk0XgNdS20vHkxAa0fSyKweT2O7m48xbcusZVSybgNsKDmSlNPO2j20wdGVIJS/9PuKSK0WfWsnZJNFabhUFnP4k9LrIWygW9+rCntoa31q3Bqqr4DIMxOR14dGzbaYCTOnRk9TU3sGzPcqR3HuvKXLywKZyE4PL7uW/BXAxpIuGIgnrgmmF8p6Nn1sjG18Csorm5J5DCpO5+sJ/CkMxshkTIvMba7MTajs2UXlgHQsqCphSvtU0f2F86jgf2o4QQAuIeblJ9bBIGQ6N3ogufaaG10Kpd0xjdLge0biDiIhRnHNz83PkMPqsDs16dh8/rZ9zFJ3DKJWNCTD9CoXIwi9Yn5gAnvreM7eucGH6B1a7y4ldVWG07g0EdAnnwSJAEDKjfWPcDQ7OySe+QglAEzvx66oalsP+WXthKXKR+sDvCTFpEVKP0eRRWzotj6k0VnDa1mi2rojHN0L2tiskJIw6wYr4P57IK9GQb5VM6IHwGMWsqiK5oZOjJtdw9rTDg16PC5UoZb+8s45nNJokOB9YqH9XOANPAubWG1BnhY0zNisyUAFi7OAarZuDxqliqvBG54Yow6RpXhWESxjU3pWBhcXsEJql2N48NWYwQ4DcFF+Ru5e2dfcM/MRO0Wh9alQc92d5q1i45u30kQ4xACujESU3+ovq/kUZXZMNT4Pk20ADjOBcRfRNC2Ll71AncMHgou6uryYiOIS360PpGADZ9CSdF3wrRPtyNOagiNUTDJjA6qPd523omhUAVAquqYkrJP04ZH2JAc8Q4qAcfBj/ou8FybI1Sh4MQArQfScn+H+N4YD8GCNuJkDQjMKPQC8A6kPSUq7mw91Y+2rwxuAy2KArxdjtTe/cN/FgSpiGrLifgIu0DLGAbiXBOYdTZKqPOHnrI8x5ESckgFr6RiLtBUl+rsn2NA18Td9zdYOJprA9rN1/4eQJ9RzTgiAp9XRWSvAPpeKTOX7+bz9cXXkJCWjw+9wFiV5YTk1eBP9nWNuUgIiRCBGRqx55TwztPpFNRYgna0tkcBoNOqidr5wG2/ieNsks74+qdECgQS6gbm8GVCau546TCVvx/g0u7bGZecQe21bbjEmc8S/62CG+9t41Wf9i3y94mAcYwmouuis8kfn4xNSdntsjfS2yKQaLdg89UcbRqxfUYGssP5JDpbOCtE2cFz1FQH8/yskwidbrarRYS97ipK/egp4Tmw2MsPm7vs+rwDWb+PGTFmSDrAD1wmsa3kb41kPg+QghibfYjb8qRJrLuXg7WfqQQGG34wkb6lO2ahiIEumFiURVSo6K5qHdfom02TunYiaRjFeRS4iLLK0kDlOM+qYfC8cB+jBCWnoj4p0Je++sJ6fROTePNdWuo93o5JbcTNw4eRmxTnk9Y+kDK4sAsy6wMmG5Y+h0VM2D+B0t46pqXMPV0dD2QC5WtGlykDATWljfhoi/jGTelmt7DAsHd7xcYOvzrtmzSFm6g4rwO7OsHF3zyIa/O/ytPX/kiG5dsBaBjSirFXSTepYVhM2JVk5gmIekYgB3rHVx/clf+9dkuLr2jhOnPp1FbYcEZY9B9kJs9W20smx2Pu1tcIKi3KIZKBBVxCeh+0Sqwg03ROTtnJw+sSae4/RqGDi/nh++i8PgjBSLBnA+SWPt9LCW7rMQmGlxw4wGmXF+OEDB4bB2GkR3cOmFOEWqNj7pTM9FSVPqnlDGh3W4eWDMGm2Jw34BlQc55lMXKLvkQL5/0Jf1iFweDcbXXxtQFk6n3W4lULTi3Tx/Sfx/Pe+vW4u4cCy246BOydxFrPaR8ZBNMvHodVkVv8RDwgn8z0jMf6c8D/yawdA/oJx2OSmjsD6kdfVrQjbbU+OOsHq7utp7hKcXsaYjjjR19US09uXPkGHZVV9E5MZFR7XJQfgLNFuG8IiAdEiIfoIGl9zGnSAzTZENZKSaSfmkZQTe13xqOB/afEEIIpvToxZRDaEELJRqc5x7T8RtrG3nqmpfwuQ8uT8P9Nw9CtWgoih9fk9+naQgevTaHR9/bTUOdSv5mJ/M+TKB0nw0NP6nv7aI4zsoezcLS+jL+OfevNNa50H06+3eUcPNfXyZSV70R1EZpKYQl8HlUigpsPH1nNgNPrOeFb3egavDhtFzef6pZeqChX2LEDk2py6Y2/lAIQOgmGgY5sbXc9upu1n8fzYsPZLFna3heVfcLivMDD9baSo13/pVGQ53KFXeXYosTxF0Wy4E3GhFSgglxaysYllbMhHtdfFOUywNrxmBIhRm7ezJ7fy4jU4swsTJt8mP4527j65csfFzfibHnVHHS5Go+K+jWFPwjB7Yvd2xj7dN/YEr52Zz/xUeU+9x4m2oF22qzgoGm0a+xqjwTi2IwNKUEqxqaStNNOHPOVF4ePYdOsQeppQbU3kYgHegH/1qk+5PALN5yCH1yJZqWU+NIDVOBz17yzyELGZ2+H5tq0iexnNOzC1hcm8OYjBrGJG9pUkHN5icJLfYzwL8VXG8S8F3VQeuAiH/+mA73Q0kR1//nSzx64IGoCoVpEyYyqt3PTMD4GXA8sP+KsGbeRlTtyNgFmblp3PzYNt78m8LeHTYycnxc/qdSMnJ8XDGyR0jbP4Dwm8QvKKasQwwLCnZxfs/e1Fc18KdTHqbmQC2az3+YXpzwQKb7A7n2+17di6pCTYXGpy+Fph9EG8YVi/Znow4JP6HXrZB3n5/ojAP87sytCAH9Rzcw8rQG9m63ha0cwvdX+eyVFPpcavDv/MGs6plN7DkHGFsdTdcxtfQdn0dqopuV5ZnkVaSHpCTq6i3MdnXk4v4DeP2eT/jqxTlNBe9oNq10MO/jBMQdcWG+qS1R7/Oimwap8bv55qIuvL9NYebOQmKsVi7vNwmLo5qvdu7jL6tHoIqAjZ4qTF4ePYfBKc0WfwUNcextiOXi785iycT3A/K36ITWePxNqokPIpI+Dr4aWNG16IlQEpHWIU0iXIEV0brKtBDDEQCnZjIirQhb00NGUySaojM+6Xlk1YuBcwsNRCwkzUCoP64LXAiBiL0DGf178G8GJYViTyYVFS66JvpxWI5cB6je6+WKLz5tIcQWwLVffcGiK675xem3/1gcD+y/Iiiq0nb+VRVgSGxOG5pF5Z4P/khu7nSeHvY+Ll1ywO0kzeFm7wYrFiv4W/dgAJZyD4oQQVf5+yc9TtmeA5imDG5ztJBmU0lBhXmfxIcVUWNXl1M/IjWkPR6gzm/nqbvaccc/9iGUgOaO3yf49sMENi+LIlndg+8mPzR52Q4YU8MnLyYR3lUePmq3oXHxzAnoSXZsqs7Zv6/l/oE/gCzh4My1XXQdE9rtYtK351GxGVJn7MZS7gYh0CfofDF3I35Pc5DwuFS25EXRdW49iQ2l1HZOxIgJVxnNiYtCrRyLlA3YEVyVrXNVz3tRnFMB2FPzIPesfotW/VlcteQMlk96F6em49ZV/rVhGBIFj66xtDSbsZlFRE5IA/4NgaY4fRuy7uHATF44KDfPYF75OWTGJjMm+wmUmutB38Fp2eV8XFBJo27jxh55DE0pptrnwGc6Q6iYzZ+wn6DSmfSCdCNr7kAkfRB5PEcJoSRSK4dw41dfsqbkGyyqimFK7hgxiisHHFnz0Te7dkZc3ZoS/rNjG1f0H/iTjPWXguOB/RcMKSX+Jr6wEIKBp/aNyEIxrQo147PQfJJTB/fmnrt+R1RcFIZxHY8v38f7+R1RhImUgsuyNqP7/YQ59Kjg7hSLVVW5qE8/CrcXUbK7LBjUQ7Zt9XfbAV/Sc2gjB+nTlaWWoFzvQdj2u0ictZ+qM9vRJDKONcpCt9nrWfRdHFuWdeeEs2qwOUxWzosNKktKQ/LtjESu+WsJ7/4rjU9eTEXXD+aKRHCcbY3NEi9QFZ2HBi7hnI47wi7KokiiND83567i5ZtiWrhaSdbOXo+MQPXzuhW2vGwSr+4lzthL9enZ1JzSbNBi1zTu7zcn4MbVEnV/Q1r6ICw9+XzrNnSp0JpdJRF8W9SBnvGV/GP9cJaWBfLmhhRUeKICMhv+tQT0ylvDhjSKoeri5l4K6SLG+IJU/QdumX0W8XY7H533BjN2LOK1dflkOuv5ZNyHOFUdVZEk2T1ATYRjR4IJ/vVIsw7xExU5b5r9FXnFRfhNM8ivf3L5UjomJDC2w+H14Ws87mBzVUt4DZ0qdxsyAr9iHA/sv0BIKXl3wzqeW7WcanfAFPiuEaOZ0rM39390Ow+d9yQmEq/PDwJqx6RT3RRAPlXrudUiWDXje57989vUl9aSlO6h8qz2eLrE8U7xICZfVs3G97bjcTU5LAmQVhXZMY4RH5by0vvP0H1YlzaLuoLAMjkgjRC+jWYxGT6+jpyuHiZcUhl8vc/wRj5/VYYVe+O+K8GxqYrqCe1o7B2PUdJI0oFqeoz14WlU+PzVlLCZPgj2brfy8QspfPJiCt7gA0MgBfjS7NSOSSf5i30hTVamVaHd2SpXD1vBuMy9JNvD1Q+D16FIRibv5xW9W8jrht628qWhi6B/bMKc/fg7RyM7OemdmsRtg9sxxPlWBGqJD+magYh7mDqfN6RV/iD8po2/5I0P03UxsTOkx8soCemY9U+iN7yJJpopgh5DZVnFQMY63gLpDfm27JrBqLRCEqwVFDfEcfFnH1PW2IBbN7miyw/Ym4J6M45WF+fIO1MPhbKGhmBQbwm3rvPneXPonJjE8Ox2nN6pK7kJCYF+klYYnt0eLUJDldNiYeTxHPtx/Bx4d8M6/vH94iBt8kBjI3/9bj5WVeOs0wcwvfAl7vj7m6wq2Ie7Rxz+FrQ5VRG8+NjfWfnUdrzuANvdvq+RjFe2U3Z9V/516UpGn1XEwt6JPJU3gJLcNCw2C71Wu3B8ms8+VyCXkb9uT4sibShMAa5ecVgqvFhL3U0l3EDIiEvy8+x/8olL1LFHmShKs95W/9H1XHJHKV+8lkJ9TfNPTwB6ioPGgYHuwXOH7OCBa3eiewJvel0K916cy+4tLfPzkjVL4shbFActHhRGlMb+23pjRFuQdhU92UHy53uwlLoxozSqx2Uy+rIKpuZuO6Lvoq7CjohQBpCKwN05FtOm4NhRi+oN30jVTS4vXc7tj1yJEnUh0jMHWRvxE21qxIGxHXL5eMsmXK1ywYoQtI+Lp7C2Fk+TrotD05jUtQe5CQGGSIPlOr4vmseJGXvwGQpWxeT7sizuzuvL1wmLSbOGB1qfoZIbU8P+xlj21DbPyAcll4YIuTWjafkl7CB9IOJBVhEmOq/lIpS2Da+PBlUeN5qihnTCHkS5y0W5y8Xy/YU8vWIZsVYbd48+gQt7h/YQ9ElN45TczszbvSsoT+zQLAzLymZYVnbYcX/tOB7Yf2GQUvLcquVhOhxuXeepFd9zVrfuxCREkz2xNwvW+2gdToThYtW0bXjdoTNcxW/Sec5ORt+1H5ti8HLsaEoHxwfMp+t8VHy5BdFC+bHNoK4Jah7sQb3FjiFVUj7cTfS6ymAb/w0PF5OS6SNC9zrOaMn5N1dw6kU13Di2Cw21zT8/tdJD3PxiOg70cv+5y7BpJramFLUjyuTvM3Zx0YBeQWVFIOT/B1F5Zjv0eCs06c64u8VR+Od+waeLVdFJtReG7LNjvQNVk7Tv6qFlPc6la0xf2B2LzY/f2/x9eLOjKL6+O2iBR5pzSw2p7+8KMyeRUuDST0WJujDwgmVwgNkR/skg7AG1xdHtcxie1Y4VRYXB4O7QLFzatz9/HDaC9zas48sd27BrGhf36c/kbs02jiuLSrk7bwJxWiWdYqspqI+nsDEWkCzcb+fcHCWMXePU/LSPrsWiGPjN5jrHvoZYOsdUR7AkFJD0OcKsAjUdlHhk5flgHmhqvrOD0BBx/4xwnceG3PiEI962zufl0cULSXI4GN+pS8h7T582gf/s2MbHWzZhSMl5PXoxqVuPX5yd3k+B44H9FwbdNCOaIwAU19cF/392t+7M2LQhOHsLwm2guyJnl/UiP05NZ15RDnsaYvGaga/fXtCAqSqoTSkGb3YU5efk4M2JRvEYxH9XQvz8YnoPbeDuN4pIiN2Az1R55ru+LNpohGizjDyjNiSot24OsllM4pJ1Tr/fxSd3xuJLtVN+XseAsYghOb/jPCyt8v+KAlabpP/oBtYsOihPEPlmbOyXGAzqIWgahCIkZ3fYGRzbB8+k8uG0NFTV4K9v7KXfyMYQTvqy5CxUbS/CH+iylYqg5IYemFHNt467e3zE4VidVk48v1k8SqhJyOiboOEFgpaOwgFalwC1j8DM/OWJk5mzaydfbt+GTdM4v2dvRrcPpAuuGTSEawYNiXjtB52NilwxFLmaZRwE8PLWnkxstxFNMYPCZVIG0k139lnFJZ23cO78C3E1pa1e3tafUWlFOJSWvy8b2E5EsYQGTJL/A545gQYpLQfhOBuhBITGDNNEESIseJpSHjHX3aZp3DvmRB5ppYHTFty6znMrl4cFdkUIJnXrwaRuxya/+2vC8cD+C4NFVUmNiqassSHsvZz4eHweHy/e/jbfvrWQqFEp+M5oh8WqNd0kHp4bu4BnbPFEEoxMzAwEzPVVqbj0ZsaGEaUFGfG+VDtFN/cMNgyZ0QrVp2aROlDyt+u+xeEI3PhW1WRIXQHLLO0w9EA2xNUzPsxCzm+GzxLtmsmw8fVMj0um6NbeAQleRYAiiM4ErY1fZXTsoW9qd8fogNtRJEhJjMXLsyPnk+EMFBDLCq1Mfy4Nv1fBESXp1CvQeXkw3qQ7XEyfPItTqs9l4lobm7/bgrdPEprTgtHiGk2nRsU5OSR/sQ+hmwgJpk2hw5BcRp8b2k2sRF+HtA5Auj4Asw5hnwCOSSE+vaqiMKFLNyZ0Cc3tHw7Ds9shIjxh7JpGg5HMBQsm88CA7xmaUhJyndEWnXZKPU+PqeWPixPw6DprK9O5bcXJPDxoKUl2PaC2YD8dEfdw2PGFsILjrBDnsrziIh74bj7bKsqxaxYu6tOXu0aOYUHBLh5bupjCuloSHQ5uGjKcy/sNOOys+Xe9+5ITF88ra/LYX1dLQXVV2Gq1JYobIhWR///geGD/BeJPo8Zw74K5IbrSdk3j7lEn8Nglz7Fq1lp8Hj/x84uJzivH3SMBTcK/H1xJn+w69t/g48NpqXjdLbo5rQo9rsrAZBdZzgYcqj/IU/bkxmA4NITPR824TKTWqpPVprI7Kwu3YsVxUHIYSM/wYlEMdFWj+MYeeLOimFNSxBnZu4PBvC2KWYXHie2apMC5WjQizS3J5aSsQqIsoUHc7pRs/iGypyoAFqgbnRZZO8CUJDfU8u0FnxAb27waWL0gJrj5iZNrsFjNkNSDqkjsqs6kIXs586oHGe1x8/yqFfgqw82Y60el4+0QQ8yKA6huA3VIBs88ey9q0yxaSj8QkKoV1qEIa2jAl9ILRjEoKYEmtmOATdN49ayzuerLzwGJKQOiY1cNGIxVVXn5By+ySX209cdkUw3Gpc5j+pRPeGbF92yvqMCtnsh+292kpdpAiQnaOB4OOysrufyLT4Kza7fu5/2N69l0oIz1ZaXB33WV280Ty5bgNXSuHTgE/KvBtzogsmefEMaoGdGufVAI73effMgPJUVtGnv3Sf11inf9VDge2H+BOKd7T6yKypPLl1LcUE9OXDx/GjWGvpZ4npq1Bl8L/rRW6ydmxQEAvnkjjs5/r+fCPx5AVeGjf6ficSvEJQmueeJGxl08All1JadlbuGfG4bhNppSNoqg5MYepL22HXfHmHC1K8CqGuxtiCXR1hzY+wxvJD5Jp7R9Jt52UUiryt/WjaRf4gGS7W4cqp9SVxSZUQ0hhTiPofH+7p6k9FfYWRHKX/9mf0cu7bSRnrEVOOxmU3ASfPxiCpXFkTsiNauJOzmKhkhWglKiVXuJe2Eny2NiOW1qdYv9mjXT07J9YTo6ADZVZ0L2Lj7ctJZ5BXtxRVoKNcGXFUXllI7EWK28e875WDQV0/0V1D8RoDiKeGT0jQjnZcEZqpQS2fgyNL7UNF4D6TgHEXs/QhyFEXcThmZls/Lq61lYsJsGn5fROR3IionFbxjY9bn0SSo/hBaNRr+0dN6cPOWoz9sSL/+wKoy949F1VhXtD3vQu3WdV/KWc3XHlxD+NQH5AGGH+n9AwpsIa/+I53h+wkSunPkZ2yvKw6SyHZrGXSNH/6hr+LVDyDaeeP9NDB48WObl5f3s5/21Y+OSrdw/6XEaa9vi3UpueKSYCRdX4vcJ7NF2/P4s7FlvoGgBBxgpTVZ89hyrNs/lPftgqrRopCoCU7iWv4VWd79V0Vky8X2S7B6khK0/OFn2TRw1jRamNwzAvqMW4TdpGJyMZ2ACY9sV0sVZyf7Fkssv3kfX+Gr8poKmSB5bN5xv93Xg6u7reWLtMMxWTkZWv49LjA1cfn4dabZ8DFMyqX1fIiWyJSAmJlI4pgM+W3hDEIakw19Wo3pNhBVSbo5j8rlldCwr5p/XZVNTbgEEw8fXcve0fTijW1PqVFRhUuaO4fwFk6nwtN2hqABDsrJ5c/K52DUL0vMtsuZOaLHKAQfE3IpfuYg375/BN6/Pwef20n90Azc8UkRWRx8BY+cLUWLvafNcxwKz+mbwzmnjXQtEXY8Sc/OPPs/E6e+ypfzAEW9/caetPDhoFSLkcwKUNETKIoRoW89la0U5c/J38O2ufMoaG+mTmsadI0fTO/XHOR79UiGE+EFKeVjnkOMz9qNESX09L+atZFXRfrJiY7lu0FCG/kx0qXbdM/F5DyUSJXjx/iw++ncq3Qa4uffDv+OIGhjaPi4U2vW5gEcvXUW8ZwvRsRb2/qU/2NXIaQzArvg5o93upiaVgP7Kd1/EMfPNFAAS1b1BL1XH7nrceRXMvbY78/T2xGwr46tZA2if2ECi0sj26iS8HoX2r22h+sQ6lO4GpiKCqwThN7GUebn2jucoqTqfLIdEUyAu0aC2Kvzneu5tlVx+2yauWRLFivJMWgd/rdobpCJKH+x/z8dDOaOQ+71k1GxBacrUrpwXS8leK+06ebE2mZEYEuyqgRCQ7qzn4YFLuHHZaW1++v0zMoJBHUDWPw2tgxVuaHiBB68oYsOiLU2rL4W8hTHcMqELry/dRnySB1zTkTF3tWnwIKVk5vZtvLNhLY0+HxO6dOXK/oNCjCVm7dzBtFXLKW1soG9qOnf2i6Onow2ZTi0XEX1tm9d2NOiTmsb2ivKwNElbAqHn524PD+oQaLbSd4Cle/h7TeiRnEKP5BRuHd7snlbn9fDuhnXsr6tlQHom4zrmHpPRx68Zv01ps0NAN01mbt/KVV9+xh9mfcWSfXuOeN+iujomfPA2MzZvZEdVJQv3FPD7mZ/yxbYt/70BN6HS5WJeeSHZt4xCSW7OdSam+TnlvCpGnVGDtclIo7LUQmXFQKzRgyIWpbK7ZjLh6nHYo2xodX4UXxsNN1JiF34u7bKJvw9ZHHzZYpVcdNuBJvc3EWKQrfhMHLvrcWyvRVg13OOycagayl/2UvSCi7jXC8h5YA3KXg/z34ln7PyVnJBaiCYNrKafsY58ltzwJcn2F+gZFyjyLZoZ1+RiFxoW0rJ9/P6PxTg0g7/0X45T01EOks5NifAZpHxSELKPVuXFbVrwJjuoG5HSfKmm4Pbzu7A73xFcuBxcyECgE/XkzL1hYziIKIuF6wYOZem+vfx53hweW7KI/OrwAjjAnm1+Ni7eGpJSk1Lg8yrMeu9gOklvpWoYige+m8+9C75lXWkJO6sqeSlvFWd/+D7uJorkO+vXctfc2WyrrKDG42Hxvj1MnZ3E1ppIuedoSPw04Mv5E+DaQUOwtaqA2zWNE3I6Ym/1ukPTSIsKN2I5VmwtP8CYN1/j8aWLeHVNHnfNnc2kGe/R4ItM3/2t4v/VjN2Ukqu+/Iy84uJgk8J3ewq4rF9/7h51wmH3f3blMhp8vpCZiFvXeXjRQiZ27X5EEqBSSrau2MHONQWkd0hh8Gn9Dyvs9dnWzdy7YC6qUJDtwbi1F9n353HhzWVcdGsZhi4CwUjCfZfmsndHCne8dv0hj3njs1cy+LT+zHptPvsV2dq+FwjMnj89/XO6JVeHvRcda6BZJLo/gjmFzyRhWx0vP3Qz6dHRrPtyDc9516OsqQzdUED7GA/3jPsm7BjS/Ql+aWPZf2w8eXv7gF9q87uAYNiptRx0ku6ZUMkXp3zKv7cOZENVKvYKN3XTKrAXhtoR+pMDCpDSqtLYP5n4xc3t/a4Gld8/fwLLXgofDwSokgIZUdLWlPBi3ip2VFXg8vtRheBd5SweGriYKR1DzTP27UyM+J37PAo71h1M9ZjIxg8g+qqwWXtRXR0fb9kU0rDjNQxKG+r5YtsWzuvZmyeXLw2jBnp0k6e3nssrI14NmHMAoCESXkMoEdJYx4iO8Ql8OGUqjyz+jrWlxcTYbFzRbwA3DB7Gkn17eWzpIgpqqklxRnHLsBGkJKVC/eMEKaAHIWIDBjVHgVvnzKK+hWBQo99PQU01L6xewZ+O4B7/reD/VWBfuGc3P5Q0B3UIVOzfWreGi/v0Izv2EKwLYNn+fRGr8D7DYH9dLR0O00jh8/j4y4S/s311PqYpUTUV/6AUzMt7UepupGtSMneNHBOS2impr+feBXObbuKmG9mh0GOYiwtvKWvyMG0e06PvFqDHvU1CWgqHghCCYWcOYtiZgyj5+BPm7S8IK5oKE5LamMU11qsRgzqAVODEXl3pnpTE3LcXMf3xz3E3hi+1BeCMQACp8dpYWZ5B6Y4MFj5W3SqoH9wzoB7Z8uvIja3lyWEL8ZuC1+d059PSUBlf3alR/ruOwb8Vdyv6pBC002rR/WCJEOdqfQ4cmhW/aeI3TRQEVpdmk3AAACAASURBVFXBZ5p4dT/ryprtEg0pMQyVB9aM4bTsAqItTb854SCrz1QMY3HY8S1Wk9yeB4ObCY3/RuqbEAmhMrVrS4vRFJUorYFxmXuRwILiHKq8Dpbs28sJOR0jyhJIYGOlDZGyCHzLQUSBbVQI1fKnQq/UNGacNzXs9ZM6dOSkDh1DXpOyB9I7P8CKkV7ABkJFxD9/VM1D5Y2N7K0N17PxGQZf7th2PLAfKYQQ5wMPAj2AoVLKX3RFdEHB7rBWbQBFKCwr3McFvfoccv9kZxTF9eH8WF2axNsP77E4/fHP2bpiR3AJXj8oifIJacjawIz4h5Jirpj5KW9NnhIM7rPzd2DoZmjqWFUY90d3mAkFgM1pITp+C9tWZTLnzYV4XF5OPH8kQycMaNNq755Tx7H47dfRWxgw21U/13ddzw+LRnDiGTOxO5oDhdul8P7T4QbcQZjwZVoDJTc9Q/l7a4Nerq1hdVjpPkRgmgRphq9t68PTm4aiKSYut0rHfXltqs5/PzuW6x8qCnvdMBU+pS/l5/tJ/KYQrdZPzag0qibnNNvfSYk9v7nhSyqgx1tJ3F6JoUcO7AlxZ/HBlKnsqq7k5bzV7KyqxNNk/NwWBUFTrKyqGszJaXmgZkDULXQZNomug4rYtioff7BmIrHYJGde1nJV4wHvd0h9F0LrFHw1xRnF5Pab+Uv/RRhNGjoPDlzKA3knEBczkESHo02J5cyYGISaBI6JbYz454cQGiS8Cv488OWBkgz2M46a9hnQL4r8nnqIAuxvET/2ajcB5wLh04+fGVLqSM+3mLWPYja8ijTCucbxdnvEdIkixBG5mV83aAiOVjlCq6pyUk5H4u2H5/jOeXNhMKhLoHJSTphcrUfXeXzpouDfW/Ly0SNoZCgdbCgRMjiaRWPpZ4u58+QH+frVecx7dzF/u/BpHrngKdpiQHWIi+friUs5o91u0hwN9Ess45nh8/nD8E0U7vTy2MOd2d8QuMnKPQ6e/awnn7+d0mYwM60KjXEa36d4IgZ1RRHYHFY0q8bnr2h43QqmCavKM3h28xC8pkajbkVaVPyJbX8vtZUWnrq9HV6PwOVRcfk1PIbKPzcMo6A+noahKRT+uR+V4zOpOicn1NNUCKpPy8a0Nv0eTLCWeyjY4qBgqzNM1hgciOjriTNV/rpgPjuqKoPXf2hemYFd8YClPyLmXhTnJAD+9vU9jLt4DBabBaEIeg2z8vTMfJLSWq8itIDZRAsMSVe4t/9i7KpBlEUnyqJjVw0eGryYS3tn4LBYOL9nr7B8NkBhbS0lrSYn/wtmXGsE+P1DENE3IJznHxOXP9nppEdyclhHq13VOL/nIYxGfoP4UTN2KeVW4H+utSClG1l5MRi7m/QqbMjGf0PCawhrMzNoSo9evLluTdgyVRGCsa2Wh5FwRueu7K2p4blVywNKcYbByHbt+df4M45onLq/OUBLm4IRHfnj315ZyZqSYmZu38qGaYsQl3ZA6iaOnXUoPgN35zhm7enIhJxilFZ5SWn6ef7WjXhdLXjjjV7y5qwn79v1DDktAi/YLKVj1C6eHRFeBD7prC3cdnYKJ3cbg3SqgfyyFbR7PCR9WkDUltqQebsEXH0CKSl/QuSg3GNEVzr168CcNxdSsNnObZM6c9ldpbxr643HaPG0MiX+JBuWck+b8rvfzUxg7dJoet9rw98lmgXFOZS5WxgnS/BnOAO5eC38KI29E4hZUxly/Iev6sCfX9hL3xGNAaqd2gmv5X7+OXUGsxqLcJ2aCRFcnyJBE16GJq0Dv0TWbELG3IYSdQWOaAd3vHYDt796fSCwNj4LjRsiHEGCmhnyivB+gzVCr4FFUWlvWwl05/4TxrKutIRNrWiHtV4Pd86dzfvnXsDWinIeWDifNaXF2JukC/486oSwwuf/CnVeLx9t3khecRGdEhO5qE8/smIOLQP87OkTueCTGbj8PnxNHqx9U9O5ZmBkGYbfKn62b1AIcS1wLUD79j+tA7hsfAf0fJqpZV6QIGtug5TFwQdPbkIij48bzz3zvw1Ie8pAC//rk84JUtQOh+sHD+XSvv3ZXVNNqjPqiBzgD+KE84bzn5fnYvgNhM9E+EykI1IzkMqln3+MR9fpsL2SpC8E0esqg4VCYUg2do9Djh0CMq/pYaYCFrZunoLbtTPwGbSAp9HD95+vjBzYUWhr3tm5D1xw+2Re+vZ7yia3C2wnBHqSnQNXdyf++zISPt+DMAMCYdKqUnVm4Pu11IYzETSryq51e9i2Kh+j6UFXsNXBQ1d2pPiGJGS35s8jdlkZjl31EYN6cLasQk21lf15KqcNLmVMWiGz9+cyuzAXXaqB/iuPEdLd2vKyTUfgFrA6rGR1Sicq3skpl4yi94T+CGssAgOhxPC3sx5jzbyNuM9rH9HKL+w6hcSh+XhtzGy0YHOWGxqeRjqnBrs4RZOOinReiHS9HeBkNh8lYDVnGdDq4nVEhIZ6VUgOqixaVJWyxsawbQwpWV1cxK6qSi74eAaN/sD5XH4/MzZtoLC2ltcmnXPY6/tv40BjA5Omv0edz4tH17HsUXl7/VreOfs8BmZkhm2vmyarivbj8vuYddGlrCouori+nn5p6QzKyPyfTz5/bhw2sAsh5gGROFL3SilnHumJpJSvAK9AoEHpiEd4JHB/RThfGJB1YOwCrXPwpUndenBKbmdWF+3HpmkMzsw6akPbKKuVPsfQAHH+HWfx9SvzABAS4hYWUzsuK8TI2aZquPy+oG60N9NB3LLwZg/n5hp273uMzr0KmfvWp3z5aj1ul5MOvaIi8tEVVcER42DNvA28ePvb7Nu6n/iUWH53zzmcfdMZoHUCfRuhAd4OjvO4/MELWPWpxjf7C1oq5CIVQeOJmQy1JbM5byeNHaOpG52OEWPBrmmkLtgXMgapQG2nGIwYC84t1Wityx1+I0Q1LG5RaYiWekuIpuN5cmO47NRdXHp7KRa7RFVgVNp+puZu5Ypvz8C2uZaYvEoa+yYh7a1zVwLH9oCOrt1pZdrqx7Hawh/wlSXVrJ2/Eb/Xj1rnA8OM2J0LgUekRVX5y4DNnJezFJvaOo2mgr4LLL1DR6KmQ8JbyNo/g9GkPmkdgYj/Z3hQsp0MDf8mXO9cAdvY4F+GbFtN5d0N6/C1EpDzGgbfF+5jT031YYkAh4L0b0a6poNZibCdCo6JR12gfXL591S5XcGuUr9p4DcN7p43h7mX/j5k280Hyrhi5qd4dQME+A2De8ecxFVH6K70W8RhA7uU8pSfYyA/Cm00cQQqKeE3qtNi4cQjSL381Pj6lXkh9cbEucUIIag5ORNpU4m12RiR1Y4Fe3YHt2nsm4R9T2PYrFUg+OaNBRh+k4Uzappy2S7K9tai+8LFsjSrRsc+7fnr5H/gbZLkrSqt4Y17PsBV5+aiu59FVl0I0tOks60F8sJRVwCQ765DRpj02DSN65+8kO0V5Ty94nuMxkYyomO4a+Roeo90cN+Zj1FXVY+e4mDvtV0xrQpSgJjYjpxH1wVWLoDhUJtUEptPorT2h2sNEzKqqrjsztImdlAAURadPvHlTClbzep3AiszR34d7i6xwYeo8BrELD9ArFsSk53EI1/9OWJQB6gqqUazapSclEbtmPTA7L/FA8iiKDgsFhLsDoZmZfOHIcPINm6HSP0B0h8oDkaAsPZHpHyDNKsAa5t5ZmHpgoy6AhrfAg7O8K0QdVVIkXVC5658uHljiLmEAHomp5JfVRVmOgFgVRV2Vx97YDddH0PdI03jMpG+ZeB6D5KmHxFP/qAf6/yCXWFSAQD7amupcrtIdARooX7D4PKZn1LVShH1saWL6J+e8ZvtQD0cfhnJtB8Lx9QIPFgRyE2qP23a58dgwfSl6C10vYWExDlFxC0spvi+gVx16khi7Ta+27uHg9ZovlQ79YOSsVR5sRc0pyWklFQUVbNm7vqQRhe/14/VbkFKicUW+Ff3G9zw1OXMeXNhMKgfhMflZcY/vuCCuyahpSwC73wwSsHSD7/sw6ov1lBRVEVavI18whM2PsMgKyaWfmnpXNCrT1COta6qnntOe5TGOheqprL7ko7o0VowJSJtKkW39CLjxa2ojTqNvRPCDu7qFkfMDxUhssAtIYC+IxrR/SIksANEWXUuGbuLjS/k4nWrpL++ncb+SdQPSkLoktgVB7j2d+MZce8gugzMDZsVe3UdQ0qcFgvZ3TJxxVuoPTkDWhW7kTC5Ww8eGXtKMDctjQpkVajmewBWsA4JzM4PgSMxqFBibkfaxyPdXwMC4ZiIsPQM2eb2EaNYum8vB1yNuPx+HJoFq6rwr/Gn88mWzawuLsJvhj58fIZBp4RjM8iQZiPUPUrI6lm6Qd+FdH2OiPpdm/su2beHRxYvJL+qini7PaIlY9MBsarNYWvF/kJ8ERytvIbB9E0b+NvJpx7Ttfza8WPpjucAzwMpwNdCiHVSyrZ7rv9LEM4LkL7vwbuYQA5YA2yIhGm/qNyaZonciCTMQEzTpckZnbvy+NIlwffcvRICWuUItDofmf/eilbrwx5lI7NTGuu/08ATmtPwefyMPncop156Ej6Pj4Gn9CU2KYa3H/gw4vmlaVJzoI6U7CSwn0FVaTWzX1/Ah088g4FE+gy82VEoN3TDaMEssasap3XqHOLwfpCR8K8rX2D3hr3ofgN/ki3AbmmV5/ZlR1F4d19S38un+uTwgmT1me2I2lqD8Boh3a0t4W6InBKRErr0cfFe3lbuPKcze3fYiV5bSfTaAJ3QEW2jx5AudB3UKWS/CpeLm2Z9SV5xMRJJ+7h4XphwFl1vHs0uEc60UhVBTnxCi6Beiqy+rjmd0hJaZ0T8sxHHeywQlt6IVimdloi3O5h98eXM2bWTjQfKyImLZ1K3HsTabFzebwAfbFqPv8WqwqZqjMnJISc+/tgG5F8HQo1QrnGDZxa0Edh/KCniuv/MDKo+1ng8aIqCKkRI34imKIxun0O0tTmtU+/ztqEgKqn1REjP/j/Bj2XFfA58/hON5ZghhIpImIb0bwHfGlBTwDb2v9J48WNw+pXjeOehj0Lc7aUAX7oTNdHBaZ26kBoVzb9OPZ3b5nwdWIoKgbQHvia/xUbp77vQ+dV8eozoyrCJA5n9+vyw82gWlYzcdEZODmUCtOuWRc2BurDthRDEpQTYBrPfmM+0m16nYkACVXf2wIi2oDb6SZhVSNbb+bh+35NqRceiqlzQszf3jD4x7Hgel5fV36wLsoBMi9LamzkIM8ZC6TXdIzJW9AQbhXf3JeHrfbTf7SUtJ4Wa8jqKa+uoGp+Fu3s8JR4fd8lCWp9ACNAsEB1n8JeX9nDdyaF6I36fQaf+HULHIiUTPnibClezyNre2homzXiPP44ZwbwVy/G3CiOqULCqCtIoR9bcAv6NNKdHWkFEIZQYDMMgb8569m3ZT7vuWQw5o39Q3venhk3TIppLZMTE8PH5F/Lgd/PJKy7CoVmY2rsPd474EaqIIpo2yZ9K281/z6xYFiJRDYFiqCoEVkXFqqqYSHLi4nni1NNDthuW1Q7dDJ+xOy0WTu/cJez1/y/4baRimiAsPaHVcvSXhCm3ncm6BRtZt3Qrul/HVBWkVaHmqu5c0qc/PVNSATizazemrV7B9ta636qC3j6Gq169mrPOG4NQBNFJMbgbvbTsSFEtKhOvC1+CXv7wVO6d8PewdIzfq/P2gx8yYGwfnv/Da1T2i6fi7JxgPtqIsVJ5dgfEp3vo88punl/3BDZVjWgaDIEO25awlrlR/AZGWPGSQAS2tL2qMmKtVJzbkftj+nPW9eMpdzUyetoL+DVAU9CxcfmSibx1wiziovworXw6FQUyOvhIzvBRURJ40FttgvvfsZGgTcUst4BzKsJ5EQsL9oQE9eAYpGTzgQMoqgKtegqEgAmduiKrL2liZh2iLmBWU1dZz61j7qeiqBKfJ5A2S0iL59nvHyU+JRD8Nh4o4+W8VRTUVDMkM4trBw3BqmqY0iQ1KjTvXtbQwHOrlrNobwFxNjtXDxjM2d2PzO6tW1Iy06eEd4ceMyx9AjIA0kVogHcgnBe1udvOqsqIr9s0jZcnTqbG7SE7Npa+aelh15XkdHLr8JE8t3I5Hl0PWBVqFnqlpDI+x4ZZ9zgY+8E6ssnZqW11zt8SflOB/ZcOi9XCY9/cx7ZV+SxasJZ8XCQOb8+ZPXvSLy0079p6BnMQiqbQ97S+qJqKR/ez74ZuKNNcWEpcoAikTcW8cSAZueFFo34n9uKBT+/k8cuep66iuUnF0A0++sdMPvrnzEAmy2MEpHxbQNpUqiZko79bgtNyaGpobGIMGblpFG4LdIUKCanv5lN6VTewqMijbIvTdEnCKZ0wTJO31q0BpwYtCn+bG9MYPvsyFk+cTpoznOInEIGgLCA2KYpXFu4lIakCjCZKaP2/kL4VLNp7bptj2FR+gMdPOY0/z5sTfKAZpskjY08h01mErNrHIYM6VrCfzIs3vUXJrtLgasbtN/B5ypl28xvcN+M25hfs4pbZ/wkGqZ2VFby3cT1qEy2yfWwcz5x+Jj1TUql0uZg4/R1qvV5006S4vp77Fs5je2UFfx7987fPC6FA4uvIqitANgIiUIiPvgFhG9Hmfl2TkjkQgZoppWRQRuZhqcjXDRrKoIwspm/aQL3Py4TO3ZjQ/gBa9WQCzCEdvEuQrjcg6bMwA4/fIo4H9p8ZQgh6DOtCj2GHXiaemtuZ19aGKzT4TTMo7vT1zh3URAtcd/RBq/YivCb+VDtRNgvL9xcyssltxjBNVhbtp8rtYtCozhgRik1AcJIVtbWG5E/3UDE1N+RtI87K2ItHRdgxHHe+cSN3j38E3aej+3QSCj0kvbqbpL+dzPelRegRqHgZ0dFUud2hbvQSlBgbN86d1cQ8sUdkcxiovL+rBzf0WItDa97flKA5OzAt7y2i4pxYzLnI2r8QyvP3gHcpneLOxKro+MzQ20ITBqlRUUzu1oMT2ndgfsEuJHByh1ySnE6kdzGBPoLIkNio9Tk48ys/zhmrwyicht/g+y9WYZom9y2YFyLedZAZYjb9m19dxYWffsSS31/N2+vX0uDzhTTcuXU/b61fw7WDBgeZI0eD7wv38tjSxeRXVZIWFc0tQ0cw5Si6NoXWGVIWBaQBZE2gWHyYYvBtw0aSV1wUMplxaBpXDxx8xP0lgzOzGJyZBQQ8B2T5lYRSoN1glCIbX0PE3H7E1/NrxfHA/gtFz5SUiBokAvh0yyZ6nXgyW8oPBLVv9BZdnn7TZEdlBSPbtaegpppLPvuYOq+36T0DxwnJJH25r81uTsUviVldTuW5HUKacewNBlPvnHxk4x/eldc2PsVXL85h37Yieo/qzhlXjwOnheGvv4Sut/ZB1XhhwiTeWr+W2fk7UIUIBDgBPtPAZxo0+H3UetsuiL2+ox8nZRTSNa4Kp+bHrWv4pcrK2ps4o3scfp+fhdPnsn1VHJkdHIw9p5rouIPjkFzYcSWztjWyoSo1aPStCYNEm4c/jQjkqBMcDs7r2apgaemN228yt6gzlV4HQ1JK6J1QASigZPBefhembcyl0muhYxvt+9I0qXC5qPG0LdV7EH7TYOb2bawoKgx9CDbBpqpsrShnVLucwx6rJVbuL+Sar74IBtjCulr++t08GvxeLu838IiPI4SKtA4A3wrwrUVahx1SImBARiavnXUOjy5eyI6qShLtDq4dNOTYeehGQVM6qDV84JkDxwP7cfzvIHBoGq5WKRkJbC4/QLmrkc6JSTg0S4haJQR41R3iE5BScvWXn1HaUB/ygNDHpGMvqCd6Y7gcb0soHgOjKbBbUfjHuWfhiD4y30uAtJwUrn78EgzTZFnhPuaU7KF/egbTJpzFTbO+QhECU0pMKbl12Ej6pWfwdHoG9405iQcWzeebnTvCaq6qEFgUJeKs3WtoTF0wmVFp++mfdIBSdxSzCjsxvpPC6LQGbh7+F6pKDuBuSMXmMHjrH+k8NTOfnK5eEBqauZc3T1jF81sG8UlBN/ymyrisPdzaex3Zyf2BAWHnBNhS4efi/1yCYer4TYEqJCdl7OeZkRtY73uBJ9bPDT6AXb0SiNpYhWgxfEVVGDphINE222F0ZwLw6DqFtTW0i41jTUlxcDZ/EH7TJD3q6LVWnli+NCwF6NZ1nl6xjEv69G+zptIa0rcKWX0DwWmJ1JGxjwZ1ciJhZLv2zLr48qMec0SIKGirOUscz7EfRxvYVlHOE8uWsLa0hGRnFDcOHsbZ3XscfsejwKh27ds06l1fVsoJb77KHwYPw2HR8Bp68ObWFIWUqCjGtM9he3k5+yqqw8ypTYtC/YkZhwzsiqKgNPoxHSqWKi+TrJlMbHGN1Qdqmf36fPZuLqT7sC6Mv/wkomLDb5r9dbX87pMPqfV6MaVEIhmyVzBoTgF7Yk1SclO57epJjB7UN7hPktOJ6TciEmkUoXBp3768s35dxHSORLC0rB1Ly9qRbHdxYnohI9NSePuB6ZTtLUdvovd53So+j+SJW9oz7Zv8wA2v5WLXVnJX31Xc1XdV80FFFLRh5Cyl5Pqvv6TOJ2jZDPddaQe+qLgUv/SFcLIrzu2AbW8DqltH8ZrYnQaOGIObnp0SYHJ06sKcXTsjzsQPIspiYWBGFu3j4pidvyMkGFsUhZ7JqXRKTKLB52NlUSFWVWVYVjush2He5LdRxPToOrVezxGldqTZgKy+NnzGXHcv0toPoR3dKuJYINR0pNYN9E2EsKWEA+G89L9+/l8Cjgf2o0R+VSXnfTwdtz9AfKvxeLh3wbeUNdRz3eChh93/SJESFcUfh43k+VXN1f6DOGgU/MIPq3jmtAm8s34dK4sKEUIwrmMuj4w9FVVRWDFvHVI3IYKpQ0JuCkIRyDYaQVRT0v7xZlGqrc5tzM7MYcLV4yjYuJdbx9yP7tPxefws+2I10//+GU9+9xCblm6jtOAA3YZ0ZtiZA/nDrK8obWwIPnhiF5Ww/+tCFJ+JDahbWsLjn2zjqUUPBTnlX78yl/Wvf4s4r0OI3AKAbhpcN2goE7p04+LPPmozAN7aazXXdF+P31RwWr6n+kqdvFm5FO1uTllJKSjY6qChsTsxOc+BbES6PyVcnkKAbUzE8+yoqqQ6QvrErQs+3LqbP40cg6KIYF3ViLNSeG9/EjYeYLSxiwnDihhxRh3/1955h0dVpX/8c+6dmklvEAhJqAm9I4p0UECUomLBFVEW6xZ3/alr3XVxV9d111XXde2uiqKi2EAFASlWkB6qIaGTQnqZcuf8/pghZDKTHlKG+3kennXKPfedu5l3zn3Pe75fW9hfgFf4y8SLKHE42Hg4C6OiUOp0olTp5zarKonhEUzq1h2DovD0lEv4w1crKXN6DGBGJnbhqYunsWxPOvetXumRy5CefvuXLpvF0ITObDtxnI/37UFKmN4rtVJ7JSk8wk80DDziYuHmumWpAbCvIbCcs4YsX4YI+039xmkiIuoZ5Knrwe3tLJMusFwG1tbXwWkJ9MTeQKq2VZ2m3OXimR+/44ZBQ5pVGe+WYR4/1ae+28i3Rw77zeAdmsb6Q1m8OftKXG7P1vyqt8tHP98Nff0Tt3C46VGicCrUQllR4JquVq0GXlFm590nPmLagok8ueB5n+MqyuzYKxwsHHgXBqNKRakda6iF2ORY9s/rjNvk/aJrbqJXHEFx+I5tL7Pz6gPv8NcV9/PdZ5t56pYXsCoC88g47Ik2T3KXEqvRyC3DRhBnsxFns/H6zCt4dP1a9uTmEG0N4fK0Pnyyfw89bDu5MXU7ZlXzarU4iYqDRW9kMH9UGr6Jx4Aa9z+EwTMbzTf+Cpv9nzg0BSEERlXFFP0CQli810Vj2bMr+OyFlTjtLnpfNwJq2LXu0jSGd+pM18go9uXlVpaPhAk4P4I/TdtL2GkDDsf3SHcRIcZwXrpsFseLizlWUkRyRCRLdu3gvfSdVDhd2ExGypxO7lj+Cb8aMZJJ3XowoWt3jhQVEmYyE2W1kpF/ivtWr/Qrq8z/6AOu6dufN3dsq3xtya7tzO0/kPtGj+PO80dx+/JP/BYxFw4dVn89JVkKMtCPrcvjYdpCCDUBYr8E52bQssE4AGFoGW/itoCe2BvI1hPH/Wqa4EkVR4oK6R4d439QExiS0Ikr+/Zny4njlFYzCXFLWVm7DfTFi4wMI37Jbk5e283TvqgqCLuGKd/BjO4Deb7cX0bfZDXhtDsDzuQLc4uwl9vZ/1OG32vSLSs7YADKSyo4vv8EEV9IKi7tAoBa5ERoge8QDmw5iJSSv173NADCLen03G6Kh8RQMjgGo1Nyx0XjWDDiTNvciM6JfHT1dT7j3DVqNGU587FUE7hSVIiKd9GtbwUZu86UVboOSK4sIR0tLuKS9+0Ymct58UeocBnYnJfCLcPc3OxVf37kyifZvHIb9jJPr37uE1/gfGgQhPjeWVgMBmb17osQgrdmz2HRujV8sCcdgcaYjof545ANZ5K65xN73YM8JISFkRDm8QK9ffhIRnROZN6ypeSUlSLx/K2tP5TJq15TlqSIM7tFl+7eFdBBSXNLXt22pVoXjYs3d2xjVu++jE/pxt8nT+HR9V9zsrSEUJOJW4aO4OahDZC8NV1A4E1KVkQVgbKWQAgBVWS7zyX0xN5AkiIiOFLsv3vT6XYTZ7MFOKLpjOqSFPCLGmI0Mq1HrxqPm7pgIite/grjkzsoOr8DrkgTIekFdMwoZ9N5KlLzH1MICI2yUZznb8Qc2ymak1k5KIogwGY/Pzoll/DwL/fT9/wyipxmXt/dj9U1CL8YO4byyZc/UFFypgwiNEn4j7mE/5iLoiokj7vI77js0hK+zsqk4mQxB5du4WPLKf4z7wBDAzgDujUIsfkG7rQ7+fuN/yahewe2DbJS5nDgkhaWHz6jCPr0D9/yi4GDObbrzlRTTAAAIABJREFUiE9SB3BVOOm8OIPjC1JBFdg1jRCjkdSYWJIiIsjIP0W3qGj+NnkKiyZMJj3jVnrb1mFUql17NaFGcTCAR75e4zOTlpz2213Np9de7/Peogp7wL8Xl1sLaKrh1DRWH/yZ3rFxTOuZyrSeqTg0DaOiNFiSQxiSkCHzoOwNzmg3hYB5lDfp67QEemJvILcPH8lPJ477fMksBgPTe6bWvw5ZBSklGw5lsSR9Bw6XxmWpaUzt0cunpBJtDeHeC8fw+Mb1OFwabjziVBckJjG+a7cax+42IJnbn76Rf//6FSJXZ5+WUsdhd/L9Zz8FPMZe5kC6JYoqcFebXR/Zf5xbh9xNfPd4Dh3OoXhEHOW9IjDmVRD6fTbFIztQMjQWVEH340d56ubPsdrcKIpHcfHXAzfT68UEnv9lLKJKOcZtVNhzYSQPr/icaBWUQKKIbjdDJg3wee71bT/x2IZ1mE45iHl8C0fu7Isr2szyI93pE5Xr088OIBTYu813ATBz12Eydx7GZDGS8bs+uOL9F0lVoXAw/xSZPxwIeM3UnXn8Yq+VDjcNI7u0lMzCfL7OPMhvP/8Mp9tNv7h4Xrx0FhEWCwO7/QWZNwvcRXgSnxGEARHxWK1JdHduTo3Pn1ZEPM3Ebt35cG+6nw2klJ7PolUrlahCwaz6poK6FlprQwm/C2m+0LNeIe0I63QwT2pTuk3Bjp7YG8j5XZJ4fNLFLFq3hkK7HQXB7LQ+PDimcbeZf1n/NYt3bq9sWfzm8CE+3JPOi5fO8rH4mjdwCMM7JfJ++k5KHA6m9OjFuJSufjZg1Zl600TGzrmAHevScdidPPaLZ3DaA+9qPY2jwonJaqJr/yT2/Xig0kfydJnl5MEcsm/sSUXXUKRJpVyTFF3QwbsbyPODNHdGJsYQSdUKUYjBxYxJJyi55wqW/mMjdrsTV4hK7oxkSnqFIewa0TX0+yWmdSYm4YyU7P68PB7fuB67phH+6UEqkm1oYUZQFZZkpHF5yl6SwwqxGVy43AKXHZ6+uzNOe7WSlTzzmdXcCoiz+OnZO9wacTYbJZ2jUQIkPJPFSLekDlw9ZBjv7trBmzu24nC7cTg8M/ttJ0/w+y9X8NJlsxBqLMSu8CQ9x49g6IoIuQah+ptHVCXcbKYggKhVhNnikzBPlpTw2taf/JK61WBkTt9+vLNzh98YQgim9qz5zq8xCPNIhHlks46pU3/0xN4ILu2VxiU9UzlVXk6YydToBdOsggLe3LHVp7OjzOXk+6NH2Hgoi9HJKT7v7xMXz0NjJzT4PCFhVs67ZCjLnllR6/ukAM1mwFDiwlHuwBpqwWKzUF7im1A0pwvrngLKU73CTqflB6qoNw6Mycak+mdpVbEw7/5hXPfALxn41FOUm6hMpNKsknt5CrFLM/12Z144y7fjaNnedJze62bdX0RZ36jK81doRq74aiaXJh1gQqcscipsZL3bne++OE6NAl1A6A85lPUMB2NV4xOVUV2SibeFEnPxIGwRVuylFT4tjKpB5eIbPD/sL2/Z7LNzFDxlug2HsyiyVxButiCUUHZvH8n7/8glK+M4ed2e5+CICKLiIrh52Aiu7tvfb3Z70+ChPPfj9z5jWw0Gbhx8ZuOQy+3mivfe5kSJ7yKlSVX597TpjEvpxpCOnbh71Reo3mulud0sGj+xTss5nfaFntgbiSKEj1xtY9h4OCvgjLvM6WR1ZoZfYm8qFaUVlXZ0gZAGheML0+j42n7MhU4MRjVgTVZKz+JmbRwojKZPZF4VW7jTB9tBTcSgGnBaVR/NF4DikfHYO4bQ5amdlc+ZrCZGzTzP531215nefS3MiOlYqc+ancNtYGlmGksPpoIQRPY2YbzASej6EwiXRDEqSLtXfdIgODmvJ+VpkQgpkXg2Qnm8cLtVKgqqBpV/fP0Ij179TzK2H0JRBNEJUfzyhQXsdhSRUqxQVMPOWEUIih0Ows0WVr21jqdu/q9HjE2Ce4cg+ksDR+7uz6KyNZwoKebOkb7SDbcOO4+8sjLe3rkdo6ri1DTm9O3PrcPOXJc1BzMoqKjw654yCKVSi+XS1DRGJyez5uBBJJLxKd2IstZ/09nZ5nBhIX9ev4YNh7Iwqwbm9O3H70aOajM+rO0F/Wq1ImFmc8DEblQUIhpRr6+LYVMG8eai9/20YiQec+2Tc3vg6BRC4egOdP7yBFffO4v7pj7qN44wqpQMqr3758W9A5naJQODUnX2agbzGE8rGp5F4fWHsny6jIQEk0NDCLDHWdB6RdF3WC96DOmKlBKHpmFSVab06MXbO3dQ7nJSMLETcYt/xny01NMeedoMQ5OVdxOFTgfy0iRyLumCcLoxGlVSHtuOll1G3qVJlKVG+OjBGxSFuf0H8kC1EltC1w48+/1j5B3Pp6ysgsf3bmLh7rWY9qk4NI24EBsK/irFbinZdOwoMSkWnr3jZZ8FWMUlocxF5Kpj5M1K4cWfNrFwyHBsVXTHFSF4aOwE7hw5iqPFRXQKCyfc7OtIlFVY4Gd3B567wIz8U5WPIy1WZvVueyqo+eXlzFjyJkXezWwVLhf/27aFvbm5vDbz8tYOr13RQJ09neZkYtfuBNrMoSoKs8/CF6/HoK5Mvn4cFpsnIUjAbRSUDoji4F+GUTYgGgwKzsRQFjx+HQPH9uW2f83HZDFiMKooisAcYqLfVcNQetYu7LS/KJqb1k/FJZLxCGSZwToTEflk5XseGTeJCLMFq3c2ZjUYiLBaeHz6dMRDozh27yDyruzKqh4ag//7bwY8/wx9//M0Y157ibyyMi5LTcNqMFI2KIbCixLp8NI+wr/NRil2Iio8/qkCfDV3vAqYwqhiu6ofJquJ4vPj/ZyR7JrGB7vTa/x8MQlRLD6xn9WZGdg1jWKHA7umkVNWiklVMVerxTs0jftXr+SG994L6EWqaJKQ3QWAZ4Z9uKgw4HnDzGbSYuP8kjpAamxswEVPm9FI33ZgEbdk1w4qqtyJgef/hx+OHfGXsNapFX3G3oqEGI28OmM2Cz9Z5mlPE56a598mTfHpS25Ofv3vBVw4awQfvbKKVRk/Uzg0xlMr9945GBBcM2MMMydNBOCSX05m8IT+rH33GxwVDkbNGEGPwV3509erWbJrB0ZVpdThCNi5vLMgme/tv+LCxHgQJkQ1b9ouERGsmXcjH+xOZ1f2SfrEx3N57758sm8vR+IEmstryFxts9TR4iLu/HI5L06fyay0Pqw4sA/zkGFMebobhuwylh77mU9zMnFobnpER5Oek+23Q9WuafSdOZh5N13J5LVLA16rUmfN9XiAxVU2+lQd12Y0MjutL2/u2OpzXcqcTnYU5RKTbMO8w79ko9m8hipujY6hDdd6GdUlmeSISA6cysPhLXEZFIWYkBAu7t72TSe2nzwRUK5aFQr78nJJjam5HVTHFz2xtzLDOnXm+wW3ePwnNY3hnRPr1DtvCkIIhk4eyNDJA8lf8QlfHcwA75dJACFmEwvP9+1m6NS9I9f+wVer/I/jJnLrsPNYuuIbnlu/kbIe/otvCgKzaqjV3CDcbOGGQb7KgW9u3+q3AFmdCpeLf33/Le9eeTUjOlfZUZgC/UjlYe/D9Jxsrnjvbb/jQ4xGRnROpHuvFAakd2R79gm/95yWga2J6hvGTlPucpEWG4vFYPD7HOUuF8bxyRj2FFRqsgO4TQoF4zthMRi4pGcqkZaG170VIXj78qt44pv1fLJvD24pubh7T+69cEyT2hdbitTYWNZ474Cq4pbuRptrn6vopZg2gFFVuaBLEmNTup7VpF6df1w0jZuHDCfGGoLVYGBC1+4su2puvTdadQgNZefjqwj/6hjC7r8oa1JVhibU3sYXiJpMRqqTWVi7OiV4OonGJKdUlnvA0+mSFB5ROYt9ZPxEQozGyt27RkXBZjTV2YE0uGNCwOcHdehIuNlS4/rJxGld6DWoBJPFjTlUIo2CwsmdcQ+JZ06f/k0yYA4zm3lk/CS23HwH2275FX+bPKVRuuytwbX9BmKs9gNkUlXSYuPo3w5KSW0JEajr4WwzbNgwuWmTv4mETvtjVvQNFBeUkndZEkWjO3p62d2e2vbrMy5nVFqPOseozpPfbODFLZsqxc5q4sKkZP4384o6x3O53by9czuLd2zDoWnMSO3NTYOH+ixOHios4JUtm0nPzWFAfAfmDx5aZwvgruyTXLV0CQ6XC5eUGITAZDDwzuVX0T0qmpEvP0+xw7ecYzEY+Ozi90i2ZXM0w8SpbCNde5cjrTas8a9hsvSv8/MEM3tyc3hg9Uq2njyBKgTTe6Xxx7ETCAuwpnAuIoTYLKWsUyehSYldCPEEcCme5uCfgflSyoK6jtMTe8tztLiIcqeTrpFR9dbVrg+/vfABdn2zFwBntJnyXuEoZRrxxxx8cPylRpk0F9ntXP7uYo4XF1PmcqIgcFer4lsMBhbPnsOgKrNm6cpEljwDjs2gdkSE3oIwj2vS56uLQ4UFvPTTJnblZNM3Lp6bBg8jOdKzPrLl+DFu+uRDnJob4V0/eWxMRy6Je8xrHVcVBaxzUSIePKvxthecmoaqKHVuwDvXaKnEfhGwWkrpEkI8DiClvKeu4/TE3nIcLS7i1s8+Zn9eHqoisBqM/H3yFMamdG2W8bevS+e+qY/6GGSbQ8wseOxaZt4xrdHj2l0uPt2/l28OHyIxLByr0chbO7aRW1ZKWmwc940ey/BOZ2rr0pWJzJvt1QE/vdhqhfD7UEKa0bC5gTg1jU3HjuLwrp9YXSuQhQ8BARx+rFeiRPi3l+ronKZFEnu1E84CrpBSzq3rva2V2KWUfJ2VyUd70lEUhVlpfRjVJSloNSzcUjLh9Zc5Wlzks2nFajDw2bXXN9uC1NY1O3nh7jfI2nWYmE5R/OKhOUy+fmyzjF1f3AW/g4rl+HWQizBE/HcI0XJrF7UhtTxkzjh8PVcBEYKIfBphbnkTap32Q30Te3N2xdwILGnG8ZoVKSV3r/qcFQf2V+pofH5gP1f26ccfxzV8m357YNOxo+SVl/ntRHS53SzesZ37RjdP8h00vh/P/fh4s4zVaByb8d8WhMdgQTsOhqQWDykQQo1Bhv0fFP8dTwXT7XFvMo0BU2AzDx2dhlJnYhdCrAI6BnjpfinlR9733A+4gLdqGWchsBAgKanlv2TbT55g+f59Pu1n5S4n76bv4Nr+A+gVhD2y2aX+0rvg0S45Uhx4A0y7Re0A7uMBXtBAqfvOREoHSCdCOTvSy1VRbNcjTSOQ5R+CLENYLgbTqKC9c9RpeepM7FLKSbW9LoSYB0wHJspa6jpSyheAF8BTimlgnE1mbebBgG10mtvNuqzMdpXYy51Otp44TqjZTL+4+BoTwqCOCQF1ua0GA6Mb6GDf1hG2W5AFd3JGAxzADJaLEEpYjcdJdwmy6GGo+BxwI9VuiIhFCFNg4+pmi9eYhjD+oUHHSOdeZOlL4MoA0yCEbUGlPIOOTlWaVIoRQkwB7gHGSlndvbZtYTOZMKqqXwudQVFatHe8qbyfvpOH167GoHh8MGOsIbw6Yzbdovy3+CeGRzC7dx+W7dlTKQtsUlU62EKZmdb2tEKagrBMQIbdDSVPAm6vx+VFiDoWI2X+zeDcBng3G2n7kfk3QMynCEOXsx22byzuAnDngdoFIUy+r9k3IvNvpbJ849qNLF8GMe8jDM2zEK4TPDS1K+YAYAZO25t/J6W8pa7jWmPx9ERJMRP+94rfrN1qMLBh/sI2pXBXE7uyT3Ll++/4fAaBZ6PQhvkLA7aGSSn5cE86/9u2hVKnk6k9erJgyPCAWiPBgJQO0I6BEo1Qau9Dl879yLzL8TevNkLIXJTw+85anD5xyHJkwb1g/wpOL/KG3oliu977ukTmTgLtcLUjBZgnokQ91yJx6rQ+LbJ4KqVs+O6TVqJjaBj/mDyV369cgSo8fdxuJM9OvbRdJHWAt7wbbKoigWK7gx+PHuG8RP8ZphCC2b37Mrt33xaKsnURwgSGlPq9WTsMwhDAotMJrsBuSWcDWXgf2FcDDpDettHiJ5FqZ4RloscEWgu0fiDB8X2LxanTfjintGKm9OzF6OQUvjl8CEUILuiShLUdlWFyy8tqNNIuqEEHvKHYXS7e2L6VZXvSMSgKV/cbwJV9+jXrpqY2g6EXyEB6L2Yw+tbYCysq2JuXS7zN1qy6JdJdBBUr8TcAKUeWPu9J7MJCjeofSkSzxaITPJxTiR08tfbJ3dvNjYYPk7p2Z+OhQ5X18tM43RpDE2oXrKoPmtvN3A/fIz0nu7Lcs3/dGtZlZfLcJZc1efy2hjAkIi2ToWIVZ8oxCggLIuRawFMGefLbjby8ZRMmVcWpuRnQoSP/nT6DCEszaOa7C/DIGgdAy/bEKUxI6yVQvhzf/ncrhNzQ9Bh0go4gnIYFLzNSe9M1MtJH0MpqMHDb8POa7OYEsDbrIHtyc3xq+OUuF19nHWRH9sl6jZF/soBTJ+oW56ovX7/7Db8c8DtmxdzAPRf/mf0/ZTTb2AAi4m8Qeiso8SBCwTwZEfMBQvUYiXy8bw+vbt1cqbleobnYcuIYd36xvHkCUBM85SA/FDCdKaWKsIfBfAFgBhHm+d+QKxAh1zVPHDpBRVDO2DPyT7H6YIbXaacn8baGa1u3BY4UFbL95AnibaEMTeiE2WDg/TnX8O6unSw/sI8Is5nrBgxidFJKs5zv+yOH/UyQATSv+09tCnuH9x7l0Wuf4lD6UQASeyVw3+LfktK38Z0ly55dzkv3LsZe5pml/rRyO3du3MtTG/5Mj0HN0wlSWmhn+Uud2PLVFDp178iMO6aSFHXm7uelnzYF9DD95sgh8svLm7w+I4QRGXYPFC3C964hBBH66zPvU0IQUf9FasdAOwqG7gildrMTnXOXoEvs//xuIy9s3oRbulEVhb9uWMfjky7istTerR1avTleXMz8jz9gf14uqlAwqAqdQsN4a/YcOoSGcv3AwVw/sPn7rDvYQjGrBuzV7NWMikJcLXcEFWV2fjv6QYrzSio9UjN3HuJ3Yx7kraz/YA1tePJzOV289uCSyqR+Gnu5nVcfeIdHP21YD3gg8rMLuW3o3RSfKsFe7kBRd/DFa2t5+P3fM3yK5/oWVJQHPFYVgiK7vVkW3pWQOUg1AVnyX3AfA+NwROjtiAC7ZYXaCdSGSyHrnFsEVSlm+8kTvPjTJuyaC6fbTYXLhV1zcc+qL8kvD/wFbWtkFRQw8Y1X2JeXiwRc0vM5MgvyuWPFJ2f13DPS+lS611fFqKhM6ta9xuM2fPA9zgqnj/G1lOB0uPj63W8bFcup4/k+RhRnBvYIjzUHixctpSC7sFLAzK25sZfZ+ftNz+H2buwak5yCIUAbqdVgJDG89nbKhiDMo1Fi3kSJW40S+XjApA6w/lAm1y59l7GvvcTvv1xBVkGdYqo65yBBldg/3bcHRyBrLUWwOrN5a7Nniye+XR9wh6wbzw9XTml1udfmIzbEs9kp3mYjxGjEajCQFBHB4svnYDHU3D2UczjPR93xNBWldnIO5wU4om7CY8ORAXbNAtjL7KR/t69R41bl2083BfzxKC0s42RmDgC/GnE+4RZLpQORwCMZvGj8pBbvFFqavpNbPv2I744e5nBRIR/t3c2l77zBwYLmW9PQCQ6CqhSjSRnQexOgNQxFGsM3hw7V+JoqRMAaeHMyvFMi39x4M/tP5WEQgm5R0TVKFhTnl/Dcb19l7TsbcWsBpAtCLfQaXvNMvzYsIWYGjuvLj59v9XtNuiUr/7eWPiN7NWrs04SEBy6juDU3llBPx0vH0DBWzJ3H61t/YuPhQyRFRHDT4GEM6BBIPuns4XK7WbR+rU+93y0lJQ4Ht332MUuuuIpwczN06egEBUE1Y7+0VxoWg/9vleZ2Mz6lWytE1HBq2xEabrbQJeLs9y0rQpAaE0v36Jgak7rb7eb34x5m7ZJvAs56jRYjiamdGHbxwEbHMXHuGFRj4FZAl71+9nm1MetX07CE+F5v1aDS5/xUouLPXOe4EBt3XTCaD6+ay7+mTPdL6pkF+dy98nMmv/EqCz9ZxtYTgTYTNY3jxcU1OkrtzcvlksVvUNRMexl02j9BldgHdUxgbv+BWAwGVCEwKgpmVeWRcROJaYZ2wJZg/qAhAX+cVCH4+0VT24yjzNY1uzhxMBuXo1qCFRAWHcqc/5vBk2v/1CgHpdMMnzoIVfX/Ex0xqYKb7l2NO+di3IUPILWjjRp/yo0TmHzDOIwWIyHhViw2M8l9Ern/nTvrPcb+vDwuffsNPtiTzs/5p/jq4M9c+8G7fJXxc6NiqolIi8VPfrkquWWlvLZ1S7OeU6f9EpSep7tzc1iVcQCzamBaz14khref3XluKfnj2q94N30nAk9rXUJoGC9On0laXFxrh1fJsmdX8OLdb+Co8C8Nzbj9Yu54ZkGznOfzV1fzzB0vo7k03C43U+cWctuiwxhNp39QVE9rYMyyRot25R7NY9/mDOISY+gxuGuD5HMXfPwhazIz/EqACaFhbJj/y2aV4r3zi+Us378PpzvwzL1/fAc+ulrvaw9mWsNoo83QOzaO3rFtJwk2BEUIHhk/iV+ddz4H8vLoHB5OUkRka4flR1LvRFD9k5Y5xEy3ASnNdp4p8yfQ78LerF68noqyUub/+mlUtepdggayFFn6LCKicWYfsZ1jiO0c06hjNx0/GnBdJ7esjIKKisp2SOn4EVnyH9CywDjI287YsPLgXydOpqCigq+zDgZ8vTk2qekEB0GZ2IOBuBAbcSFn3/ShsUQO7kRphAG1womieVKbFFBhkIy9elSzniuxZwLXPzzH62v6rwCiXW6wf9es56yLH44e4YPdu9Bq6NxRBJVy0O7yL6HwLio3IGlHkfavIHoJwpha73NaDEZenTGbaW+9zr5TeT66QVaDgRsHD23059EJLoKqxq7TcizZtZMTv+5HyeAY3AaBVARlfSLJ+b+B7CjMPTsnVSI9OuuBUOPPzjkD8LeN65j/0VLeS99JaYAuJbOqMjOtD2aDwdONVfwIvtLAbpDlyOK/N+r8r864nLSYWKwGA6EmExaDgd+dfyGjgsw8Rafx6DN2nUZxtLgIh0Uh57oe5Fx3RlTNZjTWaMnXVIQSiTSPB/safNQQhRVhu/msnLM6BwvyeXXrFr/dueCZNWtSMrlbD/441uujK/O9Ql/VkeBs3GJnh9BQPr32evbn5ZFfUU6fuHhCTaa6D9Q5Z9ATu06jGNUliTWZGX46Ki63myEJjdvybne5OHAqj2hrCAlhge3sRMRjyMK7wL7ea0rhhtDfIiy1Ojg2G+uyDhKgFoQApvVM5Z5RY3xr3SLU+2oAlMbV9U/TM6Zpx+sEL3pi12kUM9P68NKWzZ6Zu7e/2mowMjMtrVFdSG9t38pfN65DQVTKED87bTqRFt9NREKxIaL+g9RywZ0LhhSEaLmNOVaD0bPjNIDFYkpklN8CpkdydzaUf4hvOcYKLXSXoXPuodfYdRqF1Wjkw6vmcvPQ4XSPimZAfAf+PH4ii8ZPbvBYGw5l8ZcNX1PmdFLidGDXNH48doTbl9esjSPUWI8hdAsmdYCLu/cMuItZVRRmpKYFPEaE3w+WKXgkd22AFUJvRlhnnd1gdc5ZgrKPXad9MW/ZUtYfyvR73qyqfHX9jXQKaz6xreZgXVYmty3/GAWBxLOz+a+TLmJGHQqi0l0AWg4YEhGifdgx6rQtzuk+dp32xcmS4oDPG1WVvPLyNpfYxySn8MOCW9l4KAuXdDOqS3K9zMGFEunp7NHROcvoiV3nrHKwIJ/nfvyOrSdO0C0qmtuGn8fAaloro5NTOFiQj7NaT7jmlvSMbptmEiFGY7u1WNQJfppUYxdC/FkIsV0IsVUI8aUQQncA0KlkT24Ol779Bh/u2c3P+adYlXGAa5YuYW2m787JhUOGE242Y6wig2s1GLh71IW1ygXr6OgEpqmLp09IKQdIKQcBnwIPNUNMOkHCYxvWUeZ0Vu6QlECFy8VDa1f5LEDG2Wwsv3YevxgwiB7R0VzYJYn/XDKDeQOHtFLkOjrtmyaVYqSURVUe2gjU4KtzzrL5+LGAz58oKaHY4fCpS8fZbDwwZnxLhaajE9Q0ucYuhHgUuB4oBGr8ZgohFgILAZKSAtt+6QQXUVYLpU5/ZyWDomANIE2so6PTPNRZihFCrBJC7AzwbwaAlPJ+KWUX4C3gjprGkVK+IKUcJqUcFteG5Gd1zh43Dx3hl8AtBgOX9+6LsQk67W0Rze0mr6wMZw1mGDo6LUmd0yYpZX33ai8GPgMeblJEOkHDtf0GcKSokNe2/oRRVXFoGhd368EDo8e1dmjNyts7t/PEN+spczoxKArzBw3hzpGj2owpis65R5Puh4UQPaWU+70PLwP2ND0knWBBCME9o8Zw27DzOFRYQMfQsHbjZFVflu/fy6J1ayo1cxyaxitbNiMQ/O785pUv1tGpL03tinnMW5bZDlwE/KYZYtIJMsLMZvrGdwi6pA7wr++/9RNCK3e5eGXrZlw1aLXr6JxtmtoVc3lzBaKj0x45HmjXrJSYVx/hmkcWUphTRFLvztzy5A0Mu6jxxt46Og1BFwHT0WkCaTH+jQCRK48R9UkWBScLkW5J1q4j/HHW39i+Lr0VItQ5F9ETu45OE7jnwtFYqnb+uNxEfXUU4fAtw9jLHbz24DstHJ3OuYqe2HV0msDQhM68OetKRnRKJNxsJs0UjkkEbuU8tOdoC0enc66i7xLR0WkiQxI68c4VVwHgdDiZfd8GXPh7oXZJ1aWUdFoGfcauo9OMGE1G5tx1GZYQXxlfs9XEDX++upWi0jnX0GfsOjrNzHUPXoE1zMI7jy2jMLeILqmduPUfNzBwbN/WDk3nHEF3UNLROYtIKRH6DlSdZqK+Dkp6KUZH5yyiJ3Wd1kBP7Do6Ojr742VDAAAEDElEQVRBhp7YdXR0dIIMPbHr6OjoBBl6YtfR0dEJMvTErqOjoxNktEq7oxAiB8hqxKGxQG4zh3O20WNuGfSYWwY95pahppiTpZR1WtC1SmJvLEKITfXp4WxL6DG3DHrMLYMec8vQ1Jj1UoyOjo5OkKEndh0dHZ0go70l9hdaO4BGoMfcMugxtwx6zC1Dk2JuVzV2HR0dHZ26aW8zdh0dHR2dOmh3iV0I8WchxHYhxFYhxJdCiDbvXiCEeEIIsccb94dCiMjWjqkuhBBXCiF2CSHcQog23VEghJgihNgrhDgghLi3teOpCyHEK0KIbCHEztaOpb4IIboIIdYIIXZ7/y5+09ox1YUQwiKE+EEIsc0b859aO6b6IoRQhRBbhBCfNub4dpfYgSeklAOklIOAT4GHWjugerAS6CelHADsA/7QyvHUh53AbGBdawdSG0IIFfg3MBXoA1wjhOjTulHVyWvAlNYOooG4gN9LKXsDI4Hb28F1tgMTpJQDgUHAFCHEyFaOqb78Btjd2IPbXWKXUhZVeWgD2vwigZTySymly/vwOyCxNeOpD1LK3VLKva0dRz0YARyQUmZIKR3AO8CMVo6pVqSU64BTrR1HQ5BSHpdS/uT972I8Sadz60ZVO9JDifeh0fuvzecLIUQicAnwUmPHaHeJHUAI8agQ4jAwl/YxY6/KjcCK1g4iiOgMHK7y+AhtPOG0d4QQKcBg4PvWjaRuvCWNrUA2sFJK2eZjBp4C7gbcjR2gTSZ2IcQqIcTOAP9mAEgp75dSdgHeAu5o3Wg91BWz9z3347mlfav1Ij1DfWJuBwRysmjzs7L2ihAiFFgK/Lba3XObREqpecu2icAIIUS/1o6pNoQQ04FsKeXmpozTJj1PpZST6vnWxcBnwMNnMZx6UVfMQoh5wHRgomwjPaYNuM5tmSNAlyqPE4FjrRRLUCOEMOJJ6m9JKT9o7XgagpSyQAixFs/aRltetB4FXCaEmAZYgHAhxJtSyusaMkibnLHXhhCiZ5WHlwF7WiuW+iKEmALcA1wmpSxr7XiCjB+BnkKIrkIIE3A18HErxxR0CI/H38vAbinlP1o7nvoghIg73YEmhLACk2jj+UJK+QcpZaKUMgXP3/LqhiZ1aIeJHXjMWy7YDlyEZ/W4rfMsEAas9LZpPt/aAdWFEGKWEOIIcD7wmRDii9aOKRDeRek7gC/wLOi9K6Xc1bpR1Y4Q4m3gWyBVCHFECHFTa8dUD0YBvwAmeP+Gt3pnlW2ZBGCNN1f8iKfG3qj2wfaGvvNUR0dHJ8hojzN2HR0dHZ1a0BO7jo6OTpChJ3YdHR2dIENP7Do6OjpBhp7YdXR0dIIMPbHr6OjoBBl6YtfR0dEJMvTErqOjoxNk/D+QfPL5jL1xFQAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"from sklearn.decomposition import PCA\n",
"pca = PCA(n_components=2)\n",
"pca.fit(X)\n",
"print (pca.explained_variance_ratio_)\n",
"print (pca.explained_variance_)\n",
"X_new = pca.transform(X)\n",
"plt.scatter(X_new[:, 0], X_new[:, 1],marker='o',c=y)\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
"lda = LinearDiscriminantAnalysis(n_components=2)\n",
"lda.fit(X,y)\n",
"X_new = lda.transform(X)\n",
"plt.scatter(X_new[:, 0], X_new[:, 1],marker='o',c=y)\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 95,
"metadata": {
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[ 0.43377069 0.3716351 0.1945942 ]\n",
"[ 1.20962365 1.03635081 0.5426502 ]\n"
]
}
],
"source": [
"from sklearn.decomposition import PCA\n",
"pca = PCA(n_components=3)\n",
"pca.fit(X)\n",
"print pca.explained_variance_ratio_\n",
"print pca.explained_variance_"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.5"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
| Jupyter Notebook | 3 | xihushui599/machinelearning | classic-machine-learning/lda.ipynb | [
"MIT"
]
|
? $_mt->wrapper_file("wrapper.mt")->(sub {
<title>Frequently Asked Questions - JSX</title>
?= $_mt->render_file("header.mt")
<div id="main">
<h2>Frequently Asked Questions</h2>
<h3>Q. Who are you?</h3>
<div>
JSX has been developed as a research project at <a href="http://dena.com/intl/" target="_blank">DeNA Co., Ltd.</a>, one of the leading social game providers in Japan and in the world. The main developers are <a href="http://twitter.com/kazuho/" target="_blank">Kazuho Oku</a> and <a href="http://twitter.com/__gfx__/">Goro Fuji (a.k.a. gfx)</a>.
</div>
<h3>Q. What are the license terms?</h3>
<div>
JSX is provided under <a href="http://www.opensource.org/licenses/mit-license.php" target="_blank">The MIT License</a>.
</div>
<h3>Q. Can JSX be used together with the <a href="https://developers.google.com/closure/" target="_blank">Google Closure Compiler</a>?</h3>
<div>
Yes.
</div>
<div>
The current optimizer of JSX focuses on expanding short functions inline. It often inline-expands functions that are not handled by the Google Closure Compiler. On the other hand, Google Closure Compiler is very good at optimizing the expressions within a JavaScript statement.
</div>
<div>
JSX emits JavaScript fully-annotated by type hints understood by Google Closure Compiler, so that the code generated by the JSX compiler can be further optimized by the Google Closure Compiler.
</div>
<h3>Q. What are the future plans?</h3>
<div>
Some of the features we might add to the language are: generic programming, asynchronous error handling, default arguments, named arguments, node.js support.
