repo_name
stringlengths 4
136
| issue_id
stringlengths 5
10
| text
stringlengths 37
4.84M
|
|---|---|---|
ianjdarrow/protocol-legal | 396721964 | Title: Cannot "Agree and Submit" with uBlock Origin active
Question:
username_0: I went through the agreement process in both Brave and Chrome, and the final "Agree and Submit" button was not responsive. I deactivated uBlock Origin for that page, and it worked (in Brave) to complete registration. I do not know which mechanism was interfered with by uBlock Origin.
Answers:
username_1: I confirm that the [invitation website form](https://protocol-legal.com/filecoin-invite) breaks when using uBlock Origin with Firefox with these JS errors:
```
Cross-Origin Request Blocked: The Same Origin Policy disallows reading the remote resource at https://ipinfo.io/json. (Reason: CORS request did not succeed).[Learn More]
Unhandled promise rejection TypeError: "NetworkError when attempting to fetch resource." es6.promise.js:110
e es6.promise.js:110
exports _perform.js:3
M es6.promise.js:104
exports _invoke.js:5
<anonymous> _task.js:35
b _task.js:21
w
```
Sure enough, https://ipinfo.io/ is blocked by default by uBlock as a tracking service.
The error for this fetch should get caught and handled - either by leaving the appropriate form fields null, or by telling the user to disable their adblocker for the submission. |
dunst-project/dunst | 502795857 | Title: SIGCONT / SIGTSTP handling
Question:
username_0: Hello,
Currently to pause and resume `dunst` is SIGUSR1 (pause) and SIGUSR2 (resume)
I greatly appreciate this feature, but I believe the signals `SIGTSTP` (keyboard stop) and `SIGCONT` would be better suited for this task.
I don't really have a user story for this, but I could imagine tools would be able to take advantage of this more semantically correct signal handling.
Current behaviour of `dunst` is to ignore `SIGTSTP` signals
Answers:
username_1: `SIGUSR*` and `SIGTSTP`/`SIGCONT` signals differ in a simple and easy way.
- `SIGUSR*`: User defined signal, does whatever the process wants to do. If used, always documented. For server daemons, it's most often re-reading the configuration file. We at dunst use it to start/stop _the functionality_ and start/stop displaying notifications.
- `SIGTSTP`/`SIGCONT`: This is necessary for job control. `Ctrl+Z` uses this signal. But all in all, _the process_ stops by getting a signal of the terminal.
So in the semantics differ. We usually don't get spawned in a terminal and dunst has to run all the time. If we really stop the dunst process, many clients may fuck up and run into timeouts until finally DBus will probably start a new dunst instance.
I don't see where the semantics TSTP and CONT are "better suited" for dunst. Also, I'd like to remind you of #445. We plan to implement the functionality by a proper command line control.
username_0: Alright, thanks for your consideration.
Have a nice day
Status: Issue closed
|
pesto-students/batch-8-CODERHOOD | 473095497 | Title: Invitation Approval works locally and on server but fails while testing
Question:
username_0: `acceptAllWorkspaceInvites()` function in `invitations/util.js` file works while local testing works on the server but fails any tests written in which this function is used. Need to isolate the issue and fix it. |
Azure/ServerlessLibrary | 369243572 | Title: Tagging system improvements
Question:
username_0: I'll add here a better summary of what exactly we need, but the filtration/tagging system needs improvement.
Answers:
username_0: To elaborate, this work item tracks adding support for tags of different categories on the API level. Design still pending. |
GME-392/sanhak-server | 858013518 | Title: [유저 DB] - 모델 추가 요청
Question:
username_0: 
안녕하세요! 혹시 회원 DB 항목 중 'homepage' 라는 항목 추가 부탁드려도 될까요?! 😆
@username_1
Status: Issue closed
Answers:
username_1: 항목 추가했고 그때그때 요청하시면 됩니다. |
damain/ecowatchtt-web | 1046025935 | Title: Users unable to open petition
Question:
username_0: It was reported that a few users were unable to open the petition.
It was confirmed that at least one user had this issue with a Samsung phone, However this point neither the model or browser can be confirmed. |
Azure/azure_preview_modules | 379120813 | Title: storage_mb parameter is not working in azure_rm_postgresqlserver
Question:
username_0: I can see the storage_mb is always configured with a default value for 5GB.
I'm executing the following provision in the module:
```
TASK [Create ADP server instance] *************************************************************************************************************************************************
task path: /home/baikal/delivery/ansible/playbook_infra_create_15_dbaas_azure.yml:40
Using module file /usr/local/lib/python2.7/dist-packages/ansible/modules/cloud/azure/azure_rm_postgresqlserver.py
<localhost> ESTABLISH LOCAL CONNECTION FOR USER: root
<localhost> EXEC /bin/sh -c 'AZURE_SUBSCRIPTION_ID=4b6ccc83-033c-412c-8285-7812ece487c4 python && sleep 0'
[WARNING]: Azure API profile latest does not define an entry for PostgreSQLManagementClient
changed: [localhost] => {
"changed": true,
"fully_qualified_domain_name": "baikal-adptest-db.postgres.database.azure.com",
"id": "/subscriptions/4b6ccc83-033c-412c-8285-7812ece487c4/resourceGroups/baikal-adptest-rg/providers/Microsoft.DBforPostgreSQL/servers/baikal-adptest-db",
"invocation": {
"module_args": {
"ad_user": null,
"adfs_authority_url": null,
"admin_password": "<PASSWORD>",
"admin_username": "postgres",
"api_profile": "latest",
"auth_source": null,
"cert_validation_mode": null,
"client_id": null,
"cloud_environment": "AzureCloud",
"create_mode": "Default",
"enforce_ssl": false,
"location": "northeurope",
"name": "baikal-adptest-db",
"password": <PASSWORD>,
"profile": null,
"resource_group": "baikal-adptest-rg",
"secret": null,
"sku": {
"capacity": "4",
"name": "GP_Gen5_4",
"tier": "GeneralPurpose"
},
"state": "present",
"storage_mb": 307200,
"subscription_id": null,
"tenant": null,
"version": "10"
}
},
"state": "Ready",
"version": "10"
}
```
After the ansible module execution I can check the postgres server configuration and I can find the following:
```
mac-510894:baikal adgp$ az postgres server show --resource-group baikal-adptest-rg --name baikal-adptest-db
{
"administratorLogin": "postgres",
"earliestRestoreDate": "2018-11-09T11:07:05.180000+00:00",
"fullyQualifiedDomainName": "baikal-adptest-db.postgres.database.azure.com",
"id": "/subscriptions/4b6ccc83-033c-412c-8285-7812ece487c4/resourceGroups/baikal-adptest-rg/providers/Microsoft.DBforPostgreSQL/servers/baikal-adptest-db",
"location": "northeurope",
[Truncated]
"resourceGroup": "baikal-adptest-rg",
"sku": {
"capacity": 4,
"family": "Gen5",
"name": "GP_Gen5_4",
"size": null,
"tier": "GeneralPurpose"
},
"sslEnforcement": "Disabled",
"storageProfile": {
"backupRetentionDays": 7,
"geoRedundantBackup": "Disabled",
"storageMb": 307200
},
"tags": null,
"type": "Microsoft.DBforPostgreSQL/servers",
"userVisibleState": "Ready",
"version": "10"
}
```
Answers:
username_1: @username_2
username_1: @username_2
username_1: @username_2 this duplicate of #48415, may be we can close this and pay more attention on another. Thanks!
username_1: @username_2
username_1: @username_2
username_1: @username_2
username_2: this is already fixed in ansible / devel, so it should be propagated here very soon
username_1: to_be_closed , It has been fixed on latest version. Thanks!
username_1: ping
username_2: just closing as fixed
Status: Issue closed
username_0: yes, right, I have tested and it is working fine now.
Great job! |
jmschrei/pomegranate | 635466051 | Title: IndependentComponentsDistribution does not fit DiscreteDistribution components
Question:
username_0: Hello,
Trying to fit an IndependentComponentsDistribution object on two distributions, one is normal and the other is discrete, only the normal component is fitted. Moreover, after running such fitting, trying to fit the discrete component "externally" (i.e. directly fit the same object used in the construction of the multivariate distribution) produces the error "TypeError: object of type 'NoneType' has no len()".
The following code generates this behavior:
`
import pomegranate
import numpy as np
np.random.seed(0)
x = np.column_stack([np.random.randint(5, size=20), np.random.normal(size=20)])
normal_dist = pomegranate.distributions.NormalDistribution(0, 1)
discrete_dist = pomegranate.distributions.DiscreteDistribution(
dict(zip(range(5), range(5))))
multivariate_dist = pomegranate.distributions.IndependentComponentsDistribution(
[discrete_dist, normal_dist])
print('\nUnfitted multivariate distribution:'
'\n-----------------------------------\n', multivariate_dist.parameters)
multivariate_dist.fit(x)
print('\nFitted multivariate distribution:'
'\n--------------------------------\n', multivariate_dist.parameters)
# Trying to fit the discrete distribution at this point results with an error:
# >>> discrete_dist.fit(x[:, 0])
# Traceback (most recent call last):
# File "<input>", line 2, in <module>
# File "pomegranate/distributions/DiscreteDistribution.pyx", line 214, in pomegranate.distributions.DiscreteDistribution.DiscreteDistribution.fit
# File "pomegranate/distributions/DiscreteDistribution.pyx", line 269, in pomegranate.distributions.DiscreteDistribution.DiscreteDistribution.from_summaries
# TypeError: object of type 'NoneType' has no len()
# This does not happen for the normal distribution (running 'normal_dist.fit(x[:, 1])').
# Also, fitting a new identical discrete distribution works fine:
discrete_dist = pomegranate.distributions.DiscreteDistribution(
dict(zip(range(5), range(5))))
print('\nFitting discrete distribution outside the multivariate distribution:')
print('Unfitted discrete distribution:'
'\n-------------------------------\n', discrete_dist.parameters)
discrete_dist.fit(x[:, 0])
print('\nFitted discrete distribution:'
'\n-----------------------------\n', discrete_dist.parameters)
`
Answers:
username_1: Yes, unfortunately pomegranate does not support this right now. Sorry about that.
Status: Issue closed
|
fpinscala/fpinscala | 211263240 | Title: Referencing smart constructor from the companion object in trait definition doesn't compile
Question:
username_0: So I was unable to get the answer for ex 5.3. I checked the answer key and pasted that into my trait def and it wouldn't compile! The compiler complained that it couldn't resolve the symbols `empty` and `cons`. So I tried instead using `Stream.cons` and `Stream.empty` and that worked.
Why did I have to do this? Why isn't this in the answer key?
Answers:
username_1: do you have import Stream._ at top of file ?
how did the take function above it compile since that also uses empty and cons.
i take it your using sbt, try reloading it (reload) ?
username_0: Yeah, at the top of my file I did import Stream, but to prevent it from being a default import (which I believe it is?). I thought this would be relatively harmless.
`import scala.collections.immutable.{Stream => _}`
There might be a typo there, I don't have the code in front of me..
Not sure which function you mean when you say the one "above it," it's very possible that I skipped implementing another function that required some of the things defined in the companion object.
I'm using IntelliJ with a worksheet.
username_1: have a look in the answers folder for that code (on github), the take function for exercise 5.2 is the function I meant. If that compiles, then you should have no problem with takeWhile.
they just use
import Stream._
this means that instead of having to prefix Stream.empty or Stream.cons you can drop the Stream._ and just use import Stream._ instead.
the stream that you have prevented is the collections Stream. (not the same thing)
i tend to use sbt now (for compiling and running), and use intelliJ for browsing code only. |
graphql-compose/graphql-compose | 423040505 | Title: Documentation refers to TypeComposer, however it does not exist
Question:
username_0: Was it renamed to objectTypeComposer?
https://github.com/graphql-compose/graphql-compose/commit/ea23d8719edb278c00ea88b57364b6d7e6d90d78#diff-76e3058e046eb9a31144aac58f761654
Answers:
username_1: Yep it was renamed 3 days ago in v6.0.0.
https://github.com/graphql-compose/graphql-compose/releases/tag/v6.0.0
I will update documentation in couple of days. Any help very appreciated. Thanks.
username_2: @username_1 Just a thought, I think it could be best if you kept the `TypeComposer` alongside the new `ObjectTypeComposer`, i believe there's a lot of legacy code requiring this, as we slowly update.
username_1: v6 is a major release with small changes in typescript definitions.
So I decide for major version clean up API and old methods.
`TypeComposer` very easy to rename if you use VSCode. There are `search` by files with `whole word` option (looks like ).
<img width="392" alt="Screen Shot 2019-03-27 at 11 18 30 AM" src="https://user-images.githubusercontent.com/1946920/55052158-16342f00-5082-11e9-8c12-83922867a6dd.png">
It helps to replace all occuriencies very easy. |
siconos/siconos | 245424972 | Title: RockingBlock_RockingBlockED_NewMarkAlpha fails
Question:
username_0: x0: 0.0334936490538904
y0: 0.558012701892219
theta0: 1.0471975511966
PI: 3.14159265358979
Initial position of the rocking block:
[3](3.349365e-02,5.580127e-01,1.047198e+00)
Initial velocity of the rocking block:
[3](0.000000e+00,0.000000e+00,0.000000e+00)
Mass matrix of the rocking block:
1.000000e+00, 0.000000e+00, 0.000000e+00,
0.000000e+00, 1.000000e+00, 0.000000e+00,
0.000000e+00, 0.000000e+00, 1.041667e-01,
External force applied on the rocking block:
[3](0.000000e+00,-9.810000e+00,0.000000e+00)
====> Simulation initialisation ...
Size of IndexSet0: 2
Size of IndexSet1: 1
Size of IndexSet2: 0
====> Start computation ...
0% 10 20 30 40 50 60 70 80 90 100%
|----|----|----|----|----|----|----|----|----|----|
In EventDriven::advanceToEvent, some events are detected!!!
a0: [3](3.349365e-02,5.580127e-01,1.047198e+00)
a1: [3](0.000000e+00,0.000000e+00,0.000000e+00)
a2: [3](0.000000e+00,-4.905000e-06,0.000000e+00)
a3: [3](0.000000e+00,3.921237e-16,0.000000e+00)
a4: [3](0.000000e+00,-6.722087e-16,0.000000e+00)
a5: [3](0.000000e+00,2.800884e-16,0.000000e+00)
A new event occurs at time: 4.882813e-07
Runtime Exception : NewMarkAlphaOSI::computeFreeOutput, this OSNSP does not exist
Status: Issue closed
Answers:
username_1: fixed in 3842041d4f4cf9788dc113f5e6ff4ee97f61fbdf |
knighty/fimfiction-issues | 393939007 | Title: Mathjax popup on exactly two chapters
Question:
username_0: This weirded me the hell out. I was reading The Maretian (totally recommend), and on exactly two chapters I get a MathJax diagnostic popup in the lower left corner of my screen. No other part of the site and no other chapters in that very long story seem to do that.
https://www.fimfiction.net/story/396744/135/the-maretian/sol-231
https://www.fimfiction.net/story/396744/136/the-maretian/sol-232
Using Chrome. The MathJax <script> element seems to be getting added by the site's static scripts.js. Did I accidentally Pinkie myself into a site debug mode somehow? I don't _think_ I did anything unusual myself unless you count...like...scrolling down. But did I accidentally Pinkie myself into a site debug mode somehow? |
bitovi/testee | 174606298 | Title: Use local phantom
Question:
username_0: Testee uses `which` to find the phantom path but the install in the projects node_modules folder won't be reported by this. Rather than depending on a global module, it should look to see if it is installed in local project first *then* fallback to global.<issue_closed>
Status: Issue closed |
living481/Blog | 999107408 | Title: Unity Input System使用
Question:
username_0: 
 |
call4paperz/call4paperz | 166310479 | Title: Edit proposal name
Question:
username_0: Hello I would like to change the proposal name at this link:
http://call4paperz.com/events/2-tech-day-do-guru-pr/proposals/2341
Actually I thought that the field "Name" was really related to my my name, but after that I figured out that should be the proposal title then, please, change to **Testes Automatizados com NodeJS, Selenium e o framework TheIntern**
And my suggestion would be to change the label with the text "Name" to a more descriptive text like "Proposal Name" or "Proposal Title" this way is more clear to the user what the field is.
Answers:
username_1: @username_0 For security reasons, you cannnot change a proposal name after 30min.
You are not the only one suffering about it.
What I think we need to do is: put a better warning about it, and also, as you said, improve this `proposal name`.
@username_2 what do you think about?
username_2: We never had that problem, but we'll take this suggestion for the next release. The proposal name is updated.
Status: Issue closed
username_2: Never knew that other people had that problem! Anyway... it's partially solved. Let's take the label name suggestion to the next release. |
weblabdeusto/weblabdeusto | 62080256 | Title: Boolean experiment parameters
Question:
username_0: When a boolean experiment parameter is changed, it seems that the change is not applied. However, in the database it *is* applied and it has an effect, it simply is not displayed properly in the administrator GUI. Also, when saving again, the change is reset.<issue_closed>
Status: Issue closed |
nih-cfde/cfde-deriva | 993690594 | Title: Other favorites to add?
Question:
username_0: In Quarter 1 we merged the favorites feature for anatomy, assay type, and DCC terms. In Quarter 2 we plan on adding disease to this list.
What other favorites should we add? Currently the model includes the following favorites on the `user_profie` table:
- added already
- Favorite Anatomy
- Favorite Assay Type
- Favorite DCC
- pending
- Favorite Disease
- could expose UI features already
- Favorite Data Type
- Favorite File Format
- Favorite Subject Granularity
- Favorite Subject Role
- Favorite Taxon
There could be more terms (vocabularies) that we might want to consider too.
Answers:
username_1: I think Taxon, Data Type, and File Format would be fine to add. I'm not opposed to the others, but I think I've heard something about subject role and subject granularity being changed in the model. @username_2 can you remind me what the deal with those two is and whether it's worth making them options?
username_2: they're not changing in this quarter's dev model; i don't yet know concretely when, how or in what way they'll change, but in any event, i vote _not_ to include them in the favorites feature for the foreseeable horizon (two quarters' worth of horizon, by my best guess) -- if/when they do change, it will be to make them more straightforward to grok. they're hard enough for DCCs to understand and use, let alone for portal users to configure searches around, so i propose leaving them out of it for the time being.
username_1: We are officially doing Taxon, Data Type, and File Format
username_3: The checked favorites have been enabled in the app-dev catalog "1" and the app-dev registry's user\_profile detailed view also provides a filtered list of the same favorites (plus saved queries)
username_0: Linking favorites disease issue:
https://github.com/nih-cfde/dashboard/issues/103
Status: Issue closed
|
visjs/vis-timeline | 788335685 | Title: (7.4.6) onInitialDrawComplete not always called
Question:
username_0: 
onInitialDrawComplete is not always called.
workaround is an empty object of the rollingMode option:
rollingMode: {
// empty object as a fix for missing onInitialDrawComplete event
}
If you need a reproducible example, please let me know =) |
cityofaustin/atd-data-tech | 622614801 | Title: None
Question:
username_0: Currently we have a Days Under Review field to determine how long a submission has been under review. We also have date fields when actions are completed or started. Is there a certain # of days a submission must be reviewed in? We already determined that a case could be open for many years before being completed. Is the Days Under Review field good enough for reviewers to know how close to their deadline they are or do we need another field to display that warning message or deadline?
Discuss at next meeting?
Answers:
username_0: @username_1 Lets identify with stakeholders if the Review start and end dates plus the Days Under Review field meet this need or if they do need to assign Deadlines to submissions.
Status: Issue closed
username_1: Stakeholder Feedback: There's no code required review deadlines. Not MVP. Closing. |
JasXSL/GoThongs | 739350797 | Title: PanRi Free Roam - Teleporter at Sa'lad
Question:
username_0: **Describe the bug**
Teleporter is missing near Sa'lad. I won the fight against Sa'lad and the teleporter worked. But now as I tried to use the teleporter, I did not see the destination. I walked to Sa'lad and the telepoerter is not glowing. I used the memory shared and did the fight again, but the teleporter don't glow now.
**Settings:**
- GoThongs HUD Version - 2.0.8
- PanRi Version - 0.13.0
- no other Mod
- Firestorm 6.3.9.58205
Answers:
username_1: closing as duplicate of #334
Status: Issue closed
|
filesender/filesender | 199810663 | Title: Allow getting cURL-able links
Question:
username_0: The Download button uses JavaScript to start the download. This is great for encrypted files (these must be decrypted by JavaScript) but not so great for normal files.
Users may want to open the link to the download page on their desktop, but then do the actual download on another server using curl or wget, to avoid having to download a large file to their desktop, only to send it somewhere else.
Answers:
username_1: We have already solved this on our installation. I did it by adding the download url next to the Download button for unencrypted files.
If you like i can add it as a pull request.
username_2: This was already fixed at one point in the pre-1.x cycle. A functional test to prevent regressions would be nice...
username_0: Proposed a solution.
username_1: So did I, was something our users were asking for a while back, as they need to curl from clusters.
https://github.com/filesender/filesender/pull/20
username_3: The functionality is good to have. We now have two pull requests that do the same only slightly differently. I believe it's in the user's interest to see this possibility explicitly, as in: there's a link you clearly see in the UI. That's in @username_1 's pull request. @username_0 's request creates the URL all server side which when he explained it sounded like the better way to do that. @username_1 and @username_0, could the two of you perhaps agree on one common pull request? I'll happily approve that.
username_4: it seems that @username_1 closed their PR a while ago. I'm lining up to merge https://github.com/filesender/filesender/pull/19 now and then update the disclaimer message to let folks know that right click + copy link works.
Status: Issue closed
|
nextgenusfs/funannotate | 220840019 | Title: SignalIP to funannotate Ubuntu
Question:
username_0: Because of its academic license SignalP doesn't come with funannotate, right?
But I haven't found any info on Wiki or in issues on how to link installed SignalP to funannotate.
```
2017-04-11 01:16:04,340: Predicting secreted and transmembrane proteins using Phobius
2017-04-11 01:35:28,255: SignalP not installed, secretome prediction less accurate using only Phobius
2017-04-11 01:35:28,256: 0 secretome and 0 transmembane annotations added
```
Answers:
username_0: is it something I am looking for?
```
#run signalP if installed, have to manually install, so test if exists first, then run it if it does, parse results
signalp_out = os.path.join(outputdir, 'annotate_misc', 'signalp.results.txt')
secreted_out = os.path.join(outputdir, 'annotate_misc', 'annotations.secretome.txt')
membrane_out = os.path.join(outputdir, 'annotate_misc', 'annotations.transmembrane.txt')
```
username_1: You just need to have it in your $PATH. So you can download, extract, and then you add the location containing `signalp` perl script in your $PATH variable.
Status: Issue closed
username_0: got it! thanks. |
tablacus/TablacusExplorerAddons | 613960324 | Title: Force refresh - enhance to refresh in 3 situations
Question:
username_0: Please enhance the add-on so that the active file list can also be refreshed upon:
* switch pane
* bring up TE window from minimized
* switch to TE from another application
Maybe let user opt for each of the above situations.
Thanks.
 |
moment/luxon | 1015153200 | Title: Strange results when endOf and minus methods are used in different order
Question:
username_0: **Describe the bug**
I came across a problem, where i seem to be getting different results for a "previous month" time range, based on the order of methods.
**Actual vs Expected behavior**
```js
luxon.DateTime
.fromISO('2021-09-25T12:00:00Z')
.setZone('Europe/Amsterdam')
.minus({months:1})
.endOf('month')
.toISO();
```
results in `2021-08-31T23:59:59.999+02:00`, as expected.
In the next example I only switch the order in which `minus` and `endOf` are called.
```js
luxon.DateTime
.fromISO('2021-09-25T12:00:00Z')
.setZone('Europe/Amsterdam')
.endOf('month')
.minus({months:1})
.toISO();
```
results in `2021-08-30T23:59:59.999+02:00`, where I would expect the 31th of august.
**Desktop:**
- OS: MacOS
- Browser: MS Edge v 94.0.992.31
- Luxon version 2.0.2
- Timezone "Europe/Amsterdam"
Status: Issue closed
Answers:
username_0: I think I get the point, still it seems a bit strange that when I'm subtracting a month from the end of a date in oktober (31th) it'll return the 30th of september. That gives me the impression that there is some kind of awareness of month lengths relative to eachother.
username_1: That's true; there is. The goal of month plus/minus is to keep the day number the same but change the month. But it has to make _some_ choice when that day doesn't exist in the target month, and it does that by using the last day of the month. So subtracting a month of Oct 31 gives you Sept 30 rather than, say, Oct 1. But you're asking for the reverse, that Sep 30 minus a month would give Aug 31 instead of--per the goal above--Aug 30, which would be really weird!
username_0: Ok, thanks for clearing that up. |
r-lib/fs | 287110136 | Title: Usage on Solaris
Question:
username_0: Unfortunately libbsd is not available on opencsw and solaris does not define `getmode / strmode` functions.
Considering very few people actually use Solaris I am tempted to provide stub functions and disable the relevant tests on Solaris, I am not sure it is worth the effort to port the libbsd functions.<issue_closed>
Status: Issue closed |
ralphje/imagemounter | 158416158 | Title: Unable to mount images from NIST Hacking Case scenario
Question:
username_0: Hello! I am trying to use imagemounter on the EnCase images provided for [NIST's "Hacking Case" scenario](http://www.cfreds.nist.gov/Hacking_Case.html). I have been unable to mount the files so far. What is the best way to help debug this?
Here is the output I get, using `-v`:
Answers:
username_1: $ umount /tmp/im_2_WK7a4q_
[+] Found unallocated space: block offset: 0009510480, length: 0000003780
[-] Skipped 1.85 MiB 3:Unallocated unalloc volume
[+] Parsed all volumes!
[+] Analysis complete, unmounting...
$ fusermount -u /tmp/image_mounter_cQwDwb
[+] All cleaned up
```
Perhaps you are mounting using `affuse` or `xmount`, which I couldn't get to work either; `ewfmount` does work for me. The error also suggests you perhaps may not have downloaded the .E02 file from the NIST website?
Could you include the output of `sudo imount 4Dell\ Latitude\ CPi.E01 -vvvv` and `imount --check` in your bug report?
username_1: Closing due to lack of response. Please re-open when you can provide the requested output.
Status: Issue closed
username_0: Hello, and sorry for the delay. I do have both the .E01 and .E02 files.
Here is `sudo imount 4Dell\ Latitude\ CPi.E01 -vvvv`:
```shell
$ sudo imount 4Dell\ Latitude\ CPi.E01 -vvvv
[+] Mounting image 4Dell Latitude CPi.E01 using auto...
$ ewfmount -X allow_other 4Dell Latitude CPi.E01 /tmp/image_mounter_SolnJ5
Unable to open EWF file(s).
libmfdata_file_list_get_file_by_index: missing file.
libewf_handle_open_read_segment_files: unable to retrieve segment file: 1 from list.
libewf_handle_open_file_io_pool: unable to read segment files.
libewf_handle_open: unable to open handle using a file IO pool.
mount_handle_open_input: unable to open file(s).
[-] Could not mount 4Dell Latitude CPi.E01, trying other method
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/imagemounter/disk.py", line 217, in mount
_util.check_call_(cmd, stdout=subprocess.PIPE)
File "/usr/local/lib/python2.7/dist-packages/imagemounter/_util.py", line 110, in check_call_
return subprocess.check_call(cmd, *args, **kwargs)
File "/usr/lib/python2.7/subprocess.py", line 541, in check_call
raise CalledProcessError(retcode, cmd)
CalledProcessError: Command '[u'ewfmount', u'-X', u'allow_other', u'4Dell Latitude CPi.E01', u'/tmp/image_mounter_SolnJ5']' returned non-zero exit status 1
$ ewfmount 4Dell Latitude CPi.E01 /tmp/image_mounter_SolnJ5
Unable to open EWF file(s).
libmfdata_file_list_get_file_by_index: missing file.
libewf_handle_open_read_segment_files: unable to retrieve segment file: 1 from list.
libewf_handle_open_file_io_pool: unable to read segment files.
libewf_handle_open: unable to open handle using a file IO pool.
mount_handle_open_input: unable to open file(s).
[-] Could not mount 4Dell Latitude CPi.E01, trying other method
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/imagemounter/disk.py", line 217, in mount
_util.check_call_(cmd, stdout=subprocess.PIPE)
File "/usr/local/lib/python2.7/dist-packages/imagemounter/_util.py", line 110, in check_call_
return subprocess.check_call(cmd, *args, **kwargs)
File "/usr/lib/python2.7/subprocess.py", line 541, in check_call
raise CalledProcessError(retcode, cmd)
CalledProcessError: Command '[u'ewfmount', u'4Dell Latitude CPi.E01', u'/tmp/image_mounter_SolnJ5']' returned non-zero exit status 1
$ xmount --in ewf 4Dell Latitude CPi.E01 /tmp/image_mounter_SolnJ5
[-] Could not mount 4Dell Latitude CPi.E01, trying other method
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/imagemounter/disk.py", line 217, in mount
_util.check_call_(cmd, stdout=subprocess.PIPE)
File "/usr/local/lib/python2.7/dist-packages/imagemounter/_util.py", line 110, in check_call_
return subprocess.check_call(cmd, *args, **kwargs)
File "/usr/lib/python2.7/subprocess.py", line 541, in check_call
raise CalledProcessError(retcode, cmd)
CalledProcessError: Command '[u'xmount', u'--in', u'ewf', u'4Dell Latitude CPi.E01', u'/tmp/image_mounter_SolnJ5']' returned non-zero exit status 1
$ ewfmount -X allow_other 4Dell Latitude CPi.E01 4Dell Latitude CPi.E02 /tmp/image_mounter_SolnJ5
Unable to open EWF file(s).
libmfdata_file_list_get_file_by_index: missing file.
libewf_handle_open_read_segment_files: unable to retrieve segment file: 1 from list.
libewf_handle_open_file_io_pool: unable to read segment files.
libewf_handle_open: unable to open handle using a file IO pool.
mount_handle_open_input: unable to open file(s).
[-] Could not mount 4Dell Latitude CPi.E01, trying other method
Traceback (most recent call last):
File "/usr/local/lib/python2.7/dist-packages/imagemounter/disk.py", line 217, in mount
_util.check_call_(cmd, stdout=subprocess.PIPE)
[Truncated]
INSTALLED mmls
MISSING pytsk3 install using pip
INSTALLED parted
-- Detecting volume types (all recommended, first two highly recommended) --
INSTALLED fsstat
INSTALLED file
MISSING python-magic install using pip
INSTALLED disktype
-- Enhanced mounting and detecting disks (install when needed) --
INSTALLED mdadm
INSTALLED cryptsetup
INSTALLED mountavfs
-- Mounting volumes (install when needed) --
MISSING mount.xfs needed for XFS volumes, part of the xfsprogs package
INSTALLED mount.ntfs
INSTALLED lvm
INSTALLED vmfs-fuse
MISSING mount.jffs2 needed for JFFS2 volumes, part of the mtd-tools package
MISSING mount.squashfs needed for SquashFS volumes, part of the squashfs-tools package
```
username_0: I got this working, finally. I must have just had bad copies of the files -- worked perfectly after I re-downloaded them. |
xJon/The-1.7.10-Pack | 573708446 | Title: My backup will not load. I have deleted chickenchunks. No difference
Question:
username_0: will provide necessary info
Answers:
username_1: Please post your latest log file using a [paste-tool](http://paste.ubuntu.com/).