</div>
<h3>Q. How can I contribute?</h3>
<div>
Your contribution is welcome in all areas, from contribution to the language development to tweeting about JSX :-)
</div>
<div>
We desperately need more libraries, including a high-level interface for web programming (like jQuery). Although it would not be included as part of the JSX distribution, such a library would help us and others using JSX a lot!
</div>
<div>
It would also be great if we could have syntax coloring support in editors other than vim (.vim file for syntax coloring of JSX code is included in the etc/ directory of the distribution).
</div>
</div>
? })
| Mathematica | 3 | monkpit/JSX | doc/src/faq.mt | [
"MIT"
]
|
package org.xtendroid.xtendroidtest.fragments
import android.app.AlertDialog
import android.app.DialogFragment
import android.os.Bundle
import org.xtendroid.annotations.AndroidDialogFragment
import org.xtendroid.app.OnCreate
import org.xtendroid.xtendroidtest.R
@AndroidDialogFragment(R.layout.fragment_dialog) class CustomDialog extends DialogFragment {
override onCreateDialog(Bundle savedInstanceState) {
new AlertDialog.Builder(activity)
.setView(contentView) // contentView is inflated by the @AndroidDialogFragment annotation
.setPositiveButton("Ok", [ dismiss ])
.create
}
@OnCreate
def init() {
title.text = "Hello, Dialog!"
message.text = "This is a dialog fragment, created using an AlertDialog.Builder, and managed using a regular DialogFragment"
}
} | Xtend | 4 | kusl/Xtendroid | XtendroidTest/src/org/xtendroid/xtendroidtest/fragments/CustomDialog.xtend | [
"MIT"
]
|
REBOL [
Title: "Red/System unicode conversion library"
Author: "Qingtian Xie"
File: %unicode.r
Tabs: 4
Rights: "Copyright (C) 2011-2018 Red Foundation. All rights reserved."
License: "BSD-3 - https://github.com/red/red/blob/master/BSD-3-License.txt"
]
utf8-to-utf16: func [s [string!] /length /local m cp result cnt][
result: make string! (1 + length? s) * 2
cnt: 0
while [not tail? s][
cp: first s
cnt: cnt + 1
either cp < 128 [
unless length [repend result [cp #"^@"]]
][
m: 8 - length? find enbase/base to binary! cp 2 #"0"
cp: cp xor pick [0 192 224 240 248 252] m
loop m - 1 [cp: 64 * cp + (63 and first s: next s)] ;-- code point
either cp < 65536 [
unless length [append result to-bin16 cp]
][
;-- multiple UTF16 characters with surrogates
either length [cnt: cnt + 1][
cp: cp - 65536
append result to-bin16 55296 + shift/logical cp 10
append result to-bin16 cp and 1023 + 56320
]
]
]
s: next s
]
either length [cnt][result]
] | R | 5 | 0xflotus/red | system/utils/unicode.r | [
"BSL-1.0",
"BSD-3-Clause"
]
|
DAFETF NAIF DAF ENCODED TRANSFER FILE
'DAF/CK '
'2'
'6'
'M01 SC CK; EXT (2018-07...2018-09); NAIF/JPL; Ver 1.0; 2018-'
BEGIN_ARRAY 1 123
'Merged CHTLM Qs - Made-up AVs '
'48C1832F97^A'
'48C18374A5^A'
'-CF08'
'10'
'3'
'1'
123
'AEEB23383FA388^0'
'-9945D9BBC20758^0'
'-69BDC28CB25918^0'
'104D2A8BA0B3E7^0'
'1CE46572234872^-2'
'-32A527E35EFF6E^-2'
'-30DD75A7ED7FD4^-3'
'AEF1CC609642D^0'
'-9974A7F2A2CA48^0'
'-697A9F587562^0'
'FFF91C6109B03^-1'
'1CE27EDA448481^-2'
'-32A9E6E21A94D2^-2'
'-31D3C6A045BDE4^-3'
'AEF83789AF15D8^0'
'-99A3DD146DA04^0'
'-6936F3DD0701A8^0'
'FB1244FC0879A^-1'
'1D0608EBDB2D56^-2'
'-3296B87F3FDCD6^-2'
'-2EE279BD72C768^-3'
'AEFB7B8BE77F58^0'
'-99BD250F032828^0'
'-6912DA224AD898^0'
'F86CB0AD5C6F18^-1'
'1CF84A72BCE2C4^-2'
'-32993424FE19D6^-2'
'-2E53AEC2A44EB^-3'
'AF05349D40982^0'
'-9A051817CED3B8^0'
'-68AAA5B07E3794^0'
'F0E20D50CB7488^-1'
'1CEC76BC83B917^-2'
'-329ACFEFFAAECE^-2'
'-2F07357EF4CAA6^-3'
'AF0E2CE96D168^0'
'-9A4CDB7AE57968^0'
'-6842DDA02EB41C^0'
'E9538E1C8DD54^-1'
'1CF2076FE47C3E^-2'
'-3297DF097AC984^-2'
'-2F858620AF7ED^-3'
'AF17245EDA001^0'
'-9A9466FFBE66B^0'
'-67DA38F10D48E8^0'
'E1C9DF06B62798^-1'
'1CEEC32B2AAE1B^-2'
'-329DDE3443698E^-2'
'-3012B7E093778E^-3'
'AF1F9013330DA^0'
'-9ADBA4560B2478^0'
'-6771D5DB3E54A4^0'
'DA3BF3762EDB6^-1'
'1CF9BDF5077E84^-2'
'-32AD2C915ACFC^-2'
'-30A53E2A7647EC^-3'
'AF27BC06ABEBA8^0'
'-9B21BE6BB6069^0'
'-670A2AFC2D0CA8^0'
'D2CE20B4D660F8^-1'
'1EB363D1E9E1D^-2'
'-2EBDBBA8B3942A^-2'
'CF12ECA155F1^-5'
'AF279F5647098^0'
'-9B22DED232E838^0'
'-6708EB36196C8C^0'
'D2ADFA79AE41C8^-1'
'1E966D9C73D372^-2'
'-2EA3DF67172276^-2'
'15A022137AAA4^-4'
'AF2FC8F07446D^0'
'-9B6944922B3B4^0'
'-669FD979AE6CBC^0'
'CB284F7F6F20F8^-1'
'1CF33FAAA926DC^-2'
'-32845178781ED8^-2'
'-2FE62ED2B71C1C^-3'
'AF36C995B5B458^0'
'-9BAFF8ABA2DB9^0'
'-663735F732D1E4^0'
'C3987353DFA618^-1'
'1CFDAE15310A4E^-2'
'-32AAFBD30031A^-2'
'-2F1F541ABBF9D4^-3'
'AF3E89D37DEBE8^0'
'-9BF5E49E24C3B^0'
'-65CD565F8825D4^0'
'BC0472D6614B2^-1'
'1CE223EC568146^-2'
'-32D1FEE5C050EA^-2'
'-303BB2D2D5CB54^-3'
'AF45A2CC7335B^0'
'-9C3BCB074FD58^0'
'-656369DCBD85B4^0'
'B473AFD011D0B8^-1'
'1CFFD163093C34^-2'
'-32A2BB1E498122^-2'
'-303EB58A7E2732^-3'
'AF4B0C818B0CD8^0'
'-9C7817CA8DEFE^0'
'-650850955FBC84^0'
'ADECA866EE5A4^-1'
'1CF5DC823F1A8A^-2'
'-325F82FFCABEB2^-2'
'-2D34670353068^-3'
'48C1832F97^A'
'48C1833376^A'
'48C183375D^A'
'48C1833976^A'
'48C1833F76^A'
'48C1834576^A'
'48C1834B76^A'
'48C1835176^A'
'48C183575D^A'
'48C1835776^A'
'48C1835D76^A'
'48C1836376^A'
'48C1836976^A'
'48C1836F76^A'
'48C18374A5^A'
'48C1832F97^A'
'1^1'
'F^1'
END_ARRAY 1 123
TOTAL_ARRAYS 1
~NAIF/SPC BEGIN COMMENTS~
This CK is for testing with the image: /tmp/themis_kernels/I74199019RDR.cub
This CK was generated using the following command: {}
~NAIF/SPC END COMMENTS~
| XC | 2 | ladoramkershner/ale | tests/pytests/data/I74199019RDR/m01_sc_ext56_1_sliced_-53000.xc | [
"Unlicense"
]
|
$$ MODE TUSCRIPT
secondsrange=2
PRINT "Sleeping ",secondsrange," seconds "
WAIT #secondsrange
PRINT "Awake after Naping ",secondsrange, " seconds"
| Turing | 3 | LaudateCorpus1/RosettaCodeData | Task/Sleep/TUSCRIPT/sleep.tu | [
"Info-ZIP"
]
|
# argv.fy
# Example of fancy's interface for command line arguments
"This will always get printed, even when required from another file" println
__FILE__ if_main: {
"This will get printed, if this file is directly run with fancy" println
}
| Fancy | 3 | bakkdoor/fancy | examples/argv.fy | [
"BSD-3-Clause"
]
|
#!/bin/bash
# The script prints the total number of the examples in the test process.
#
# It's a part of the test process
#
# TODO: Write stats to Firebase and display the stats at the site.
count=$( cat tmp/tests_count.txt 2> /dev/null || echo 0 )
printf "\n%bSUITE: %s%b\n" "\x1B[32m" "$TEST_SUITE" "\x1B[0m"
printf "\n%bTOTAL TESTS COMPLETED: %s%b\n" "\x1B[32m" "$count" "\x1B[0m"
rm -f tmp/tests_count.txt
| Shell | 4 | taco-tues-on-a-fri/date-fns | scripts/test/countTests.sh | [
"MIT"
]
|
<$mt:HTTPContentType type="application/atom+xml"$><?xml version="1.0" encoding="<$mt:PublishCharset$>"?>
<feed xmlns="http://www.w3.org/2005/Atom">
<title><$mt:BlogName remove_html="1" escape="xml"$></title>
<link rel="alternate" type="text/html" href="<$mt:BlogURL escape="xml"$>" />
<link rel="self" type="application/atom+xml" href="<$mt:Link template="feed_recent"$>" />
<id>tag:<$mt:BlogHost exclude_port="1" escape="xml"$>,<$mt:TemplateCreatedOn format="%Y-%m-%d"$>:<$mt:BlogRelativeURL escape="xml"$><$mt:BlogID$></id>
<updated><mt:Entries blog_ids="children" include_with_website="1" lastn="1"><$mt:EntryModifiedDate utc="1" format="%Y-%m-%dT%H:%M:%SZ"$></mt:Entries></updated>
<mt:If tag="BlogDescription"><subtitle><$mt:BlogDescription remove_html="1" escape="xml"$></subtitle></mt:If>
<generator uri="https://movabletype.net/"><$mt:ProductName version="1"$></generator>
<mt:Entries blog_ids="children" include_with_website="1" lastn="15">
<entry>
<title><$mt:EntryTitle remove_html="1" escape="xml"$></title>
<link rel="alternate" type="text/html" href="<$mt:EntryPermalink escape="xml"$>" />
<id><$mt:EntryAtomID$></id>
<published><$mt:EntryDate utc="1" format="%Y-%m-%dT%H:%M:%SZ"$></published>
<updated><$mt:EntryModifiedDate utc="1" format="%Y-%m-%dT%H:%M:%SZ"$></updated>
<summary><$mt:EntryExcerpt remove_html="1" escape="xml"$></summary>
<author>
<name><$mt:EntryAuthorDisplayName escape="xml"$></name>
<mt:If tag="EntryAuthorURL"><uri><$mt:EntryAuthorURL escape="xml"$></uri></mt:If>
</author>
<mt:EntryCategories><category term="<$mt:CategoryLabel escape="xml"$>" scheme="http://www.sixapart.com/ns/types#category" />
</mt:EntryCategories>
<content type="html" xml:lang="<$mt:BlogLanguage ietf="1"$>" xml:base="<$mt:BlogURL escape="xml"$>">
<$mt:EntryBody escape="xml"$>
<$mt:EntryMore escape="xml"$>
</content>
</entry>
</mt:Entries>
</feed>
| MTML | 2 | kanpapa/mt-theme-rimo | themes/rimo/templates/feed_recent.mtml | [
"MIT"
]
|
defmodule FrameworkBenchmarks.Handlers.PlainText do
@moduledoc """
This is the handle for the /plaintext route
"""
def handle(conn) do
conn
|> Plug.Conn.put_resp_content_type("text/plain")
|> Plug.Conn.send_resp(200, "Hello, World!")
end
end
| Elixir | 4 | xsoheilalizadeh/FrameworkBenchmarks | frameworks/Elixir/plug/lib/framework_benchmarks/handlers/plain-text.ex | [
"BSD-3-Clause"
]
|
FROM ubuntu:18.04
# Enable source repositories, which are disabled by default on Ubuntu >= 18.04
RUN sed -i 's/^# deb-src/deb-src/' /etc/apt/sources.list
COPY scripts/cross-apt-packages.sh /tmp/
RUN bash /tmp/cross-apt-packages.sh
# Required for cross-build gcc
RUN apt-get update && \
apt-get install -y --no-install-recommends \
libgmp-dev \
libmpfr-dev \
libmpc-dev
COPY scripts/illumos-toolchain.sh /tmp/
RUN bash /tmp/illumos-toolchain.sh x86_64 sysroot
RUN bash /tmp/illumos-toolchain.sh x86_64 binutils
RUN bash /tmp/illumos-toolchain.sh x86_64 gcc
COPY scripts/sccache.sh /scripts/
RUN sh /scripts/sccache.sh
COPY scripts/cmake.sh /scripts/
RUN /scripts/cmake.sh
ENV \
AR_x86_64_unknown_illumos=x86_64-illumos-ar \
CC_x86_64_unknown_illumos=x86_64-illumos-gcc \
CXX_x86_64_unknown_illumos=x86_64-illumos-g++
ENV HOSTS=x86_64-unknown-illumos
ENV RUST_CONFIGURE_ARGS --enable-extended --disable-docs
ENV SCRIPT python3 ../x.py dist --host $HOSTS --target $HOSTS
| Dockerfile | 3 | mbc-git/rust | src/ci/docker/host-x86_64/dist-x86_64-illumos/Dockerfile | [
"ECL-2.0",
"Apache-2.0",
"MIT-0",
"MIT"
]
|
static const uint32_t in_default_val[100] = {
0xbcf53bc7, 0x3e91ab21, 0x3f2b2c74, 0x3e9d1f6b,
0x3e5b1bd3, 0x3f09c922, 0x3f4428f3, 0x3f103015,
0x3de599a3, 0xbca6e341, 0xbd2e1214, 0xbc87968f,
0xbe2cd052, 0xbf1c68c2, 0xbf55dea8, 0xbf235faa,
0xbea2e717, 0xbe899a6f, 0xbecd218b, 0xbec298f8,
0xbcad67b9, 0x3f090e58, 0x3f0c6656, 0x3ed75328,
0x3ec7bfac, 0x3f0fd65f, 0x3f2f9107, 0x3f093857,
0x3e232195, 0x3b76b844, 0x3e1fd2af, 0x3e3a5613,
0xbebca9b8, 0xbf1afad6, 0xbf47003b, 0xbf0e7990,
0xbee72a14, 0xbea58e41, 0xbeed2177, 0xbeb4be03,
0x3e331239, 0x3ecc8678, 0x3ecf0e59, 0x3e825328,
0x3ee94d66, 0x3f5c5a80, 0x3f15de38, 0x3f11777b,
0xbdd445fa, 0xbd36f3c5, 0x3e8dc1e1, 0xbe39fdc0,
0xbec37d10, 0xbf1e1615, 0xbf025288, 0xbee8c1b9,
0xbe852759, 0xbf11d5a3, 0xbe896cfa, 0xbe889cbe,
0x3e0cc29d, 0x3eedb816, 0x3e41a0d4, 0x3f015b2f,
0x3eecfa87, 0x3f4f110d, 0x3f33430d, 0x3f15c127,
0xbda65fed, 0xbb8b684e, 0xbc6d801f, 0xbe990fb9,
0xbf1e907d, 0xbf800000, 0xbf0778bf, 0xbed8543d,
0xbe605a08, 0xbef34813, 0xbeae7557, 0xbe56a0a4,
0x3e98a4c2, 0x3eee73cf, 0x3eb95539, 0x3e0baadd,
0x3ebf8ced, 0x3f09388c, 0x3f1fa2cb, 0x3ede4653,
0x3df02037, 0x3e4c6717, 0xbda54f14, 0xbe0e8987,
0xbf1739b0, 0xbf2d733a, 0xbef4a308, 0xbe930645,
0xbef3abd9, 0xbe380acc, 0xbeaf6a8e, 0x3dc64ce8
};
static const uint32_t in_default_coeff[15] = {
0x3ca3d70a, 0xbcb10bd8, 0x3ca3d70a, 0x3fdf3cc8,
0xbf4f5c29, 0x3f800000, 0xbfcbcc26, 0x3f800000,
0x3fdf995a, 0xbf70ded3, 0x3f800000, 0xbffff2e5,
0x3f800000, 0x3ffa9104, 0xbf75dcc6
};
static const uint32_t ref_default[100] = {
0xba1cf30f, 0x3bab3ad5, 0x3c896159, 0x3cca4ba9,
0x3d34eb78, 0x3d98474d, 0x3ddffb9d, 0x3e156293,
0x3e3a8e79, 0x3e5f64c3, 0x3e78947c, 0x3e80a895,
0x3e72c5d4, 0x3e46d3d5, 0x3e00f3a8, 0x3d10671d,
0xbd981b15, 0xbe482bae, 0xbea111c9, 0xbed53e19,
0xbef9a5b1, 0xbf0455b7, 0xbf015e56, 0xbee35bf0,
0xbeac923f, 0xbe4645ec, 0xbcf45f15, 0x3e0ba33a,
0x3e93132a, 0x3ed22a4d, 0x3efc9a2d, 0x3f06219a,
0x3efd2b97, 0x3ed6e1a9, 0x3e997914, 0x3e1cc471,
0xbc2fb69f, 0xbe2e418b, 0xbea13de4, 0xbed9ebfe,
0xbefb45e4, 0xbf0361b8, 0xbef9d0de, 0xbed6d520,
0xbe9ed526, 0xbe2fd0ef, 0xbcdfb793, 0x3e005ddd,
0x3e82009b, 0x3ebb3497, 0x3ee2b0ff, 0x3eef06e1,
0x3ee7aa3f, 0x3ec8c4f8, 0x3e98116d, 0x3e2f24f4,
0x3d045273, 0xbdece40e, 0xbe77e9f8, 0xbeb39607,
0xbed5f3cd, 0xbee2a9b4, 0xbedbd167, 0xbeb9586a,
0xbe8755f2, 0xbe06b345, 0x3c3d616e, 0x3e23af4b,
0x3e92e4fb, 0x3ecb49ba, 0x3eee1fbb, 0x3ef7f6fe,
0x3ee8e8ca, 0x3ebfb232, 0x3e867787, 0x3de8c715,
0xbd393c7a, 0xbe540d98, 0xbeb00880, 0xbee62ce6,
0xbf012e85, 0xbf02c62a, 0xbeedc2b5, 0xbebe2501,
0xbe71f4af, 0xbda79e2f, 0x3da2110d, 0x3e6975ac,
0x3eb39d35, 0x3ee29319, 0x3ef7a07e, 0x3ef72790,
0x3edc7ab2, 0x3eb051b3, 0x3e6c7fec, 0x3dcda67d,
0xbd1d8fcf, 0xbe23b57d, 0xbe8a1b97, 0xbeaeb3a0
};
static const uint32_t in_rand_coeff[330] = {
0x3ca3d70a, 0x3c4537e4, 0x3ca3d70a, 0x3e4be208,
0xbc499fe1, 0x3f800000, 0x3ff1cdec, 0x3f800000,
0xbf731e19, 0xbe725f8a, 0x3f800000, 0x3f84c2d6,
0x3f800000, 0xbefd9b34, 0xbea085a6, 0x3ca3d70a,
0xbcc888ab, 0x3ca3d70a, 0xbef1a105, 0xbda88ab3,
0x3f800000, 0xbf5ff8c6, 0x3f800000, 0xbf5d20a3,
0xbe55d122, 0x3f800000, 0xbf53bc2c, 0x3f800000,
0xbfb133d9, 0xbef66cd1, 0x3ca3d70a, 0x3d1bc833,
0x3ca3d70a, 0xbd0355e1, 0xb98bf1f3, 0x3f800000,
0x3fbccc2c, 0x3f800000, 0x3be8a4a2, 0xbac68db9,
0x3f800000, 0xbff19732, 0x3f800000, 0x3f662d36,
0xbec4b721, 0x3ca3d70a, 0xbc622b92, 0x3ca3d70a,
0xbdf0a4fa, 0xbc45833d, 0x3f800000, 0x3f8a01cb,
0x3f800000, 0xbe04cb2b, 0xbd079165, 0x3f800000,
0x3f7f1d51, 0x3f800000, 0xbe46fc7e, 0xbd1c8ce6,
0x3f800000, 0xbf2feada, 0x3f800000, 0x3d82109a,
0xbd2f34ea, 0x3f800000, 0x3fda1cdb, 0x3f800000,
0xbf0c54e7, 0xbdaca1a4, 0x3f800000, 0xbf20eeb1,
0x3f800000, 0x3e4c4735, 0xbe206b56, 0x3f800000,
0xbffba4b8, 0x3f800000, 0x3f2dc202, 0xbe2d66a5,
0x3f800000, 0x3d590906, 0x3f800000, 0xbeb1fce7,
0xbef8b263, 0x3ca3d70a, 0x3d236345, 0x3ca3d70a,
0xbd72a875, 0xba66091c, 0x3f800000, 0x3ffd6291,
0x3f800000, 0xbd3a9e51, 0xbc20c6a3, 0x3f800000,
0xbe6ebc3f, 0x3f800000, 0x3e6b5871, 0xbc84f62a,
0x3f800000, 0xbee06e37, 0x3f800000, 0x3e516193,
0xbcaf42f2, 0x3f800000, 0x3fd1ee78, 0x3f800000,
0xbeaea6cd, 0xbdb3237b, 0x3f800000, 0xbf46d727,
0x3f800000, 0x3e8fe7d1, 0xbe2063e0, 0x3f800000,
0xbfccd506, 0x3f800000, 0x3f583767, 0xbe448ba6,
0x3f800000, 0x3d23e128, 0x3f800000, 0xbe57c043,
0xbec0e285, 0x3ca3d70a, 0x3d1ee0b2, 0x3ca3d70a,
0x3dba8964, 0xbb22da4f, 0x3f800000, 0x3ff302fc,
0x3f800000, 0x3dbf4b31, 0xbbc33824, 0x3f800000,
0x3fea7ef0, 0x3f800000, 0xbd62fbcc, 0xbbcb1d49,
0x3f800000, 0x3fe76d9e, 0x3f800000, 0xbce05a56,
0xbcb1908b, 0x3f800000, 0x3f83ef38, 0x3f800000,
0x3ea2e05d, 0xbce9f862, 0x3f800000, 0x3fed8262,
0x3f800000, 0xbf1ce115, 0xbde09de6, 0x3f800000,
0xbfc616d5, 0x3f800000, 0x3f1f602c, 0xbe017d15,
0x3f800000, 0xbd394a0c, 0x3f800000, 0x3e797d54,
0xbea4f84a, 0x3ca3d70a, 0x3d1ef3ca, 0x3ca3d70a,
0x3dab83b1, 0xbb80940d, 0x3f800000, 0x3fe3ff5d,
0x3f800000, 0xbe40cb7e, 0xbc4713ed, 0x3f800000,
0x3fb58cab, 0x3f800000, 0x3e3fda77, 0xbd4c7967,
0x3f800000, 0x3ff663b4, 0x3f800000, 0xbee73997,
0xbd702bc4, 0x3f800000, 0x3f46d974, 0x3f800000,
0x3f0d62f9, 0xbd9f9f9a, 0x3f800000, 0x3f97994d,
0x3f800000, 0xbecb7be9, 0xbdcfccc5, 0x3f800000,
0x3fff7741, 0x3f800000, 0xbf2553d1, 0xbdd58c04,
0x3f800000, 0xbfee8488, 0x3f800000, 0x3f256d44,
0xbde819e8, 0x3f800000, 0xbfe05432, 0x3f800000,
0x3f21b227, 0xbe0938f1, 0x3f800000, 0xbfcfc90b,
0x3f800000, 0x3f1a43cf, 0xbe26c338, 0x3f800000,
0xbffd2247, 0x3f800000, 0x3f7bcc55, 0xbe80ab96,
0x3ca3d70a, 0xbc92a917, 0x3ca3d70a, 0xbd0efc03,
0xbb4e25cb, 0x3f800000, 0x3eb91cd3, 0x3f800000,
0xbdb8c052, 0xbb9614fd, 0x3f800000, 0x3f4497fc,
0x3f800000, 0xbda3be0c, 0xbc4a203f, 0x3f800000,
0x3f6fafa5, 0x3f800000, 0xbe36cc00, 0xbd0be585,
0x3f800000, 0x3e6d7857, 0x3f800000, 0xbcd49373,
0xbd74a00a, 0x3f800000, 0xbf856b67, 0x3f800000,
0x3f3a92ed, 0xbe08fde1, 0x3f800000, 0x3f7527c5,
0x3f800000, 0xbf2e9558, 0xbe365144, 0x3f800000,
0xbfafe47a, 0x3f800000, 0x3f7bfc50, 0xbe9731c6,
0x3f800000, 0xbfc0461f, 0x3f800000, 0x3f8a3254,
0xbea3ebb3, 0x3f800000, 0xbe347d51, 0x3f800000,
0x3d9b9805, 0xbeac1463, 0x3f800000, 0xbfffa3b5,
0x3f800000, 0x3fa154f6, 0xbed00ffb, 0x3ca3d70a,
0x3b077d7d, 0x3ca3d70a, 0xbd5fd22f, 0xba9bb410,
0x3f800000, 0x3d82227c, 0x3f800000, 0x3d1d349f,
0xbb18b85a, 0x3f800000, 0xbfa78858, 0x3f800000,
0x3e4e3fb8, 0xbc265e7a, 0x3f800000, 0x3ea61d65,
0x3f800000, 0xbe4f335c, 0xbd1022a9, 0x3f800000,
0x3e3b9525, 0x3f800000, 0xbd18ebdd, 0xbd104721,
0x3f800000, 0x3ebe215e, 0x3f800000, 0xbed4f50b,
0xbd385402, 0x3f800000, 0x3f1ee0a2, 0x3f800000,
0xbee0b2f1, 0xbd49f354, 0x3f800000, 0x3fed7316,
0x3f800000, 0xbf055cd5, 0xbdad3318, 0x3f800000,
0xbda8ed71, 0x3f800000, 0x3d227c25, 0xbdd6b75f,
0x3f800000, 0x3f35da0f, 0x3f800000, 0xbeb5293e,
0xbe7b0caf, 0x3f800000, 0x3ffb0d63, 0x3f800000,
0xbfa9d672, 0xbeefb2fb
};
static const uint16_t in_rand_config[18] = {
0x0003, 0x0003, 0x0003, 0x0008, 0x0003, 0x000B, 0x0008, 0x0003,
0x0008, 0x0008, 0x0008, 0x000B, 0x000B, 0x0003, 0x000B, 0x0008,
0x000B, 0x000B
};
static const uint32_t in_rand_mono_val[66] = {
0x3f800000, 0x3f034193, 0xbf286d19, 0xbee540a5,
0x3e26e33b, 0xbf2064a6, 0x3f70f85c, 0xbe8b287e,
0xbf800000, 0x3e7579ed, 0xbebc4f72, 0xbf2b4a68,
0xbf104907, 0x3ca0f100, 0xbf3dd896, 0xbf2fde5c,
0x3f21e0b3, 0x3f1f8754, 0x3f800000, 0xbe09bfba,
0xbd4457e9, 0x3f6238c4, 0xbf4004c9, 0xbf800000,
0x3f74d2d6, 0x3f800000, 0x3e92c4b5, 0x3e8da5fa,
0xbd2a1c65, 0xbdb14ac5, 0xbe8b6825, 0xbd99fec1,
0xbe0b24db, 0x3e03246e, 0xbeca6126, 0xbefcb439,
0x3e779dad, 0xbb74140c, 0xbedf52bc, 0x3f800000,
0xbea21297, 0xbeb65edc, 0xbf0988f0, 0x3ecd5598,
0x3e0b2e78, 0xbeafcb0f, 0xbf800000, 0xbd643694,
0xbe1980a7, 0x3ee174e4, 0x3e82ae0e, 0x3e78118b,
0xbeb45ecd, 0x3f6c0927, 0xbf800000, 0xbe59d39d,
0xbeedce8c, 0xbf382234, 0x3ee9bde9, 0xbf800000,
0x3df0c786, 0x3f063be1, 0xbebccb5b, 0xbea9671c,
0xbe0c57d3, 0x3dac7612
};
static const uint32_t ref_rand_mono[66] = {
0x3ca3d70a, 0x3d64f6e7, 0x3d9543cc, 0xbc12b8bc,
0x3d5c7346, 0xbe45faed, 0x3f0378e3, 0xbf8df559,
0x40009718, 0xc0464621, 0x40829547, 0xbc5b4085,
0xbd2fb6ba, 0xbd29b07c, 0x3b2c3898, 0x3c7bc19a,
0xba7fed63, 0x3d160196, 0x3dbed0f0, 0x3d383dcf,
0xbda8f659, 0xbdf54d25, 0xbc75c8af, 0xbc43f9a7,
0xbbf2b03c, 0x3ca3d70a, 0x3d9b78d8, 0x3e25fff0,
0x3e9a34d1, 0x3ef32f44, 0x3f1f1b8a, 0x3f3f04e3,
0x3f550dde, 0x3b27dcb6, 0x3c8569dd, 0x3cbcd362,
0xbdf73a80, 0xbf349c5a, 0xbff3c2f2, 0xc05b3d2e,
0xc092f716, 0xc09bce97, 0xc08a0adb, 0xc06806b5,
0x3b322700, 0x3c12386f, 0xbcd23f35, 0xba920e74,
0xbb11c874, 0x3b9ab905, 0xbc52df41, 0x3d38eee2,
0xbd5ba88b, 0x3e2f01be, 0xbe3955ce, 0xbb8b68b7,
0xbc886a6a, 0xbd80d8c2, 0xbe074ee7, 0xbea20e09,
0xbeee9221, 0xbf58c7f2, 0xbf7e3707, 0xbfb6e6a5,
0xbfbd1fa2, 0xbfd7c425
};
static const uint32_t in_rand_stereo_val[132] = {
0x3f800000, 0x3f0fca96, 0x3f034193, 0x3e8b72e4,
0xbf286d19, 0xbf800000, 0xbee540a5, 0xbca5c85b,
0x3e26e33b, 0xbf09ebf0, 0xbf2064a6, 0xbe5a3c55,
0x3f70f85c, 0xbef1939c, 0xbe8b287e, 0xbec023d9,
0xbf800000, 0xbd8c5657, 0x3e7579ed, 0xbf800000,
0xbebc4f72, 0xbe8f8384, 0xbf2b4a68, 0xbece16c6,
0xbf104907, 0xbea0cf77, 0x3ca0f100, 0x3f35f9fe,
0xbf3dd896, 0x3f800000, 0xbf2fde5c, 0xbe98e9f8,
0x3f21e0b3, 0xbf403348, 0x3f1f8754, 0xbda8470e,
0x3f800000, 0xbefdafc7, 0xbe09bfba, 0xbecd3997,
0xbd4457e9, 0x3edb9052, 0x3f6238c4, 0x3e826ac4,
0xbf4004c9, 0xbf800000, 0xbf800000, 0x3e1d94a4,
0x3f74d2d6, 0x3e08a537, 0x3f800000, 0xbf800000,
0x3e92c4b5, 0x3f3d9883, 0x3e8da5fa, 0xbe0a81cb,
0xbd2a1c65, 0x3f668a53, 0xbdb14ac5, 0x3f0e28e4,
0xbe8b6825, 0x3f46d398, 0xbd99fec1, 0x3f7ba66b,
0xbe0b24db, 0x3f6ec2a8, 0x3e03246e, 0xbdbdb8aa,
0xbeca6126, 0xbef74e68, 0xbefcb439, 0xbf3cb8bb,
0x3e779dad, 0xbebfb60b, 0xbb74140c, 0x3d9ab9ce,
0xbedf52bc, 0x3bdc4632, 0x3f800000, 0xbf270982,
0xbea21297, 0xbef701ef, 0xbeb65edc, 0xbf800000,
0xbf0988f0, 0xbe1db24a, 0x3ecd5598, 0xbd837835,
0x3e0b2e78, 0xbe536194, 0xbeafcb0f, 0xbf800000,
0xbf800000, 0x3f4f8d9f, 0xbd643694, 0x3f800000,
0xbe1980a7, 0xbf319706, 0x3ee174e4, 0xbe07352a,
0x3e82ae0e, 0x3f70b3d7, 0x3e78118b, 0x3f15a0f3,
0xbeb45ecd, 0x3f0addc1, 0x3f6c0927, 0xbe956eb6,
0xbf800000, 0x3ed77e8b, 0xbe59d39d, 0xbe0154f6,
0xbeedce8c, 0xbf41fcf3, 0xbf382234, 0x3e9d1fa7,
0x3ee9bde9, 0x3d987640, 0xbf800000, 0x3f4fcc49,
0x3df0c786, 0xbbf43bd0, 0x3f063be1, 0x3ec2ae50,
0xbebccb5b, 0xbe86fecd, 0xbea9671c, 0xbeb5ea0e,
0xbe0c57d3, 0xbe8e0803, 0x3dac7612, 0x3f800000
};
static const uint32_t ref_rand_stereo[132] = {
0x3ca3d70a, 0x3c380d8d, 0x3d64f6e7, 0x3cfea6d3,
0x3d9543cc, 0x3ce238d1, 0xbc12b8bc, 0xb9d433a7,
0x3d5c7346, 0xbc0b195a, 0xbe45faed, 0x3d49a9e3,
0x3f0378e3, 0xbe283bef, 0xbf8df559, 0x3ec1e4fc,
0x40009718, 0xbf2e90e4, 0xc0464621, 0x3f7b7ebf,
0x40829547, 0xbf92c72f, 0xbc5b4085, 0xbc03e59d,
0xbd2fb6ba, 0xbccf519f, 0xbd29b07c, 0xbc1fed41,
0x3b2c3898, 0x3d834c90, 0x3c7bc19a, 0x3dc1da96,
0xba7fed63, 0xbc8bd574, 0x3d160196, 0xbe11a765,
0x3dbed0f0, 0xbdd7e7aa, 0x3d383dcf, 0x3ca34d54,
0xbda8f659, 0x3da93489, 0xbdf54d25, 0x3dabd24b,
0xbc75c8af, 0xbca3d70a, 0xbc43f9a7, 0x3c6203de,
0xbbf2b03c, 0xbd473d5d, 0x3ca3d70a, 0xbca3d70a,
0x3d9b78d8, 0xbd62ca5f, 0x3e25fff0, 0xbdb28ef9,
0x3e9a34d1, 0xbe091d90, 0x3ef32f44, 0xbe0b278f,
0x3f1f1b8a, 0xbd36f73d, 0x3f3f04e3, 0x3de23de8,
0x3f550dde, 0x3ec4415f, 0x3b27dcb6, 0xbaf2d7e4,
0x3c8569dd, 0xbcde7c77, 0x3cbcd362, 0xbe3c3722,
0xbdf73a80, 0xbf44f1b4, 0xbf349c5a, 0xc00fa974,
0xbff3c2f2, 0xc09c138c, 0xc05b3d2e, 0xc10426c0,
0xc092f716, 0xc136328f, 0xc09bce97, 0xc1594e8b,
0xc08a0adb, 0xc175a67f, 0xc06806b5, 0xc18ffbfd,
0x3b322700, 0xbb8748b1, 0x3c12386f, 0xbd3427cc,
0xbcd23f35, 0xbe1b9858, 0xba920e74, 0x3ca3d70a,
0xbb11c874, 0xbce36786, 0x3b9ab905, 0x3df0fd01,
0xbc52df41, 0xbe106c81, 0x3d38eee2, 0x3eb7d7e5,
0xbd5ba88b, 0xbeae2fd1, 0x3e2f01be, 0x3f2e3967,
0xbe3955ce, 0xbef46dc6, 0xbb8b68b7, 0xbb258b78,
0xbc886a6a, 0xbc9ff714, 0xbd80d8c2, 0xbd21368c,
0xbe074ee7, 0xbe05a51d, 0xbea20e09, 0xbe21933c,
0xbeee9221, 0xbeac37ca, 0xbf58c7f2, 0xbe837de3,
0xbf7e3707, 0xbec73d0a, 0xbfb6e6a5, 0xbdda8855,
0xbfbd1fa2, 0xbd9cbe2c, 0xbfd7c425, 0x3e8eedc2
};
| Max | 2 | maxvankessel/zephyr | tests/lib/cmsis_dsp/filtering/src/biquad_f32.pat | [
"Apache-2.0"
]
|
[Files]
Source: "RevitPythonShell\bin\Release\2015\PythonConsoleControl.dll"; DestDir: "{app}"; Flags: replacesameversion
Source: "RevitPythonShell\bin\Release\2015\RevitPythonShell.dll"; DestDir: "{app}"; Flags: replacesameversion
Source: "RevitPythonShell\bin\Release\2015\RpsRuntime.dll"; DestDir: "{app}"; Flags: replacesameversion
Source: "RevitPythonShell\bin\Release\2015\RevitPythonShell.addin"; DestDir: "{userappdata}\Autodesk\Revit\Addins\2015"; Flags: replacesameversion
Source: "RevitPythonShell\bin\Release\2015\ICSharpCode.AvalonEdit.dll"; DestDir: "{app}"
Source: "RevitPythonShell\bin\Release\2015\IronPython.dll"; DestDir: "{app}"
Source: "RevitPythonShell\bin\Release\2015\IronPython.Modules.dll"; DestDir: "{app}"
Source: "RevitPythonShell\bin\Release\2015\Microsoft.Scripting.Metadata.dll"; DestDir: "{app}"
Source: "RevitPythonShell\bin\Release\2015\Microsoft.Dynamic.dll"; DestDir: "{app}"
Source: "RevitPythonShell\bin\Release\2015\Microsoft.Scripting.dll"; DestDir: "{app}"
Source: "RevitPythonShell\bin\Release\2015\DefaultConfig\RevitPythonShell.xml"; DestDir: "{userappdata}\RevitPythonShell\2015"; Flags: onlyifdoesntexist
Source: "RevitPythonShell\bin\Release\2015\DefaultConfig\init.py"; DestDir: {userappdata}\RevitPythonShell\2015; Flags: confirmoverwrite;
Source: "RevitPythonShell\bin\Release\2015\DefaultConfig\startup.py"; DestDir: {userappdata}\RevitPythonShell\2015; Flags: confirmoverwrite;
[code]
{ HANDLE INSTALL PROCESS STEPS }
procedure CurStepChanged(CurStep: TSetupStep);
var
AddInFilePath: String;
LoadedFile : TStrings;
AddInFileContents: String;
ReplaceString: String;
SearchString: String;
begin
if CurStep = ssPostInstall then
begin
AddinFilePath := ExpandConstant('{userappdata}\Autodesk\Revit\Addins\2015\RevitPythonShell.addin');
LoadedFile := TStringList.Create;
SearchString := 'Assembly>RevitPythonShell.dll<';
ReplaceString := 'Assembly>' + ExpandConstant('{app}') + '\RevitPythonShell.dll<';
try
LoadedFile.LoadFromFile(AddInFilePath);
AddInFileContents := LoadedFile.Text;
{ Only save if text has been changed. }
if StringChangeEx(AddInFileContents, SearchString, ReplaceString, True) > 0 then
begin;
LoadedFile.Text := AddInFileContents;
LoadedFile.SaveToFile(AddInFilePath);
end;
finally
LoadedFile.Free;
end;
end;
end;
[Setup]
AppName=RevitPythonShell for Autodesk Revit 2015
AppVerName=RevitPythonShell for Autodesk Revit 2015
RestartIfNeededByRun=false
DefaultDirName={pf32}\RevitPythonShell\2015
OutputBaseFilename=Setup_RevitPythonShell_2015
ShowLanguageDialog=auto
FlatComponentsList=false
UninstallFilesDir={app}\Uninstall
UninstallDisplayName=RevitPythonShell for Autodesk Revit 2015
AppVersion=2015.0
VersionInfoVersion=2015.0
VersionInfoDescription=RevitPythonShell for Autodesk Revit 2015
VersionInfoTextVersion=RevitPythonShell for Autodesk Revit 2015
| Inno Setup | 4 | PavelAltynnikov/revitpythonshell | Setup_RevitPythonShell_2015.iss | [
"MIT"
]
|
import React from 'react';
import path from 'path';
import fs from 'fs';
import { remote } from 'electron';
import { Flexbox } from 'nylas-component-kit';
import displayedKeybindings from './keymaps/displayed-keybindings';
import CommandItem from './keymaps/command-item';
class PreferencesKeymaps extends React.Component {
static displayName = 'PreferencesKeymaps';
static propTypes = {
config: React.PropTypes.object,
};
constructor() {
super();
this.state = {
templates: [],
bindings: this._getStateFromKeymaps(),
};
this._loadTemplates();
}
componentDidMount() {
this._disposable = NylasEnv.keymaps.onDidReloadKeymap(() => {
this.setState({bindings: this._getStateFromKeymaps()});
});
}
componentWillUnmount() {
this._disposable.dispose();
}
_getStateFromKeymaps() {
const bindings = {};
for (const section of displayedKeybindings) {
for (const [command] of section.items) {
bindings[command] = NylasEnv.keymaps.getBindingsForCommand(command) || [];
}
}
return bindings;
}
_loadTemplates() {
const templatesDir = path.join(NylasEnv.getLoadSettings().resourcePath, 'keymaps', 'templates');
fs.readdir(templatesDir, (err, files) => {
if (!files || !(files instanceof Array)) return;
let templates = files.filter((filename) => {
return path.extname(filename) === '.json';
});
templates = templates.map((filename) => {
return path.parse(filename).name;
});
this.setState({templates: templates});
});
}
_onShowUserKeymaps() {
const keymapsFile = NylasEnv.keymaps.getUserKeymapPath();
if (!fs.existsSync(keymapsFile)) {
fs.writeFileSync(keymapsFile, '{}');
}
remote.shell.showItemInFolder(keymapsFile);
}
_onDeleteUserKeymap() {
const chosen = remote.dialog.showMessageBox(NylasEnv.getCurrentWindow(), {
type: 'info',
message: "Are you sure?",
detail: "Delete your custom key bindings and reset to the template defaults?",
buttons: ['Cancel', 'Reset'],
});
if (chosen === 1) {
const keymapsFile = NylasEnv.keymaps.getUserKeymapPath();
fs.writeFileSync(keymapsFile, '{}');
}
}
_renderBindingsSection = (section) => {
return (
<section key={`section-${section.title}`}>
<div className="shortcut-section-title">{section.title}</div>
{
section.items.map(([command, label]) => {
return (
<CommandItem
key={command}
command={command}
label={label}
bindings={this.state.bindings[command]}
/>
);
})
}
</section>
);
}
render() {
return (
<div className="container-keymaps">
<section>
<Flexbox className="container-dropdown">
<div>Shortcut set:</div>
<div className="dropdown">
<select
style={{margin: 0}}
tabIndex={-1}
value={this.props.config.get('core.keymapTemplate')}
onChange={(event) => this.props.config.set('core.keymapTemplate', event.target.value)}
>
{this.state.templates.map((template) => {
return <option key={template} value={template}>{template}</option>
})}
</select>
</div>
<div style={{flex: 1}} />
<button className="btn" onClick={this._onDeleteUserKeymap}>Reset to Defaults</button>
</Flexbox>
<p>
You can choose a shortcut set to use keyboard shortcuts of familiar email clients.