You can find it using [this](https://github.com/username_1/The-1.12.2-Pack/blob/master/log_location.md#finding-modpack-installation-folder) guide.
username_1: Closed due to inactivity.
Will reopen if requested.
Status: Issue closed
|
iot-nextop/IOT | 488411299 | Title: B팀
Answers:
username_1: https://docs.google.com/document/d/1fxje_PNgaBGz_vYkY4CisGmdCi3nN9d6zinNYomgfOE/edit
username_2: int redLed = 12;
int greenLed = 11;
int buzzer = 10;
int smokeA0 = A5;
int sensorThres = 800;
const int buzzerPin = 12;
const int flamePin = 11;
int Flame = HIGH;
void setup() {
Serial.begin(9600);
pinMode(buzzerPin, OUTPUT);
pinMode(flamePin, INPUT);
pinMode(redLed, OUTPUT);
pinMode(greenLed, OUTPUT);
pinMode(buzzer, OUTPUT);
pinMode(smokeA0, INPUT);
}
void loop() {
int analogSensor = analogRead(smokeA0);
Serial.print("Pin A0: ");
Serial.println(analogSensor);
// Checks if it has reached the threshold value
if (analogSensor > sensorThres)
{
digitalWrite(redLed, HIGH);
digitalWrite(greenLed, LOW);
tone(buzzer, 1000, 20000);
}
else
{
digitalWrite(redLed, LOW);
digitalWrite(greenLed, HIGH);
noTone(buzzer);
}
delay(1000);
Flame = digitalRead(flamePin);
if (Flame== LOW)
{
Serial.println("Fire!!!");
digitalWrite(buzzerPin, HIGH);
}
else
{
Serial.println("No worries");
digitalWrite(buzzerPin, LOW);
}
delay(1000);
}
Status: Issue closed
|
mbdavid/LiteDB | 252473001 | Title: about null value
Question:
username_0: class:
```cs
public class Person
{
public int Id { get; set; }
public string Fullname { get; set; }
public DateTime? Birthday { get; set; }
}
```
Test:
```cs
col.Insert(new Person { Fullname = "John" });
col.Insert(new Person { Fullname = "Doe" });
col.Insert(new Person { Fullname = "Joana" });
col.Insert(new Person { Fullname = "Marcus" });
```
delete result:
```cs
var delResult = col.Delete(x => x.Birthday < DateTime.UtcNow);
```
OR
find result:
```cs
var findResult = col.Find(x => x.Birthday < DateTime.UtcNow);
```
I think `delResult = 0` or `findResult.Count() = 0`, but result of test is `delResult = 4` or `findResult.Count() = 4`.
Is it a bug?
Answers:
username_1: Hi, that happend because `null` are a type of value (like int or string). And, when you compare 2 diferent types (like DateTime with null) LiteDB comprare data type order only. And null are less than Date. This will happend if you add another datatype too, like
`db.Engine.Insert("col", new BsonDocument { ["_id"] = 5, ["Fullname"] = "Demo", ["Birthday"] = 5 });`
So, when I compare `5` with `DateTime.Now` I just compare order of each data type:
https://github.com/username_1/LiteDB/blob/master/LiteDB/Document/BsonType.cs
I will take a look on how MongoDB resolve this and return here.
username_1: { "_id" : 3, "a" : null }
{ "_id" : 4 }
```
Mongo does not return different types when using `>` `<`
It's not too much clear to me after read this:
https://docs.mongodb.com/manual/reference/bson-type-comparison-order/#bson-types-comparison-order
username_0: Thank you for your job:)
Status: Issue closed
|
kentcdodds/webpack-validator | 134197892 | Title: [idea] - Use property-validator
Question:
username_0: hey, @username_1. This project sounds like an awesome idea.
I recently built a library that helps with property validation in javascript objects. (https://github.com/username_0/property-validator)
Maybe we could use it validate the webpack config options?
Here's an idea of how it could look like:
```javascript
import { validate, assert, optional, presence, isArray } from 'property-validator'
const validationResult = validate(webpackConfig, [
presence('entry'),
optional(isArray('resolve.root')),
isArray('resolve.alias')
...
])
```
or use assert
this would raise a ValidationError (my preferred option)
```javascript
assert(webpackConfig, [ ... ])
```
validation functions are all stateless and are super easy to extend.
All you need to do is to create a function with this signature:
```javascript
function(paramName) {
return function(params) {
return { valid: true, message: 'bla' }
}
}
```
Answers:
username_0: Let me know if you think this makes sense.
Status: Issue closed
username_1: I actually have built my own version of this as well ([api-check](http://npm.im/api-check)) and I considered using [joi](https://github.com/hapijs/joi). And there's nothing stopping us from using these for this project. However, I don't think that it's necessary to require it for every validator. I like the API that I've created for defining validators and I don't see a reason to change it to this (still open to talk about changes).
So I think we'll stick with the library agnostic approach for now. You might consider writing a wrapper that converts your validator's output to this tool's format.
Thanks for the idea and the interest!
username_0: Sure! No problem.
This is a super interesting idea, so yeah. I should send some PRs and ideas your way soon. |
gobridge/workshops | 136128398 | Title: Wichita, KS - Date June 25th-26th 2016 - Course: TBD - Venue: TBD
Question:
username_0: Organizers: @username_0 and the [devICT](http://devict.org/) community.
Venue: TBD
Date: June 25th and 26th
Curriculum: TBD
## For organizer/s or sponsor
### Logistic items
- [ ] Has venue and it sits this many people: ___
- [ ] Has sponsor for food and drinks
- [x] Has organizer(s)
- [x] Has teacher(s) @username_0
- [x] Has date
- [ ] [Course selected](https://github.com/gobridge/workshops/blob/master/available_courses.md#course-descriptions-and-corresponding-levels)
- [ ] Draft of event started on [Bridge Troll](https://www.bridgetroll.org/events)
- [ ] Included in the event description if bikes are allowed inside or not (ask venue)
- [ ] Included in the event description if child care is available or not
(Issue ready :dart: to be labeled as `scheduled` :boom::v::tada::balloon::rocket: after all tasks above are completed:exclamation:)
### A couple weeks before the event
- [ ] Order placed for catering, if applicable
- [ ] Child care arranged for, if applicable
## For GoBridge
### First steps
- [ ] Self-assigned the issue
- [ ] Pinged the person who opened the issue
- [ ] Google Hangout scheduled (if this is a first timer)
Continue taking the actions necessary to bring issue to a `scheduled` status according to our [issue workflow](https://github.com/gobridge/workshops/blob/master/issue_workflow.md).
### After event is labeled as `scheduled`
- [ ] Bridge Troll event made public; added event link to this thread
- [ ] Opened a meetup event to point to the Bridge Troll event (RSVP on meetup should be closed)
- [ ] Added a PR to add event to GoBridge's website (remove past events while at it)
- [ ] First social media announcement's gone out
- [ ] Conducted organizer training
- [ ] Conducted teacher training
- [ ] Assignee added event date -1 week to calendar as reminder
## For lead organizer
### Presentations
- [ ] [Opening presentation](https://github.com/gobridge/workshop-material/blob/master/organizer-welcome.slide) forked, edited, and ready to present
- [ ] [Closing presentation](https://github.com/gobridge/workshop-material/blob/master/organizer-closing.slide) forked, edited, and ready to present
### Event date -1 week
- [ ] Confirmed venue and date
- [ ] More rounds of announcements on social media. Mention how many spots left
Answers:
username_0: I put out a survey within my local community to gauge workshop interest. Here were some highlights.
* There were 44 respondents.
* 31 stated they were very likely to attend (4 or 5 on a scale of 1-5).
* Of those 31 the average self-reported level of programming experience was 6.13 on a scale of 1-10.
* Of those same 31 respondents 3 indicated they would use childcare services if they were provided.
username_0: So this workshop is in June but in May I'm doing a 2 hour intro workshop for the Wichita chapter of Women Who Code. http://www.meetup.com/WWCWichita/events/229977852/
I'm going to focus on getting the environment set up and go through some basics.
Does anyone know where I could get some swag to hand out?
Status: Issue closed
username_0: Done :) |
grafana/grafana-image-renderer | 1084272643 | Title: CVE-2021-3918, CVE-2021-3807
Question:
username_0: <!-- Please only use this template for submitting enhancement requests -->
**What would you like to be added**:
- Upgrade package json-schema to version 0.4.0 or above.
- Upgrade package ansi-regex to version 5.0.1 or above.
**Why is this needed**:
Critical and high vulnerabilities during security (Aqua) scanning.
Answers:
username_1: Hello! Thank you for reporting this. Both are coming from `npm` that is included in our base Docker image but that we don't use. Closing this issue as a duplicate of https://github.com/grafana/grafana-image-renderer/issues/302 |
aws/amazon-ecs-agent | 139943252 | Title: Better reporting about OOMed containers.
Question:
username_0: Currently if container gets OOM killed due to hitting the memory limit the UI just says that the container exited. It's possible to detect that the container ran out of memory using `docker inspect`.
I think such situation should be shown in the events/task info.
Also in my case the container initially didn't get killed, but just got stuck not responding on ports. But it got killed as soon as I tried to start `bash` inside the container for debugging. Then by using `docker inspect` I was able to find out the actual reason of crash. Simple in retrospect, but took me couple days to find especially since only on some deployments the container hit the memory limit.
Answers:
username_1: Which version of Docker are you using? There have been some problems with [unreliability of the OOM status flag](https://github.com/docker/docker/issues/18510) (especially in Docker 1.9.x), which is what the Agent reads in order to report exit reasons.
username_0: I'm using the last ECS optimized image which has docker `1.9.1`. So the reporting actually checks that flag, but in my case docker just didn't set it properly?
username_1: Correct, we read the flag [here](https://github.com/aws/amazon-ecs-agent/blob/v1.8.1/agent/engine/docker_container_engine.go#L533-L535).
username_0: I'll close it since it's clearly an issue upstream. Thanks.
Status: Issue closed
|
jpuri/react-draft-wysiwyg | 256802147 | Title: How do i display the html code in react-draft-wysiwyg style?
Question:
username_0: There may be some conflict of style.
I want to display html with react-draft-wysiwyg component.
What is your input, what is you see. Wysiwyg.
Answers:
username_1: Hey @username_0 do you need to display html generated at some other place in editor while also allowing user to edit it ?
Can you make your use case more clear ?
username_2: @username_1 I need to do exactly what you described! How can I make that?
username_3: Hi, any progress about this question?
username_4: Hi, any progress about this question?
username_5: Hi, any progress about this question?
username_6: Please help me with this ?
username_7: You might want to check this:
https://username_1.github.io/react-draft-wysiwyg/#/docs
(scroll to the end of the page, to the "HTML" heading) |
alphagov/govuk-design-system | 420956713 | Title: Consider adding a layout example for header with service name.
Question:
username_0: I feel like the current example headers in the layout section do not cover the majority use case, which is GOV.UK together with a service name.
The current examples:
<img width="778" alt="Screenshot 2019-03-14 at 11 01 24" src="https://user-images.githubusercontent.com/2204224/54351938-943b1380-4648-11e9-8b5c-f3832747968b.png">
<img width="953" alt="Screenshot 2019-03-14 at 11 01 29" src="https://user-images.githubusercontent.com/2204224/54351943-969d6d80-4648-11e9-9781-e546dd1ba6cf.png">
From my experience assessing services, the most common example is:
<img width="872" alt="Screenshot 2019-03-14 at 11 02 31" src="https://user-images.githubusercontent.com/2204224/54352075-b2a10f00-4648-11e9-9428-cf2684d85180.png">
---
I'd like to add the above example because:
* We should have examples of the most common use case
* To use as a 'reference implementation' when services are testing their layout.
I feel like the example without a service name comes from a time when we used to recommend services avoid having a name. Is this still the case?
FWIW In practice I don't think many services are using a navigation style like the one in the example.
Answers:
username_1: We need (somewhere) to show everything that's possible.
If we were going to make any changes, it'd be good to understand what issues you're seeing and what you mean by a 'reference implementation'?
Thanks,
Ollie
username_2: Hi @username_0,
We believe that @username_1 has answered your question. As you have not posted anything further for over two weeks we consider this issue resolved and will close it.
If this is not the case, please feel free to re-open this issue.
Status: Issue closed
|
fimad/scalpel | 149394941 | Title: Scraper for inner HTML
Question:
username_0: Currently there's the `text` scraper that grabs the textual content of the selected node, and there's the `html` scraper that get's the whole selected node. I'd like a scraper that gives me only the inner HTML, without the selected tag itself.
Answers:
username_1: What would that return for something like ` <a> foo <b> bar </b> baz </a> `?
Would it be `foo <b> bar </b> baz` or `<b> bar </b>`?
username_0: The former.
username_1: This has been released in 0.3.1
Status: Issue closed
|
godotengine/godot | 946819702 | Title: Regression in the GDNative API generator: Creates invalid JSON for `AESContext::update()`
Question:
username_0: ### Godot version
3.4.beta1
### System information
Windows 10
### Issue description
```shell
.\godot.windows.tools.64.exe --gdnative-generate-json-api api.json
```
The above command generates a file `api.json` which can then be used in GDNative plugins that are currently working on a custom Godot build or unreleased Godot versions (like 3.x).
However, this command is currently generating an artefact which later on causes `godot-cpp` to fail to compile and be used in a plugin.
This is specifically the case with the JSON created for the `AESContext::start()` function as shown below.
```json
{
"name": "start",
"return_type": "enum.Error",
"is_editor": false,
"is_noscript": false,
"is_const": false,
"is_reverse": false,
"is_virtual": false,
"has_varargs": false,
"is_from_script": false,
"arguments": [
{
"name": "mode",
"type": "int",
"has_default_value": false,
"default_value": ""
},
{
"name": "key",
"type": "PoolByteArray",
"has_default_value": false,
"default_value": ""
},
{
"name": "iv",
"type": "PoolByteArray",
"has_default_value": true,
"default_value": "[PoolByteArray]" // <- Notice the square brackets
}
]
},
```
This then converts is converted into the code below in the godot-cpp API generator, which is not actually valid C++.
```cpp
Error start(const int64_t mode, const PoolByteArray key, const PoolByteArray iv = [PoolByteArray]);
```
### Steps to reproduce
1. Generate api.json
```shell
.\godot.windows.tools.64.exe --gdnative-generate-json-api api.json
```
2. Open api.json, search for `[PoolByteArray]`
3. You should find it in `AESContext > start`
### Minimal reproduction project
N/A
Answers:
username_1: I don't know much about the JSON API generator.
The code in `AESContext` is valid:
```
Error start(Mode p_mode, PoolByteArray p_key, PoolByteArray p_iv = PoolByteArray());
```
We could change it to be `Vector<uint8_t>()` (although they're not really the same in `3.x`).
username_1: Turns out it's an error in the `godot-cpp` binding generator: See godotengine/godot-cpp#590
username_0: @username_1 Correct me if I am wrong, if the problem was solely in godot-cpp, shouldn't we have not seen `[PoolByteArray]` in api.json in the first place?
The api.json has the square bracket symbols right after getting generated from the Godot binary and AFAIK it doesn't look like it belongs there.
username_1: The default argument is written by the parser as a `Variant.operator String()`. In fact, it generates `[]` also for other things, like string array, int array, ecc. The reason why it writes the name is that `Variant.operator String()` is not implemented for PoolByteArray. We could also implement that, but still wouldn't fix the issue in the godot-cpp bindings.
In fact, in some places you also get `[Object:null]` as default arguments in the generated json api.
Status: Issue closed
|
LuckSiege/PictureSelector | 635227153 | Title: 相册点击相机拍照出现问题
Question:
username_0: 
Answers:
username_0: java.lang.IllegalArgumentException: Failed to find configured root that contains /storage/emulated/0/DCIM/Camera/IMG_20200609_16365822image/png
at androidx.core.content.FileProvider$SimplePathStrategy.getUriForFile(FileProvider.java:744)
at androidx.core.content.FileProvider.getUriForFile(FileProvider.java:418)
at com.luck.picture.lib.tools.PictureFileUtils.parUri(PictureFileUtils.java:571)
at com.luck.picture.lib.PictureBaseActivity.startOpenCamera(PictureBaseActivity.java:934)
at com.luck.picture.lib.PictureSelectorActivity.startCamera(PictureSelectorActivity.java:690)
at com.luck.picture.lib.PictureSelectorActivity.onTakePhoto(PictureSelectorActivity.java:1274)
at com.luck.picture.lib.adapter.PictureImageGridAdapter.lambda$onBindViewHolder$0(PictureImageGridAdapter.java:153)
at com.luck.picture.lib.adapter.-$$Lambda$PictureImageGridAdapter$qvbHL3G2CY7hZL16_NieLEof5nI.onClick(Unknown Source:2)
at android.view.View.performClick(View.java:6637)
at android.view.View.performClickInternal(View.java:6614)
Status: Issue closed
|
dask/fastparquet | 406842556 | Title: Papermill import breaks fastparquet writer
Question:
username_0: Hi,
I am experiencing issues with writing data to parquet (with dask and fastparquet) when I import papermill in my notebook. The issue is the result of the combination of a couple of packages so I'll be cross posting this issue on the different repositories.
- fastparquet
- [papermill](https://github.com/nteract/papermill/issues/304)
- [dask-adlfs](https://github.com/eriklangenborg-rs/dask-adlfs) (unfortunately this repo has no issues)
Here is a code sample to reproduce the issue:
```
from random import randint
import papermill as pm
import pandas as pd
import dask.dataframe as dd
# a function to generate a sizeable dataframe
def get_large_df(cols, rows):
data = {}
for c in range(cols):
data['rand{}'.format(c)] = [randint(0, 1000000) for x in range(rows)]
return pd.DataFrame(data=data)
# Write the data frame to Azure Data Lake
df = get_large_df(10, 1_000)
ddf = dd.from_pandas(df, npartitions=4)
ddf.reset_index()
ddf.to_parquet('adl://my_container/test_parquet', engine='fastparquet', storage_options=STORAGE_OPTIONS)
# Read the data frame
dd.read_parquet('adl://prodstate/mh-webops/mh-webops/test/btan03/test_parquet', engine='fastparquet', storage_options=STORAGE_OPTIONS)
```
Reading the dataframe fails with the following exception:
```
Exception FileNotFoundError('test_parquet/_metadata/_metadata',)
```
It is unclear to me why dask looks for a `_metadata/_metadata` file because that does not exist. Now here is the funny thing:
- Simply removing the `import papermill as pm` resolves the issue,
- Writing to a local file instead of one on Azure works as expected,
- Writing and reading with `pyarrow` works (although I have to set `partition_on` when writing, which I didn't expect)
It is unclear to me which one of the packages is not playing nice.
Answers:
username_1: I'm sorry, I have simply no clue here! I'm not aware of any other similar issue.
username_0: Ok thanks. We'll find a way around the issue.
username_1: Since this was closed on the papermill tracker, I'll assume as they do that this was only for a very specific set of version/circumstances.
Status: Issue closed
|
karenlmasters/gz-hubbleseq | 257030507 | Title: Help finding references on the common use of.....
Question:
username_0: I'd like to give a good sense of how common various assumptions/lazy approximations about morphology are in the extragalactic community, so in your reading over the next few weeks could you remember to comment below with references (ideally copying the full \bibitem code, but the ADS URL will do) which:
1. Use B/T (or some proxy for it) to place spiral galaxies in a sequence
2. Use colour or spectral time to define a late/early split
3. Claim that colour and morphology are equivalent
4. Define ETG as anything without a spiral (or have a different definition)
5. Define late type as only late-type spirals (e.g. excluding Sas).
Thanks for the help.
Answers:
username_1: So I dug through my thesis to try and find some that fit the bill for each of these, some might not be quite what you were looking for though:
1) Brinchamann + (2004) use concentration to map to T-types
Shimasaku + (2001) classify their galaxies into T-types to place spirals in sequence
Ravindranath + (2004) use a cut Sersic index to identify Sbs-Sds
Kelvin + (2012) use Sersic index to split late and early types
Ball + (2007) use concentration, sersic index and T-types to put galaxies in sequence - (incidentally to show that colour does not equal morphology)
2) Strateva + (2001) was the first paper to statistically show the different colour-colour distributions of early and late-type galaxies - I'd expect papers that use colour to split these will all reference this paper
but that's all I've got for now - hope it helps!