To edit a shortcut, click it in the list below and enter a replacement on the keyboard.
</p>
{displayedKeybindings.map(this._renderBindingsSection)}
</section>
<section>
<h2>Customization</h2>
<p>You can manage your custom shortcuts directly by editing your shortcuts file.</p>
<button className="btn" onClick={this._onShowUserKeymaps}>Edit custom shortcuts</button>
</section>
</div>
);
}
}
export default PreferencesKeymaps;
| JSX | 5 | cnheider/nylas-mail | packages/client-app/internal_packages/preferences/lib/tabs/preferences-keymaps.jsx | [
"MIT"
]
|
# Copyright 2019 gRPC authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# This is for the gRPC build system. This isn't intended to be used outsite of
# the BUILD file for gRPC. It contains the mapping for the template system we
# use to generate other platform's build system files.
#
# Please consider that there should be a high bar for additions and changes to
# this file.
# Each rule listed must be re-written for Google's internal build system, and
# each change must be ported from one to the other.
#
load("@rules_proto//proto:defs.bzl", "proto_library")
load(
"//bazel:generate_objc.bzl",
"generate_objc",
"generate_objc_hdrs",
"generate_objc_non_arc_srcs",
"generate_objc_srcs",
)
def proto_library_objc_wrapper(
name,
srcs,
deps = [],
use_well_known_protos = False):
"""proto_library for adding dependencies to google/protobuf protos
use_well_known_protos - ignored in open source version
"""
proto_library(
name = name,
srcs = srcs,
deps = deps,
)
def grpc_objc_examples_library(
name,
srcs = [],
hdrs = [],
textual_hdrs = [],
data = [],
deps = [],
defines = [],
sdk_frameworks = [],
includes = []):
"""objc_library for testing, only works in //src/objective-c/exmaples
Args:
name: name of target
hdrs: public headers
srcs: all source files (.m)
textual_hdrs: private headers
data: any other bundle resources
defines: preprocessors
sdk_frameworks: sdks
includes: added to search path, always [the path to objc directory]
deps: dependencies
"""
native.objc_library(
name = name,
srcs = srcs,
hdrs = hdrs,
textual_hdrs = textual_hdrs,
data = data,
defines = defines,
includes = includes,
sdk_frameworks = sdk_frameworks,
deps = deps + [":RemoteTest"],
)
def grpc_objc_testing_library(
name,
srcs = [],
hdrs = [],
textual_hdrs = [],
data = [],
deps = [],
defines = [],
includes = []):
"""objc_library for testing, only works in //src/objective-c/tests
Args:
name: name of target
hdrs: public headers
srcs: all source files (.m)
textual_hdrs: private headers
data: any other bundle resources
defines: preprocessors
includes: added to search path, always [the path to objc directory]
deps: dependencies
"""
additional_deps = [
":RemoteTest",
"//src/objective-c:grpc_objc_client_internal_testing",
]
if not name == "TestConfigs":
additional_deps += [":TestConfigs"]
native.objc_library(
name = name,
hdrs = hdrs,
srcs = srcs,
textual_hdrs = textual_hdrs,
data = data,
defines = defines,
includes = includes,
deps = deps + additional_deps,
)
def local_objc_grpc_library(name, deps, testing = True, srcs = [], use_well_known_protos = False, **kwargs):
"""!!For local targets within the gRPC repository only!! Will not work outside of the repo
"""
objc_grpc_library_name = "_" + name + "_objc_grpc_library"
generate_objc(
name = objc_grpc_library_name,
srcs = srcs,
deps = deps,
use_well_known_protos = use_well_known_protos,
**kwargs
)
generate_objc_hdrs(
name = objc_grpc_library_name + "_hdrs",
src = ":" + objc_grpc_library_name,
)
generate_objc_non_arc_srcs(
name = objc_grpc_library_name + "_non_arc_srcs",
src = ":" + objc_grpc_library_name,
)
arc_srcs = None
if len(srcs) > 0:
generate_objc_srcs(
name = objc_grpc_library_name + "_srcs",
src = ":" + objc_grpc_library_name,
)
arc_srcs = [":" + objc_grpc_library_name + "_srcs"]
library_deps = ["@com_google_protobuf//:protobuf_objc"]
if testing:
library_deps += ["//src/objective-c:grpc_objc_client_internal_testing"]
else:
library_deps += ["//src/objective-c:proto_objc_rpc"]
native.objc_library(
name = name,
hdrs = [":" + objc_grpc_library_name + "_hdrs"],
non_arc_srcs = [":" + objc_grpc_library_name + "_non_arc_srcs"],
srcs = arc_srcs,
defines = [
"GPB_USE_PROTOBUF_FRAMEWORK_IMPORTS=0",
"GPB_GRPC_FORWARD_DECLARE_MESSAGE_PROTO=0",
],
includes = ["_generated_protos"],
deps = library_deps,
)
| Python | 5 | arghyadip01/grpc | src/objective-c/grpc_objc_internal_library.bzl | [
"Apache-2.0"
]
|
<html>
<body>
<pre>
A world of dew,
and within every dewdrop
a world of struggle
</pre>
</body>
</html>
| DIGITAL Command Language | 1 | kernelrich/h2o-3 | h2o-docs/src/front/assets/addthis/s7.addthis.com | [
"Apache-2.0"
]
|
Red [
Title: "Red Tips Widget"
Author: "Xie Qingtian"
File: %tips.red
Tabs: 4
Rights: "Copyright (C) 2014-2018 Red Foundation. All rights reserved."
License: {
Distributed under the Boost Software License, Version 1.0.
See https://github.com/red/red/blob/master/BSL-License.txt
}
]
tips!: make face! [
type: 'panel color: 0.0.128 offset: 0x0 size: 150x200
actors: object [
on-key-down: func [face [object!] event [event!]][
probe event/key
]
]
] | Red | 4 | 0xflotus/red | environment/console/GUI/tips.red | [
"BSL-1.0",
"BSD-3-Clause"
]
|
[[features.task-execution-and-scheduling]]
== Task Execution and Scheduling
In the absence of an `Executor` bean in the context, Spring Boot auto-configures a `ThreadPoolTaskExecutor` with sensible defaults that can be automatically associated to asynchronous task execution (`@EnableAsync`) and Spring MVC asynchronous request processing.
[TIP]
====
If you have defined a custom `Executor` in the context, regular task execution (that is `@EnableAsync`) will use it transparently but the Spring MVC support will not be configured as it requires an `AsyncTaskExecutor` implementation (named `applicationTaskExecutor`).
Depending on your target arrangement, you could change your `Executor` into a `ThreadPoolTaskExecutor` or define both a `ThreadPoolTaskExecutor` and an `AsyncConfigurer` wrapping your custom `Executor`.
The auto-configured `TaskExecutorBuilder` allows you to easily create instances that reproduce what the auto-configuration does by default.
====
The thread pool uses 8 core threads that can grow and shrink according to the load.
Those default settings can be fine-tuned using the `spring.task.execution` namespace, as shown in the following example:
[source,yaml,indent=0,subs="verbatim",configprops,configblocks]
----
spring:
task:
execution:
pool:
max-size: 16
queue-capacity: 100
keep-alive: "10s"
----
This changes the thread pool to use a bounded queue so that when the queue is full (100 tasks), the thread pool increases to maximum 16 threads.
Shrinking of the pool is more aggressive as threads are reclaimed when they are idle for 10 seconds (rather than 60 seconds by default).
A `ThreadPoolTaskScheduler` can also be auto-configured if need to be associated to scheduled task execution (using `@EnableScheduling` for instance).
The thread pool uses one thread by default and its settings can be fine-tuned using the `spring.task.scheduling` namespace, as shown in the following example:
[source,yaml,indent=0,subs="verbatim",configprops,configblocks]
----
spring:
task:
scheduling:
thread-name-prefix: "scheduling-"
pool:
size: 2
----
Both a `TaskExecutorBuilder` bean and a `TaskSchedulerBuilder` bean are made available in the context if a custom executor or scheduler needs to be created.
| AsciiDoc | 4 | techAi007/spring-boot | spring-boot-project/spring-boot-docs/src/docs/asciidoc/features/task-execution-and-scheduling.adoc | [
"Apache-2.0"
]
|
#!/bin/sh
# Copyright 2018 The Kubernetes Authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
echo "Generating self-signed cert"
mkdir -p /certs
openssl req -x509 -sha256 -nodes -days 365 -newkey rsa:2048 \
-keyout /certs/privateKey.key \
-out /certs/certificate.crt \
-subj "/C=UK/ST=Warwickshire/L=Leamington/O=OrgName/OU=IT Department/CN=example.com"
# If we're running on Windows, skip loading the Linux .so modules.
if [ "$(uname)" = "Windows_NT" ]; then
sed -i -E "s/^(load_module modules\/ndk_http_module.so;)$/#\1/" conf/nginx.conf
sed -i -E "s/^(load_module modules\/ngx_http_lua_module.so;)$/#\1/" conf/nginx.conf
sed -i -E "s/^(load_module modules\/ngx_http_lua_upstream_module.so;)$/#\1/" conf/nginx.conf
# NOTE(claudiub): on Windows, nginx will take the paths in the nginx.conf file as relative paths.
cmd /S /C "mklink /D C:\\openresty\\certs C:\\certs"
fi
echo "Starting nginx"
nginx -g "daemon off;"
| Shell | 4 | 767829413/kubernetes | test/images/echoserver/run.sh | [
"Apache-2.0"
]
|
{% set result = 'yes' | to_bool() %}
{% include 'jinja_filters/common.sls' %}
| SaltStack | 2 | byteskeptical/salt | tests/integration/files/file/base/jinja_filters/jinja_to_bool.sls | [
"Apache-2.0"
]
|
import {NgModule} from '@angular/core';
@NgModule({jit: true})
export class NoAotModule {
}
| TypeScript | 3 | John-Cassidy/angular | packages/compiler-cli/test/compliance/test_cases/r3_compiler_compliance/ng_modules/no_aot.ts | [
"MIT"
]
|
data {
int<lower=0> N;
int<lower=0> M;
array[N] int<lower=0, upper=1> y;
array[N] row_vector[M] x;
}
parameters {
vector[M] beta;
}
model {
for (m in 1 : M)
beta[m] ~ cauchy(0.0, 2.5);
for (n in 1 : N)
y[n] ~ bernoulli(inv_logit(x[n] * beta));
}
| Stan | 4 | sthagen/stan-dev-stan | src/test/test-models/performance/logistic.stan | [
"CC-BY-3.0",
"BSD-3-Clause"
]
|
--TEST--
"Reference Unpacking - From Functions" list()
--FILE--
<?php
$arr = [1, 2];
function no_ref($a) {
return $a;
}
function no_ref_by_ref(&$a) {
return $a;
}
function &ref_return(&$a) {
return $a;
}
function &ref_return_global() {
global $arr;
return $arr;
}
$a = [1, 2];
[&$var] = no_ref($a);
var_dump($var);
var_dump($a);
$a = [1, 2];
[&$var] = no_ref_by_ref($a);
var_dump($var);
var_dump($a);
$a = [1, 2];
[&$var] = ref_return($a);
var_dump($var);
var_dump($a);
[,&$var] = ref_return_global();
var_dump($var);
var_dump($arr);
?>
--EXPECTF--
Notice: Attempting to set reference to non referenceable value in %s on line %d
int(1)
array(2) {
[0]=>
int(1)
[1]=>
int(2)
}
Notice: Attempting to set reference to non referenceable value in %s on line %d
int(1)
array(2) {
[0]=>
int(1)
[1]=>
int(2)
}
int(1)
array(2) {
[0]=>
&int(1)
[1]=>
int(2)
}
int(2)
array(2) {
[0]=>
int(1)
[1]=>
&int(2)
}
| PHP | 3 | thiagooak/php-src | Zend/tests/list/list_reference_003.phpt | [
"PHP-3.01"
]
|
package com.baeldung.reducingIfElse;
public class AddCommand implements Command {
private int a;
private int b;
public AddCommand(int a, int b) {
this.a = a;
this.b = b;
}
@Override
public Integer execute() {
return a + b;
}
}
| Java | 4 | DBatOWL/tutorials | patterns/design-patterns-creational/src/main/java/com/baeldung/reducingIfElse/AddCommand.java | [
"MIT"
]
|
//
// Copyright (c) 2008, Brian Frank and Andy Frank
// Licensed under the Academic Free License version 3.0
//
// History:
// 25 Apr 08 Brian Frank Creation
//
**
** This represents a method return type which is
** always [this type]`docLang::Methods#thisReturns`.
**
const final class This
{
**
** Private constructor.
**
private new make() {}
} | Fantom | 3 | fanx-dev/fanx | library/sys/fan/lang/This.fan | [
"AFL-3.0"
]
|
"Ejemplos gráficos" by "Xavi C." (in spanish)
[una imagen para cada localidad]
Section 1 - Extensiones
Include Basic Screen Effects Sp by Emily Short.
Include Basic Help Menu SP by Emily Short.
Include Location Images by Emily Short.
Section 2 - Definición de gráficos y sonidos
The graphics background color is "$FFFFFF".
The graphics window pixel count is 200.
The graphics window position is g-above.
Section - Definicion de imagenes
Figure of Paisaje1 is the file "29985__151_a_4.jpg".
Part 2 - Localidades
Mundo is a room. The description is "Un mundo con imágenes." The room-illustration is Figure of Paisaje1.
Mar is south of mundo.
| Inform 7 | 3 | brtl-fhc/I7-Spanish | EJEMPLOS/XaviTutorial/Cap07_30.ni | [
"Artistic-2.0"
]
|
$onExternalInput
set dd 'DD Files' /base,elc_techs,elc_plants2020,elc_importexport,elc_dh-pipes,ind_techs,res_app_techs,res_techs,res_heatsav,sup_h2_chain,sup_bioref,
sup_biogasplants,tra_techs,elc_excessheat,ldc_wasteheat,elc_trade,syssettings,ind_demandproj,res_demandproj,elc_baseconstraints,
ind_baseconstraints,res_baseconstraints,tra_baseconstraints,sys_deliverycosts,elc_maxelcexports,elc_dh-pipesdata,sys_subannual_data,
elc_excessheat_pots,ind_ee-low,res_buildingstockproj,res_restrictheatsav,res_app_effproj,sup_northseaminingproj,sup-elc_renewablepotentials,
tra_xassumptions,elc_taxessubsidies,ind_taxessubsidies,res_taxessubsidies,tra_xtaxessubsidies,sup_taxessubsidies,ets-nets_emicoeff,
tra_demandproj,tra_minimum_shares,tra_ttb,tra_shiftpotential,tra_infrastructure_structure,sys_fuelinfrastructure,fuel_constraints,
tra_ev_share,sys_elc_ie,ldc_increaseddemand_varprofile,sup_fuelprice_bf18,sys_dr,vat,elc_ccs_techs,sys_carbon_budget,sys_carbon_budget_1-5,
sys_co2_2040_ship_ts,sys_co2_2040_ts,sys_co2_2050,tra_ev_share_aa,tra_ev_share_cb,tra_int_shipping,z_cb_no_tax,z_ccs_storagepotentials_cb_15,
z_ea_2018,z_ea_aa,z_ea_b,z_ea_dccc,z_ea_regeringen,z_ea_s/;
set offeps(dd) 'dd read under offeps' / sup_northseaminingproj /;
set scenddmap(scenario<,ddorder,dd) /
Alternativet.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.tra_ev_share_aa
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.z_ea_aa
)
CB_15.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_ccs_techs
17.elc_trade
18.syssettings
19.ind_demandproj
20.res_demandproj
21.elc_baseconstraints
22.ind_baseconstraints
23.res_baseconstraints
24.tra_baseconstraints
25.sys_deliverycosts
26.elc_maxelcexports
27.elc_dh-pipesdata
28.sys_subannual_data
29.elc_excessheat_pots
30.ind_ee-low
31.res_buildingstockproj
32.res_restrictheatsav
33.res_app_effproj
34.sup_northseaminingproj
35.sup-elc_renewablepotentials
36.tra_xassumptions
37.elc_taxessubsidies
38.ind_taxessubsidies
39.res_taxessubsidies
40.tra_xtaxessubsidies
41.sup_taxessubsidies
42.ets-nets_emicoeff
43.tra_demandproj
44.tra_minimum_shares
45.tra_ttb
46.tra_shiftpotential
47.tra_infrastructure_structure
48.sys_fuelinfrastructure
49.fuel_constraints
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.sys_carbon_budget_1-5
56.tra_ev_share_cb
57.z_cb_no_tax
58.z_ccs_storagepotentials_cb_15
)
CB_20.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.sys_elc_ie
50.ldc_increaseddemand_varprofile
51.sup_fuelprice_bf18
52.sys_dr
53.vat
54.sys_carbon_budget
55.tra_ev_share_cb
56.z_cb_no_tax
)
DCCC.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.tra_ev_share
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.z_cb_no_tax
56.z_ea_dccc
)
EA_2018.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.tra_ev_share
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.z_ea_2018
)
Fossilfri_2040.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.tra_ev_share
49.fuel_constraints
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.sys_co2_2040_ts
)
Fossilfri_2050.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.tra_ev_share
49.fuel_constraints
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_co2_2050
54.sys_dr
55.vat
)
Frozen_policy_scenarie.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.tra_ev_share
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
)
International_bunkering.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.tra_ev_share
49.fuel_constraints
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.tra_int_shipping
55.vat
)
International_bunkering_2040.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.tra_ev_share
49.fuel_constraints
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.tra_int_shipping
55.vat
56.sys_co2_2040_ship_ts
)
Radikale_Venstre.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.tra_ev_share
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.z_ea_b
)
Regeringen.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.tra_ev_share
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.z_ea_regeringen
)
Socialdemokratiet.(
1.base
2.elc_techs
3.elc_plants2020
4.elc_importexport
5.elc_dh-pipes
6.ind_techs
7.res_app_techs
8.res_techs
9.res_heatsav
10.sup_h2_chain
11.sup_bioref
12.sup_biogasplants
13.tra_techs
14.elc_excessheat
15.ldc_wasteheat
16.elc_trade
17.syssettings
18.ind_demandproj
19.res_demandproj
20.elc_baseconstraints
21.ind_baseconstraints
22.res_baseconstraints
23.tra_baseconstraints
24.sys_deliverycosts
25.elc_maxelcexports
26.elc_dh-pipesdata
27.sys_subannual_data
28.elc_excessheat_pots
29.ind_ee-low
30.res_buildingstockproj
31.res_restrictheatsav
32.res_app_effproj
33.sup_northseaminingproj
34.sup-elc_renewablepotentials
35.tra_xassumptions
36.elc_taxessubsidies
37.ind_taxessubsidies
38.res_taxessubsidies
39.tra_xtaxessubsidies
40.sup_taxessubsidies
41.ets-nets_emicoeff
42.tra_demandproj
43.tra_minimum_shares
44.tra_ttb
45.tra_shiftpotential
46.tra_infrastructure_structure
47.sys_fuelinfrastructure
48.fuel_constraints
49.tra_ev_share
50.sys_elc_ie
51.ldc_increaseddemand_varprofile
52.sup_fuelprice_bf18
53.sys_dr
54.vat
55.z_ea_s
) /;
set TimeSlice 'ALL_TS' /ANNUAL,R,S,F,W,RWD,RNW,SWD,SNW,FWD,FNW,WWD,WNW,RWDA,RWDC,RWDD,RWDB,RNWA,RNWC,RNWD,RNWB,SWDA,SWDC,SWDD,SWDB,SNWA,SNWC,SNWD,SNWB,FWDA,FWDC,FWDD,FWDB,FNWA,FNWC,FNWD,FNWB,WWDA,WWDC,WWDD,WWDB,WNWA,WNWC,WNWD,WNWB/;
set MILESTONYR 'Years for this model run' / 2010,2012,2015,2020,2025,2030,2035,2040,2045,2050/;
scalar gmsBOTime 'Adjustment for total available time span of years available in the model' / 1970 /;
set gmsRunScenario(scenario) 'name of the model run' / Frozen_policy_scenarie /;
set extensions(*,*,*) 'TIMES Extensions' / ''.(VALIDATE.NO, REDUCE.YES, DSCAUTO.YES, DEBUG.NO, DUMPSOL.NO, SOLVE_NOW.YES, MODEL_NAME.TIMES
XTQA.YES, VAR_UC.YES, OBLONG.YES, DAMAGE.NO, STAGES.NO, SOLVEDA.YES, DATAGDX.YES, VDA.YES, VEDAVDD.YES) /;
singleton set gmsObj(*) 'Choice of objective function formulations' / 'AUTO' /; // ALT, AUTO, LIN, MOD, STD
$onMulti
singleton set gmsddlocation(*) 'Location of DD files' / '' 'TIMES-DK_COMETS/model/'/;
singleton set gmsrunlocation(*) 'Location of Run file' / '' 'TIMES_Demo/model/demo12.run'/;
$offMulti
$offExternalInput
| GAMS | 3 | MaREI-EPMG/TIMES_MIRO | dkdata.gms | [
"MIT"
]
|
/*##############################################################################
HPCC SYSTEMS software Copyright (C) 2012 HPCC Systems®.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
############################################################################## */
#option ('targetClusterType', 'roxie');
import lib_stringlib,std.system.thorlib;
prefix := 'hthor';
useLayoutTrans := false;
useLocal := true;
usePayload := false;
useVarIndex := false;
//define constants
DG_GenFlat := true; //TRUE gens FlatFile
DG_GenChild := true; //TRUE gens ChildFile
DG_GenGrandChild := true; //TRUE gens GrandChildFile
DG_GenIndex := true; //TRUE gens FlatFile AND the index
DG_GenCSV := true; //TRUE gens CSVFile
DG_GenXML := true; //TRUE gens XMLFile
DG_GenVar := true; //TRUE gens VarFile only IF MaxField >= 3
DG_MaxField := 3; // maximum number of fields to use building the data records
DG_MaxChildren := 3; //maximum (1 to n) number of child recs
// generates (#parents * DG_MaxChildren) records
DG_MaxGrandChildren := 3; //maximum (1 to n) number of grandchild recs
// generates (#children * DG_MaxGrandChildren) records
DG_FileOut := '~REGRESS::' + prefix + '::DG_';
DG_ParentFileOut := '~REGRESS::' + prefix + '::DG_Parent.d00';
DG_ChildFileOut := '~REGRESS::' + prefix + '::DG_Child.d00';
DG_GrandChildFileOut := '~REGRESS::' + prefix + '::DG_GrandChild.d00';
DG_FetchFileName := '~REGRESS::' + prefix + '::DG_FetchFile';
DG_FetchIndex1Name := '~REGRESS::' + prefix + '::DG_FetchIndex1';
DG_FetchIndex2Name := '~REGRESS::' + prefix + '::DG_FetchIndex2';
DG_FetchIndexDiffName:= '~REGRESS::' + prefix + '::DG_FetchIndexDiff';
DG_MemFileName := '~REGRESS::' + prefix + '::DG_MemFile';
//record structures
DG_FetchRecord := RECORD
INTEGER8 sequence;
STRING2 State;
STRING20 City;
STRING25 Lname;
STRING15 Fname;
END;
DG_FetchFile := DATASET(DG_FetchFileName,{DG_FetchRecord,UNSIGNED8 __filepos {virtual(fileposition)}},FLAT);
#IF (useLayoutTrans=false)
#IF (usePayload=false)
DG_FetchIndex1 := INDEX(DG_FetchFile,{Lname,Fname,__filepos},DG_FetchIndex1Name);
DG_FetchIndex2 := INDEX(DG_FetchFile,{Lname,Fname, __filepos}, DG_FetchIndex2Name);
#ELSE
#IF (useVarIndex=true)
DG_FetchIndex1 := INDEX(DG_FetchFile,{Lname,Fname},{STRING fn := TRIM(Fname), state, STRING100 x {blob}:= fname, __filepos},DG_FetchIndex1Name);
DG_FetchIndex2 := INDEX(DG_FetchFile,{Lname,Fname},{STRING fn := TRIM(Fname), state, STRING100 x {blob}:= fname, __filepos},DG_FetchIndex2Name);
#ELSE
DG_FetchIndex1 := INDEX(DG_FetchFile,{Lname,Fname},{state ,__filepos},DG_FetchIndex1Name);
DG_FetchIndex2 := INDEX(DG_FetchFile,{Lname,Fname},{state, __filepos}, DG_FetchIndex2Name);
#END
#END
#ELSE
// Declare all indexes such that layout translation is required... Used at run-time only, not at setup time...
#IF (usePayload=false)
DG_FetchIndex1 := INDEX(DG_FetchFile,{Fname,Lname,__filepos},DG_FetchIndex1Name);
DG_FetchIndex2 := INDEX(DG_FetchFile,{Fname,Lname, __filepos}, DG_FetchIndex2Name);
#ELSE
#IF (useVarIndex=true)
DG_FetchIndex1 := INDEX(DG_FetchFile,{Fname,Lname},{STRING fn := TRIM(Fname), state, STRING100 x {blob}:= fname, __filepos},DG_FetchIndex1Name);
DG_FetchIndex2 := INDEX(DG_FetchFile,{Fname,Lname},{STRING fn := TRIM(Fname), state, STRING100 x {blob}:= fname, __filepos},DG_FetchIndex2Name);
#ELSE
DG_FetchIndex1 := INDEX(DG_FetchFile,{Fname,Lname},{state ,__filepos},DG_FetchIndex1Name);
DG_FetchIndex2 := INDEX(DG_FetchFile,{Fname,Lname},{state, __filepos}, DG_FetchIndex2Name);
#END
#END
#END
DG_OutRec := RECORD
unsigned4 DG_ParentID;
string10 DG_firstname;
string10 DG_lastname;
unsigned1 DG_Prange;
END;
DG_OutRecChild := RECORD
unsigned4 DG_ParentID;
unsigned4 DG_ChildID;
string10 DG_firstname;
string10 DG_lastname;
unsigned1 DG_Prange;
END;
#if(DG_GenVar = TRUE)
DG_VarOutRec := RECORD
DG_OutRec;
IFBLOCK(self.DG_Prange%2=0)
string20 ExtraField;
END;
END;
#end
//DATASET declarations
DG_BlankSet := dataset([{0,'','',0}],DG_OutRec);
#if(DG_GenFlat = TRUE OR DG_GenIndex = TRUE)
DG_FlatFile := DATASET(DG_FileOut+'FLAT',{DG_OutRec,UNSIGNED8 filepos{virtual(fileposition)}},FLAT);
DG_FlatFileEvens := DATASET(DG_FileOut+'FLAT_EVENS',{DG_OutRec,UNSIGNED8 filepos{virtual(fileposition)}},FLAT);
#end
#if(DG_GenIndex = TRUE)
DG_indexFile := INDEX(DG_FlatFile,
RECORD
#if(useLayoutTrans=false)
DG_firstname;
DG_lastname;
#else
DG_lastname;
DG_firstname;
#end
#if(usePayload = TRUE)
END , RECORD
#end
DG_Prange;
filepos
END,DG_FileOut+'INDEX');
DG_indexFileEvens := INDEX(DG_FlatFileEvens,
RECORD
#if(useLayoutTrans=false)
DG_firstname;
DG_lastname;
#else
DG_lastname;
DG_firstname;
#end
#if(usePayload = TRUE)
END , RECORD
#end
DG_Prange;
filepos
END,DG_FileOut+'INDEX_EVENS');
#end
#if(DG_GenCSV = TRUE)
DG_CSVFile := DATASET(DG_FileOut+'CSV',DG_OutRec,CSV);
#end
#if(DG_GenXML = TRUE)
DG_XMLFile := DATASET(DG_FileOut+'XML',DG_OutRec,XML);
#end
#if(DG_GenVar = TRUE)
DG_VarOutRecPlus := RECORD
DG_VarOutRec,
unsigned8 __filepos { virtual(fileposition)};
END;
DG_VarFile := DATASET(DG_FileOut+'VAR',DG_VarOutRecPlus,FLAT);
DG_VarIndex := INDEX(DG_VarFile,{
#if(useLayoutTrans=false)
DG_firstname;
DG_lastname;
#else
DG_lastname;
DG_firstname;
#end
__filepos},DG_FileOut+'VARINDEX');
DG_VarVarIndex := INDEX(DG_VarFile,{
#if(useLayoutTrans=false)
DG_firstname;
DG_lastname;
#else
DG_lastname;
DG_firstname;
#end
__filepos},{ string temp_blob1 := TRIM(ExtraField); string10000 temp_blob2 {blob} := ExtraField },DG_FileOut+'VARVARINDEX');
#end
#if(DG_GenChild = TRUE)
DG_ParentFile := DATASET(DG_ParentFileOut,{DG_OutRec,UNSIGNED8 filepos{virtual(fileposition)}},FLAT);
DG_ChildFile := DATASET(DG_ChildFileOut,{DG_OutRecChild,UNSIGNED8 filepos{virtual(fileposition)}},FLAT);
#if(DG_GenGrandChild = TRUE)
DG_GrandChildFile := DATASET(DG_GrandChildFileOut,{DG_OutRecChild,UNSIGNED8 filepos{virtual(fileposition)}},FLAT);
#end
#end
//define data atoms - each set has 16 elements
SET OF STRING10 DG_Fnames := ['DAVID','CLAIRE','KELLY','KIMBERLY','PAMELA','JEFFREY','MATTHEW','LUKE',
'JOHN' ,'EDWARD','CHAD' ,'KEVIN' ,'KOBE' ,'RICHARD','GEORGE' ,'DIRK'];
SET OF STRING10 DG_Lnames := ['BAYLISS','DOLSON','BILLINGTON','SMITH' ,'JONES' ,'ARMSTRONG','LINDHORFF','SIMMONS',
'WYMAN' ,'MORTON','MIDDLETON' ,'NOWITZKI','WILLIAMS','TAYLOR' ,'DRIMBAD' ,'BRYANT'];
SET OF UNSIGNED1 DG_PrangeS := [1, 2, 3, 4, 5, 6, 7, 8,
9,10,11,12,13,14,15,16];
SET OF STRING10 DG_Streets := ['HIGH' ,'CITATION' ,'MILL','25TH' ,'ELGIN' ,'VICARAGE','YAMATO' ,'HILLSBORO',
'SILVER','KENSINGTON','MAIN','EATON','PARK LANE','HIGH' ,'POTOMAC','GLADES'];
SET OF UNSIGNED1 DG_ZIPS := [101,102,103,104,105,106,107,108,
109,110,111,112,113,114,115,116];
SET OF UNSIGNED1 DG_AGES := [31,32,33,34,35,36,37,38,
39,40,41,42,43,44,45,56];
SET OF STRING2 DG_STATES := ['FL','GA','SC','NC','TX','AL','MS','TN',
'KY','CA','MI','OH','IN','IL','WI','MN'];
SET OF STRING3 DG_MONTHS := ['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG',
'SEP','OCT','NOV','DEC','ABC','DEF','GHI','JKL'];
//----------------------------- Child query related definitions ----------------------------------
// Raw record definitions:
sqHouseRec :=
record
string addr;
string10 postcode;
unsigned2 yearBuilt := 0;
end;
sqPersonRec :=
record
string forename;
string surname;
udecimal8 dob;
udecimal8 booklimit := 0;
unsigned2 aage := 0;
end;
sqBookRec :=
record
string name;
string author;
unsigned1 rating100;
udecimal8_2 price := 0;
end;
// Nested record definitions
sqPersonBookRec :=
record
sqPersonRec;
dataset(sqBookRec) books;
end;
sqHousePersonBookRec :=
record
sqHouseRec;
dataset(sqPersonBookRec) persons;
end;
// Record definitions with additional ids
sqHouseIdRec :=
record
unsigned4 id;
sqHouseRec;
end;
sqPersonIdRec :=
record
unsigned4 id;
sqPersonRec;
end;
sqBookIdRec :=
record
unsigned4 id;
sqBookRec;
end;
// Same with parent linking field.
sqPersonRelatedIdRec :=
record
sqPersonIdRec;
unsigned4 houseid;
end;
sqBookRelatedIdRec :=
record
sqBookIdRec;
unsigned4 personid;
end;
// Nested definitions with additional ids...
sqPersonBookIdRec :=
record
sqPersonIdRec;
dataset(sqBookIdRec) books;
end;
sqHousePersonBookIdRec :=
record
sqHouseIdRec;
dataset(sqPersonBookIdRec) persons;
end;
sqPersonBookRelatedIdRec :=
RECORD
sqPersonBookIdRec;
unsigned4 houseid;
END;
sqNestedBlob :=
RECORD
udecimal8 booklimit := 0;
END;
sqSimplePersonBookRec :=
RECORD
string20 surname;
string10 forename;
udecimal8 dob;
//udecimal8 booklimit := 0;
sqNestedBlob limit{blob};
unsigned1 aage := 0;
dataset(sqBookIdRec) books{blob};
END;
sqNamePrefix := '~REGRESS::' + prefix + '::';
sqHousePersonBookName := sqNamePrefix + 'HousePersonBook';
sqPersonBookName := sqNamePrefix + 'PersonBook';
sqHouseName := sqNamePrefix + 'House';
sqPersonName := sqNamePrefix + 'Person';
sqBookName := sqNamePrefix + 'Book';
sqSimplePersonBookName := sqNamePrefix + 'SimplePersonBook';
sqHousePersonBookIndexName := sqNamePrefix + 'HousePersonBookIndex';
sqPersonBookIndexName := sqNamePrefix + 'PersonBookIndex';
sqHouseIndexName := sqNamePrefix + 'HouseIndex';
sqPersonIndexName := sqNamePrefix + 'PersonIndex';
sqBookIndexName := sqNamePrefix + 'BookIndex';
sqSimplePersonBookIndexName := sqNamePrefix + 'SimplePersonBookIndex';
sqHousePersonBookIdExRec := record
sqHousePersonBookIdRec;
unsigned8 filepos{virtual(fileposition)};
end;
sqPersonBookRelatedIdExRec := record
sqPersonBookRelatedIdRec;
unsigned8 filepos{virtual(fileposition)};
end;
sqHouseIdExRec := record
sqHouseIdRec;
unsigned8 filepos{virtual(fileposition)};
end;
sqPersonRelatedIdExRec := record
sqPersonRelatedIdRec;
unsigned8 filepos{virtual(fileposition)};
end;
sqBookRelatedIdExRec := record
sqBookRelatedIdRec;
unsigned8 filepos{virtual(fileposition)};
end;
sqSimplePersonBookExRec := record
sqSimplePersonBookRec;
unsigned8 filepos{virtual(fileposition)};
end;
// Dataset definitions:
sqHousePersonBookDs := dataset(sqHousePersonBookName, sqHousePersonBookIdExRec, thor);
sqPersonBookDs := dataset(sqPersonBookName, sqPersonBookRelatedIdRec, thor);
sqHouseDs := dataset(sqHouseName, sqHouseIdExRec, thor);
sqPersonDs := dataset(sqPersonName, sqPersonRelatedIdRec, thor);
sqBookDs := dataset(sqBookName, sqBookRelatedIdRec, thor);
sqHousePersonBookExDs := dataset(sqHousePersonBookName, sqHousePersonBookIdExRec, thor);
sqPersonBookExDs := dataset(sqPersonBookName, sqPersonBookRelatedIdExRec, thor);
sqHouseExDs := dataset(sqHouseName, sqHouseIdExRec, thor);
sqPersonExDs := dataset(sqPersonName, sqPersonRelatedIdExRec, thor);
sqBookExDs := dataset(sqBookName, sqBookRelatedIdExRec, thor);
sqSimplePersonBookDs := dataset(sqSimplePersonBookName, sqSimplePersonBookExRec, thor);
sqSimplePersonBookIndex := index(sqSimplePersonBookDs, { surname, forename, aage }, { sqSimplePersonBookDs }, sqSimplePersonBookIndexName);
//related datasets:
//Don't really work because inheritance structure isn't preserved.
relatedBooks(sqPersonIdRec parentPerson) := sqBookDs(personid = parentPerson.id);
relatedPersons(sqHouseIdRec parentHouse) := sqPersonDs(houseid = parentHouse.id);
sqNamesTable1 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable2 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable3 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable4 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable5 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable6 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable7 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable8 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesTable9 := dataset(sqSimplePersonBookDs, sqSimplePersonBookName, FLAT);
sqNamesIndex1 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex2 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex3 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex4 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex5 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex6 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex7 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex8 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
sqNamesIndex9 := index(sqSimplePersonBookIndex,sqSimplePersonBookIndexName);
//----------------------------- Text search definitions ----------------------------------
TS_MaxTerms := 50;
TS_MaxProximity := 10;
TS_MaxWildcard := 1000;
TS_MaxMatchPerDocument := 1000;
TS_MaxFilenameLength := 255;
TS_MaxActions := 255;
TS_sourceType := unsigned2;
TS_wordCountType := unsigned8;
TS_segmentType := unsigned1;
TS_wordPosType := unsigned8;
TS_docPosType := unsigned8;
TS_documentId := unsigned8;
TS_termType := unsigned1;
TS_distanceType := integer8;
TS_indexWipType := unsigned1;
TS_wipType := unsigned8;
TS_stageType := unsigned1;
TS_dateType := unsigned8;
TS_sourceType TS_docid2source(TS_documentId x) := (x >> 48);
TS_documentId TS_docid2doc(TS_documentId x) := (x & 0xFFFFFFFFFFFF);
TS_documentId TS_createDocId(TS_sourceType source, TS_documentId doc) := (TS_documentId)(((unsigned8)source << 48) | doc);
boolean TS_docMatchesSource(TS_documentId docid, TS_sourceType source) := (docid between TS_createDocId(source,0) and (TS_documentId)(TS_createDocId(source+1,0)-1));
TS_wordType := string20;
TS_wordFlags := enum(unsigned1, HasLower=1, HasUpper=2);
TS_wordIdType := unsigned4;
TS_NameWordIndex := '~REGRESS::' + prefix + '::TS_wordIndex';
TS_wordIndex := index({ TS_wordType word, TS_documentId doc, TS_segmentType segment, TS_wordPosType wpos, TS_indexWipType wip } , { TS_wordFlags flags, TS_wordType original, TS_docPosType dpos}, TS_NameWordIndex);
TS_wordIndexRecord := recordof(TS_wordIndex);
DG_MemFileRec := RECORD
unsigned2 u2;
unsigned3 u3;
big_endian unsigned2 bu2;
big_endian unsigned3 bu3;
integer2 i2;
integer3 i3;
big_endian integer2 bi2;
big_endian integer3 bi3;
END;
DG_MemFile := DATASET(DG_MemFileName,DG_MemFileRec,FLAT);
//UseStandardFiles
//tidyoutput
//nothor
//UseIndexes
#option ('checkAsserts',true);
import lib_stringLib;
MaxTerms := TS_MaxTerms;
MaxProximity := TS_MaxProximity;
MaxWildcard := TS_MaxWildcard;
MaxMatchPerDocument := TS_MaxMatchPerDocument;
MaxFilenameLength := TS_MaxFilenameLength;
MaxActions := TS_MaxActions;
sourceType := TS_sourceType;
wordCountType := TS_wordCountType;
segmentType := TS_segmentType;
wordPosType := TS_wordPosType;
docPosType := TS_docPosType;
documentId := TS_documentId;
termType := TS_termType;
distanceType := TS_distanceType;
stageType := TS_stageType;
dateType := TS_dateType;
wordType := TS_wordType;
wordFlags := TS_wordFlags;
wordIdType := TS_wordIdType;
wordIndex := TS_wordIndex;
//May want the following, probably not actually implemented as an index - would save having dpos in the index, but more importantly storing it in the candidate match results because the mapping could be looked up later.
wordIndexRecord := TS_wordIndexRecord;
MaxWipIndexEntry := 4;
MaxWordsInDocument := 1000000;
MaxWordsInSet := 20;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
actionEnum := ENUM(
None = 0,
//Minimal operations required to implement the searching.