P.S. Here[refs.txt](https://github.com/username_0/gz-hubbleseq/files/1330088/refs.txt) is a .txt file with the bibtex references in for you to copy in to your .bib file. The codes for them are the first author followed by two numbers denoting the year e.g. \cite{ball07} |
facebook/fresco | 735014539 | Title: version change in handle bitmap shaders
Question:
username_0: Hey! Unfortunately this is a limitation for bitmap shaders, they repeat edges. In your case, you can use the overlay rounding mode, see https://frescolib.org/docs/rounded-corners-and-circles.html#caveats for more information.
_Originally posted by @username_1 in https://github.com/facebook/fresco/issues/2546#issuecomment-710000373_
Thank you very much for your reply, because this problem occurred when I upgraded from version 1.9.0 to version 2.3.0. I want to know why, under the same conditions, the bitmap boundary was not drawn repeatedly in version 1.9.0.
Answers:
username_1: Hey! I don't believe we changed anything in version 1.9.0. This caveat has always been documented, even before we had Fresco version 1.0.0.
Did your image change or maybe aspect ratio etc?
username_0: Thank you very much for your reply. I'll write a demo and try again。
After the reappearance, I sent out the demo
username_0: The attachment is the corresponding test demo。
modify the config.gradle file with 1.9.0 to compile apk
1.install and run,the display is normal
2.click the top-right set menu and adjust the fontSize too max and return back(dialog choose sure)
3.the display is still normal
modify the version to 2.3.0 and do the above steps again，step1 is normal，step3 is error。
// facebook,fresco\stetho
'facebook-fresco' : 'com.facebook.fresco:fresco:2.3.0',
'facebook-fresco-imagepipeline-okhttp3' : 'com.facebook.fresco:imagepipeline-okhttp3:2.3.0',
'facebook-fresco-animated-base' : 'com.facebook.fresco:animated-base:2.3.0',
'facebook-fresco-animated-gif' : 'com.facebook.fresco:animated-gif:2.3.0',
'facebook-fresco-animated-webp' : 'com.facebook.fresco:animated-webp:2.3.0',
'facebook-fresco-webpsupport' : 'com.facebook.fresco:webpsupport:2.3.0',
// 'facebook-fresco' : 'com.facebook.fresco:fresco:1.9.0',
// 'facebook-fresco-imagepipeline-okhttp3' : 'com.facebook.fresco:imagepipeline-okhttp3:1.3.0',
// 'facebook-fresco-animated-base' : 'com.facebook.fresco:animated-base:1.9.0',
// 'facebook-fresco-animated-gif' : 'com.facebook.fresco:animated-gif:1.9.0',
// 'facebook-fresco-animated-webp' : 'com.facebook.fresco:animated-webp:1.9.0',
// 'facebook-fresco-webpsupport' : 'com.facebook.fresco:webpsupport:1.9.0',
[TestAll.zip](https://github.com/facebook/fresco/files/5514805/TestAll.zip)
username_0: The http address of the picture in demo above cannot be accessed, and the http address in this attachment should be OK
Special note: the placeholderImage has the same problem，so both demo can be used for testing。
[TestAll.zip](https://github.com/facebook/fresco/files/5514889/TestAll.zip)
username_0: Is there any progress?
looking forward to your reply. 😊
username_0: The question still exists. Do you need any other details？
username_2: 问题解决了? |
Azure/azure-cli | 562441286 | Title: Creating PHP WebApp Fails
Question:
username_0: ### **This is autogenerated. Please review and update as needed.**
## Describe the bug
**Command Name**
`az webapp create`
**Errors:**
```
update_site_config() takes 2 positional arguments but 3 were given
Traceback (most recent call last):
python3.6/site-packages/knack/cli.py, ln 206, in invoke
cmd_result = self.invocation.execute(args)
cli/core/commands/__init__.py, ln 608, in execute
raise ex
cli/core/commands/__init__.py, ln 666, in _run_jobs_serially
results.append(self._run_job(expanded_arg, cmd_copy))
cli/core/commands/__init__.py, ln 657, in _run_job
cmd_copy.exception_handler(ex)
...
cli/core/commands/__init__.py, ln 306, in __call__
return self.handler(*args, **kwargs)
azure/cli/core/__init__.py, ln 493, in default_command_handler
return op(**command_args)
cli/command_modules/appservice/custom.py, ln 135, in create_webapp
match['setter'](cmd, match, site_config)
TypeError: update_site_config() takes 2 positional arguments but 3 were given
```
## To Reproduce:
Steps to reproduce the behavior. Note that argument values have been redacted, as they may contain sensitive information.
- _Put any pre-requisite steps here..._
- `az webapp create --resource-group {} --name {} --plan {} --runtime {} --deployment-local-git`
## Expected Behavior
## Environment Summary
```
Linux-4.15.0-1067-azure-x86_64-with-debian-stretch-sid
Python 3.6.5
Shell: bash
azure-cli 2.0.81
```
## Additional Context
Possibly related to:
https://github.com/Azure/azure-cli/issues/12109
<!--Please don't remove this:-->
<!--auto-generated-->
Answers:
username_1: @username_0 , thanks for reporting, we are looking at it.
username_2: Same as #12159, please use the EdgeBuilds the fix will be in 2.0.82
Status: Issue closed
|
Adamor/FEWD3.28 | 234009805 | Title: Michelle's Final Project
Question:
username_0: https://github.com/username_0/FEWD0328_michelle/tree/master/final_project
this is the portfolio site to showcase my previous work and for job hunting.
i wish i have more time to figure out parallax and masonry, hopefully I will use them in my future update. a few things i'm still working on:
1. nav bar or dots navigation (like vertical carousel according to the page)
2. responsive layouts
Comfort: 3
Completeness: 3
Wins:
I find some of the javascript plugin very useful. I'd say i'm now pretty comfortable with HTML and CSS. I love styling in CSS, it feels great when I build my own design to work!
Challenges:
still need more practice on @mediaquery and responsive layout, what if on a more complicated structure site. do we style each page separately for different screen sizes?
Thanks for a great class! I enjoy so much learning all the cool stuff!
-Michelle |
cdnjs/cdnjs | 205003044 | Title: [Request] Add request-promise
Question:
username_0: **Library name:** request-promise
**Git repository url:**
**npm package url(optional):**
**License(s):**
**Official homepage:**
**Wanna say something? Leave message here:**
=====================
Notes from cdnjs maintainer:
You are welcome to add a library via sending pull request,
it'll be faster then just opening a request issue,
and please don't forget to read the guidelines for contributing, thanks!!<issue_closed>
Status: Issue closed |
OutSystems/WebView | 726186626 | Title: Link on package in NuGet
Answers:
username_1: Can you provide more details?
username_2: I think he is looking for a link to a nuget package of this project in the Readme. That would make it easier to start. I just searched for the same.
Status: Issue closed
username_1: Here:
https://www.nuget.org/packages/WebViewControl-Avalonia/
and
https://www.nuget.org/packages/WebViewControl-WPF/
username_2: Hi @username_1
Thank you very much for providing Nuget packages!
Currently the build requirements are not linked in the package. I tried to solve it manually, but didn't succeed yet. You are using https://gitlab.com/username_1/cefglue for cefglue. This is a fork of https://gitlab.com/xiliumhq/chromiumembedded/cefglue. As far as I see only the original cefglue has nuget package. Do you have one for your fork too? Do I need any other requirements?
Does the current version of Webview require the 0.10-preview like the nuget package expects or could I also use the latest stable version?
Thanks and regards!
username_1: Hi @username_2
Yes I'm using my (https://gitlab.com/username_1/cefglue) cefglue fork which supports Avalonia.
You can find the cefglue packages on nuget too: https://www.nuget.org/packages/CefGlue.Avalonia/
Regarding Avalonia version used, you might have problems (probably building) using the stable one and so I recommend using the latest preview (if possible).
Feel free to submit a PR with updated build information if you find something missing or not clear.
username_2: Hi @username_1
Thanks for the quick help.
I've installed the second package too.
```
<PackageReference Include="WebViewControl-Avalonia" Version="2.75.66" />
<PackageReference Include="CefGlue.Avalonia" Version="75.1.87" />
```
After adding a basic webview to my app (copied from the example):
```
<webview:WebView Address="{Binding CurrentAddress}" />
```
following exception occurs:
```
Unhandled exception. System.Reflection.TargetInvocationException: Exception has been thrown by the target of an invocation.
---> System.IO.FileNotFoundException: Could not load file or assembly 'Xilium.CefGlue.Avalonia, Version=75.1.87.0, Culture=neutral, PublicKeyToken=null'. The system cannot find the file specified.
File name: 'Xilium.CefGlue.Avalonia, Version=75.1.87.0, Culture=neutral, PublicKeyToken=null'
---> System.IO.FileNotFoundException: Unable to locate
File name: 'Xilium.CefGlue.Avalonia.dll'
```
Do you have any ideas about this? Looks like it cannot find the CefGlue package I have installed.
When it works I'll add a PR with updated Getting started instructions.
username_1: It looks like the packages weren't properly installed and the binaries copied to the output... can you confirm?
username_2: The output folder looks like this:
Debug/
- netcoreapp3.1/
-- Avalonia.*.dll
-- WebViewControl.Avalonia.dll
-- x64/
--- Xilium.CefGlue.*.dll
--- Xilium.CefGlue.ProwserProcess.exe
So all Avalonia DLLs, the WebView DLL and CefGlue DLLs are there, but XefGlue is in a x64 subfolder unlike the Avalonia DLLs.
Is this supposed to run only under Windows and not cross platform? I am wondering because I see this BrowserProcess.exe under macOS.
username_1: It runs on both platforms but on macOS it requires proper packaging due to the way CEF works.