ReadWord, // termNum, source, segment, word, wordFlagMask, wordFlagCompare,
ReadWordSet, // termNum, source, segment, words, wordFlagMask, wordFlagCompare,
AndTerms, //
OrTerms, //
AndNotTerms, //
PhraseAnd, //
ProximityAnd, // distanceBefore, distanceAfter
MofNTerms, // minMatches, maxMatches
RankMergeTerms, // left outer join
RollupByDocument, // grouped rollup by document.
NormalizeMatch, // Normalize proximity records.
//The following aren't very sensible as far as text searching goes, but are here to test the underlying functionality
AndJoinTerms, // join on non-proximity
AndNotJoinTerms, //
MofNJoinTerms, // minMatches, maxMatches
RankJoinTerms, // left outer join
ProximityMergeAnd, // merge join on proximity
PassThrough,
Last,
//The following are only used in the production
FlagModifier, // wordFlagMask, wordFlagCompare
QuoteModifier, //
Max
);
// FAIL(stageType, 'Missing entry: ' + (string)action));
boolean definesTerm(actionEnum action) :=
(action in [actionEnum.ReadWord, actionEnum.ReadWordSet]);
stageRecord := { stageType stage };
wordRecord := { wordType word; };
wordSet := set of wordType;
stageSet := set of stageType;
searchRecord :=
RECORD
stageType stage;
actionEnum action;
//termType term;
dataset(stageRecord) inputs{maxcount(maxTerms)};
distanceType maxWip;
distanceType maxWipChild;
distanceType maxWipLeft;
distanceType maxWipRight;
//The item being searched for
wordType word;
dataset(wordRecord) words{maxcount(maxWordsInSet)};
wordFlags wordFlagMask;
wordFlags wordFlagCompare;
sourceType source;
segmentType segment;
//Modifiers for the connector/filter
distanceType maxDistanceRightBeforeLeft;
distanceType maxDistanceRightAfterLeft;
unsigned1 minMatches;
unsigned1 maxMatches;
END;
childMatchRecord := RECORD
wordPosType wpos;
wordPosType wip;
END;
simpleUserOutputRecord :=
record
unsigned2 source;
unsigned6 subDoc;
wordPosType wpos;
wordPosType wip;
dataset(childMatchRecord) words{maxcount(MaxProximity)};
end;
StageSetToDataset(stageSet x) := dataset(x, stageRecord);
StageDatasetToSet(dataset(stageRecord) x) := set(x, stage);
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Matches
matchRecord := RECORD
documentId doc;
segmentType segment;
wordPosType wpos;
wordPosType wip;
dataset(childMatchRecord) children{maxcount(MaxProximity)};
END;
createChildMatch(wordPosType wpos, wordPosType wip) := transform(childMatchRecord, self.wpos := wpos; self.wip := wip);
SetOfInputs := set of dataset(matchRecord);
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Functions which are helpful for hand constructing queries...
CmdReadWord(wordType word, sourceType source = 0, segmentType segment = 0, wordFlags wordFlagMask = 0, wordFlags wordFlagCompare = 0) :=
TRANSFORM(searchRecord,
SELF.action := actionEnum.ReadWord;
SELF.source := source;
SELF.segment := segment;
SELF.word := word;
SELF.wordFlagMask := wordFlagMask;
SELF.wordFlagCompare:= wordFlagCompare;
SELF.maxWip := 1;
SELF := []);
defineCmdTermCombineTerm(actionEnum action, stageSet inputs, distanceType maxDistanceRightBeforeLeft = 0, distanceType maxDistanceRightAfterLeft = 0) :=
TRANSFORM(searchRecord,
SELF.action := action;
SELF.inputs := StageSetToDataset(inputs);
SELF.maxDistanceRightBeforeLeft := maxDistanceRightBeforeLeft;
SELF.maxDistanceRightAfterLeft := maxDistanceRightAfterLeft;
SELF.maxWip := 1;
SELF.maxWipLeft := 1;
SELF.maxWipRight := 1;
SELF := []);
CmdTermAndTerm(stageType leftStage, stageType rightStage) :=
defineCmdTermCombineTerm(actionEnum.AndTerms, [leftStage, rightStage]);
CmdAndTerms(stageSet stages) :=
defineCmdTermCombineTerm(actionEnum.AndTerms, stages);
CmdTermAndNotTerm(stageType leftStage, stageType rightStage) :=
defineCmdTermCombineTerm(actionEnum.AndNotTerms, [leftStage, rightStage]);
CmdTermAndNotTerms(stageSet stages) :=
defineCmdTermCombineTerm(actionEnum.AndNotTerms, stages);
CmdMofNTerms(stageSet stages, unsigned minMatches, unsigned maxMatches = 999999999) :=
TRANSFORM(searchRecord,
SELF.action := actionEnum.MofNTerms;
SELF.inputs := StageSetToDataset(stages);
SELF.minMatches := minMatches;
SELF.maxMatches := maxMatches;
SELF.maxWip := 1;
SELF := []);
CmdPhraseAnd(stageSet stages) :=
defineCmdTermCombineTerm(actionEnum.PhraseAnd, stages);
CmdProximityAnd(stageType leftStage, stageType rightStage, distanceType maxDistanceRightBeforeLeft, distanceType maxDistanceRightAfterLeft) :=
defineCmdTermCombineTerm(actionEnum.ProximityAnd, [leftStage, rightStage], maxDistanceRightBeforeLeft, maxDistanceRightAfterLeft);
CmdTermOrTerm(stageType leftStage, stageType rightStage) :=
defineCmdTermCombineTerm(actionEnum.OrTerms, [leftStage, rightStage]);
CmdOrTerms(stageSet stages) :=
defineCmdTermCombineTerm(actionEnum.OrTerms, stages);
//-------------------------------------------------------------------------------------------------------------
//-------------------------------------------------------------------------------------------------------------
//---------------------------------------- Code for executing queries -----------------------------------------
//-------------------------------------------------------------------------------------------------------------
//-------------------------------------------------------------------------------------------------------------
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Matching helper functions
matchSingleWordFlags(wordIndex wIndex, searchRecord search) :=
keyed(search.segment = 0 or wIndex.segment = search.segment, opt) AND
((wIndex.flags & search.wordFlagMask) = search.wordFlagCompare);
matchSingleWord(wordIndex wIndex, searchRecord search) :=
keyed(wIndex.word = search.word) AND
matchSingleWordFlags(wIndex, search);
matchManyWord(wordIndex wIndex, searchRecord search) :=
keyed(wIndex.word in set(search.words, word)) AND
matchSingleWordFlags(wIndex, search);
matchFirstWord(wordIndex wIndex, searchRecord search) :=
keyed(search.source = 0 OR TS_docMatchesSource(wIndex.doc, search.source), opt);
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// ReadWord
doReadWord(searchRecord search) := FUNCTION
matches := sorted(wordIndex, doc, segment, wpos, wip)(
matchSingleWord(wordIndex, search) AND
matchFirstWord(wordIndex, search));
matchRecord createMatchRecord(wordIndexRecord ds) := transform
self := ds;
self.children := []
end;
steppedMatches := stepped(matches, doc, segment, wpos);
projected := project(steppedMatches, createMatchRecord(left));
return projected;
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// ReadWord
doReadWordSet(searchRecord search) := FUNCTION
matches := sorted(wordIndex, doc, segment, wpos, wip)(
matchManyWord(wordIndex, search) AND
matchFirstWord(wordIndex, search));
matchRecord createMatchRecord(wordIndexRecord ds) := transform
self := ds;
self.children := []
end;
steppedMatches := stepped(matches, doc, segment, wpos);
projected := project(steppedMatches, createMatchRecord(left));
return projected;
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// OrTerms
doOrTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return merge(inputs, doc, segment, wpos, dedup); // MORE option to specify priority?
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// AndTerms
doAndTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return mergejoin(inputs, STEPPED(left.doc = right.doc), doc, segment, wpos, dedup); // MORE option to specify priority?
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// AndNotTerms
doAndNotTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return mergejoin(inputs, STEPPED(left.doc = right.doc), doc, segment, wpos, left only);
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// RankMergeTerms
doRankMergeTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return mergejoin(inputs, STEPPED(left.doc = right.doc), doc, segment, wpos, left outer);
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// M of N
doMofNTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return mergejoin(inputs, STEPPED(left.doc = right.doc), doc, segment, wpos, wip, dedup, mofn(search.minMatches, search.maxMatches)); // MORE option to specify priority?
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Join varieties - primarily for testing
//Note this testing transform wouldn't work correctly with proximity operators as inputs.
matchRecord createDenormalizedMatch(matchRecord l, dataset(matchRecord) matches) := transform
wpos := min(matches, wpos);
wend := max(matches, wpos + wip);
self.wpos := wpos;
self.wip := wend - wpos;
self.children := normalize(matches, 1, createChildMatch(LEFT.wpos, LEFT.wip));
self := l;
end;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// AndJoinTerms
doAndJoinTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return join(inputs, STEPPED(left.doc = right.doc) and (left.wpos <> right.wpos), createDenormalizedMatch(LEFT, ROWS(left)), sorted(doc, segment, wpos));
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// AndNotJoinTerms
doAndNotJoinTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return join(inputs, STEPPED(left.doc = right.doc), createDenormalizedMatch(LEFT, ROWS(left)), sorted(doc, segment, wpos), left only);
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// RankJoinTerms
doRankJoinTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return join(inputs, STEPPED(left.doc = right.doc), createDenormalizedMatch(LEFT, ROWS(left)), sorted(doc, segment, wpos), left outer);
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// M of N
doMofNJoinTerms(searchRecord search, SetOfInputs inputs) := FUNCTION
return join(inputs, STEPPED(left.doc = right.doc), createDenormalizedMatch(LEFT, ROWS(left)), sorted(doc, segment, wpos), mofn(search.minMatches, search.maxMatches));
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// PhraseAnd
steppedPhraseCondition(matchRecord l, matchRecord r, distanceType maxWip) :=
(l.doc = r.doc) and (l.segment = r.segment) and
(r.wpos between l.wpos+1 and l.wpos+maxWip);
doPhraseAnd(searchRecord search, SetOfInputs inputs) := FUNCTION
steppedCondition(matchRecord l, matchRecord r) := steppedPhraseCondition(l, r, search.maxWipLeft);
condition(matchRecord l, matchRecord r) :=
(r.wpos = l.wpos + l.wip);
matchRecord createMatch(matchRecord l, dataset(matchRecord) allRows) := transform
self.wip := sum(allRows, wip);
self := l;
end;
matches := join(inputs, STEPPED(steppedCondition(left, right)) and condition(LEFT, RIGHT), createMatch(LEFT, ROWS(LEFT)), sorted(doc, segment, wpos));
return matches;
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// ProximityAnd
steppedProximityCondition(matchRecord l, matchRecord r, distanceType maxWipLeft, distanceType maxWipRight, distanceType maxDistanceRightBeforeLeft, distanceType maxDistanceRightAfterLeft) := function
// if maxDistanceRightBeforeLeft is < 0 it means it must follow, so don't add maxWipRight
maxRightBeforeLeft := IF(maxDistanceRightBeforeLeft >= 0, maxDistanceRightBeforeLeft + maxWipRight, maxDistanceRightBeforeLeft);
maxRightAfterLeft := IF(maxDistanceRightAfterLeft >= 0, maxDistanceRightAfterLeft + maxWipLeft, maxDistanceRightAfterLeft);
return
(l.doc = r.doc) and (l.segment = r.segment) and
(r.wpos + maxRightBeforeLeft >= l.wpos) and // (right.wpos + right.wip + maxRightBeforeLeft >= left.wpos)
(r.wpos <= l.wpos + (maxRightAfterLeft)); // (right.wpos <= left.wpos + left.wip + maxRightAfterLeft)
end;
doProximityAnd(searchRecord search, SetOfInputs inputs) := FUNCTION
steppedCondition(matchRecord l, matchRecord r) := steppedProximityCondition(l, r, search.maxWipLeft, search.maxWipRight, search.maxDistanceRightBeforeLeft, search.maxDistanceRightAfterLeft);
condition(matchRecord l, matchRecord r) :=
(r.wpos + r.wip + search.maxDistanceRightBeforeLeft >= l.wpos) and
(r.wpos <= l.wpos + l.wip + search.maxDistanceRightAfterLeft);
overlaps(wordPosType wpos, childMatchRecord r) := (wpos between r.wpos and r.wpos + (r.wip - 1));
createMatch(matchRecord l, matchRecord r) := function
wpos := if(l.wpos < r.wpos, l.wpos, r.wpos);
wend := if(l.wpos + l.wip > r.wpos + r.wip, l.wpos + l.wip, r.wpos + r.wip);
rawLeftChildren := IF(exists(l.children), l.children, dataset(row(createChildMatch(l.wpos, l.wip))));
rawRightChildren := IF(exists(r.children), r.children, dataset(row(createChildMatch(r.wpos, r.wip))));
leftChildren := sorted(rawLeftChildren, wpos, wip, assert);
rightChildren := sorted(rawRightChildren, wpos, wip, assert);
anyOverlaps := exists(join(leftChildren, rightChildren,
overlaps(left.wpos, right) or overlaps(left.wpos+(left.wip-1), right) or
overlaps(right.wpos, left) or overlaps(right.wpos+(right.wip-1), left), all));
//Check for any overlaps between the words, should be disjoint.
matchRecord matchTransform := transform, skip(anyOverlaps)
self.wpos := wpos;
self.wip := wend - wpos;
self.children := merge(leftChildren, rightChildren, dedup);
self := l;
end;
return matchTransform;
end;
matches := join(inputs, STEPPED(steppedCondition(left, right)) and condition(LEFT, RIGHT), createMatch(LEFT, RIGHT), sorted(doc, segment, wpos));
return matches;
END;
doProximityMergeAnd(searchRecord search, SetOfInputs inputs) := FUNCTION
steppedCondition(matchRecord l, matchRecord r) := steppedProximityCondition(l, r, search.maxWipLeft, search.maxWipRight, search.maxDistanceRightBeforeLeft, search.maxDistanceRightAfterLeft);
condition(matchRecord l, matchRecord r) :=
(r.wpos + r.wip + search.maxDistanceRightBeforeLeft >= l.wpos) and
(r.wpos <= l.wpos + l.wip + search.maxDistanceRightAfterLeft);
overlaps(wordPosType wpos, childMatchRecord r) := (wpos between r.wpos and r.wpos + (r.wip - 1));
anyOverlaps (matchRecord l, matchRecord r) := function
wpos := if(l.wpos < r.wpos, l.wpos, r.wpos);
wend := if(l.wpos + l.wip > r.wpos + r.wip, l.wpos + l.wip, r.wpos + r.wip);
rawLeftChildren := IF(exists(l.children), l.children, dataset(row(createChildMatch(l.wpos, l.wip))));
rawRightChildren := IF(exists(r.children), r.children, dataset(row(createChildMatch(r.wpos, r.wip))));
leftChildren := sorted(rawLeftChildren, wpos, wip, assert);
rightChildren := sorted(rawRightChildren, wpos, wip, assert);
return exists(join(leftChildren, rightChildren,
overlaps(left.wpos, right) or overlaps(left.wpos+(left.wip-1), right) or
overlaps(right.wpos, left) or overlaps(right.wpos+(right.wip-1), left), all));
end;
matches := mergejoin(inputs, STEPPED(steppedCondition(left, right)) and condition(LEFT, RIGHT) and not anyOverlaps(LEFT,RIGHT), sorted(doc, segment, wpos));
return matches;
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Normalize denormalized proximity records
doNormalizeMatch(searchRecord search, SetOfInputs inputs) := FUNCTION
matchRecord createNorm(matchRecord l, unsigned c) := transform
hasChildren := count(l.children) <> 0;
curChild := l.children[NOBOUNDCHECK c];
self.wpos := if (hasChildren, curChild.wpos, l.wpos);
self.wip := if (hasChildren, curChild.wip, l.wip);
self.children := [];
self := l;
end;
normalizedRecords := normalize(inputs[1], MAX(1, count(LEFT.children)), createNorm(left, counter));
groupedNormalized := group(normalizedRecords, doc, segment);
sortedNormalized := sort(groupedNormalized, wpos);
dedupedNormalized := dedup(sortedNormalized, wpos, wip);
return group(dedupedNormalized);
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Rollup by document
doRollupByDocument(searchRecord search, dataset(matchRecord) input) := FUNCTION
groupByDocument := group(input, doc);
dedupedByDocument := rollup(groupByDocument, group, transform(matchRecord, self.doc := left.doc; self.segment := 0; self.wpos := 0; self.wip := 0; self := left));
return dedupedByDocument;
END;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
processStage(searchRecord search, SetOfInputs allInputs) := function
inputs:= RANGE(allInputs, StageDatasetToSet(search.inputs));
result := case(search.action,
actionEnum.ReadWord => doReadWord(search),
actionEnum.ReadWordSet => doReadWordSet(search),
actionEnum.OrTerms => doOrTerms(search, inputs),
actionEnum.AndTerms => doAndTerms(search, inputs),
actionEnum.AndNotTerms => doAndNotTerms(search, inputs),
actionEnum.RankMergeTerms => doRankMergeTerms(search, inputs),
actionEnum.MofNTerms => doMofNTerms(search, inputs),
actionEnum.PhraseAnd => doPhraseAnd(search, inputs),
actionEnum.ProximityAnd => doProximityAnd(search, inputs),
// actionEnum.ProximityMergeAnd => doProximityMergeAnd(search, inputs),
actionEnum.AndJoinTerms => doAndJoinTerms(search, inputs),
actionEnum.AndNotJoinTerms => doAndNotJoinTerms(search, inputs),
actionEnum.RankJoinTerms => doRankJoinTerms(search, inputs),
actionEnum.MofNJoinTerms => doMofNJoinTerms(search, inputs),
actionEnum.RollupByDocument => doRollupByDocument(search, allInputs[search.inputs[1].stage]), // more efficient than way normalize is handled, but want to test both varieties
actionEnum.NormalizeMatch => doNormalizeMatch(search, inputs),
dataset([], matchRecord));
//check that outputs from every stage are sorted as required.
sortedResult := sorted(result, doc, segment, wpos, assert);
return sortedResult;
end;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// Code to actually execute the query:
convertToUserOutput(dataset(matchRecord) results) := function
simpleUserOutputRecord createUserOutput(matchRecord l) := transform
self.source := TS_docid2source(l.doc);
self.subDoc := TS_docid2doc(l.doc);
self.words := l.children;
SELF := l;
END;
return project(results, createUserOutput(left));
end;
ExecuteQuery(dataset(searchRecord) queryDefinition, dataset(matchRecord) initialResults = dataset([], matchRecord)) := function
#if (useLocal=true)
executionPlan := thisnode(global(queryDefinition, opt, few)); // Store globally for efficient access
results := ALLNODES(LOCAL(graph(initialResults, count(executionPlan), processStage(executionPlan[NOBOUNDCHECK COUNTER], rowset(left)), parallel)));
#else
executionPlan := global(queryDefinition, opt, few); // Store globally for efficient access
results := graph(initialResults, count(executionPlan), processStage(executionPlan[NOBOUNDCHECK COUNTER], rowset(left)), parallel);
#end
userOutput := convertToUserOutput(results);
return userOutput;
end;
executeReadWord(wordType word, sourceType source = 0, segmentType segment = 0, wordFlags wordFlagMask = 0, wordFlags wordFlagCompare = 0) :=
doReadWord(row(CmdReadWord(word, source, segment, wordFlagMask, wordFlagCompare)));
executeAndTerms(SetOfInputs stages) :=
doAndTerms(row(CmdAndTerms([])), stages);
executeAndNotTerms(SetOfInputs stages) :=
doAndNotTerms(row(CmdTermAndNotTerms([])), stages);
executeMofNTerms(SetOfInputs stages, unsigned minMatches, unsigned maxMatches = 999999999) :=
doMofNTerms(row(CmdMofNTerms([], minMatches, maxMatches)), stages);
executeOrTerms(SetOfInputs stages) :=
doOrTerms(row(CmdOrTerms([])), stages);
executePhrase(SetOfInputs stages) :=
doPhraseAnd(row(CmdPhraseAnd([])), stages);
executeProximity(SetOfInputs stages, distanceType maxDistanceRightBeforeLeft, distanceType maxDistanceRightAfterLeft) :=
doProximityAnd(row(CmdProximityAnd(0,0, maxDistanceRightBeforeLeft, maxDistanceRightAfterLeft)), stages);
///////////////////////////////////////////////////////////////////////////////////////////////////////////
// A simplified query language
parseQuery(string queryText) := function
searchParseRecord :=
RECORD(searchRecord)
unsigned numInputs;
END;
productionRecord :=
record
unsigned termCount;
dataset(searchParseRecord) actions{maxcount(MaxActions)};
end;
unknownTerm := (termType)-1;
PRULE := rule type (productionRecord);
ARULE := rule type (searchParseRecord);
///////////////////////////////////////////////////////////////////////////////////////////////////////////
pattern ws := [' ','\t'];
token number := pattern('-?[0-9]+');
//pattern wordpat := pattern('[A-Za-z0-9]+');
pattern wordpat := pattern('[A-Za-z][A-Za-z0-9]*');
pattern quotechar := '"';
token quotedword := quotechar wordpat quotechar;
///////////////////////////////////////////////////////////////////////////////////////////////////////////
PRULE forwardExpr := use(productionRecord, 'ExpressionRule');
ARULE term0
:= quotedword transform(searchParseRecord,
SELF.action := actionEnum.ReadWord;
SELF.word := $1[2..length($1)-1];
SELF := []
)
| 'CAPS' '(' SELF ')' transform(searchParseRecord,
SELF.wordFlagMask := wordFlags.hasUpper;
SELF.wordFlagCompare := wordFlags.hasUpper;
SELF := $3;
)
| 'NOCAPS' '(' SELF ')' transform(searchParseRecord,
SELF.wordFlagMask := wordFlags.hasUpper;
SELF.wordFlagCompare := 0;
SELF := $3;
)
| 'ALLCAPS' '(' SELF ')' transform(searchParseRecord,
SELF.wordFlagMask := wordFlags.hasUpper+wordFlags.hasLower;
SELF.wordFlagCompare := wordFlags.hasUpper;
SELF := $3;
)
;
ARULE term0List
:= term0 transform(searchParseRecord,
SELF.action := actionEnum.ReadWordSet;
SELF.words := dataset(row(transform(wordRecord, self.word := $1.word)));
SELF.word := '';
SELF := $1;
)
| SELF ',' term0 transform(searchParseRecord,
SELF.words := $1.words + dataset(row(transform(wordRecord, self.word := $3.word)));
SELF := $1;
)
;
PRULE termList
:= forwardExpr transform(productionRecord, self.termCount := 1; self.actions := $1.actions)
| SELF ',' forwardExpr transform(productionRecord, self.termCount := $1.termCount + 1; self.actions := $1.actions + $3.actions)
;
PRULE term1
:= term0 transform(productionRecord, self.termCount := 1; self.actions := dataset($1))
| 'SET' '(' term0List ')' transform(productionRecord, self.termCount := 1; self.actions := dataset($3))
| '(' forwardExpr ')'
| 'AND' '(' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.AndTerms;
// self.numInputs := count($3.actions) - sum($3.actions, numInputs);
self.numInputs := $3.termCount;
self := [];
)
)
)
| 'ANDNOT' '(' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.AndNotTerms;
self.numInputs := $3.termCount;
self := [];
)
)
)
| 'RANK' '(' forwardExpr ',' forwardExpr ')'
transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.RankMergeTerms;
self.numInputs := 2;
self := []
)
)
)
| 'MOFN' '(' number ',' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.MOfNTerms;
self.numInputs := $5.termCount;
SELF.minMatches := (integer)$3;
SELF.maxMatches := $5.termCount;
self := [];
)
)
)
| 'MOFN' '(' number ',' number ',' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $7.actions + row(
transform(searchParseRecord,
self.action := actionEnum.MOfNTerms;
self.numInputs := $7.termCount;
SELF.minMatches := (integer)$3;
SELF.maxMatches := (integer)$5;
self := [];
)
)
)
| 'OR' '(' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.OrTerms;
self.numInputs := $3.termCount;
self := [];
)
)
)
| 'PHRASE' '(' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.PhraseAnd;
self.numInputs := $3.termCount;
self := [];
)
)
)
| 'PROXIMITY' '(' forwardExpr ',' forwardExpr ',' number ',' number ')'
transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.ProximityAnd;
self.numInputs := 2;
self.maxDistanceRightBeforeLeft := (integer)$7;
self.maxDistanceRightAfterLeft := (integer)$9;
self := []
)
)
)
| 'PRE' '(' forwardExpr ',' forwardExpr ')'
transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.ProximityAnd;
self.numInputs := 2;
self.maxDistanceRightBeforeLeft := -1;
self.maxDistanceRightAfterLeft := MaxWordsInDocument;
self := []
)
)
)
| 'AFT' '(' forwardExpr ',' forwardExpr ')'
transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.ProximityAnd;
self.numInputs := 2;
self.maxDistanceRightBeforeLeft := MaxWordsInDocument;
self.maxDistanceRightAfterLeft := -1;
self := []
)
)
)
| 'PROXMERGE' '(' forwardExpr ',' forwardExpr ',' number ',' number ')'
transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.ProximityMergeAnd;
self.numInputs := 2;
self.maxDistanceRightBeforeLeft := (integer)$7;
self.maxDistanceRightAfterLeft := (integer)$9;
self := []
)
)
)
| 'ANDJOIN' '(' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.AndJoinTerms;
self.numInputs := $3.termCount;
self := [];
)
)
)
| 'ANDNOTJOIN' '(' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.AndNotJoinTerms;
self.numInputs := $3.termCount;
self := [];
)
)
)
| 'MOFNJOIN' '(' number ',' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.MOfNJoinTerms;
self.numInputs := $5.termCount;
SELF.minMatches := (integer)$3;
SELF.maxMatches := $5.termCount;
self := [];
)
)
)
| 'MOFNJOIN' '(' number ',' number ',' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $7.actions + row(
transform(searchParseRecord,
self.action := actionEnum.MOfNJoinTerms;
self.numInputs := $7.termCount;
SELF.minMatches := (integer)$3;
SELF.maxMatches := (integer)$5;
self := [];
)
)
)
| 'RANKJOIN' '(' forwardExpr ',' forwardExpr ')'
transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + $5.actions + row(
transform(searchParseRecord,
self.action := actionEnum.RankJoinTerms;
self.numInputs := 2;
self := []
)
)
)
| 'ROLLAND' '(' termList ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.AndTerms;
self.numInputs := $3.termCount;
self := [];
)
) + row(
transform(searchParseRecord,
self.action := actionEnum.RollupByDocument;
self.numInputs := 1;
self := [];
)
)
)
| 'NORM' '(' forwardExpr ')' transform(productionRecord,
self.termCount := 1;
self.actions := $3.actions + row(
transform(searchParseRecord,
self.action := actionEnum.NormalizeMatch;
self.numInputs := 1;
self := [];
)
)
)
;
PRULE expr
:= term1 : define ('ExpressionRule')
;
infile := dataset(row(transform({ string line{maxlength(1023)} }, self.line := queryText)));
resultsRecord := record
dataset(searchParseRecord) actions{maxcount(MaxActions)};
end;
resultsRecord extractResults(dataset(searchParseRecord) actions) :=
TRANSFORM
SELF.actions := actions;
END;
p1 := PARSE(infile,line,expr,extractResults($1.actions),first,whole,skip(ws),nocase,parse);
pnorm := normalize(p1, left.actions, transform(right));
//Now need to associate sequence numbers, and correctly set them up.
wipRecord := { wordPosType wip; };
stageStackRecord := record
stageType prevStage;
dataset(stageRecord) stageStack{maxcount(MaxActions)};
dataset(wipRecord) wipStack{maxcount(MaxActions)};
end;
nullStack := row(transform(stageStackRecord, self := []));
wordPosType _max(wordPosType l, wordPosType r) := if(l < r, r, l);
assignStageWip(searchParseRecord l, stageStackRecord r) := module
shared stageType thisStage := r.prevStage + 1;
shared stageType maxStage := count(r.stageStack);
shared stageType minStage := maxStage+1-l.numInputs;
shared thisInputs := r.stageStack[minStage..maxStage];
shared maxLeftWip := r.wipStack[minStage].wip;
shared maxRightWip := r.wipStack[maxStage].wip;
shared maxChildWip := max(r.wipStack[minStage..maxStage], wip);
shared sumMaxChildWip := sum(r.wipStack[minStage..maxStage], wip);
shared thisMaxWip := case(l.action,
actionEnum.ReadWord=>MaxWipIndexEntry,
actionEnum.AndTerms=>maxChildWip,
actionEnum.OrTerms=>maxChildWip,
actionEnum.AndNotTerms=>maxLeftWip,
actionEnum.PhraseAnd=>sumMaxChildWip,
actionEnum.ProximityAnd=>_max(l.maxDistanceRightBeforeLeft,l.maxDistanceRightAfterLeft) + sumMaxChildWip,
actionEnum.MofNTerms=>maxChildWip,
maxChildWip);
export searchParseRecord nextRow := transform
self.stage := thisStage;
self.inputs := thisInputs;
self.maxWip := thisMaxWip;
self.maxWipLeft := maxLeftWip;
self.maxWipRight := maxRightWip;
self.maxWipChild := maxChildWip;
self := l;
end;
export stageStackRecord nextStack := transform
self.prevStage := thisStage;
self.stageStack := r.stageStack[1..maxStage-l.numInputs] + row(transform(stageRecord, self.stage := thisStage));
self.wipStack := r.wipStack[1..maxStage-l.numInputs] + row(transform(wipRecord, self.wip := thisMaxWip;));
end;
end;
sequenced := process(pnorm, nullStack, assignStageWip(left, right).nextRow, assignStageWip(left, right).nextStack);
return project(sequenced, transform(searchRecord, self := left));
end;
inputRecord := { string query{maxlength(2048)}; };
MaxResults := 10000;
processedRecord := record(inputRecord)
dataset(searchRecord) request{maxcount(MaxActions)};
dataset(simpleUserOutputRecord) result{maxcount(MaxResults)};
end;
processedRecord doBatchExecute(inputRecord l) := transform
request := parseQuery(l.query);
self.request := request;
self.result := ExecuteQuery(request);
self := l;
end;
doSingleExecute(string queryText) := function
request := parseQuery(queryText);
result := ExecuteQuery(request);
return result;
end;
q1 := dataset([
#if (0)
'AND("black","sheep")',
'ANDNOT("black","sheep")',
'MOFN(2,"black","sheep","white")',
'MOFN(2,2,"black","sheep","white")',
//Word tests
'("nonexistant")',
'("one")',
'CAPS("one")',
'NOCAPS("one")',
'ALLCAPS("one")',
'"ibm"', // simple word, and an alias
//Or tests
'OR("nonexistant1", "nonexistant2")', // neither side matches
'OR("nonexistant1", "sheep")', // RHS matches
'OR("twinkle", "nonexistant2")', // LHS matches
'OR("twinkle", "twinkle")', // should dedup
'OR("sheep", "black")', // matches in same document
'OR("sheep", "twinkle")', // matches in different documents
'OR("one", "sheep", "sheep", "black", "fish")', // matches in different documents
'OR(OR("one", "sheep"), OR("sheep", "black", "fish"))', // matches in different documents
//And tests
'AND("nonexistant1", "nonexistant2")', // neither side matches
'AND("nonexistant1", "sheep")', // RHS matches
'AND("twinkle", "nonexistant2")', // LHS matches
'AND("twinkle", "twinkle")', // should dedup
'AND("sheep", "black")', // matches in same document
'AND("sheep", "twinkle")', // matches in different documents
'AND("in", "a")', // high frequencies
'AND("twinkle", "little", "how", "star")', // Nary
'AND(AND("twinkle", "little"), AND("how", "star"))', // Nested
'AND(AND("twinkle", "little"), AND("how", "wonder"))', // Nested
//MORE: Should also test segment restriction....
'ANDNOT("nonexistant1", "nonexistant2")', // neither side matches
'ANDNOT("nonexistant1", "sheep")', // RHS matches
'ANDNOT("twinkle", "nonexistant2")', // LHS matches
'ANDNOT("twinkle", "twinkle")', // should dedup
'ANDNOT("sheep", "black")', // matches in same document
'ANDNOT("sheep", "twinkle")', // matches in different documents
'ANDNOT("one", "sheep", "black", "fish")', // matches one, but none of the others
//Phrases
'PHRASE("nonexistant1", "nonexistant2")', // words don't occour
'PHRASE("in", "are")', // doesn't occur, but words do
'PHRASE("baa", "black")', // occurs, but
'PHRASE("x", "y", "x", "x", "y")', // a partial match, first - doesn't actually make it more complicatied to implement
'PHRASE("james","arthur","stuart")', // check that next occurence of stuart takes note of the change of document.
'PHRASE(OR("black","white"),"sheep")', // proximity on a non-word input
'PHRASE("one", "for", OR(PHRASE("the","master"),PHRASE("the","dame"),PHRASE("the","little","boy")))',
// more complex again
//M of N
'MOFN(2, "humpty", "horses", "together", "beansprout")', // m<matches
'MOFN(3, "humpty", "horses", "together", "beansprout")', // m=matches
'MOFN(4, "humpty", "horses", "together", "beansprout")', // m>matches
'MOFN(2,2, "humpty", "horses", "together", "beansprout")', // too many matches
'MOFN(2, "nonexistant", "little", "bo")', // first input fails to match any
'MOFN(2, "little", "bo", "nonexistant")', // lose an input while finising candidates
'MOFN(2, "one", "two", "three", "four", "five")',
'MOFN(2, "nonexistant", "two", "three", "four", "five")',
'MOFN(2, "one", "nonexistant", "three", "four", "five")',
'MOFN(2, "nonexistant1", "nonexistant2", "three", "four", "five")',
'MOFN(2, "nonexistant1", "nonexistant2", "nonexistant3", "four", "five")',
'MOFN(2, "nonexistant1", "nonexistant2", "nonexistant3", "nonexistant4", "five")',
'MOFN(2, PHRASE("little","bo"),PHRASE("three","bags"),"sheep")', // m of n on phrases
'MOFN(2, PHRASE("Little","Bo"),PHRASE("three","bags"),"sheep")', // m of n on phrases - capital letters don't match
'MOFN(2, OR("little","big"), OR("king", "queen"), OR("star", "sheep", "twinkle"))',
//Proximity
'PROXIMITY("nonexistant1", "nonexistant2", -1, 1)',
'PROXIMITY("black", "nonexistant2", -1, 1)',
'PROXIMITY("nonexistant1", "sheep", -1, 1)',
//Adjacent checks
'PROXIMITY("ship", "sank", 0, 0)', // either order but adjacent
'NORM(PROXIMITY("ship", "sank", 0, 0))',
'PROXIMITY("ship", "sank", -1, 0)', // must follow
'PROXIMITY("ship", "sank", 0, -1)', // must preceed
'PROXIMITY("sank", "ship", 0, 0)', // either order but adjacent
'PROXIMITY("sank", "ship", -1, 0)', // must follow
'PROXIMITY("sank", "ship", 0, -1)', // must preceed
//Within a distance of 1
'PROXIMITY("ship", "sank", 1, 1)', // either order but only 1 intervening word
'PROXIMITY("ship", "sank", -1, 1)', // must follow
'PROXIMITY("ship", "sank", 1, -1)', // must preceed
'PROXIMITY("sank", "ship", 1, 1)', // either order but only 1 intervening word
'PROXIMITY("sank", "ship", -1, 1)', // must follow
'PROXIMITY("sank", "ship", 1, -1)', // must preceed
'PROXIMITY("ship", "sank", 0, 2)', // asymetric range
//Within a distance of 2
'PROXIMITY("ship", "ship", 2, 2)', // either order but only 2 intervening word, no dups
// *** currently fails because of lack of duplication in lowest merger
'PROXIMITY("zx", "zx", 5, 5)', // "zx (za) zx", "zx (za zx zb zc zd) zx" and "zx (zb zc zd zx)"
'PROXIMITY(PROXIMITY("zx", "zx", 5, 5), "zx", 1, 1)', // "zx (za) zx (zb zc zd) zx" - obtained two different ways.
'NORM(PROXIMITY(PROXIMITY("zx", "zx", 5, 5), "zx", 1, 1))', // as above, but normalized
'PROXIMITY(PROXIMITY("zx", "zx", 5, 5), "zx", 0, 0)', // "zx (za) zx (zb zc zd) zx" - can obly be obtained from first
// you could imagine -ve left and right to mean within - would need -1,0 in stepping, and appropriate hard condition.
'PROXIMITY("ibm", "business", 2, 2)', // alias doesn't allow matching within itself.
'PROXIMITY("ibm", "business", 3, 3)', // alias should match now with other word
'PROXIMITY("ibm", "ibm", 0, 0)', // aliases and non aliases cause fun.
//More combinations of operators
'AND(OR("twinkle", "black"), OR("sheep", "wonder"))',
'OR(AND("twinkle", "sheep"), AND("star", "black"))',
'OR(AND("twinkle", "star"), AND("sheep", "black"))',
//Silly queries
'OR("star","star","star","star","star")',
'AND("star","star","star","star","star")',
'MOFN(4,"star","star","star","star","star")',
//Other operators
'PRE("twinkle", "twinkle")',
'PRE(PHRASE("twinkle", "twinkle"), PHRASE("little","star"))',
'PRE(PHRASE("little","star"), PHRASE("twinkle", "twinkle"))',
'PRE(PROXIMITY("twinkle","twinkle", 3, 3), PROXIMITY("little", "star", 2, 2))',
'AFT("twinkle", "twinkle")',
'AFT(PHRASE("little","star"), PHRASE("twinkle", "twinkle"))',
'AFT(PHRASE("twinkle", "twinkle"), PHRASE("little","star"))',
'AFT(PROXIMITY("twinkle","twinkle", 3, 3), PROXIMITY("little", "star", 2, 2))',
// Left outer joins for ranking.
'RANK("sheep", OR("peep", "baa"))',
'RANK("three", OR("bags", "full"))',
'RANK("three", OR("one", "bags"))',
//Non standard variants - AND, generating a single record for the match. Actually for each cross product as it is currently (and logically) implemented
'ANDJOIN("nonexistant1", "nonexistant2")', // neither side matches
'ANDJOIN("nonexistant1", "sheep")', // RHS matches
'ANDJOIN("twinkle", "nonexistant2")', // LHS matches
'ANDJOIN("twinkle", "twinkle")', // should dedup
'ANDJOIN("sheep", "black")', // matches in same document
'ANDJOIN("sheep", "twinkle")', // matches in different documents
'ANDJOIN("in", "a")', // high frequencies
'ANDJOIN("twinkle", "little", "how", "star")', // Nary
'ANDNOTJOIN("nonexistant1", "nonexistant2")', // neither side matches
'ANDNOTJOIN("nonexistant1", "sheep")', // RHS matches
'ANDNOTJOIN("twinkle", "nonexistant2")', // LHS matches
'ANDNOTJOIN("twinkle", "twinkle")', // should dedup
'ANDNOTJOIN("sheep", "black")', // matches in same document
'ANDNOTJOIN("sheep", "twinkle")', // matches in different documents
'ANDNOTJOIN("one", "sheep", "black", "fish")', // matches one, but none of the others
'MOFNJOIN(2, "humpty", "horses", "together", "beansprout")', // m<matches
'MOFNJOIN(3, "humpty", "horses", "together", "beansprout")', // m=matches
'MOFNJOIN(4, "humpty", "horses", "together", "beansprout")', // m>matches
'MOFNJOIN(2,2, "humpty", "horses", "together", "beansprout")', // too many matches
'MOFNJOIN(2, "nonexistant", "little", "bo")', // first input fails to match any
'MOFNJOIN(2, "little", "bo", "nonexistant")', // lose an input while finising candidates
'MOFNJOIN(2, "one", "two", "three", "four", "five")',
'MOFNJOIN(2, "nonexistant", "two", "three", "four", "five")',
'MOFNJOIN(2, "one", "nonexistant", "three", "four", "five")',
'MOFNJOIN(2, "nonexistant1", "nonexistant2", "three", "four", "five")',
'MOFNJOIN(2, "nonexistant1", "nonexistant2", "nonexistant3", "four", "five")',
'MOFNJOIN(2, "nonexistant1", "nonexistant2", "nonexistant3", "nonexistant4", "five")',
'MOFNJOIN(2, PHRASE("little","bo"),PHRASE("three","bags"),"sheep")', // m of n on phrases
'MOFNJOIN(2, PHRASE("Little","Bo"),PHRASE("three","bags"),"sheep")', // m of n on phrases - capital letters don't match
'MOFNJOIN(2, OR("little","big"), OR("king", "queen"), OR("star", "sheep", "twinkle"))',
'RANKJOIN("sheep", "black")',
'RANKJOIN("sheep", OR("peep", "baa"))',
'RANKJOIN("three", OR("bags", "full"))',
'RANKJOIN("three", OR("one", "bags"))',
//ROLLAND - does AND, followed by a rollup by doc. Should also check that smart stepping still works through the grouped rollup
'ROLLAND("nonexistant1", "nonexistant2")', // neither side matches
'ROLLAND("nonexistant1", "sheep")', // RHS matches
'ROLLAND("twinkle", "nonexistant2")', // LHS matches
'ROLLAND("twinkle", "twinkle")', // should dedup
'ROLLAND("sheep", "black")', // matches in same document
'ROLLAND("sheep", "twinkle")', // matches in different documents
'ROLLAND("in", "a")', // high frequencies
'ROLLAND("twinkle", "little", "how", "star")', // Nary
'AND(ROLLAND("twinkle", "little"), ROLLAND("how", "star"))', // Nary
/*
//Same tests as proximity above, but not calling a transform - merging instead
'PROXMERGE("ship", "sank", 0, 0)', // either order but adjacent
'PROXMERGE("ship", "sank", -1, 0)', // must follow
'PROXMERGE("ship", "sank", 0, -1)', // must preceed
'PROXMERGE("sank", "ship", 0, 0)', // either order but adjacent
'PROXMERGE("sank", "ship", -1, 0)', // must follow
'PROXMERGE("sank", "ship", 0, -1)', // must preceed
'PROXMERGE("ship", "sank", 1, 1)', // either order but only 1 intervening word
'PROXMERGE("ship", "sank", -1, 1)', // must follow
'PROXMERGE("ship", "sank", 1, -1)', // must preceed
'PROXMERGE("sank", "ship", 1, 1)', // either order but only 1 intervening word
'PROXMERGE("sank", "ship", -1, 1)', // must follow
'PROXMERGE("sank", "ship", 1, -1)', // must preceed
'PROXMERGE("ship", "sank", 0, 2)', // asymetric range
*/
//SET should be equivalent to OR
'SET("nonexistant1", "nonexistant2")', // neither side matches
'SET("nonexistant1", "sheep")', // RHS matches
'SET("twinkle", "nonexistant2")', // LHS matches
'SET("twinkle", "twinkle")', // should dedup
#end
'SET("sheep", "black")', // matches in same document
'SET("sheep", "twinkle")', // matches in different documents
'SET("one", "sheep", "sheep", "black", "fish")', // matches in different documents
'OR(SET("one", "sheep"), SET("sheep", "black", "fish"))', // matches in different documents
//MORE:
// SET("a","b","c") to test stepping merging.
// STEPPED flag on merge to give an error if input doesn't support stepping.
// What about the duplicates that can come out of the proximity operators?
// where the next on the rhs is at a compatible position, but in a different document
// What about inverse of proximity x not w/n y
// Can inverse proximity be used for sentance/paragraph. Can we combine them so short circuited before temporaries created.
//MORE: What other boundary conditions can we think of.
''
], inputRecord);
p := project(q1, doBatchExecute(LEFT));
output(p);
| ECL | 4 | miguelvazq/HPCC-Platform | ecl/regress/localbug.ecl | [
"Apache-2.0"
]
|
-- test cases for like any
CREATE OR REPLACE TEMPORARY VIEW like_any_table AS SELECT * FROM (VALUES
('google', '%oo%'),
('facebook', '%oo%'),
('linkedin', '%in'))
as t1(company, pat);
SELECT company FROM like_any_table WHERE company LIKE ANY ('%oo%', '%in', 'fa%');
SELECT company FROM like_any_table WHERE company LIKE ANY ('microsoft', '%yoo%');
select
company,
CASE
WHEN company LIKE ANY ('%oo%', '%in', 'fa%') THEN 'Y'
ELSE 'N'
END AS is_available,
CASE
WHEN company LIKE ANY ('%oo%', 'fa%') OR company LIKE ANY ('%in', 'ms%') THEN 'Y'
ELSE 'N'
END AS mix
FROM like_any_table;
-- Mix test with constant pattern and column value
SELECT company FROM like_any_table WHERE company LIKE ANY ('%zz%', pat);
-- not like any test
SELECT company FROM like_any_table WHERE company NOT LIKE ANY ('%oo%', '%in', 'fa%');
SELECT company FROM like_any_table WHERE company NOT LIKE ANY ('microsoft', '%yoo%');
SELECT company FROM like_any_table WHERE company NOT LIKE ANY ('%oo%', 'fa%');
SELECT company FROM like_any_table WHERE NOT company LIKE ANY ('%oo%', 'fa%');
-- null test
SELECT company FROM like_any_table WHERE company LIKE ANY ('%oo%', NULL);
SELECT company FROM like_any_table WHERE company NOT LIKE ANY ('%oo%', NULL);
SELECT company FROM like_any_table WHERE company LIKE ANY (NULL, NULL);
SELECT company FROM like_any_table WHERE company NOT LIKE ANY (NULL, NULL);
-- negative case
SELECT company FROM like_any_table WHERE company LIKE ANY ();
| SQL | 4 | kesavanvt/spark | sql/core/src/test/resources/sql-tests/inputs/like-any.sql | [
"BSD-2-Clause",
"Apache-2.0",
"CC0-1.0",
"MIT",
"MIT-0",
"ECL-2.0",
"BSD-3-Clause-No-Nuclear-License-2014",
"BSD-3-Clause"
]
|
<html>
<body>
<script type="text/javascript" charset="utf-8">
window.addEventListener('unload', function (e) {
require('fs').writeFileSync(__dirname + '/unload', 'unload');
}, false);
</script>
</body>
</html>
| HTML | 3 | lingxiao-Zhu/electron | spec/fixtures/api/unload.html | [
"MIT"
]
|
.class public Lconditions/TestBooleanToShort;
.super Ljava/lang/Object;
.field private showConsent:Z
.method public writeToParcel(Lconditions/TestBooleanToShort;)V
.locals 0
iget-boolean p1, p0, Lconditions/TestBooleanToShort;->showConsent:Z
int-to-short p1, p1
invoke-virtual {p0, p1}, Lconditions/TestBooleanToShort;->write(S)V
return-void
.end method
.method public write(S)V
.locals 0
return-void
.end method
| Smali | 2 | Dev-kishan1999/jadx | jadx-core/src/test/smali/conditions/TestBooleanToShort.smali | [
"Apache-2.0"
]
|
= jcstress Tests
:Author: LMAX Development Team
:Email:
:Date: {docdata}
https://github.com/openjdk/jcstress/
> The Java Concurrency Stress (jcstress) is the experimental harness and a suite of tests to aid the research in the correctness of concurrency support in the JVM, class libraries, and hardware.
The Disruptor has some jcstress tests to experiment and validate the correctness of the concurrency promises that the
Disruptor makes.
== Running jcstress Tests
[source,shell script]
----
$ ./gradlew jcstress
----
| AsciiDoc | 3 | Demoncloudy/disruptor | src/docs/asciidoc/en/developer-guide/25_jsctress_tests.adoc | [
"Apache-2.0"
]
|
#![crate_name = "foo"]
pub mod a {
pub struct Foo;
pub enum Bar {
Baz,
}
}
// @count 'foo/index.html' '//*[code="pub use a::Foo;"]' 1
#[doc(no_inline)]
pub use a::Foo;
// @count 'foo/index.html' '//*[code="pub use a::Bar::Baz;"]' 1
#[doc(no_inline)]
pub use a::Bar::Baz;
| Rust | 3 | Eric-Arellano/rust | src/test/rustdoc/constructor-imports.rs | [
"ECL-2.0",
"Apache-2.0",
"MIT-0",
"MIT"
]
|
#include "__emit.inc"
stock test__shift(&local_refvar, local_refarray[])
{
const local_const = 0;
new local_var = 0;
static local_static_var = 0;
local_label:
// ok
__emit shl.c.pri global_const;
__emit shl.c.pri local_const;
__emit shl.c.pri 0;
__emit shl.c.pri 1;
__emit shl.c.pri 15;
__emit shl.c.pri (cellbits / charbits);
#if cellbits == 16
__emit shl.c.pri 15;
__emit shl.c.pri 0xF;
#elseif cellbits == 32
__emit shl.c.pri 31;
__emit shl.c.pri 0x1F;
#else // cellbits == 64
__emit shl.c.pri 63;
__emit shl.c.pri 0x3F;
#endif
// should trigger an error
__emit shl.c.pri global_var;
__emit shl.c.pri global_func;
__emit shl.c.pri local_refvar;
__emit shl.c.pri local_refarray;
__emit shl.c.pri local_var;
__emit shl.c.pri local_static_var;
__emit shl.c.pri local_label;
__emit shl.c.pri -1;
__emit shl.c.pri -0x1;
#if cellbits == 16
__emit shl.c.pri 16;
__emit shl.c.pri 0x10;
#elseif cellbits == 32
__emit shl.c.pri 32;
__emit shl.c.pri 0x20;
#else // cellbits == 64
__emit shl.c.pri 64;
__emit shl.c.pri 0x40;
#endif
}
stock test__label(&local_refvar, local_refarray[])
{
const local_const = 0;
new local_var = 0;
static local_static_var = 0;
local_label:
// ok
__emit jump local_label;
__emit jump local_label2;
__emit jump :local_label2;
local_label2:
// should trigger an error
__emit jump global_const;
__emit jump global_var;
__emit jump global_func;
__emit jump local_refvar;
__emit jump local_refarray;
__emit jump local_const;
__emit jump local_var;
__emit jump local_static_var;
__emit jump 0;
}
main()
{
new t, a[1];
test__shift(t, a); // 11
test__label(t, a); // 9
}
| PAWN | 4 | pawn-lang/pawn | source/compiler/tests/__emit_p2.pwn | [
"Zlib"
]
|
0 reg32_t "dword"
1 code_t "proc*"
2 num32_t "int"
3 uint32_t "size_t"
4 ptr(num8_t) "char*"
5 ptr(ptr(num8_t)) "char**"
6 ptr(struct(0:num32_t,4:ptr(num8_t),8:ptr(num8_t),12:ptr(num8_t),16:ptr(num8_t),20:ptr(num8_t),24:ptr(num8_t),28:ptr(num8_t),32:ptr(num8_t),36:ptr(num8_t),40:ptr(num8_t),44:ptr(num8_t),48:ptr(struct(0:ptr(TOP),4:ptr(struct(0:num32_t,4:ptr(reg8_t),8:ptr(reg8_t),12:ptr(reg8_t),16:ptr(reg8_t),20:ptr(reg8_t),24:ptr(reg8_t),28:ptr(reg8_t),32:ptr(reg8_t),36:ptr(reg8_t),40:ptr(reg8_t),44:ptr(reg8_t),48:ptr(TOP),52:ptr(TOP),56:num32_t,60:num32_t,64:num32_t,68:uint16_t,70:int8_t,71:array(num8_t,1),72:ptr(TOP),76:num64_t,84:ptr(TOP),88:ptr(TOP),92:ptr(TOP),96:ptr(TOP),100:uint32_t,104:num32_t,108:array(num8_t,40))),8:num32_t)),52:ptr(struct(0:num32_t,4:ptr(num8_t),8:ptr(num8_t),12:ptr(num8_t),16:ptr(num8_t),20:ptr(num8_t),24:ptr(num8_t),28:ptr(num8_t),32:ptr(num8_t),36:ptr(num8_t),40:ptr(num8_t),44:ptr(num8_t),48:ptr(struct(0:ptr(TOP),4:ptr(TOP),8:num32_t)),52:ptr(TOP),56:num32_t,60:num32_t,64:num32_t,68:uint16_t,70:int8_t,71:array(num8_t,1),72:ptr(TOP),76:num64_t,84:ptr(TOP),88:ptr(TOP),92:ptr(TOP),96:ptr(TOP),100:uint32_t,104:num32_t,108:array(num8_t,40))),56:num32_t,60:num32_t,64:num32_t,68:uint16_t,70:int8_t,71:array(num8_t,1),72:ptr(TOP),76:num64_t,84:ptr(TOP),88:ptr(TOP),92:ptr(TOP),96:ptr(TOP),100:uint32_t,104:num32_t,108:array(num8_t,40))) "FILE*"
7 ptr(TOP) "void*"
3 uint32_t "__gid_t"
8 ptr(uint32_t) "__gid_t*"
3 uint32_t "__uid_t"
3 uint32_t "unsigned int"
9 num8_t "char"
10 ptr(array(reg8_t,16)) "unknown_128*"
11 ptr(array(reg8_t,56)) "unknown_448*"
12 ptr(array(reg8_t,177)) "unknown_1416*"
13 ptr(array(reg8_t,58)) "unknown_464*"
14 union(ptr(num8_t),ptr(struct(0:reg16_t,2:num8_t))) "Union_0"
15 ptr(num32_t) "int*"
16 ptr(struct(0:reg16_t,2:num8_t)) "StructFrag_0*"
17 ptr(reg32_t) "dword*"
4 ptr(num8_t) "char[]"
18 ptr(reg16_t) "word*"
17 ptr(reg32_t) "dword[]"
19 ptr(struct(0:ptr(num8_t),4:ptr(num8_t),8:uint32_t,12:ptr(ptr(num8_t)))) "group*"
20 ptr(struct(0:num32_t,4:ptr(ptr(num8_t)),4294967292:reg32_t)) "Struct_4*"
21 ptr(struct(0:ptr(num8_t),4:num32_t,8:ptr(num32_t),12:num32_t)) "option*"
22 ptr(ptr(TOP)) "void**"
23 ptr(struct(0:ptr(num8_t),4:ptr(num8_t),8:uint32_t,12:uint32_t,16:ptr(num8_t),20:ptr(num8_t),24:ptr(num8_t))) "passwd*"
15 ptr(num32_t) "int[]"
24 ptr(ptr(num32_t)) "int[]*"
25 ptr(ptr(struct(0:reg32_t,4:num8_t))) "StructFrag_1**"
26 ptr(struct(0:array(reg8_t,6),6:num8_t)) "StructFrag_3*"
27 ptr(struct(0:reg32_t,40:ptr(num8_t),44:ptr(num8_t))) "Struct_1*"
28 ptr(struct(40:ptr(num8_t),44:ptr(num8_t))) "Struct_2*"
8 ptr(uint32_t) "size_t*"
29 array(reg8_t,3) "unknown_24"
30 ptr(struct(0:reg32_t,4:ptr(TOP))) "Struct_0*"
31 reg64_t "qword"
32 array(reg8_t,32) "unknown_256"
33 int32_t "signed int"
8 ptr(uint32_t) "unsigned int*"
34 union(ptr(struct(0:reg32_t,40:ptr(num8_t),44:ptr(num8_t))),ptr(reg32_t)) "Union_3"
35 ptr(struct(0:reg32_t,4:reg32_t)) "StructFrag_6*"
36 ptr(struct(0:array(reg8_t,40),40:num8_t)) "StructFrag_14*"
37 ptr(ptr(uint16_t)) "unsigned short**"
38 ptr(union(ptr(num8_t),ptr(struct(0:reg16_t,2:num8_t)))) "Union_0*"
39 ptr(struct(0:array(reg8_t,16),16:uint32_t)) "StructFrag_13*"
40 ptr(code_t) "proc**"
41 ptr(uint16_t) "unsigned short*"
42 union(ptr(num8_t),ptr(struct(0:reg32_t,4:num8_t))) "Union_1"
43 ptr(struct(0:array(reg8_t,66458),66458:reg32_t)) "StructFrag_8*"
44 ptr(struct(0:array(reg8_t,536870908),4294967292:reg32_t)) "StructFrag_9*"
45 ptr(struct(0:array(reg8_t,29116),29116:reg32_t)) "StructFrag_11*"
46 ptr(struct(0:array(reg8_t,648),648:reg32_t)) "StructFrag_12*"
47 array(reg8_t,4096) "unknown_32768"
48 array(reg8_t,135168) "unknown_1081344"
49 array(reg8_t,30) "unknown_240"
50 array(reg8_t,5) "unknown_40"
51 array(reg8_t,21) "unknown_168"
52 array(reg8_t,16) "unknown_128"
53 array(reg8_t,41) "unknown_328"
54 array(reg8_t,7) "unknown_56"
55 array(reg8_t,51) "unknown_408"
56 array(reg8_t,13) "unknown_104"
57 array(reg8_t,27) "unknown_216"
58 array(reg8_t,17) "unknown_136"
59 reg16_t "word"
60 array(reg8_t,18) "unknown_144"
61 array(reg8_t,142) "unknown_1136"
62 array(reg8_t,53) "unknown_424"
63 array(reg8_t,55) "unknown_440"
64 array(reg8_t,12) "unknown_96"
65 array(reg8_t,86) "unknown_688"
66 array(reg8_t,33) "unknown_264"
67 array(reg8_t,92) "unknown_736"
68 array(reg8_t,9) "unknown_72"
69 array(reg8_t,79) "unknown_632"
70 array(reg8_t,31) "unknown_248"
71 array(reg8_t,26) "unknown_208"
72 array(reg8_t,10) "unknown_80"
73 array(reg8_t,64) "unknown_512"
74 array(reg8_t,22) "unknown_176"
75 array(reg8_t,24) "unknown_192"
76 array(reg8_t,83) "unknown_664"
77 array(reg8_t,77) "unknown_616"
78 array(reg8_t,222) "unknown_1776"
79 array(reg8_t,11) "unknown_88"
80 array(reg8_t,177) "unknown_1416"
81 array(reg8_t,29) "unknown_232"
82 array(reg8_t,81) "unknown_648"
83 array(reg8_t,47) "unknown_376"
84 array(reg8_t,52) "unknown_416"
85 array(reg8_t,42) "unknown_336"
86 array(reg8_t,71) "unknown_568"
87 array(reg8_t,60) "unknown_480"
88 array(reg8_t,56) "unknown_448"
89 array(reg8_t,46) "unknown_368"
90 array(reg8_t,14) "unknown_112"
91 array(reg8_t,105) "unknown_840"
92 array(reg8_t,62) "unknown_496"
93 array(reg8_t,114) "unknown_912"
94 array(reg8_t,28) "unknown_224"
95 array(reg8_t,35) "unknown_280"
96 array(reg8_t,38) "unknown_304"
97 array(reg8_t,111) "unknown_888"
98 array(reg8_t,39) "unknown_312"
99 array(reg8_t,58) "unknown_464"
100 array(reg8_t,15) "unknown_120"
101 array(reg8_t,40) "unknown_320"
102 array(reg8_t,95) "unknown_760"
103 array(reg8_t,34) "unknown_272"
104 array(reg8_t,23) "unknown_184"
105 array(reg8_t,19) "unknown_152"
106 array(reg8_t,20) "unknown_160"
107 array(reg8_t,91) "unknown_728"
108 array(reg8_t,36) "unknown_288"
109 array(reg8_t,48) "unknown_384"
110 array(reg8_t,126) "unknown_1008"
111 array(reg8_t,59) "unknown_472"
112 array(reg8_t,45) "unknown_360"
113 array(reg8_t,70) "unknown_560"
114 array(reg8_t,6) "unknown_48"
115 array(reg8_t,89) "unknown_712"
116 array(reg8_t,61) "unknown_488"
117 array(reg8_t,25) "unknown_200"
118 array(reg8_t,50) "unknown_400"
119 array(reg8_t,190) "unknown_1520"
120 array(reg8_t,49) "unknown_392"
121 array(reg8_t,173) "unknown_1384"
122 array(reg8_t,149) "unknown_1192"
123 array(reg8_t,146) "unknown_1168"
124 array(reg8_t,37) "unknown_296"
125 array(reg8_t,54) "unknown_432"
126 array(reg8_t,118) "unknown_944"
127 array(reg8_t,94) "unknown_752"
128 array(reg8_t,140) "unknown_1120"
129 array(reg8_t,57) "unknown_456"
130 array(reg8_t,101) "unknown_808"
131 array(reg8_t,102) "unknown_816"
132 array(reg8_t,75) "unknown_600"
133 array(reg8_t,98) "unknown_784"
134 array(reg8_t,113) "unknown_904"
135 array(reg8_t,69) "unknown_552"
136 array(reg8_t,100) "unknown_800"
137 array(reg8_t,44) "unknown_352"
138 array(reg8_t,73) "unknown_584"
139 array(reg8_t,129) "unknown_1032"
140 array(num8_t,23) "char[23]"
141 array(num8_t,39) "char[39]"
142 array(num8_t,14) "char[14]"
143 array(num8_t,4) "char[4]"
144 array(num8_t,69) "char[69]"
145 array(num8_t,65) "char[65]"
146 struct(0:ptr(num8_t),4:num32_t,8:ptr(num32_t),12:num32_t) "option"
147 array(num8_t,2) "char[2]"
148 array(num8_t,17) "char[17]"
149 array(num8_t,22) "char[22]"
150 array(num8_t,31) "char[31]"
151 array(num8_t,64) "char[64]"
152 array(num8_t,50) "char[50]"
153 array(num8_t,138) "char[138]"
154 array(num8_t,45) "char[45]"
155 array(num8_t,54) "char[54]"
156 array(num8_t,90) "char[90]"
157 array(num8_t,10) "char[10]"
158 array(num8_t,15) "char[15]"
159 array(num8_t,7) "char[7]"
160 array(num8_t,16) "char[16]"
161 array(num8_t,35) "char[35]"
162 array(num8_t,6) "char[6]"
163 array(num8_t,8) "char[8]"
164 array(num8_t,3) "char[3]"
165 array(num8_t,34) "char[34]"
166 array(num8_t,33) "char[33]"
167 array(num8_t,25) "char[25]"
168 array(num8_t,12) "char[12]"
169 array(num8_t,56) "char[56]"
170 array(reg32_t,127) "dword[127]"
171 array(reg32_t,30) "dword[30]"
172 array(reg32_t,34) "dword[34]"
173 array(num8_t,203) "char[203]"
174 array(num8_t,28) "char[28]"
175 array(num8_t,32) "char[32]"
176 array(num8_t,36) "char[36]"
177 array(num8_t,40) "char[40]"
178 array(num8_t,44) "char[44]"
179 array(num8_t,48) "char[48]"
180 array(num8_t,52) "char[52]"
181 array(num8_t,60) "char[60]"
182 array(num8_t,21) "char[21]"
183 array(num8_t,20) "char[20]"
184 array(num8_t,47) "char[47]"
185 array(num8_t,38) "char[38]"
186 array(ptr(TOP),54) "void*[54]"
187 array(num8_t,9) "char[9]"
188 array(num8_t,81) "char[81]"
189 array(reg8_t,900) "unknown_7200"
190 array(reg8_t,3856) "unknown_30848"
191 array(reg8_t,5132) "unknown_41056"
1 code_t "(void -?-> dword)*"
192 array(reg8_t,232) "unknown_1856"
193 array(reg8_t,256) "unknown_2048"
| BlitzBasic | 1 | matt-noonan/retypd-data | data/chroot.decls | [
"MIT"
]
|
crypto
onchain
gwei
top
ueat
hr ETH
exit | Gosu | 0 | elan17/GamestonkTerminal | scripts/test_crypto_onchain.gst | [
"MIT"
]
|
;;; lang/org/autoload/org-babel.el -*- lexical-binding: t; -*-
;;;###autoload
(defun +org-eval-handler (beg end)
"TODO"
(save-excursion
(if (not (cl-loop for pos in (list beg (point) end)
if (save-excursion (goto-char pos) (org-in-src-block-p t))
return (goto-char pos)))
(message "Nothing to evaluate at point")
(let* ((element (org-element-at-point))
(block-beg (save-excursion
(goto-char (org-babel-where-is-src-block-head element))
(line-beginning-position 2)))
(block-end (save-excursion
(goto-char (org-element-property :end element))
(skip-chars-backward " \t\n")
(line-beginning-position)))
(beg (if beg (max beg block-beg) block-beg))
(end (if end (min end block-end) block-end))
(lang (or (org-eldoc-get-src-lang)
(user-error "No lang specified for this src block"))))
(cond ((and (string-prefix-p "jupyter-" lang)
(require 'jupyter nil t))
(jupyter-eval-region beg end))
((let ((major-mode (org-src-get-lang-mode lang)))
(+eval/region beg end))))))))
;;;###autoload
(defun +org-lookup-definition-handler (identifier)
"TODO"
(when (org-in-src-block-p t)
(let ((mode (org-src-get-lang-mode
(or (org-eldoc-get-src-lang)
(user-error "No lang specified for this src block")))))
(cond ((and (eq mode 'emacs-lisp-mode)
(fboundp '+emacs-lisp-lookup-definition))
(+emacs-lisp-lookup-definition identifier)
'deferred)
((user-error "Definition lookup in SRC blocks isn't supported yet"))))))
;;;###autoload
(defun +org-lookup-references-handler (identifier)
"TODO"
(when (org-in-src-block-p t)
(user-error "References lookup in SRC blocks isn't supported yet")))
;;;###autoload
(defun +org-lookup-documentation-handler (identifier)
"TODO"
(when (org-in-src-block-p t)
(let ((mode (org-src-get-lang-mode
(or (org-eldoc-get-src-lang)
(user-error "No lang specified for this src block"))))
(info (org-babel-get-src-block-info t)))
(cond ((string-prefix-p "jupyter-" (car info))
(and (require 'jupyter nil t)
(call-interactively #'jupyter-inspect-at-point)
(display-buffer (help-buffer))
'deferred))
((and (eq mode 'emacs-lisp-mode)
(fboundp '+emacs-lisp-lookup-documentation))
(+emacs-lisp-lookup-documentation identifier)
'deferred)
((user-error "Documentation lookup in SRC blocks isn't supported yet"))))))
;;
;;; Hooks
;;;###autoload
(defun +org-clear-babel-results-h ()
"Remove the results block for the org babel block at point."
(when (and (org-in-src-block-p t)
(org-babel-where-is-src-block-result))
(org-babel-remove-result)
t))
| Emacs Lisp | 4 | leezu/doom-emacs | modules/lang/org/autoload/org-babel.el | [
"MIT"
]
|
/*
*
* Copyright 2016 gRPC authors.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*
*/
// Hack TEST macro of gTest and make they conform XCTest style. We only
// need test name (b), not test case name (a).
#define TEST(a, b) -(void)test##b
#define ASSERT_TRUE XCTAssert
#define ASSERT_EQ XCTAssertEqual
#import <XCTest/XCTest.h>
#include <grpcpp/test/server_context_test_spouse.h>
#include <cstring>
#include <vector>
#include <grpcpp/impl/grpc_library.h>
static grpc::internal::GrpcLibraryInitializer g_initializer;
const char key1[] = "metadata-key1";
const char key2[] = "metadata-key2";
const char val1[] = "metadata-val1";
const char val2[] = "metadata-val2";
bool ClientMetadataContains(const grpc::ServerContext& context, const grpc::string_ref& key,
const grpc::string_ref& value) {
const auto& client_metadata = context.client_metadata();
for (auto iter = client_metadata.begin(); iter != client_metadata.end(); ++iter) {
if (iter->first == key && iter->second == value) {
return true;
}
}
return false;
}
@interface ServerContextTestSpouseTest : XCTestCase
@end
@implementation ServerContextTestSpouseTest
TEST(ServerContextTestSpouseTest, ClientMetadata) {
grpc::ServerContext context;
grpc::testing::ServerContextTestSpouse spouse(&context);
spouse.AddClientMetadata(key1, val1);
ASSERT_TRUE(ClientMetadataContains(context, key1, val1));
spouse.AddClientMetadata(key2, val2);
ASSERT_TRUE(ClientMetadataContains(context, key1, val1));
ASSERT_TRUE(ClientMetadataContains(context, key2, val2));
}
TEST(ServerContextTestSpouseTest, InitialMetadata) {
grpc::ServerContext context;
grpc::testing::ServerContextTestSpouse spouse(&context);
std::multimap<std::string, std::string> metadata;
context.AddInitialMetadata(key1, val1);
metadata.insert(std::pair<std::string, std::string>(key1, val1));
ASSERT_EQ(metadata, spouse.GetInitialMetadata());
context.AddInitialMetadata(key2, val2);
metadata.insert(std::pair<std::string, std::string>(key2, val2));
ASSERT_EQ(metadata, spouse.GetInitialMetadata());
}
TEST(ServerContextTestSpouseTest, TrailingMetadata) {
grpc::ServerContext context;
grpc::testing::ServerContextTestSpouse spouse(&context);
std::multimap<std::string, std::string> metadata;
context.AddTrailingMetadata(key1, val1);
metadata.insert(std::pair<std::string, std::string>(key1, val1));
ASSERT_EQ(metadata, spouse.GetTrailingMetadata());
context.AddTrailingMetadata(key2, val2);
metadata.insert(std::pair<std::string, std::string>(key2, val2));
ASSERT_EQ(metadata, spouse.GetTrailingMetadata());
}
@end
| Objective-C++ | 5 | arghyadip01/grpc | test/cpp/cocoapods/test/server_context_test_spouse_test.mm | [
"Apache-2.0"
]
|
( Generated from test_externs_in.muv by the MUV compiler. )
( https://github.com/revarbat/pymuv )
: _main[ -- ret ]
var _v
voidfoo 0 _v !
singlefoo _v !
{ multfoo }list _v !
qux _v !
{ "Fee" "Fie" "Foe" }list array_interpret _v !
"%d: %s" { 5 "Fum" }list
2 try
array_explode 1 + rotate fmtstring
depth 0 swap - rotate depth 1 - popn
catch abort
endcatch _v !
0
;
: __start
"me" match me ! me @ location loc ! trig trigger !
_main
;
| MUF | 3 | revarbat/pymuv | tests/test_externs_cmp.muf | [
"MIT"
]
|
package unit.issues;
using unit.issues.Issue8760.SortStringTools;
using unit.issues.Issue8760.SortFloatTools;
private class SortStringTools {
public static function sorted<T:String, A:Array<T>>(arr:A) {
return 'sorted<T:String>()';
}
}
private class SortFloatTools {
public static function sorted<T:Float, A:Array<T>>(arr:A) {
return 'sorted<T:Float>()';
}
}
class Issue8760 extends Test {
function test() {
var arr1 = ["abc", "def"];
eq("sorted<T:String>()", arr1.sorted());
var arr2 = [123,456];
eq("sorted<T:Float>()", arr2.sorted());
}
} | Haxe | 4 | Alan-love/haxe | tests/unit/src/unit/issues/Issue8760.hx | [
"MIT"
]
|
HAI 1.2
BTW NOT
NOT WIN
BTW AND
BOTH OF WIN AN FAIL
BTW OR
EITHER OF WIN AN FAIL
BTW XOR
WON OF WIN AN FAIL
BTW AND (infinite arity)
ALL OF WIN AN FAIL AN FAIL AN WIN MKAY
BTW OR (infinite arity)
ANY OF WIN AN FAIL AN FAIL AN WIN MKAY
KTHXBYE
| LOLCODE | 2 | Himanshu21git/Lokalise-source | lolcode-fun-post/boolean_operators.lol | [
"MIT"
]
|
{{! Copyright (c) Avanade. Licensed under the MIT License. See https://github.com/Avanade/Beef }}
CREATE TABLE [{{CdcSchema}}].[{{OutboxTableName}}] (
/*
* This is automatically generated; any changes will be lost.
*/
[OutboxId] INT IDENTITY (1, 1) NOT NULL PRIMARY KEY CLUSTERED ([OutboxId] ASC),
[CreatedDate] DATETIME NOT NULL,
[{{pascal Name}}MinLsn] BINARY(10) NOT NULL, -- Primary table: {{Schema}}.{{Name}}
[{{pascal Name}}MaxLsn] BINARY(10) NOT NULL,
{{#each CdcJoins}}
[{{pascal Name}}MinLsn] BINARY(10) NOT NULL, -- Related table: {{Schema}}.{{TableName}}
[{{pascal Name}}MaxLsn] BINARY(10) NOT NULL,
{{/each}}
[IsComplete] BIT NOT NULL,
[CompletedDate] DATETIME NULL
); | Harbour | 4 | ualehosaini/Beef | tools/Beef.CodeGen.Core/Templates/DbCdcOutboxTableCreate_sql.hb | [
"MIT"
]
|
<?xml version="1.0" encoding="utf-8"?>
<Project ToolsVersion="4.0" DefaultTargets="Build" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<PropertyGroup>
<Configuration Condition=" '$(Configuration)' == '' ">Debug</Configuration>
<Platform Condition=" '$(Platform)' == '' ">AnyCPU</Platform>
<ProductVersion>10.0.20506</ProductVersion>
<SchemaVersion>2.0</SchemaVersion>
<ProjectGuid>{ef1654e8-f2e2-43a2-888a-eb1cb177f75a}</ProjectGuid>
<ProjectTypeGuids>{89896941-7261-4476-8385-4DA3CE9FDB83};{C089C8C0-30E0-4E22-80C0-CE093F111A43};{656346D9-4656-40DA-A068-22D5425D4639}</ProjectTypeGuids>
<OutputType>Library</OutputType>
<AppDesignerFolder>Properties</AppDesignerFolder>
<RootNamespace>Sugar.Echoes.WP8.Test</RootNamespace>
<AssemblyName>Sugar.Echoes.WP8.Test</AssemblyName>
<TargetFrameworkIdentifier>WindowsPhone</TargetFrameworkIdentifier>
<TargetFrameworkVersion>v8.0</TargetFrameworkVersion>
<SilverlightVersion>$(TargetFrameworkVersion)</SilverlightVersion>
<SilverlightApplication>true</SilverlightApplication>
<SupportedCultures>
</SupportedCultures>
<XapOutputs>true</XapOutputs>
<GenerateSilverlightManifest>true</GenerateSilverlightManifest>
<XapFilename>Sugar.Echoes.WP8.Test_$(Configuration)_$(Platform).xap</XapFilename>
<SilverlightManifestTemplate>Properties\AppManifest.xml</SilverlightManifestTemplate>
<SilverlightAppEntry>Sugar.Echoes.WP8.Test.App</SilverlightAppEntry>
<ValidateXaml>true</ValidateXaml>
<MinimumVisualStudioVersion>11.0</MinimumVisualStudioVersion>
<ThrowErrorsInValidation>true</ThrowErrorsInValidation>
<Name>Sugar.Echoes.WP8.Test</Name>
<AllowLegacyCreate>False</AllowLegacyCreate>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|AnyCPU' ">
<DebugSymbols>true</DebugSymbols>
<DebugType>full</DebugType>
<Optimize>false</Optimize>
<OutputPath>..\..\Bin\Debug\WP8</OutputPath>
<DefineConstants>DEBUG;TRACE;SILVERLIGHT;WINDOWS_PHONE</DefineConstants>
<NoStdLib>true</NoStdLib>
<NoConfig>true</NoConfig>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
<XmlDocWarningLevel>WarningOnPublicMembers</XmlDocWarningLevel>
<CpuType>anycpu</CpuType>
<GeneratePDB>True</GeneratePDB>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release|AnyCPU' ">
<DebugType>pdbonly</DebugType>
<Optimize>true</Optimize>
<OutputPath>..\..\Bin\Release\WP8</OutputPath>
<DefineConstants>TRACE;SILVERLIGHT;WINDOWS_PHONE</DefineConstants>
<NoStdLib>true</NoStdLib>
<NoConfig>true</NoConfig>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|x86' ">
<DebugSymbols>true</DebugSymbols>
<DebugType>full</DebugType>
<Optimize>False</Optimize>
<OutputPath>..\..\Bin\x86\Debug\WP8</OutputPath>
<DefineConstants>DEBUG;TRACE;SILVERLIGHT;WINDOWS_PHONE</DefineConstants>
<NoStdLib>true</NoStdLib>
<NoConfig>true</NoConfig>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
<XmlDocWarningLevel>WarningOnPublicMembers</XmlDocWarningLevel>
<CpuType>anycpu</CpuType>
<GeneratePDB>True</GeneratePDB>
<GenerateMDB>True</GenerateMDB>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release|x86' ">
<DebugType>pdbonly</DebugType>
<Optimize>true</Optimize>
<OutputPath>..\..\Bin\x86\Release\WP8</OutputPath>
<DefineConstants>TRACE;SILVERLIGHT;WINDOWS_PHONE</DefineConstants>
<NoStdLib>true</NoStdLib>
<NoConfig>true</NoConfig>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|ARM' ">
<DebugSymbols>true</DebugSymbols>
<DebugType>full</DebugType>
<Optimize>false</Optimize>
<OutputPath>..\..\Bin\ARM\Debug\WP8</OutputPath>
<DefineConstants>DEBUG;TRACE;SILVERLIGHT;WINDOWS_PHONE</DefineConstants>
<NoStdLib>true</NoStdLib>
<NoConfig>true</NoConfig>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release|ARM' ">
<DebugType>pdbonly</DebugType>
<Optimize>true</Optimize>
<OutputPath>..\..\Bin\ARM\Release\WP8</OutputPath>
<DefineConstants>TRACE;SILVERLIGHT;WINDOWS_PHONE</DefineConstants>
<NoStdLib>true</NoStdLib>
<NoConfig>true</NoConfig>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<ItemGroup>
<Reference Include="RemObjects.Elements.EUnit" />
</ItemGroup>
<ItemGroup>
<Compile Include="App.xaml.pas">
<DependentUpon>App.xaml</DependentUpon>
</Compile>
<Compile Include="LocalizedStrings.pas" />
<Compile Include="MainPage.xaml.pas">
<DependentUpon>MainPage.xaml</DependentUpon>
</Compile>
<Compile Include="Properties\AssemblyInfo.pas" />
<Compile Include="Resources\AppResources.Designer.pas">
<DependentUpon>Resources\AppResources.resx</DependentUpon>
</Compile>
</ItemGroup>
<ItemGroup>
<ApplicationDefinition Include="App.xaml">
<SubType>Designer</SubType>
<Generator>MSBuild:Compile</Generator>
</ApplicationDefinition>
<Page Include="MainPage.xaml">
<SubType>Designer</SubType>
<Generator>MSBuild:Compile</Generator>
</Page>
</ItemGroup>
<ItemGroup>
<None Include="Properties\AppManifest.xml" />
<None Include="Properties\WMAppManifest.xml">
<SubType>Designer</SubType>
</None>
</ItemGroup>
<ItemGroup>
<Content Include="Assets\AlignmentGrid.png" />
<Content Include="Assets\ApplicationIcon.png">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content>
<Content Include="Assets\Tiles\FlipCycleTileLarge.png">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content>
<Content Include="Assets\Tiles\FlipCycleTileMedium.png">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content>
<Content Include="Assets\Tiles\FlipCycleTileSmall.png">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content>
<Content Include="Assets\Tiles\IconicTileMediumLarge.png">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content>
<Content Include="Assets\Tiles\IconicTileSmall.png">
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
</Content>
</ItemGroup>
<ItemGroup>
<EmbeddedResource Include="Resources\AppResources.resx">
<Generator>PublicResXFileCodeGenerator</Generator>
</EmbeddedResource>
</ItemGroup>
<ItemGroup>
<Folder Include="Assets\" />
<Folder Include="Assets\Tiles\" />
<Folder Include="Properties\" />
<Folder Include="Resources\" />
</ItemGroup>
<ItemGroup>
<ProjectReference Include="..\..\..\Sugar\Sugar.Echoes.WP8.oxygene">
<Name>Sugar.Echoes.WP8</Name>
<Project>{bd5d51e2-b11a-4356-9450-64566f8cdc38}</Project>
<Private>True</Private>
<HintPath>..\..\..\Sugar\bin\Debug\WP8\Sugar.dll</HintPath>
</ProjectReference>
</ItemGroup>
<!-- To modify your build process, add your task inside one of the targets below and uncomment it.