Check this script... I use in CefGlue:
https://gitlab.com/username_1/cefglue/-/blob/3770/CefGlue.Demo.Avalonia/bundle.sh
username_1: Check https://github.com/OutSystems/WebView/tree/macos-bundle
Status: Issue closed
username_1: MacOS bundling changed and was simpified. |
cougargrades/api | 689360401 | Title: Add Github Actions continuous deployment
Question:
username_0: Probably based on:
https://github.com/cougargrades/web/blob/afc56a8697327e46d07749cab05a528727ec93d7/.github/workflows/firebase.yml
Answers:
username_0: Working implementation in eb9206f924d24fad157627ea68e7da6a02ab385f
Status: Issue closed
|
MarimerLLC/csla | 412236203 | Title: Type is not resolved for member Csla.Security.UnauthenticatedPrincipal
Question:
username_0: **Describe the bug**
I got an exception when running unit test, but it works fine when running the application, I think it's because the ApplicationContext.User is different, but could not figure out how to resolve it
The logic first makes sure user is logged out as below
`Csla.ApplicationContext.User = new UnauthenticatedPrincipal();`
Then it logs them in again, how the exception below occurs on initialising the connection inside DataPortal_Fetch from a SystemConfig class that extends ReadOnlyListBase
`using (SqlConnection cn = new SqlConnection(connectionString))`
`Exception message: Csla.DataPortalException: DataPortal.Fetch failed (DataPortal.Fetch failed (The type initializer for 'System.Data.SqlClient.SqlConnection' threw an exception.)) ---> Csla.Reflection.CallMethodException: CustomIdentity.DataPortal_Fetch method call failed ---> Csla.DataPortalException: DataPortal.Fetch failed (The type initializer for 'System.Data.SqlClient.SqlConnection' threw an exception.) ---> Csla.Reflection.CallMethodException: SystemConfig.DataPortal_Fetch method call failed ---> System.TypeInitializationException: The type initializer for 'System.Data.SqlClient.SqlConnection' threw an exception. ---> System.Runtime.Serialization.SerializationException: Type is not resolved for member 'Csla.Security.UnauthenticatedPrincipal,Csla, Version=4.3.14.0`
we use Microsoft.VisualStudio.TestTools.UnitTesting for unit testing
When I run the application and debug the same logic above,
Csla.ApplicationContext.User is also UnauthenticatedPrincipal when the connection is initialised, but does not cause the exception
**Version and Platform**
CSLA version: 4.3.14
OS: Windows
Platform: WinForms
Answers:
username_1: You are encountering a known issue with the way unit test frameworks use threads.
The CSLA `ApplicationContext` relies on a provider model, so each runtime environment can manage things like the user identity in a way that is correct for the runtime.
For example, ASP.NET relies on HttpContext, while some smart client environments use a static field and others use System.Threading.Thread.CurrentPrincipal.
The default behavior was in CSLA 4.3 relies on the current thread's CurrentPrincipal. That never works in unit testing environments because the unit testing host (mstest) "reuses" a thread between the host and each individual test.
The solution is to implement a cleanup method in each test class (this is an mstest concept), and in that method make sure to set `ApplicationContext.User` to null. This way when the current thread returns from the test back to the mstest host .NET doesn't attempt to serialize the current principal between environments.
username_1: For future reference I've documented this in the /docs folder.
https://github.com/MarimerLLC/csla/blob/master/docs/unit-testing.md
Status: Issue closed
|
amit-mittal/TFTP | 224878632 | Title: The OP code is currently a string and should be binary numbers. Also the size of RRQ and WRQ looks wrong.
Question:
username_0: RRQ and WRQ should have 4+file_name_size + mode_name_size bytes. Because first 2 are for OP code and the other 2 are zero bytes between the file name and after the mode name.
Answers:
username_1: Likewise, strings containing \000 bytes cannot be used because the string terminates at this point, and in addition when using malloc() the size to be allocated needs one added when any of the str... functions are used due to 0 termination. |
BD2KGenomics/dcc-ops | 225573231 | Title: Persistance of redwood databases on VM shutdown and boot up
Question:
username_0: Right now, if the system is shut down, the containers for the metadata and auth databases get destroyed, along with any data in it i believe (when the VM and redwood are brought up again, the registered tokens and metadata get cleared and so the containers are empty). This makes it hard to test if we want to shut down the VM at the end of the day in order to avoid costs. A way we could easily fix this is by mounting the data folder from each db defined in the `yml`file on the system by using the `volumes` tag and letting docker-compose mount the data in the local driver.
See https://github.com/BD2KGenomics/dcc-ops/blob/develop/boardwalk/prod.yml
and https://code.wiki/page/rkfuXEOq/Persist-data-in-MongoDB-with-Docker-Compose
I assume these changes will involve editing not only the `yml` but potentially the redwood cli. I think if it only does a `docker-compose down`, then we can just add the `-v` flag at the end to take down all the defined volumes in the `yml` with the upper level tag
Answers:
username_1: You didn't call `redwood down`, but data was lost on restart?
username_1: I just tested and data persisted through restart just fine.
Status: Issue closed
username_1: This shouldn't be an issue anymore. @username_0, you can reopen this if it appears again |
vueComponent/ant-design-vue | 425891423 | Title: Tree 组件
Question:
username_0: - [ ] I have searched the [issues](https://github.com/vueComponent/ant-design-vue/issues) of this repository and believe that this is not a duplicate.
### Version
1.3.7
### Environment
macOs 10.14.3 webpack vue2.6.6
### Reproduction link
Permission denied
### Steps to reproduce
tree选择组件
两个父节点，拥有相同key的子节点
1.两个父节点同时全选
2.其中一个父节点取消全选
### What is expected?
视图正常渲染，并且解决子节点显示错误后无法点击的问题
### What is actually happening?
此时选中的key是正常的，但是节点显示错误，重复key的子节点没有取消选中，并且无法进行点击选中、取消选中
<!-- generated by issue-helper. DO NOT REMOVE -->
Answers:
username_1: 确保节点全局唯一，不分层次 |
jlippold/tweakCompatible | 430093177 | Title: `Essentials Cracked` working on iOS 12.1.2
Question:
username_0: ```
{
"packageId": "com.pulandres.essentials",
"action": "working",
"userInfo": {
"arch32": false,
"packageId": "com.pulandres.essentials",
"deviceId": "iPhone10,6",
"url": "http://cydia.saurik.com/package/com.pulandres.essentials/",
"iOSVersion": "12.1.2",
"packageVersionIndexed": false,
"packageName": "Essentials Cracked",
"category": "Tweaks",
"repository": "Pulandres Repository",
"name": "Essentials Cracked",
"installed": "1.1.2-1",
"packageIndexed": false,
"packageStatusExplaination": "This tweak has not been reviewed. Please submit a review if you choose to install.",
"id": "com.pulandres.essentials",
"commercial": false,
"packageInstalled": true,
"tweakCompatVersion": "0.1.5",
"shortDescription": "Complete settings for your device",
"latest": "1.1.2-1",
"author": "<NAME>",
"packageStatus": "Unknown"
},
"base64": "<KEY>
"chosenStatus": "working",
"notes": ""
}
```<issue_closed>
Status: Issue closed |
DenisB405/Sismec-Stefanini | 584487647 | Title: Front-End
Question:
username_0: Tela cadastro de Candidato
- [ ] Criar tela cadastro de candidato
- [ ] Criar formulário
- [ ] Validar formulário
Tela Instituição
- [ ] Criar listagem de Instituição
- [ ] Incluir botão cadastrar - excluir atualizar
- [ ] Implementar pesquisa<issue_closed>
Status: Issue closed |
firecracker-microvm/firecracker-containerd | 386531277 | Title: Determine how agent and runc should be embedded into the microvm
Question:
username_0: firecracker-microvm depends on three components to be inside the microvm: the agent (communicating across the `vsock`s for control and stdio), containerd's runc shim (supervising runc), and runc (running the requested container). We need to determine how we want those components to be made available inside the microvm: embedded in the microvm-guest image, added as a separate device, or something else.
Status: Issue closed
Answers:
username_1: Automation of image creation from #153 covers this. Closing. |
fsprojects/FSharpLint | 182247462 | Title: Wrong suggestion to replace property assigning in constructor with property name
Question:
username_0: 
Answers:
username_1: Thank you for reporting this
I'm currently unable to repeat:
Are there any errors in the file when this bug appears? I'm able to get it to occur only when the type checker is unable to determine whether or not they are properties (if I remove `open System.Diagnostics`)
Status: Issue closed
|
jiejieTop/mqttclient | 792966296 | Title: 关于mqtt连接失败的错误提示
Question:
username_0: 请问如何在连接失败时获取connack_return_codes[失败原因](https://github.com/jiejieTop/mqttclient/blob/1b81bc2bac566ed61af0334a500bbe782bdccff0/mqtt/MQTTConnect.h#L29)？
int rc = mqtt_connect(client);
while (rc != CLIENT_STATE_CONNECTED ) {
slog_error("mqtt connect %s:%s failed", HEWU_SERVER, HEWU_PORT);
sleep(5);
rc = mqtt_connect(client);
}<issue_closed>
Status: Issue closed |
gobuffalo/buffalo | 243553208 | Title: Buffalo command shows duplicate plugins
Question:
username_0: ```
help Help about any command
heroku [PLUGIN] Tools for deploying Buffalo to Heroku
heroku [PLUGIN] Tools for deploying Buffalo to Heroku
```
Assuming that a plugin is installed in GOPATH and the plugins directory.<issue_closed>
Status: Issue closed |
11ty/eleventy | 351748986 | Title: Access .eleventy.js data in templates
Question:
username_0: Is there a way to access .eleventy.js data in template files? I would like to eg. add a `data-subdir="{{ eleventyConfig.pathPrefix }}"` attribute to the DOM to read with JavaScript and prepend it to paths (eg. for AJAX calls).
I did some search and tried config, eleventyConfig, eleventy but nothing seemed to work. It would be useful as it would help avoiding duplicate declarations (and possible errors).
Answers:
username_1: Duplicate of #184.
Status: Issue closed
username_0: Thanks, closing this then.
username_2: Thanks @username_1! |
robertvazan/sourceafis-java | 556396381 | Title: Question: Is there a way to create a template for matching from multiple input images?
Question:
username_0: ... and would that even make a difference? I have often seen in commercial packages a way of combining templates from multiple images (of the same finger) when "enrolling" a fingerprint, in order to create an enhanced template with good data, for matching. Would that be possible with SourceAFIS? Or even desirable?
Answers:
username_1: This is called template fusion. It's one of the features considered for future development:
https://sourceafis.machinezoo.com/custom#fusion
I will keep this issue open as a reminder about this feature request. |
WarEmu/WarBugs | 333304213 | Title: The "Paranoid Raven Games" Quest's Last Stage Is Duplicated.
Question:
username_0: *Issues should be unique. Check if someone else reported the issue first, and please don't report duplicates.*
Expected behavior and actual behavior:
This last stage of the a quest-line for Destruction in Chaos Wastes that starts in their warcamp, then takes you through the CW RvR lake to two NPCs, before returning to the warcamp. The last stage is duplicated and can be both accepted and returned twice, getting ~7k XP and 1.5 gold for each.