Other similar extension points exist, see Microsoft.Common.targets.
<Target Name="BeforeBuild">
</Target>
<Target Name="AfterBuild">
</Target>
-->
<ProjectExtensions>
<VisualStudio>
<FlavorProperties Guid="{c089c8c0-30e0-4e22-80c0-ce093f111a43}" User="">
<SilverlightMobileCSProjectFlavor>
<FullDeploy>False</FullDeploy>
<DebuggerType>Managed</DebuggerType>
<DebuggerAgentType>Managed</DebuggerAgentType>
<Tombstone>False</Tombstone>
</SilverlightMobileCSProjectFlavor>
</FlavorProperties>
</VisualStudio>
</ProjectExtensions>
<Import Project="$(MSBuildExtensionsPath)\Microsoft\$(TargetFrameworkIdentifier)\$(TargetFrameworkVersion)\Microsoft.$(TargetFrameworkIdentifier).$(TargetFrameworkVersion).Overrides.targets" />
<Import Project="$(MSBuildExtensionsPath)\RemObjects Software\Oxygene\$(TargetFrameworkIdentifier)\$(TargetFrameworkVersion)\RemObjects.Oxygene.Echoes.$(TargetFrameworkIdentifier).targets" />
<PropertyGroup>
<PreBuildEvent />
</PropertyGroup>
<Import Project="..\..\Sugar.Shared.Test.projitems" Label="Shared" />
</Project> | Oxygene | 2 | mosh/sugar | Sugar.Tests/Main/WP8/Sugar.Echoes.WP8.Test.oxygene | [
"BSD-3-Clause"
]
|
package com.baeldung.spring.cloud.ribbon.retry.backoff;
import org.springframework.cloud.netflix.ribbon.RibbonLoadBalancedRetryFactory;
import org.springframework.cloud.netflix.ribbon.SpringClientFactory;
import org.springframework.context.annotation.Profile;
import org.springframework.retry.backoff.BackOffPolicy;
import org.springframework.retry.backoff.ExponentialRandomBackOffPolicy;
import org.springframework.stereotype.Component;
@Component
@Profile("exponential-random-backoff")
class ExponentialRandomBackoffRetryFactory extends RibbonLoadBalancedRetryFactory {
public ExponentialRandomBackoffRetryFactory(SpringClientFactory clientFactory) {
super(clientFactory);
}
@Override
public BackOffPolicy createBackOffPolicy(String service) {
ExponentialRandomBackOffPolicy exponentialRandomBackOffPolicy = new ExponentialRandomBackOffPolicy();
exponentialRandomBackOffPolicy.setInitialInterval(1000);
exponentialRandomBackOffPolicy.setMultiplier(2);
exponentialRandomBackOffPolicy.setMaxInterval(10000);
return exponentialRandomBackOffPolicy;
}
} | Java | 4 | DBatOWL/tutorials | spring-cloud/spring-cloud-ribbon-retry/ribbon-client-service/src/main/java/com/baeldung/spring/cloud/ribbon/retry/backoff/ExponentialRandomBackoffRetryFactory.java | [
"MIT"
]
|
example (A B C D : Prop) (h1 : A → B) (h2 : B → C) (h3 : C → D) : A → D :=
calc A → B : h1
... → C : h2
... → D : h3
| Lean | 4 | ericrbg/lean | tests/lean/run/calc_imp.lean | [
"Apache-2.0"
]
|
cucumber.plugin=pretty, json:target/cucumber/cucumber.json | INI | 2 | DBatOWL/tutorials | testing-modules/cucumber/src/test/resources/junit-platform.properties | [
"MIT"
]
|
<!DOCTYPE html>
<html><head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<meta charset="UTF-8">
<title>BMP Suite Image List</title>
<style>
.b { background:url(bmpsuite_files/bkgd.png); }
.q { background-color:#fff0e0; }
.bad { background-color:#ffa0a0; }
.sz1 { width:127px; height:64px; }
.szbad { width:64px; height:64px; }
</style>
</head>
<body class="vsc-initialized">
<h1>BMP Suite Image List</h1>
<p><i>For <a href="https://entropymine.com/jason/bmpsuite/">BMP Suite</a>
version 2.6</i></p>
<p>This document describes the images in <i>BMP Suite</i>, and shows what
I allege to be the correct way to interpret them. PNG and JPEG images are
used for reference.
</p>
<p>It also shows how your web browser displays the BMP images,
but that’s not its main purpose.
BMP is poor image format to use on web pages, so a web browser’s
level of support for it is arguably not important.</p>
<table cellpadding="8" border="1">
<tbody><tr>
<th>File</th>
<th>Ver.</th>
<th>Correct display</th>
<th>In your browser</th>
<th>Notes</th>
</tr>
<tr>
<td>g/pal1.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal1.png"></td>
<td class="b"><img src="bmpsuite_files/pal1.bmp"></td>
<td>1 bit/pixel paletted image, in which black is the first color in
the palette.</td>
</tr>
<tr>
<td>g/pal1wb.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal1.png"></td>
<td class="b"><img src="bmpsuite_files/pal1wb.bmp"></td>
<td>1 bit/pixel paletted image, in which white is the first color in
the palette.</td>
</tr>
<tr>
<td>g/pal1bg.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal1bg.png"></td>
<td class="b"><img src="bmpsuite_files/pal1bg.bmp"></td>
<td>1 bit/pixel paletted image, with colors other than black and white.</td>
</tr>
<tr>
<td class="q">q/pal1p1.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal1p1.png"></td>
<td class="b"><img src="bmpsuite_files/pal1p1.bmp"></td>
<td>1 bit/pixel paletted image, with only one color in the palette.
The documentation says that 1-bpp images have a palette size of 2
(not “up to 2”), but it would be silly for a viewer not to
support a size of 1.</td>
</tr>
<tr>
<td class="q">q/pal2.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal2.png"></td>
<td class="b"><img src="bmpsuite_files/pal2.bmp"></td>
<td>A paletted image with 2 bits/pixel. Usually only 1, 4,
and 8 are allowed, but 2 is legal on Windows CE.</td>
</tr>
<tr>
<td class="q">q/pal2color.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal2color.png"></td>
<td class="b"><img src="bmpsuite_files/pal2color.bmp"></td>
<td>Same as pal2.bmp, but with a color palette instead of grayscale
palette.</td>
</tr>
<tr>
<td>g/pal4.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal4.png"></td>
<td class="b"><img src="bmpsuite_files/pal4.bmp"></td>
<td>Paletted image with 12 palette colors, and 4 bits/pixel.</td>
</tr>
<tr>
<td>g/pal4gs.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal4gs.png"></td>
<td class="b"><img src="bmpsuite_files/pal4gs.bmp"></td>
<td>Paletted image with 12 grayscale palette colors, and 4 bits/pixel.</td>
</tr>
<tr>
<td>g/pal4rle.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal4.png"></td>
<td class="b"><img src="bmpsuite_files/pal4rle.bmp"></td>
<td>4-bit image that uses RLE compression.</td>
</tr>
<tr>
<td class="q">q/pal4rletrns.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal4rletrns.png"><br>
or<br><img src="bmpsuite_files/pal4rletrns-0.png"><br>
or<br><img src="bmpsuite_files/pal4rletrns-b.png"></td>
<td class="b"><img src="bmpsuite_files/pal4rletrns.bmp"></td>
<td>An RLE-compressed image that uses “delta”
codes to skip over some pixels, leaving them undefined. Some viewers
make undefined pixels transparent, others make them black, and
others assign them palette color 0 (purple, in this case).</td>
</tr>
<tr>
<td class="q">q/pal4rlecut.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal4rlecut.png"><br>
or<br><img src="bmpsuite_files/pal4rlecut-0.png"><br>
or<br><img src="bmpsuite_files/pal4rlecut-b.png"></td>
<td class="b"><img src="bmpsuite_files/pal4rlecut.bmp"></td>
<td>An RLE-compressed image that uses “delta” codes,
and early EOL & EOBMP markers, to skip over some pixels.
It’s okay if the viewer’s image doesn’t exactly match
any of the reference images.</td>
</tr>
<tr>
<td>g/pal8.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8.bmp"></td>
<td>Our standard paletted image, with 252 palette colors, and 8
bits/pixel.</td>
</tr>
<tr>
<td>g/pal8-0.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8-0.bmp"></td>
<td>Every field that can be set to 0 is set to 0: pixels/meter=0;
colors used=0 (meaning the default 256); size-of-image=0.</td>
</tr>
<tr>
<td>g/pal8gs.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8gs.png"></td>
<td class="b"><img src="bmpsuite_files/pal8gs.bmp"></td>
<td>An 8-bit image with a palette of 252 grayscale colors.</td>
</tr>
<tr>
<td>g/pal8rle.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8rle.bmp"></td>
<td>8-bit image that uses RLE compression.</td>
</tr>
<tr>
<td class="q">q/pal8rletrns.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8rletrns.png"><br>
or<br><img src="bmpsuite_files/pal8rletrns-0.png"><br>
or<br><img src="bmpsuite_files/pal8rletrns-b.png"></td>
<td class="b"><img src="bmpsuite_files/pal8rletrns.bmp"></td>
<td>8-bit version of q/pal4rletrns.bmp.</td>
</tr>
<tr>
<td class="q">q/pal8rlecut.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8rlecut.png"><br>
or<br><img src="bmpsuite_files/pal8rlecut-0.png"><br>
or<br><img src="bmpsuite_files/pal8rlecut-b.png"></td>
<td class="b"><img src="bmpsuite_files/pal8rlecut.bmp"></td>
<td>8-bit version of q/pal4rlecut.bmp.</td>
</tr>
<tr>
<td>g/pal8w126.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8w126.png"></td>
<td class="b"><img src="bmpsuite_files/pal8w126.bmp"></td>
<td rowspan="3">Images with different widths and heights.
In BMP format, rows are padded to a multiple of four bytes, so we
test all four possibilities.</td>
</tr>
<tr>
<td>g/pal8w125.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8w125.png"></td>
<td class="b"><img src="bmpsuite_files/pal8w125.bmp"></td>
</tr>
<tr>
<td>g/pal8w124.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8w124.png"></td>
<td class="b"><img src="bmpsuite_files/pal8w124.bmp"></td>
</tr>
<tr>
<td>g/pal8topdown.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8topdown.bmp"></td>
<td>BMP images are normally stored from the bottom up, but
there is a way to store them from the top down.</td>
</tr>
<tr>
<td class="q">q/pal8offs.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8offs.bmp"></td>
<td>A file with some unused bytes between the palette and the
image. This is probably valid, but I’m not 100% sure.</td>
</tr>
<tr>
<td class="q">q/pal8oversizepal.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8oversizepal.bmp"></td>
<td>An 8-bit image with 300 palette colors. This may be invalid,
because the documentation could
be interpreted to imply that 8-bit images aren’t allowed
to have more than 256 colors.</td>
</tr>
<tr>
<td>g/pal8nonsquare.bmp</td>
<td>3</td>
<td class="b">
<img src="bmpsuite_files/pal8nonsquare-v.png"><br>
or<br>
<img src="bmpsuite_files/pal8nonsquare-e.png">
</td>
<td class="b"><img src="bmpsuite_files/pal8nonsquare.bmp"></td>
<td>An image with non-square pixels: the X pixels/meter is twice
the Y pixels/meter. Image <i>editors</i> can be expected to
leave the image “squashed”; image <i>viewers</i> should
consider stretching it to its correct proportions.</td>
</tr>
<tr>
<td>g/pal8os2.bmp</td>
<td>OS/2v1</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2.bmp"></td>
<td>An OS/2-style bitmap. This format can be called OS/2 BMPv1,
or Windows BMPv2.</td>
</tr>
<tr>
<td class="q">q/pal8os2-sz.bmp</td>
<td>OS/2v1</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2-sz.bmp"></td>
<td>Some OS/2 BMP specifications say that the <i>size</i> field in the file
header should be set to the aggregate size of the file header and
<i>infoheader</i>, instead of the total file size.
For OS/2v1, that means it will always be 26.
BMP decoders usually ignore this field, so it shouldn’t
cause a problem.</td>
</tr>
<tr>
<td class="q">q/pal8os2-hs.bmp</td>
<td>OS/2v1</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2-hs.bmp"></td>
<td>Some OS/2 BMP specifications define the fields at offsets 6 and
8 to be a “hotspot” (for cursor graphics).
Though the fields are not used in BMP files, they are sometimes,
as in this file, set to nonzero values.
This should cause no problems, except that it could prevent some
programs from detecting this file as a BMP file.</td>
</tr>
<tr>
<td class="q">q/pal8os2sp.bmp</td>
<td>OS/2v1</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2sp.bmp"></td>
<td>An OS/2v1 with a less-than-full-sized palette.
Probably not valid, but such files have been seen in the wild.</td>
</tr>
<tr>
<td class="q">q/pal8os2v2.bmp</td>
<td>OS/2v2</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2v2.bmp"></td>
<td>An OS/2v2 bitmap.</td>
</tr>
<tr>
<td class="q">q/pal8os2v2-16.bmp</td>
<td>OS/2v2</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2v2-16.bmp"></td>
<td>An OS/2v2 bitmap whose header has only 16 bytes, instead of the full 64.</td>
</tr>
<tr>
<td class="q">q/pal8os2v2-sz.bmp</td>
<td>OS/2v2</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2v2-sz.bmp"></td>
<td>An OS/2v2 bitmap. Like q/pal8os2-sz.bmp, the <i>size</i> field is set to
the size of the headers (78), instead of the size of the file.</td>
</tr>
<tr>
<td class="q">q/pal8os2v2-40sz.bmp</td>
<td>OS/2v2</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8os2v2-40sz.bmp"></td>
<td>An OS/2v2 bitmap, with a 40-byte header. Like q/pal8os2-sz.bmp,
the <i>size</i> field is set to the size of the headers (54),
instead of the size of the file. Except for that, this file
cannot be distinguished from a Windows BMPv3 file.</td>
</tr>
<tr>
<td class="q">q/rgb24rle24.bmp</td>
<td>OS/2v2</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24rle24.bmp"></td>
<td>An OS/2v2 bitmap with RLE24 compression. This image uses a limited
number of colors, just to make it more compressible.</td>
</tr>
<tr>
<td class="q">q/pal1huff.bmp</td>
<td>OS/2v2</td>
<td class="b"><img src="bmpsuite_files/pal1.png"></td>
<td class="b"><img src="bmpsuite_files/pal1huff.bmp"></td>
<td>My attempt to make a BMP file with Huffman 1D compression.
It is quite possibly incorrect. Even if everything else about it is correct,
I have no way to know whether it is black/white reversed, and/or flipped
vertically.</td>
</tr>
<tr>
<td>g/pal8v4.bmp</td>
<td>4</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8v4.bmp"></td>
<td>A v4 bitmap. I’m not sure that the gamma and chromaticity values in
this file are sensible, because I can’t find any detailed documentation
of them. Note that bmpsuite v2.4 and earlier had the gamma set differently
(and probably incorrectly).</td>
</tr>
<tr>
<td>g/pal8v5.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/pal8v5.bmp"></td>
<td>A v5 bitmap. Version 5 has additional colorspace options over v4, so it
is easier to create, and ought to be more portable.</td>
</tr>
<tr>
<td>g/rgb16.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16.png"></td>
<td class="b"><img src="bmpsuite_files/rgb16.bmp"></td>
<td>A 16-bit image with the default color format: 5 bits each for red,
green, and blue, and 1 unused bit.
The whitest colors should (I assume) be displayed as pure white:
<span style="background-color:rgb(255,255,255)">(255,255,255)</span>, not
<span style="background-color:rgb(248,248,248)">(248,248,248)</span>.</td>
</tr>
<tr>
<td>g/rgb16bfdef.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16.png"></td>
<td class="b"><img src="bmpsuite_files/rgb16bfdef.bmp"></td>
<td>Same format as rgb16.bmp, but with a BITFIELDS segment.</td>
</tr>
<tr>
<td>g/rgb16-565.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16-565.png"></td>
<td class="b"><img src="bmpsuite_files/rgb16-565.bmp"></td>
<td>A 16-bit image with a BITFIELDS segment indicating 5 red, 6 green,
and 5 blue bits. This is a standard 16-bit format, even supported by
old versions of Windows that don’t support any other non-default 16-bit
formats.
The whitest colors should be displayed as pure white:
<span style="background-color:rgb(255,255,255)">(255,255,255)</span>, not
<span style="background-color:rgb(248,252,248)">(248,252,248)</span>.</td>
</tr>
<tr>
<td>g/rgb16-565pal.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16-565.png"></td>
<td class="b"><img src="bmpsuite_files/rgb16-565pal.bmp"></td>
<td>A 16-bit image with both a BITFIELDS segment and a palette.</td>
</tr>
<tr>
<td class="q">q/rgb16faketrns.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16.png"><br>
or maybe<br>
<img class="b" src="bmpsuite_files/rgb16faketrns.png">
</td>
<td class="b"><img src="bmpsuite_files/rgb16faketrns.bmp"></td>
<td>Same idea as q/rgb32fakealpha.bmp. The default 16-bit color format has
one unused bit per pixel, and in this image some of the unused bits are
set to 1. It’s possible that some viewers will interpret this image
as having transparency.
</td>
</tr>
<tr>
<td class="q">q/rgb16-231.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16-231.png"></td>
<td class="b"><img src="bmpsuite_files/rgb16-231.bmp"></td>
<td>An unusual and silly 16-bit image, with 2 red bits, 3 green bits, and 1
blue bit. Most viewers do support this image, but the colors may be darkened
with a yellow-green shadow. That’s because they’re doing simple
bit-shifting (possibly including one round of bit replication), instead of
proper scaling.</td>
</tr>
<tr>
<td class="q">q/rgb16-3103.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16-3103.png"></td>
<td class="b"><img src="bmpsuite_files/rgb16-3103.bmp"></td>
<td>Similar to q/rgb16-231.bmp, with 3 red bits, 10 green bits, and 3
blue bits.</td>
</tr>
<tr>
<td class="q">q/rgba16-4444.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba16-4444.png"></td>
<td class="b"><img src="bmpsuite_files/rgba16-4444.bmp"></td>
<td>A 16-bit image with an alpha channel. There are 4 bits for each color
channel, and 4 bits for the alpha channel.
It’s not clear if this is valid, but I can’t find anything that
suggests it isn’t.
</td>
</tr>
<tr>
<td class="q">q/rgba16-5551.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba16-5551.png"></td>
<td class="b"><img src="bmpsuite_files/rgba16-5551.bmp"></td>
<td>Similar to q/rgba16-4444.bmp, with 5 red bits, 5 green bits, 5 blue bits,
and a 1-bit alpha channel.</td>
</tr>
<tr>
<td class="q">q/rgba16-1924.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba16-1924.png"></td>
<td class="b"><img src="bmpsuite_files/rgba16-1924.bmp"></td>
<td>Similar to q/rgba16-4444.bmp, with 1 red bit, 9 green bits, 2 blue bits,
and 4 bits for the alpha channel.
</td>
</tr>
<tr>
<td>g/rgb24.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24.bmp"></td>
<td>A perfectly ordinary 24-bit (truecolor) image.</td>
</tr>
<tr>
<td>g/rgb24pal.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24pal.bmp"></td>
<td>A 24-bit image, with a palette containing 256 colors. There is little if
any reason for a truecolor image to contain a palette, but it is legal.</td>
</tr>
<tr>
<td class="q">q/rgb24largepal.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24largepal.bmp"></td>
<td>A 24-bit image, with a palette containing 300 colors.
The fact that the palette has more than 256 colors may cause some viewers
to complain, but the documentation does not mention a size limit.</td>
</tr>
<tr>
<td class="q">q/rgb24prof.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24prof.bmp"></td>
<td>My attempt to make a BMP file with an embedded color profile.</td>
</tr>
<tr>
<td class="q">q/rgb24prof2.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24prof2.bmp"></td>
<td>This image tries to test whether color profiles are fully supported.
It has the red and green channels swapped, and an embedded color profile
that tries to swap them back. Support for this is uncommon.</td>
</tr>
<tr>
<td class="q">q/rgb24lprof.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24lprof.bmp"></td>
<td>My attempt to make a BMP file with a linked color profile.
Supporting linked profiles may be a bad idea, as it can lead to security vulnerabilities.</td>
</tr>
<tr>
<td class="q">q/rgb24jpeg.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgb24.jpg"></td>
<td class="b"><img src="bmpsuite_files/rgb24jpeg.bmp"></td>
<td rowspan="2">My attempt to make BMP files with embedded JPEG and PNG images.
These are not likely to be supported by much of anything (they’re
intended for printers).<br>
These image are stored in top-down order, with a positive bV5Height field.
This might not be correct. The documentation is very confusing on this issue.</td>
</tr>
<tr>
<td class="q">q/rgb24png.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb24png.bmp"></td>
</tr>
<tr>
<td>g/rgb32.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb32.bmp"></td>
<td>A 32-bit image using the default color format for 32-bit images (no
BITFIELDS segment). There are 8 bits per color channel, and 8 unused
bits. The unused bits are set to 0.</td>
</tr>
<tr>
<td>g/rgb32bfdef.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb32bfdef.bmp"></td>
<td>Same format as rgb32.bmp, but with a BITFIELDS segment.</td>
</tr>
<tr>
<td>g/rgb32bf.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb32bf.bmp"></td>
<td>A 32-bit image with a BITFIELDS segment. As usual, there are 8 bits per
color channel, and 8 unused bits. But the color channels are in an unusual
order, so the viewer must read the BITFIELDS, and not just guess.</td>
</tr>
<tr>
<td class="q">q/rgb32h52.bmp</td>
<td>(52)</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb32h52.bmp"></td>
<td>Similar to g/rgb32bf.bmp, but with a 52-byte
“BITMAPV2INFOHEADER”. This is an uncommon version of BMP, and I
can’t confirm that this file is correct.</td>
</tr>
<tr>
<td class="q">q/rgb32-xbgr.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb32-xbgr.bmp"></td>
<td>Color channels are the same size and order as rgb32bfdef.bmp, but they use
the highest available bits, instead of the lowest (or vice versa, depending
on your byte-order perspective).</td>
</tr>
<tr>
<td class="q">q/rgb32fakealpha.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"><br>
or<br>
<img class="b" src="bmpsuite_files/fakealpha.png">
</td>
<td class="b"><img src="bmpsuite_files/rgb32fakealpha.bmp"></td>
<td>Same as g/rgb32.bmp, except that the unused bits are set to something
other than 0.
If the image becomes transparent toward the bottom, it probably means
the viewer uses heuristics to guess whether the undefined
data represents transparency.
Reportedly, in ICO icon format, a 32-bit image has transparency if any
of the could-be alpha samples are nonzero. Some BMP decoders probably
use the same algorithm for BMP.</td>
</tr>
<tr>
<td class="q">q/rgb32-111110.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb24.png"></td>
<td class="b"><img src="bmpsuite_files/rgb32-111110.bmp"></td>
<td>A 32 bits/pixel image, with all 32 bits used: 11 each for red and
green, and 10 for blue. As far as I know, this is valid, but it
is unusual.</td>
</tr>
<tr>
<td class="q">q/rgb32-7187.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb32-7187.png"></td>
<td class="b"><img src="bmpsuite_files/rgb32-7187.bmp"></td>
<td>A 32 bits/pixel image, with 7 bits for red, 18 for green, and 7 for
blue.</td>
</tr>
<tr>
<td class="q">q/rgba32-1.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba32.png"></td>
<td class="b"><img src="bmpsuite_files/rgba32-1.bmp"></td>
<td>A BMP with an alpha channel. Transparency is barely documented,
so it’s <i>possible</i> that this file is not correctly formed.
The color channels are in the usual order.</td>
</tr>
<tr>
<td class="q">q/rgba32-2.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba32.png"></td>
<td class="b"><img src="bmpsuite_files/rgba32-2.bmp"></td>
<td>Same as q/rgba32-1.bmp, but with the color channels
in an unusual order, to prevent viewers from
passing this test by making a lucky guess.</td>
</tr>
<tr>
<td class="q">q/rgba32-1010102.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba32-1010102.png"></td>
<td class="b"><img src="bmpsuite_files/rgba32-1010102.bmp"></td>
<td>A 32 bits/pixel image, with 10 bits for red, 10 for green, 10 for blue,
and 2 for alpha.</td>
</tr>
<tr>
<td class="q">q/rgba32-81284.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba32-81284.png"></td>
<td class="b"><img src="bmpsuite_files/rgba32-81284.bmp"></td>
<td>A 32 bits/pixel image, with 8 bits for red, 12 for green, 8 for blue,
and 4 for alpha.</td>
</tr>
<tr>
<td class="q">q/rgba32-61754.bmp</td>
<td>5</td>
<td class="b"><img src="bmpsuite_files/rgba32-61754.png"></td>
<td class="b"><img src="bmpsuite_files/rgba32-61754.bmp"></td>
<td>A 32 bits/pixel image, with 6 bits for red, 17 for green, 5 for blue,
and 4 for alpha.</td>
</tr>
<tr>
<td class="q">q/rgba32abf.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgba32.png"></td>
<td class="b"><img src="bmpsuite_files/rgba32abf.bmp"></td>
<td>An image of type BI_ALHPABITFIELDS. Supposedly, this was used on
Windows CE. I don’t know whether it is constructed correctly.</td>
</tr>
<tr>
<td class="q">q/rgba32h56.bmp</td>
<td>(56)</td>
<td class="b"><img src="bmpsuite_files/rgba32.png"></td>
<td class="b"><img src="bmpsuite_files/rgba32h56.bmp"></td>
<td>Similar to q/rgba32-2.bmp, but with a 56-byte
“BITMAPV3INFOHEADER”. This is an uncommon version of BMP, and I
can’t confirm that this file is correct.</td>
</tr>
<tr>
<td class="q">x/ba-bm.bmp</td>
<td>OS/2v2</td>
<td class="b"><img src="bmpsuite_files/pal8.png"></td>
<td class="b"><img src="bmpsuite_files/ba-bm.bmp"></td>
<td>This image uses the OS/2v2 “Bitmap Array” (BA) container
format. Although a BA file may contain multiple images, this file has
only one.</td>
</tr>
<tr>
<td class="bad">b/badbitcount.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badbitcount.bmp"></td>
<td>Header indicates an absurdly large number of bits/pixel.</td>
</tr>
<tr>
<td class="bad">b/badbitssize.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badbitssize.bmp"></td>
<td>Header incorrectly indicates that the bitmap is several GB in size.</td>
</tr>
<tr>
<td class="bad">b/baddens1.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img class="sz1" src="bmpsuite_files/baddens1.bmp"></td>
<td rowspan="2">Density (pixels per meter) suggests the image is <i>much</i>
larger in one dimension than the other.</td>
</tr>
<tr>
<td class="bad">b/baddens2.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img class="sz1" src="bmpsuite_files/baddens2.bmp"></td>
</tr>
<tr>
<td class="bad">b/badfilesize.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badfilesize.bmp"></td>
<td>Header incorrectly indicates that the file is several GB in size.</td>
</tr>
<tr>
<td class="bad">b/badheadersize.bmp</td>
<td>?</td>
<td class="b">N/A</td>
<td class="b"><img class="sz1" src="bmpsuite_files/badheadersize.bmp"></td>
<td>Header size is 66 bytes, which is not a valid size for any known BMP
version.</td>
</tr>
<tr>
<td class="bad">b/badpalettesize.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badpalettesize.bmp"></td>
<td>Header incorrectly indicates that the palette contains an absurdly large
number of colors.</td>
</tr>
<tr>
<td class="bad">b/badplanes.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badplanes.bmp"></td>
<td>The “planes” setting, which is required to be 1, is not 1.</td>
</tr>
<tr>
<td class="bad">b/badrle4.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badrle4.bmp"></td>
<td>An invalid RLE4-compressed image that tries to cause buffer overruns.</td>
</tr>
<tr>
<td class="bad">b/badrle4bis.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badrle4bis.bmp"></td>
<td>Another invalid RLE4-compressed image that tries to cause buffer overruns.</td>
</tr>
<tr>
<td class="bad">b/badrle4ter.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badrle4ter.bmp"></td>
<td>Another invalid RLE4-compressed image that tries to cause buffer overruns.</td>
</tr>
<tr>
<td class="bad">b/badrle.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badrle.bmp"></td>
<td>8-bit version of b/badrle4.bmp.</td>
</tr>
<tr>
<td class="bad">b/badrlebis.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badrlebis.bmp"></td>
<td>8-bit version of b/badrle4bis.bmp.</td>
</tr>
<tr>
<td class="bad">b/badrleter.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/badrleter.bmp"></td>
<td>8-bit version of b/badrle4ter.bmp.</td>
</tr>
<tr>
<td class="bad">b/badwidth.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img class="szbad" src="bmpsuite_files/badwidth.bmp"></td>
<td>The image claims to be a negative number of pixels in width.</td>
</tr>
<tr>
<td class="bad">b/pal8badindex.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/pal8badindex.bmp"></td>
<td>Many of the palette indices used in the image are not present in the
palette.</td>
</tr>
<tr>
<td class="bad">b/reallybig.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img class="szbad" src="bmpsuite_files/reallybig.bmp"></td>
<td>An image with a very large reported width and height.</td>
</tr>
<tr>
<td class="bad">b/rgb16-880.bmp</td>
<td>3</td>
<td class="b"><img src="bmpsuite_files/rgb16-880.png"><br>(?)</td>
<td class="b"><img src="bmpsuite_files/rgb16-880.bmp"></td>
<td>A 16-bit image with a BITFIELDS segment indicating 8 red, 8 green,
and 0 blue bits. The documentation doesn’t say whether undefined
channels are legal, or how they should be handled.
</td></tr>
<tr>
<td class="bad">b/rletopdown.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/rletopdown.bmp"></td>
<td>An RLE-compressed image that tries to use top-down orientation,
which isn’t allowed.</td>
</tr>
<tr>
<td class="bad">b/shortfile.bmp</td>
<td>3</td>
<td class="b">N/A</td>
<td class="b"><img src="bmpsuite_files/shortfile.bmp"></td>
<td>A file that has been truncated in the middle of the bitmap.</td>
</tr>
</tbody></table>
</body></html>
| HTML | 4 | r00ster91/serenity | Base/res/html/misc/bmpsuite.html | [
"BSD-2-Clause"
]
|
# Copyright (c) 2018-2021, Carnegie Mellon University
# See LICENSE for details
_testJam := (t) -> let(
xx1 := Last(t.params[2].params[1]),
xx2 := Last(t.params[3].params[1]),
(xx1 <> 0 or xx2 <> 0)
and let(
M := t.params[1].dims()[1],
K := t.getTag(ABB).params[1],
n := When(xx1 <> 0, t.params[4][xx1], t.params[4][xx2]),
n <= K/2
)
);
_testIxA := (t) -> let(
@par := @.cond(e -> e = GTPar),
PatternMatch(t, [GT, @(1), @par, @par, \.\.\. ], empty_cx())
and let(
M := @(1).val.dims()[1],
K := t.getTag(ABB).params[1],
m := t.dims()[1],
K > M
and m > K
)
);
_testIxAxI := (t) -> let(
@ixi := @.cond(e -> e = XChain([1,0,2])),
PatternMatch(t, [GT, @(1), @ixi, @ixi, \.\.\. ], empty_cx())
and let(
M := @(1).val.dims()[1],
K := t.getTag(ABB).params[1],
m := t.dims()[1],
n := t.params[4][1],
p := t.params[4][2],
(K/2 < p
and 1 < K/2)
or (
m > K
and K > M*p
)
)
);
_testAxI := (t) -> let(
@vec := @.cond(e -> e = GTVec),
PatternMatch(t, [GT, @(1), @vec, @vec, \.\.\. ], empty_cx())
and let(
M := @(1).val.dims()[1],
K := t.getTag(ABB).params[1],
m := t.dims()[1],
n := t.params[4][1],
K/2 < n
and 1 < K/2
)
);
NewRulesFor(GT, rec(
# splits a loop based on basic block size K
# K is passed in with the tag ABB
# I_N x A_M -> I_(N/K/M) x I_(K/M) x A_M
GT_BB_IxA := rec(
forTransposition := false,
switch := false,
applicable := (t) ->
t.hasTag(ABB)
and _testIxA(t),
children := (self, t) >> let(
K := t.getTag(ABB).params[1],
spl := t.params[1],
M := spl.dims()[1],
n := t.params[4][1],
r := (K / M),
s := n / r,
# if q <=1, just drop the tag, otherwise split the loop
[[
GT(
spl,
XChain([2,1,0]),
XChain([2,1,0]),
[r, s]
).withTags(Filtered(t.getTags(), e -> ObjId(e) <> ABB))
]]
),
apply := (self, t, C, nt) >> C[1]
),
# I_n x A_m x I_p, tagged with (K, mu)
# 1) mp < K
# 2) mp >= K
# a) m > K/mu
# b) m <= K/mu
# a) p > mu
# b) p <= mu
GT_BB_IxAxI := rec(
forTransposition := false,
switch := false,
applicable := (t) ->
t.hasTag(ABB)
and _testIxAxI(t),
children := (self, t) >> let(
K := t.getTag(ABB).params[1],
M := t.params[1].dims()[1],
spl := t.params[1],
n := t.params[4][1],
p := t.params[4][2],
m := t.dims()[1],
Cond(
# cond1
m > K and K > M*p,
let(
s := K / (M*p),
u := n/s,
[[GT(
spl, XChain([2,1,0,3]), XChain([2,1,0,3]), [s,u,p]
).withTags(t.getTags())]]
),
# cond2
# p > K/2
let(
s := K / 2,
u := p/s,
[[GT(
spl, XChain([1,0,3,2]), XChain([1,0,3,2]), [n,s,u]
).withTags(t.getTags())]]
)
)
),
apply := (self, t, C, nt) >> C[1]
),
#
GT_BB_AxI := rec(
forTransposition := false,
switch := false,
applicable := (t) ->
t.hasTag(ABB)
and _testAxI(t),
children := (self, t) >> let(
M := t.params[1].dims()[1],
K := t.getTag(ABB).params[1],
n := t.params[4][1],
spl := t.params[1],
r := K / 2,
s := n / r,
[[GT(
spl, XChain([0,2,1]), XChain([0,2,1]), [r,s]
).withTags(t.getTags())]]
),
apply := (self, t, C, nt) >> C[1]
),
# when K <= M, we just drop the tag.
GT_BB_DropTag := rec(
forTransposition := false,
switch := false,
applicable := (self, t) >>
t.hasTag(ABB)
and ObjId(t) = GT
and let(
K := t.getTag(ABB).params[1],
K = 2
),
# build a new GT object without the ABB tag.
children := (self, t) >>
[[
t.withoutTag(ABB)
]],
apply := (self, t, C, nt) >> C[1],
),
# just like nthloop except only applicable when
# the Jam version below isn't.
GT_BB_NthLoop := CopyFields(GT_NthLoop, rec(
requiredFirstTag := [AExpRight,ANoTag,ALimitNthLoop,ABB],
applicable := (t) ->
Length(t.params[4]) > 0
and (
not t.hasTag(ABB)
or (
t.hasTag(ABB)
and not _testIxA(t)
and not _testIxAxI(t)
and not _testAxI(t)
and not _testJam(t)
)
)
)),
GT_BB_NthLoopJam := rec(
requiredFirstTag := [ABB,ALimitNthLoop],
applicable := (t) ->
Length(t.params[4]) > 0
and t.hasTag(ABB)
and _testJam(t),
# choose the rightmost index.
freedoms := t -> let(
xx1 := Last(t.params[2].params[1]),
xx2 := Last(t.params[3].params[1]),
When(xx1 <> 0, [[xx1]], [[xx2]])
),
child := (t, fr) -> let(
spl := t.params[1],
M := spl.dims()[1],
g := t.params[2],
s := t.params[3],
loopid := fr[1],
newK := t.getTag(ABB).params[1]/t.params[4][loopid],
When(t.hasTag(ALimitNthLoop),
[
GT(spl,
g.without(loopid),
s.without(loopid),
ListWithout(t.params[4], loopid)).withTags(
When(newK = 2,
Filtered(t.getTags(), e -> ObjId(e) <> ABB),
Concat(
[ABB(newK)],
Filtered(t.getTags(), e -> ObjId(e) <> ABB)
)
)
),
InfoNt(loopid)
],
[
GT(spl, g.without(loopid),
s.without(loopid), ListWithout(t.params[4], loopid)),
InfoNt(loopid)
]
)
),
apply := (t, C, nt) -> let(
loopid := nt[2].params[1],
dft := nt[1].params[1],
g := t.params[2],
s := t.params[3],
loop_dims := t.params[4],
i := Ind(loop_dims[loopid]),
JamISum(i,
Scat(s.part(loopid, i, Rows(dft), loop_dims))
* C[1]
* Gath(g.part(loopid, i, Cols(dft), loop_dims))
)
)
)
));
| GAP | 4 | sr7cb/spiral-software | namespaces/spiral/paradigms/cache/breakdown/gt.gi | [
"BSD-2-Clause-FreeBSD"
]
|
[Constructor(DOMString type, optional PaymentMethodChangeEventInit eventInitDict),
SecureContext,
Exposed=Window,
Func="mozilla::dom::PaymentRequest::PrefEnabled"]
interface PaymentMethodChangeEvent : PaymentRequestUpdateEvent {
readonly attribute DOMString methodName;
readonly attribute object? methodDetails;
};
dictionary PaymentMethodChangeEventInit : PaymentRequestUpdateEventInit {
required DOMString methodName;
object? methodDetails;
};
| WebIDL | 3 | tlively/wasm-bindgen | crates/web-sys/webidls/enabled/PaymentMethodChangeEvent.webidl | [
"Apache-2.0",
"MIT"
]
|
server {
server_name a-demo.chaosgenius.io;
root /var/www/a-demo.chaosgenius.io;
index index.html index.htm;
location ~* \.(?:manifest|appcache|html?|xml|json)$ {
expires -1;
# access_log logs/static.log; # I don't usually include a static log
}
location ~* \.(?:css|js)$ {
try_files $uri =404;
expires 1y;
access_log off;
add_header Cache-Control "public";
}
# Any route containing a file extension (e.g. /devicesfile.js)
location ~ ^.+\..+$ {
try_files $uri =404;
}
# Any route that doesn't have a file extension (e.g. /devices)
location / {
try_files $uri $uri/ /index.html;
}
location /api {
include proxy_params;
proxy_pass http://0.0.0.0:5000;
}
listen 80 default_server;
}
| Io | 4 | eltociear/chaos_genius | scripts/extra/nginx/sites-available/a-demo.chaosgenius.io | [
"MIT"
]
|
# Much of this file is really testing mpow.
# Numerical powers are either passed to mpow or else
# returned unevaluated by do_Power, depending on the type of args.
# We are missing several test cases here.
# But, do_Power and mpow should now reproduce, more or less what Mma does.
T Arg(Complex(1.0,1.0)) - N(Arg(Complex(1,1))) == 0.0
T Arg(Complex(1,1)) == π/4
ClearAll(z)
T 100^(1/2) == 10 # fix bug in commit following 06ca924038273a0309b2dbb2ea7c84d540bd84dd
# Fix bug. we can't do setfixed in canonexpr!(mx::Mxpr{:Power}) or this is broken
res = 102^(1/2)
T res^2 == 102
ClearAll(res)
T 27^(1/3) == 3
# TODO: we don't want the following
T Apply(List, 26^(1/3)) == [2 ^ (1//3), 13 ^ (1//3)] # FIXME
T 27^(2/3) == 9
T (-27)^(2/3) == (-1) ^ (2//3) * 9
T (-27)^(1/3) == 3 * (-1) ^ (1//3)
# Neg. Int to float power took a long time to fix, because we do not print full error message (domain error).
T Chop((-1)^(3.2) + 0.8090169943749477 + 0.5877852522924728 * I) == 0
T 3^0 == 1
# two bug fixes in the next one
T If( BigIntInput() , Head((3^0)) == BigInt , Head((3^0)) == Int)
T Apply(List,3^(1/2)) == [3,1//2]
T 4^(1/2) == 2
T Apply(List, 28^(1/3)) == [2 ^ (2//3), 7 ^ (1//3)]
T Apply(List,(49*7*3*3)^(1/3)) == [7,3^(2//3)]
T Chop((-1.0)^(1//3) - (0.5 + 0.8660254037844387 * I)) == 0
T Chop(N((-1)^(1//3)) - (0.5 + 0.8660254037844387 * I)) == 0
T 1/I == -I
T (I + 1)^(-2) == -1/2 * I
T (I + 1)^(-3) == -1/4 - 1/4 * I
T (I + 2)^(-3) == 2/125 - 11/125 * I
# fixed 98e02317aee46e7eceaac47d98c4401ef23682f0
T (-27/64)^(2/3) == 9/16*((-1)^(2/3))
# fixed error raised in Comparison
T (((9/16)*((-1)^(2/3)) == 3/16), True)
T 27^(1/2) == 3(3^(1/2))
# fixes a bug.
T Head((a*b)^(1/2)) == Power
# Check that pi --> float(pi) when needed. This is not automatic in Julia
T Isa(J( mpow(pi,-3) ), Real)
ClearAll(z,a,b)
### End old somemath_test.sj
T Head(1//1) == Int64
T Head(3//1) == Int64
ClearAll(a,b,ar)
T Head(Re(a)) == Re
T Re(I*a)[2] == Im(a)
T Im(I*a) == Re(a)
T Im(I*a*b) == Re(a*b)
T Re(I*a*b) == -Im(a*b)
T Re(2*a) == 2*Re(a)
T Im(2*a) == 2*Im(a)
T 1/0 == DirectedInfinity()
T 1/0.0 == ComplexInfinity
T 1/big"0.0" == ComplexInfinity
T 1//0 == DirectedInfinity()
T 1/DirectedInfinity() == 0
T DirectedInfinity() == ComplexInfinity
T DirectedInfinity()^(2) == DirectedInfinity()
T DirectedInfinity()^(-2) == 0
T DirectedInfinity() * a * 0 == Indeterminate
T ComplexInfinity^(-2) == 0
T Infinity^(-2) == 0
T Apply(Times, [DirectedInfinity(),a,a,a,0]) == Indeterminate
T 0 * Indeterminate == Indeterminate
T 0 * Infinity == Indeterminate
T 0 * ComplexInfinity == Indeterminate
T 1/0 == ComplexInfinity
T 0/0 == Indeterminate
T a * 3 * ComplexInfinity == ComplexInfinity
T Infinity * (1+I) == DirectedInfinity((1 + I)*(2^(-1/2)))
T Args(a*Infinity) == [a,Infinity]
T Args(Infinity * -a) == [a,-Infinity]
T Infinity + 1 == Infinity
T Infinity + 10 == Infinity
T Infinity - 10 == Infinity
T Infinity * 10 == Infinity
T Infinity * -10 == -Infinity
# FIXME
# T - Infinity * -10 == -Infinity
# T - Infinity * 10 == -Infinity
# T -Infinity - 10
T J( mpow(1,0) == 1 )
T J( mpow(1,1) == 1 )
T Isa(J(mpow(1,-1)), Integer)
T J( mpow(2,1//2) == mxpr(:Power, 2, 1//2) )
T J( mpow(2,-1//2) == mxpr(:Power, 2, -1//2) )
T J( mpow(4,1//2) == 2 )
T J( mpow(4,-1//2) == 1//2 )
T J( mpow(7^3,1//2) == mxpr(:Times, 7, mxpr(:Power, 7, 1//2)) )
T J( mpow(8*5,1//2) == mxpr(:Times, 2, mxpr(:Power, 2, 1//2), mxpr(:Power, 5, 1//2)) )
T J( mpow(9, 2//3) == mxpr(:Times, 3, mxpr(:Power, 3, 1//3)) )
T J( mpow(-9,2//3) == mxpr(:Times, 3, mxpr(:Power, 3, 1//3), mxpr(:Power,-1,2//3)))
T J(mmul(2 + 2*im, 1//2) == Complex(1,1))
## FIXME: this behavior is strange and wrong.
n1 = 1/(2+2I)
n2 = (2+2I)
r = n1 * n2
T (Isa(r,Integer) && r == 1)
T (r === (n2 * n1))
T Isa(n2 * n1, Integer)
T Not( n2 * n1 === n1 * n2)
T Not(Isa(n1 * n2, Integer))
# We cant change these. The user must use mpow here.
# T J( 2^(1//2) ) == mxpr(:Power, 2, 1//2)
# T J( 2^(-1//2)) == mxpr(:Power, 2, -1//2) )
#T J(4^1//2) == 2
#T J(4^-1//2) == 1//2
# T J( 7^3^1//2) == mxpr(:Times, 7, mxpr(:Power, 7, 1//2))
# T J( (8*5)^1//2) == mxpr(:Times, 2, mxpr(:Power, 2, 1//2), mxpr(:Power, 5, 1//2))
# T J( 9^2//3) == mxpr(:Times, 3, mxpr(:Power, 3, 1//3))
# T J( -9^2//3) == mxpr(:Times, 3, mxpr(:Power, 3, 1//3), mxpr(:Power,-1,2//3))
T Sqrt(-1) == I # fixes bug in mpow{T<:Integer, V<:Integer}(x::T,y::Rational{V})
T Sqrt(-1)^2 == -1 # same bug
## Fix domain error bug in _mpow{T<:Integer, V<:Integer}(x::T,y::Rational{V})
## d02b2a74cc6a1ee99597486c23a6d85de292cce9
T (2^(1/3))^(-4) == (1/2)*2^(-1/3)
ClearAll(a,b,n1,n2)
| Objective-J | 5 | UnofficialJuliaMirrorSnapshots/Symata.jl-a906b1d5-d016-55c4-aab3-8a20cba0db2a | symata_test/arithmetic_test.sj | [
"MIT"
]
|
{
"log" : {
"appender" : {
"type" : "syslog"
},
"levels" : {
/*"default" : "info"*/
}
},
"db" : {
"permissions" : [
{"type":"db.role","object":"admin","caller":"com.palm.configurator","operations":{"*":"allow"}},
{"type":"db.role","object":"admin","caller":"com.palm.service.backup","operations":{"*":"allow"}},
{"type":"db.role","object":"admin","caller":"com.palm.odd.service","operations":{"*":"allow"}},
{"type":"db.role","object":"admin","caller":"com.palm.service.migrationscript","operations":{"*":"allow"}},
{"type":"db.role","object":"admin","caller":"com.palm.spacecadet","operations":{"*":"allow"}}
],
"quotas" : [
{"owner":"*","size":62914560},
{"owner":"com.palm.*","size":235929600}
],
"loadStepSize" : 173,
"purgeWindow": 7,
},
"bdb" : {
"cacheSize": 3145728,
"maxLocks" : 20000,
"maxLockers" : 1000,
"compactStepSize" : 25000
}
}
| Opal | 3 | webosce/db8 | src/db-luna/mojodb.conf.opal | [
"Apache-2.0"
]
|
// check-pass
const fn i((a, b): (u32, u32)) -> u32 {
a + b
}
fn main() {}
| Rust | 3 | Eric-Arellano/rust | src/test/ui/consts/const-fn-destructuring-arg.rs | [
"ECL-2.0",
"Apache-2.0",
"MIT-0",
"MIT"
]
|
package foo
open class A {
val x: Any = object {}
} | Groff | 2 | qussarah/declare | jps-plugin/testData/incremental/classHierarchyAffected/propertyNullabilityChanged/A.kt.new.2 | [
"Apache-2.0"
]
|
# SOME DESCRIPTIVE TITLE.
# Copyright (C) YEAR THE PACKAGE'S COPYRIGHT HOLDER
# This file is distributed under the same license as the PACKAGE package.
#
# Translators:
# Janusz Harkot <[email protected]>, 2015
# Piotr Jakimiak <[email protected]>, 2015
# m_aciek <[email protected]>, 2016
# m_aciek <[email protected]>, 2015-2016
msgid ""
msgstr ""
"Project-Id-Version: Django REST framework\n"
"Report-Msgid-Bugs-To: \n"
"POT-Creation-Date: 2020-10-13 21:45+0200\n"
"PO-Revision-Date: 2020-10-13 19:45+0000\n"
"Last-Translator: Xavier Ordoquy <[email protected]>\n"
"Language-Team: Polish (http://www.transifex.com/django-rest-framework-1/django-rest-framework/language/pl/)\n"
"MIME-Version: 1.0\n"
"Content-Type: text/plain; charset=UTF-8\n"
"Content-Transfer-Encoding: 8bit\n"
"Language: pl\n"
"Plural-Forms: nplurals=4; plural=(n==1 ? 0 : (n%10>=2 && n%10<=4) && (n%100<12 || n%100>14) ? 1 : n!=1 && (n%10>=0 && n%10<=1) || (n%10>=5 && n%10<=9) || (n%100>=12 && n%100<=14) ? 2 : 3);\n"
#: authentication.py:70
msgid "Invalid basic header. No credentials provided."
msgstr "Niepoprawny podstawowy nagłówek. Brak danych uwierzytelniających."
#: authentication.py:73
msgid "Invalid basic header. Credentials string should not contain spaces."
msgstr "Niepoprawny podstawowy nagłówek. Ciąg znaków danych uwierzytelniających nie powinien zawierać spacji."
#: authentication.py:83
msgid "Invalid basic header. Credentials not correctly base64 encoded."
msgstr "Niepoprawny podstawowy nagłówek. Niewłaściwe kodowanie base64 danych uwierzytelniających."
#: authentication.py:101
msgid "Invalid username/password."
msgstr "Niepoprawna nazwa użytkownika lub hasło."
#: authentication.py:104 authentication.py:206
msgid "User inactive or deleted."
msgstr "Użytkownik nieaktywny lub usunięty."
#: authentication.py:184
msgid "Invalid token header. No credentials provided."
msgstr "Niepoprawny nagłówek tokena. Brak danych uwierzytelniających."
#: authentication.py:187
msgid "Invalid token header. Token string should not contain spaces."
msgstr "Niepoprawny nagłówek tokena. Token nie może zawierać odstępów."
#: authentication.py:193
msgid ""
"Invalid token header. Token string should not contain invalid characters."
msgstr "Błędny nagłówek z tokenem. Token nie może zawierać błędnych znaków."
#: authentication.py:203
msgid "Invalid token."
msgstr "Niepoprawny token."
#: authtoken/apps.py:7
msgid "Auth Token"
msgstr "Token uwierzytelniający"
#: authtoken/models.py:13
msgid "Key"
msgstr "Klucz"
#: authtoken/models.py:16
msgid "User"
msgstr "Użytkownik"
#: authtoken/models.py:18
msgid "Created"
msgstr "Stworzono"
#: authtoken/models.py:27 authtoken/serializers.py:19
msgid "Token"
msgstr "Token"
#: authtoken/models.py:28
msgid "Tokens"
msgstr "Tokeny"
#: authtoken/serializers.py:9
msgid "Username"
msgstr "Nazwa użytkownika"
#: authtoken/serializers.py:13
msgid "Password"
msgstr "Hasło"
#: authtoken/serializers.py:35
msgid "Unable to log in with provided credentials."
msgstr "Podane dane uwierzytelniające nie pozwalają na zalogowanie."
#: authtoken/serializers.py:38
msgid "Must include \"username\" and \"password\"."
msgstr "Musi zawierać \"username\" i \"password\"."
#: exceptions.py:102
msgid "A server error occurred."
msgstr "Wystąpił błąd serwera."
#: exceptions.py:142
msgid "Invalid input."
msgstr ""
#: exceptions.py:161
msgid "Malformed request."
msgstr "Zniekształcone żądanie."
#: exceptions.py:167
msgid "Incorrect authentication credentials."
msgstr "Błędne dane uwierzytelniające."
#: exceptions.py:173
msgid "Authentication credentials were not provided."
msgstr "Nie podano danych uwierzytelniających."
#: exceptions.py:179
msgid "You do not have permission to perform this action."
msgstr "Nie masz uprawnień, by wykonać tę czynność."
#: exceptions.py:185
msgid "Not found."
msgstr "Nie znaleziono."
#: exceptions.py:191
#, python-brace-format
msgid "Method \"{method}\" not allowed."
msgstr "Niedozwolona metoda \"{method}\"."
#: exceptions.py:202
msgid "Could not satisfy the request Accept header."
msgstr "Nie można zaspokoić nagłówka Accept żądania."
#: exceptions.py:212
#, python-brace-format
msgid "Unsupported media type \"{media_type}\" in request."
msgstr "Brak wsparcia dla żądanego typu danych \"{media_type}\"."
#: exceptions.py:223
msgid "Request was throttled."
msgstr "Żądanie zostało zdławione."
#: exceptions.py:224
#, python-brace-format
msgid "Expected available in {wait} second."
msgstr ""
#: exceptions.py:225
#, python-brace-format
msgid "Expected available in {wait} seconds."
msgstr ""
#: fields.py:316 relations.py:245 relations.py:279 validators.py:90
#: validators.py:183
msgid "This field is required."
msgstr "To pole jest wymagane."
#: fields.py:317
msgid "This field may not be null."
msgstr "Pole nie może mieć wartości null."
#: fields.py:701
msgid "Must be a valid boolean."
msgstr ""
#: fields.py:766
msgid "Not a valid string."
msgstr ""
#: fields.py:767
msgid "This field may not be blank."
msgstr "To pole nie może być puste."
#: fields.py:768 fields.py:1881
#, python-brace-format
msgid "Ensure this field has no more than {max_length} characters."
msgstr "Upewnij się, że to pole ma nie więcej niż {max_length} znaków."
#: fields.py:769
#, python-brace-format
msgid "Ensure this field has at least {min_length} characters."
msgstr "Upewnij się, że pole ma co najmniej {min_length} znaków."
#: fields.py:816
msgid "Enter a valid email address."
msgstr "Podaj poprawny adres e-mail."
#: fields.py:827
msgid "This value does not match the required pattern."
msgstr "Ta wartość nie pasuje do wymaganego wzorca."
#: fields.py:838
msgid ""
"Enter a valid \"slug\" consisting of letters, numbers, underscores or "
"hyphens."
msgstr "Wprowadź poprawną wartość pola typu \"slug\", składającą się ze znaków łacińskich, cyfr, podkreślenia lub myślnika."
#: fields.py:839
msgid ""
"Enter a valid \"slug\" consisting of Unicode letters, numbers, underscores, "
"or hyphens."
msgstr ""
#: fields.py:854
msgid "Enter a valid URL."
msgstr "Wprowadź poprawny adres URL."
#: fields.py:867
msgid "Must be a valid UUID."
msgstr ""
#: fields.py:903
msgid "Enter a valid IPv4 or IPv6 address."
msgstr "Wprowadź poprawny adres IPv4 lub IPv6."
#: fields.py:931
msgid "A valid integer is required."
msgstr "Wymagana poprawna liczba całkowita."
#: fields.py:932 fields.py:969 fields.py:1005 fields.py:1366
#, python-brace-format
msgid "Ensure this value is less than or equal to {max_value}."
msgstr "Upewnij się, że ta wartość jest mniejsza lub równa {max_value}."
#: fields.py:933 fields.py:970 fields.py:1006 fields.py:1367
#, python-brace-format
msgid "Ensure this value is greater than or equal to {min_value}."
msgstr "Upewnij się, że ta wartość jest większa lub równa {min_value}."
#: fields.py:934 fields.py:971 fields.py:1010
msgid "String value too large."
msgstr "Za długi ciąg znaków."
#: fields.py:968 fields.py:1004
msgid "A valid number is required."
msgstr "Wymagana poprawna liczba."
#: fields.py:1007
#, python-brace-format
msgid "Ensure that there are no more than {max_digits} digits in total."
msgstr "Upewnij się, że liczba ma nie więcej niż {max_digits} cyfr."
#: fields.py:1008
#, python-brace-format
msgid ""
"Ensure that there are no more than {max_decimal_places} decimal places."
msgstr "Upewnij się, że liczba ma nie więcej niż {max_decimal_places} cyfr dziesiętnych."
#: fields.py:1009
#, python-brace-format
msgid ""
"Ensure that there are no more than {max_whole_digits} digits before the "
"decimal point."
msgstr "Upewnij się, że liczba ma nie więcej niż {max_whole_digits} cyfr całkowitych."
#: fields.py:1148
#, python-brace-format
msgid "Datetime has wrong format. Use one of these formats instead: {format}."
msgstr "Wartość daty z czasem ma zły format. Użyj jednego z dostępnych formatów: {format}."
#: fields.py:1149
msgid "Expected a datetime but got a date."
msgstr "Oczekiwano datę z czasem, otrzymano tylko datę."
#: fields.py:1150
#, python-brace-format
msgid "Invalid datetime for the timezone \"{timezone}\"."
msgstr ""
#: fields.py:1151
msgid "Datetime value out of range."
msgstr ""
#: fields.py:1236
#, python-brace-format
msgid "Date has wrong format. Use one of these formats instead: {format}."
msgstr "Data ma zły format. Użyj jednego z tych formatów: {format}."
#: fields.py:1237
msgid "Expected a date but got a datetime."
msgstr "Oczekiwano daty a otrzymano datę z czasem."
#: fields.py:1303
#, python-brace-format
msgid "Time has wrong format. Use one of these formats instead: {format}."
msgstr "Błędny format czasu. Użyj jednego z dostępnych formatów: {format}"
#: fields.py:1365
#, python-brace-format
msgid "Duration has wrong format. Use one of these formats instead: {format}."
msgstr "Czas trwania ma zły format. Użyj w zamian jednego z tych formatów: {format}."
#: fields.py:1399 fields.py:1456
#, python-brace-format
msgid "\"{input}\" is not a valid choice."
msgstr "\"{input}\" nie jest poprawnym wyborem."
#: fields.py:1402
#, python-brace-format
msgid "More than {count} items..."
msgstr "Więcej niż {count} elementów..."
#: fields.py:1457 fields.py:1603 relations.py:485 serializers.py:570
#, python-brace-format
msgid "Expected a list of items but got type \"{input_type}\"."
msgstr "Oczekiwano listy elementów, a otrzymano dane typu \"{input_type}\"."
#: fields.py:1458
msgid "This selection may not be empty."
msgstr "Zaznaczenie nie może być puste."
#: fields.py:1495
#, python-brace-format
msgid "\"{input}\" is not a valid path choice."
msgstr "\"{input}\" nie jest poprawną ścieżką."
#: fields.py:1514
msgid "No file was submitted."
msgstr "Nie przesłano pliku."
#: fields.py:1515
msgid ""
"The submitted data was not a file. Check the encoding type on the form."
msgstr "Przesłane dane nie były plikiem. Sprawdź typ kodowania formatki."
#: fields.py:1516
msgid "No filename could be determined."
msgstr "Nie można określić nazwy pliku."
#: fields.py:1517
msgid "The submitted file is empty."
msgstr "Przesłany plik jest pusty."
#: fields.py:1518
#, python-brace-format
msgid ""
"Ensure this filename has at most {max_length} characters (it has {length})."
msgstr "Upewnij się, że nazwa pliku ma długość co najwyżej {max_length} znaków (aktualnie ma {length})."
#: fields.py:1566
msgid ""
"Upload a valid image. The file you uploaded was either not an image or a "
"corrupted image."
msgstr "Prześlij poprawny plik graficzny. Przesłany plik albo nie jest grafiką lub jest uszkodzony."
#: fields.py:1604 relations.py:486 serializers.py:571
msgid "This list may not be empty."
msgstr "Lista nie może być pusta."
#: fields.py:1605
#, python-brace-format
msgid "Ensure this field has at least {min_length} elements."
msgstr ""
#: fields.py:1606
#, python-brace-format
msgid "Ensure this field has no more than {max_length} elements."
msgstr ""
#: fields.py:1682
#, python-brace-format
msgid "Expected a dictionary of items but got type \"{input_type}\"."
msgstr "Oczekiwano słownika, ale otrzymano \"{input_type}\"."
#: fields.py:1683
msgid "This dictionary may not be empty."
msgstr ""
#: fields.py:1755
msgid "Value must be valid JSON."
msgstr "Wartość musi być poprawnym ciągiem znaków JSON"
#: filters.py:49 templates/rest_framework/filters/search.html:2
msgid "Search"
msgstr "Szukaj"
#: filters.py:50
msgid "A search term."
msgstr ""
#: filters.py:180 templates/rest_framework/filters/ordering.html:3
msgid "Ordering"
msgstr "Kolejność"
#: filters.py:181
msgid "Which field to use when ordering the results."
msgstr ""
#: filters.py:287
msgid "ascending"
msgstr "rosnąco"
#: filters.py:288
msgid "descending"
msgstr "malejąco"
#: pagination.py:174
msgid "A page number within the paginated result set."
msgstr ""
#: pagination.py:179 pagination.py:372 pagination.py:590
msgid "Number of results to return per page."
msgstr ""
#: pagination.py:189
msgid "Invalid page."
msgstr "Niepoprawna strona."
#: pagination.py:374
msgid "The initial index from which to return the results."
msgstr ""
#: pagination.py:581
msgid "The pagination cursor value."
msgstr ""
#: pagination.py:583
msgid "Invalid cursor"
msgstr "Niepoprawny wskaźnik"
#: relations.py:246
#, python-brace-format
msgid "Invalid pk \"{pk_value}\" - object does not exist."
msgstr "Błędny klucz główny \"{pk_value}\" - obiekt nie istnieje."
#: relations.py:247
#, python-brace-format
msgid "Incorrect type. Expected pk value, received {data_type}."
msgstr "Błędny typ danych. Oczekiwano wartość klucza głównego, otrzymano {data_type}."
#: relations.py:280
msgid "Invalid hyperlink - No URL match."
msgstr "Błędny hyperlink - nie znaleziono pasującego adresu URL."
#: relations.py:281
msgid "Invalid hyperlink - Incorrect URL match."
msgstr "Błędny hyperlink - błędne dopasowanie adresu URL."
#: relations.py:282
msgid "Invalid hyperlink - Object does not exist."
msgstr "Błędny hyperlink - obiekt nie istnieje."
#: relations.py:283
#, python-brace-format
msgid "Incorrect type. Expected URL string, received {data_type}."
msgstr "Błędny typ danych. Oczekiwano adresu URL, otrzymano {data_type}"
#: relations.py:448
#, python-brace-format
msgid "Object with {slug_name}={value} does not exist."
msgstr "Obiekt z polem {slug_name}={value} nie istnieje"
#: relations.py:449
msgid "Invalid value."
msgstr "Niepoprawna wartość."
#: schemas/utils.py:32
msgid "unique integer value"
msgstr ""
#: schemas/utils.py:34
msgid "UUID string"
msgstr ""
#: schemas/utils.py:36
msgid "unique value"
msgstr ""
#: schemas/utils.py:38
#, python-brace-format
msgid "A {value_type} identifying this {name}."
msgstr ""
#: serializers.py:337
#, python-brace-format
msgid "Invalid data. Expected a dictionary, but got {datatype}."
msgstr "Niepoprawne dane. Oczekiwano słownika, otrzymano {datatype}."
#: templates/rest_framework/admin.html:116
#: templates/rest_framework/base.html:136
msgid "Extra Actions"
msgstr ""
#: templates/rest_framework/admin.html:130
#: templates/rest_framework/base.html:150
msgid "Filters"
msgstr "Filtry"
#: templates/rest_framework/base.html:37
msgid "navbar"
msgstr ""
#: templates/rest_framework/base.html:75
msgid "content"
msgstr ""
#: templates/rest_framework/base.html:78
msgid "request form"
msgstr ""
#: templates/rest_framework/base.html:157
msgid "main content"
msgstr ""
#: templates/rest_framework/base.html:173
msgid "request info"
msgstr ""
#: templates/rest_framework/base.html:177
msgid "response info"
msgstr ""
#: templates/rest_framework/horizontal/radio.html:4
#: templates/rest_framework/inline/radio.html:3
#: templates/rest_framework/vertical/radio.html:3
msgid "None"
msgstr "None"
#: templates/rest_framework/horizontal/select_multiple.html:4
#: templates/rest_framework/inline/select_multiple.html:3
#: templates/rest_framework/vertical/select_multiple.html:3
msgid "No items to select."
msgstr "Nie wybrano wartości."
#: validators.py:39
msgid "This field must be unique."
msgstr "Wartość dla tego pola musi być unikalna."
#: validators.py:89
#, python-brace-format
msgid "The fields {field_names} must make a unique set."
msgstr "Pola {field_names} muszą tworzyć unikalny zestaw."
#: validators.py:171
#, python-brace-format
msgid "Surrogate characters are not allowed: U+{code_point:X}."
msgstr ""
#: validators.py:243
#, python-brace-format
msgid "This field must be unique for the \"{date_field}\" date."
msgstr "To pole musi mieć unikalną wartość dla jednej daty z pola \"{date_field}\"."
#: validators.py:258
#, python-brace-format
msgid "This field must be unique for the \"{date_field}\" month."
msgstr "To pole musi mieć unikalną wartość dla konkretnego miesiąca z pola \"{date_field}\"."
#: validators.py:271
#, python-brace-format
msgid "This field must be unique for the \"{date_field}\" year."
msgstr "To pole musi mieć unikalną wartość dla konkretnego roku z pola \"{date_field}\"."
#: versioning.py:40
msgid "Invalid version in \"Accept\" header."
msgstr "Błędna wersja w nagłówku \"Accept\"."
#: versioning.py:71
msgid "Invalid version in URL path."
msgstr "Błędna wersja w ścieżce URL."
#: versioning.py:116
msgid "Invalid version in URL path. Does not match any version namespace."
msgstr "Niepoprawna wersja w ścieżce URL. Nie pasuje do przestrzeni nazw żadnej wersji."
#: versioning.py:148
msgid "Invalid version in hostname."
msgstr "Błędna wersja w nazwie hosta."
#: versioning.py:170
msgid "Invalid version in query parameter."
msgstr "Błędna wersja w parametrach zapytania."
| Gettext Catalog | 2 | scratchmex/django-rest-framework | rest_framework/locale/pl/LC_MESSAGES/django.po | [
"BSD-3-Clause"
]
|
-- {-# OPTIONS -v scope:10 -v scope.inverse:100 #-}
open import Common.Equality
open import A.Issue1635 Set
test : ∀ x → x ≡ foo
test x = refl
-- ERROR:
-- x != .#A.Issue1635-225351734.foo of type Foo
-- when checking that the expression refl has type x ≡ foo
-- SLIGHTLY BETTER:
-- x != .A.Issue1635.Foo.foo of type Foo
-- when checking that the expression refl has type x ≡ foo
-- WANT: x != foo ...
| Agda | 3 | shlevy/agda | test/Fail/Issue1635.agda | [
"BSD-3-Clause"
]
|
"""Nightscout util functions."""
import hashlib
def hash_from_url(url: str):
"""Hash url to create a unique ID."""
return hashlib.sha256(url.encode("utf-8")).hexdigest()
| Python | 4 | tbarbette/core | homeassistant/components/nightscout/utils.py | [
"Apache-2.0"
]
|
*****************************
* Sample Session from Section 7.5,
* "The Concurrency Workbench: A Semantics-Based Tool for
* the Verification of Concurrent Systems", by Rance Cleaveland,
* Joachim Parrow, and Bernhard Steffen
* ACM Transactions on Programming Languages and Systems
* January, 1993, Volume 15, Number 1
******************************
*****************************
* Entered by David Tarditi. Apr 15 1993
* Errors in the specifications of
* senders and receivers found in the captions of Figures 11 and 12
* have (hopefully) been corrected.
*****************************
*****************************
* Specification of ABP
*****************************
bi SPEC
send0.'rec0.SPEC + send1.'rec1.SPEC
*****************************
* Definition of the medium
*****************************
bi Medium
s00.(t.'r00.Medium + t.Medium) + \
s10.(t.'r10.Medium + t.Medium) + \
s01.(t.'r01.Medium + t.Medium) + \
s11.(t.'r11.Medium + t.Medium) + \
sack0.(t.'rack0.Medium + t.Medium) + \
sack1.(t.'rack1.Medium + t.Medium)
*****************************
* Definition of the sender
*****************************
bi S_0
send0.S00 + send1.S10
bi S00
's00.(rack0.S_1 + rack1.S00 + t.S00)
*******
* This state was given incorrectly in the caption for Fig. 11
* in the toplas paper
*******
bi S10
's10.(rack0.S_1 + rack1.S10 + t.S10)
bi S_1
send0.S01 + send1.S11
bi S01
's01.(rack1.S_0 + rack0.S01 + t.S01)
bi S11
's11.(rack1.S_0 + rack0.S11 + t.S11)
*****************************
* Definition of receiver
*****************************
bi R0
r00.'rec0.'sack0.R1 + r10.'rec1.'sack0.R1 + \
r01.'sack1.R0 + r11.'sack1.R0 + t.'sack1.R0
*****************************
* Definition of receiver state R1 was incorrectly in the caption
* for Fig. 12
*****************************
bi R1
r01.'rec0.'sack1.R0 + r11.'rec1.'sack1.R0 + \
r00.'sack0.R1 + r10.'sack0.R1 + t.'sack0.R1
***********************************************
* Assembly into ABP
***********************************************
bi ABP
(S_0|Medium|R0) \
\{r00,r10,r01,r11,s00,s10,s01,s11,rack0,rack1,sack0,sack1}
eq
ABP
SPEC
************************************************
* Check sorts
************************************************
sort SPEC
sort ABP
************************************************
* Check for equality
************************************************
mayeq
ABP
SPEC
***********************************************
* Check for deadlock, find deadlocked states
***********************************************
bpi
Deadlock
~<>T
cp
ABP
~Deadlock
fd
ABP
***********************************
* Corrected definition for sender
***********************************
bi S_0
send0.S00 + send1.S10
* S00 definition was wrong
*********
* S00 definition in caption for Fig. 13 was wrong
*********
bi S00
's00.(rack0.S_1 + rack1.S00 + t.S00) + rack0.S00 + rack1.S00
*********
* S10 definition in caption was wrong
*********
bi S10
's10.(rack0.S_1 + rack1.S10 + t.S10) + rack0.S10 + rack1.S10
bi S_1
send0.S01 + send1.S11
***********
* The definitions for S01 and S11 are also incorrect
***********
bi S01
's01.(rack1.S_0 + rack0.S01 + t.S01) + rack0.S01 + rack1.S01
bi S11
's11.(rack1.S_0 + rack0.S11 + t.S11) + rack0.S11 + rack1.S11
***********************************
* corrected definition for receiver
***********************************
bi R0
r00.'rec0.R0' + r10.'rec1.R0' + r01.R0" + r11.R0" + t.R0"
bi R0'
'sack0.R1 + r00.R0' + r10.R0' + r01.R0' + r11.R0'
bi R0"
'sack1.R0 + r00.R0" + r10.R0" + r01.R0" + r11.R0"
*****
* Caption for Fig. 14 was incorrect: original was ... r11.'rec.R1" ...
*****
bi R1
r01.'rec0.R1' + r11.'rec1.R1' + r00.R1" + r10.R1" + t.R1"
bi R1'
'sack1.R0 + r00.R1' + r10.R1' + r01.R1' + r11.R1'
bi R1"
'sack0.R1 + r00.R1" + r10. R1" + r01.R1" + r11.R1"
eq
ABP
SPEC
***
* Exit concurrency workbench
**
quit
| Redcode | 3 | dkuspawono/TILT-Compiler | BenchData/toplas.cw | [
"MIT-CMU"
]
|
DeprecationCopView = require '../lib/deprecation-cop-view'
describe "DeprecationCop", ->
[activationPromise, workspaceElement] = []
beforeEach ->
workspaceElement = atom.views.getView(atom.workspace)
activationPromise = atom.packages.activatePackage('deprecation-cop')
expect(atom.workspace.getActivePane().getActiveItem()).not.toExist()
describe "when the deprecation-cop:view event is triggered", ->
it "displays the deprecation cop pane", ->
atom.commands.dispatch workspaceElement, 'deprecation-cop:view'
waitsForPromise ->
activationPromise
deprecationCopView = null
waitsFor ->
deprecationCopView = atom.workspace.getActivePane().getActiveItem()
runs ->
expect(deprecationCopView instanceof DeprecationCopView).toBeTruthy()
describe "deactivating the package", ->
it "removes the deprecation cop pane item", ->
atom.commands.dispatch workspaceElement, 'deprecation-cop:view'
waitsForPromise ->
activationPromise
waitsForPromise ->
Promise.resolve(atom.packages.deactivatePackage('deprecation-cop')) # Wrapped for Promise & non-Promise deactivate
runs ->
expect(atom.workspace.getActivePane().getActiveItem()).not.toExist()
| CoffeeScript | 3 | davidbertsch/atom | packages/deprecation-cop/spec/deprecation-cop-spec.coffee | [
"MIT"
]
|
% Prolog type rules for Lark.
% a Type = type(Perm, Base)
% a Base = base(Kind, Name, [Type])
% a Kind = struct | class
% a Perm = own | var(Var) ?
access(Perm1, type(PermT, BaseT), type(PermO, BaseO)) :-
permits(PermT, Perm1),
applyAccessPerm(Perm1, BaseT, PermO),
applyOwnerPerm(Perm1, BaseT, BaseO).
applyAccessPerm(P, class, P).
applyAccessPerm(_, struct, own).
applyOwnerPerm(Perm1, [H | T], [H1 | T1]) :-
applyOwnerPerm(Perm1, H, H1),
applyOwnerPerm(Perm1, T, T1).
applyOwnerPerm(Perm1, [], []).
applyOwnerPerm(Perm1, base(Kind1, Name1, Types1), base(Kind1, Name1, Types2)) :-
applyOwnerPerm(Perm1, Types1, Types2).
applyOwnerPerm(Perm1, type(PermT, BaseT), type(PermO, BaseO)) :-
permitsConditionally(Perm1, PermT, PermO),
applyOwnerPerm(Perm1, BaseT, BaseO).
% own Vec<borrow Vec<own String>>
%
% apply share:
% - `share Vec<share Vec<share String>>` is ok
% - `share Vec<borrow Vec<own String>>` is ok, which is a bit weird
% - `share Vec<borrow Vec<share String>>` is also ok, which is also weird
% - but those are "non-minimal types" we would never *infer*, do I care?
%
% apply borrow:
% - `borrow Vec<borrow Vec<own String>>` is ok
% - `borrow Vec<borrow Vec<borrow String>>` is not ok, because
% - `borrow Vec<share Vec<share String>>` is not ok
| Prolog | 4 | tolziplohu/lark | notes/rules.prolog | [
"Apache-2.0",
"MIT"
]
|
/*
* Copyright (c) 2018-2020, Andreas Kling <[email protected]>
*
* SPDX-License-Identifier: BSD-2-Clause
*/
#include <AK/LexicalPath.h>
#include <LibCore/ArgsParser.h>
#include <stdio.h>
#include <unistd.h>
int main(int argc, char** argv)
{
if (pledge("stdio cpath", nullptr) < 0) {
perror("pledge");
return 1;
}
bool force = false;
bool symbolic = false;
const char* target = nullptr;
const char* path = nullptr;
Core::ArgsParser args_parser;
args_parser.add_option(force, "Force the creation", "force", 'f');
args_parser.add_option(symbolic, "Create a symlink", "symbolic", 's');
args_parser.add_positional_argument(target, "Link target", "target");
args_parser.add_positional_argument(path, "Link path", "path", Core::ArgsParser::Required::No);
args_parser.parse(argc, argv);
String path_buffer;
if (!path) {
path_buffer = LexicalPath::basename(target);
path = path_buffer.characters();
}
do {
if (symbolic) {
int rc = symlink(target, path);
if (rc < 0 && !force) {
perror("symlink");
return 1;
} else if (rc == 0) {
return 0;
}
} else {
int rc = link(target, path);
if (rc < 0 && !force) {
perror("link");
return 1;
} else if (rc == 0) {
return 0;
}
}
int rc = unlink(path);
if (rc < 0) {
perror("unlink");
return 1;
}
force = false;
} while (true);
return 0;
}
| C++ | 3 | r00ster91/serenity | Userland/Utilities/ln.cpp | [
"BSD-2-Clause"
]
|
# configuration file for util/mkerr.pl
#
# use like this:
#
# perl ../../../util/mkerr.pl -conf rsaref.ec \
# -nostatic -staticloader -write *.c
L RSAREF rsaref_err.h rsaref_err.c
| eC | 3 | jiangzhu1212/oooii | Ouroboros/External/OpenSSL/openssl-1.0.0e/demos/engines/rsaref/rsaref.ec | [
"MIT"
]
|
-- Copyright 2021 Jeff Foley. All rights reserved.
-- Use of this source code is governed by Apache 2 LICENSE that can be found in the LICENSE file.
name = "SpyOnWeb"
type = "scrape"
function start()
set_rate_limit(2)
end
function horizontal(ctx, domain)
local page, err = request(ctx, {url=build_url(domain)})
if (err ~= nil and err ~= "") then
log(ctx, "horizontal request to service failed: " .. err)
return
end
local pattern = "\"/go/([a-z0-9-]{2,63}[.][a-z]{2,3}([a-z]{2}|))\""
local matches = submatch(page, pattern)
if (matches == nil or #matches == 0) then
return
end
for i, match in pairs(matches) do
if (match ~= nil and #match >= 2 and match[2] ~= "") then
associated(ctx, domain, match[2])
end
end
end
function build_url(domain)
return "https://spyonweb.com/" .. domain
end
| Ada | 4 | Elon143/Amass | resources/scripts/scrape/spyonweb.ads | [
"Apache-2.0"
]
|
# Generate chromatic sequence
"k" "48 + (2 * !12)" kona
"maj" "s:0 2 4 5 7 9 11;do:{(s@(!x)!#s)+(12*((!x)%#s))+48};do 14" kona
# cycle through generated table
0.1 dmetro dup 0 pset
# frequency from tseq
0 "maj" tseq mtof
# amplitude
0.3
# carrier
1
# randomized modulator
0 p 1 4 trand floor
# randomized mod index
0 p 0 4 trand
fm
| SourcePawn | 2 | aleatoricforest/Sporth | examples/kona.sp | [
"MIT"
]
|
%%
%unicode 6.1
%public
%class UnicodeScripts_6_1_extensions_1
%type int
%standalone
%include ../../resources/common-unicode-all-enumerated-property-defined-values-only-java
%%
<<EOF>> { printOutput(); return 1; }
\p{Script_Extensions:Arabic} { setCurCharPropertyValue("Script_Extensions:Arabic"); }
\p{Script_Extensions:Armenian} { setCurCharPropertyValue("Script_Extensions:Armenian"); }
\p{Script_Extensions:Bengali} { setCurCharPropertyValue("Script_Extensions:Bengali"); }
\p{Script_Extensions:Bopomofo} { setCurCharPropertyValue("Script_Extensions:Bopomofo"); }
\p{Script_Extensions:Buhid} { setCurCharPropertyValue("Script_Extensions:Buhid"); }
\p{Script_Extensions:Cypriot} { setCurCharPropertyValue("Script_Extensions:Cypriot"); }
\p{Script_Extensions:Gujarati} { setCurCharPropertyValue("Script_Extensions:Gujarati"); }
\p{Script_Extensions:Mongolian} { setCurCharPropertyValue("Script_Extensions:Mongolian"); }
[^] { }
| JFlex | 4 | Mivik/jflex | testsuite/testcases/src/test/cases/unicode-scripts/UnicodeScripts_6_1_extensions_1.flex | [
"BSD-3-Clause"
]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.