Steps to reproduce the problem:
If you can auto-advance the quest, just go to Rugger Skullfork in the warcamp and he'll give you two of quests which can turn in to Onnister Preul just a few feet to the right under the purple tent. Otherwise, the quest starts at Onnister Preul, taking you to Daria Ravenblood outside Order warcamp, then Rhece Farseeker southwest of Statue of the Everchosen, and back to Rugger Skullfork in the warcamp.
Screenshots/Videos or archive.org evidences:
Status: Issue closed
Answers:
username_1: fixed with patch from 27.07.2018 |
kubeflow/pipelines | 705173436 | Title: Kubeflow Pipelines on GCP can't launch due to pods issues.
Question:
username_0: I created multiple clusters on GCP using the default settings through the AI platform. When I open the pipelines tab in AI Platform, I get the error : "Error occured while trying to proxy to: ...".
I am getting this error on multiple pipelines on multiple clusters. None of the sample tutorials or templates work due to the same error.
1) Is this a kubeflow issue or a GCP issue?
2) I'm not too savvy on Kubernetes. I've tried creating clusters using different approaches on GCP. I don't think I'm doing anything wrong because I've been following a bunch of different tutorials and documentation etc etc.. they all lead to these 2 pods issues: ml-pipeline-65479485c8-mcj9x and ml-pipeline-persistenceagent-67db8b8964-mlbmv. I made a S/O post here: https://stackoverflow.com/questions/63946871/gcp-ai-platform-pipelines-clusters-does-not-have-minimum-availability
Answers:
username_1: This is really strange.
Did you install using the Google Cloud Marketplace UX?
Did you create a new Kubernetes cluster in the process?
username_0: I did the following:
On AI Platform -> Pipelines -> New Instance
This takes me to a Kubeflow pipelines config page. Here I am able to create a new cluster or select an existing one. I can create it by selecting a zone and clicking the "Create cluster" button (Here is where I think it creates it on the Google Cloud Marketplace). Then I select the checkbox "Use managed storage" and click "Deploy".
Even if I create a cluster via the Kubernetes Engine, when I link the Kubeflow pipeline to it, the same problems with the 2 pods exist - "Does not have minimum availability".
Am I missing something obvious?
Below is the full pod description for one of the pods:
```
Name: ml-pipeline-65479485c8-mcj9x
Namespace: default
Priority: 0
Node: gke-cluster-3-default-pool-007784cb-qcsn/10.150.0.2
Start Time: Thu, 17 Sep 2020 22:15:19 +0000
Labels: app=ml-pipeline
app.kubernetes.io/name=kubeflow-pipelines-3
pod-template-hash=65479485c8
Annotations: kubernetes.io/limit-ranger: LimitRanger plugin set: cpu request for container ml-pipeline-api-server
Status: Running
IP: 10.4.0.8
IPs:
IP: 10.4.0.8
Controlled By: ReplicaSet/ml-pipeline-65479485c8
Containers:
ml-pipeline-api-server:
Container ID: ...
Image: ...
Image ID: ...
Ports: 8888/TCP, 8887/TCP
Host Ports: 0/TCP, 0/TCP
State: Running
Started: Fri, 18 Sep 2020 10:27:31 +0000
Last State: Terminated
Reason: Error
Exit Code: 255
Started: Fri, 18 Sep 2020 10:20:38 +0000
Finished: Fri, 18 Sep 2020 10:27:31 +0000
Ready: False
Restart Count: 98
Requests:
cpu: 100m
Liveness: exec [wget -q -S -O - http://localhost:8888/apis/v1beta1/healthz] delay=3s timeout=2s period=5s #success=1 #failure=3
Readiness: exec [wget -q -S -O - http://localhost:8888/apis/v1beta1/healthz] delay=3s timeout=2s period=5s #success=1 #failure=3
Environment:
HAS_DEFAULT_BUCKET: true
BUCKET_NAME:
PROJECT_ID: <set to the key 'project_id' of config map 'gcp-default-config'> Optional: false
POD_NAMESPACE: default (v1:metadata.namespace)
DEFAULTPIPELINERUNNERSERVICEACCOUNT: pipeline-runner
OBJECTSTORECONFIG_SECURE: false
OBJECTSTORECONFIG_BUCKETNAME:
DBCONFIG_DBNAME: kubeflow_pipelines_3_pipeline
DBCONFIG_USER: <set to the key 'username' in secret 'mysql-credential'> Optional: false
DBCONFIG_PASSWORD: <set to the key 'password' in secret 'mysql-credential'> Optional: false
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from ml-pipeline-token-77xl8 (ro)
Conditions:
[Truncated]
Type: Secret (a volume populated by a Secret)
SecretName: ml-pipeline-token-77xl8
Optional: false
QoS Class: Burstable
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning BackOff 52m (x409 over 11h) kubelet, gke-cluster-3-default-pool-007784cb-qcsn Back-off restarting failed container
Warning Unhealthy 31m (x94 over 12h) kubelet, gke-cluster-3-default-pool-007784cb-qcsn Readiness probe failed:
Warning Unhealthy 31m (x29 over 10h) kubelet, gke-cluster-3-default-pool-007784cb-qcsn (combined from similar events): Readiness probe failed: c
annot exec in a stopped state: unknown
Warning Unhealthy 17m (x95 over 12h) kubelet, gke-cluster-3-default-pool-007784cb-qcsn Liveness probe failed:
Normal Pulled 7m26s (x97 over 12h) kubelet, gke-cluster-3-default-pool-007784cb-qcsn Container image "gcr.io/cloud-marketplace/google-cloud-ai
-platform/kubeflow-pipelines/apiserver:1.0.0" already present on machine
Warning Unhealthy 75s (x78 over 12h) kubelet, gke-cluster-3-default-pool-007784cb-qcsn Liveness probe errored: rpc error: code = DeadlineExceede
d desc = context deadline exceeded
```
username_1: This is the likely issue.
I've had this problem before: https://github.com/kubeflow/pipelines/issues/3798 https://github.com/kubeflow/pipelines/issues/3795 https://github.com/kubeflow/pipelines/issues/3797
As a workaround, uncheck that box.
/cc @username_2
I think the current Marketplace UX could be more intuitive and easier to use out-of-the-box.
username_2: Yes, when you check "use managed storage", you must setup your managed storage beforehand and fill in the fields below to connect your KFP instance with your managed storages.
Status: Issue closed
username_0: Thank you so much! Starting fresh and doing everything without "using managed storage" fixed the pods issued. My pipelines are up and running successfully!
Even when "Using managed storage" I felt like I messed up the setup for managed storage somewhere because it really isn't clear for ML Ops noobs. |
ixrjog/opscloud4 | 1085664776 | Title: 关于k8s的pod容器终端无法登录问题
Question:
username_0: 登录容器终端的时候无法操作,感觉是没登录成功,而且后台没报错,有一些日志如下
新事件 Event success! message: SimpleEvent(eventType=USER_PERMISSION, action=CREATE, body=UserPermission(id=29047, userId=1, businessId=81, businessType=8, permissionRole=null, rate=null, content=null, createTime=null, updateTime=null))
2021-12-21 18:04:12.099 INFO 2924 --- [Async-Service-5] c.b.o.event.listener.NoticeListener : 监听事件 : eventType = USER_PERMISSION , action = CREATE
我用的4.0.4后端的代码,后端也没报错所以不好排查,大佬能帮我看看怎么解决吗
还有就是应用配置里面的应用key是填什么的,是不是因为这里的原因
Answers:
username_1: 请检查当前容器镜像是否支持终端登录，点击web中log看下是否有日志滚动
username_0: 怎么查看当前容器镜像是否支持终端登录,是我部署的那个应用的镜像是吗
username_0: 或者是有什么办法能设置我的容器镜像能支持终端登录,有文档吗,而且我看了我页面中Terminal和Log点击操作请求的路径都是/api/ws/kubernetes/terminal
username_1: 日志能出来吗，如果你用的是阿里云ACK，控制台可以登录容器终端
username_0: 浏览器控制台的报错信息
阿里云ack我也接入过,在阿里云的控制台是可以登录终端,但是接入到ops系统也是无法登录,是不是我现在的系统版本这个功能还没支持,或者是我设置上哪里还有问题
username_1: nginx配置需要支持websocket协议
username_0: 我是本地分别部署的前端和后端,还没用ng进行代理的,而且也是本地浏览器直接请求的,是不是要将前端后端打包在一起部署起来然后用ng进行代理才能使用终端登录的功能
username_1: 这种情况不需要配置，2种部署方式都支持web终端，你可以试下ssh-server是否能登录容器，oc4使用sh命令启动容器shell请确认是否支持sh命令
username_0: 
远程服务器是可以登录的,就是pod终端无法登录,说明是我pod容器的问题了是吧
username_1: oc4登录容器只需要与k8s-master通信即可，目前看可能是你的容器不支持sh，你可以找个centos容器镜像测试下
username_0: 感谢大佬的耐心回复,问题已经解决了哈,谢谢
username_1: 是因为容器镜像不支持`sh`命令？ |
phantomcyber/phantom-apps | 803488511 | Title: Feature: Add host and certificate information lookup actions to Passivetotal app
Question:
username_0: **Name of the app**
phpassivetotal
**Changes**
Adding 4 new actions
* lookup certificate
* lookup certificate hash
* get host pairs
* get host components
These actions retrieve information about domains / ips / certificate hashes from the [PassiveTotal API](https://api.passivetotal.org/). PR will be opened soon.
<img width="946" alt="Screenshot 2021-02-08 at 12 46 47" src="https://user-images.githubusercontent.com/2430239/107215527-b6dc2f80-6a0b-11eb-8362-d1eef4d09b71.png">
Answers:
username_1: This issue has been addressed as part of this PR(https://github.com/phantomcyber/phantom-apps/pull/478). Hence, closing the issue.
Status: Issue closed
|
dadafly1244/2021_SWTeam | 850800304 | Title: 09차 회의록_2021.03.21
Question:
username_0: # 09차 회의록
### 2021.03.21 (일) 20:00 ~ 20:30 ( 210분 ) / 비대면
참가자 ( 총 4인 )
: 양다영(팀장), 김수진, 송나민, 최예린
## 주요 안건
- 블록 다이어그램 공유/수정
- 세부 기능 조정
## 회의 내용
#### 1. 블록 다이어그램 공유/수정
- 최예린 플로우차트 베이스 + 팀원 내용 보충
- 피드백 : 슬랙 제출 / 2021. 03. 23. (화)
#### 2. 세부 기능 조정
1. ##### 관리자 지정 방식
1. 의견
- 1안 : 회원가입 시 등급 지정 ( 기본 관리자 : 단체 )
- 2안 : IoT 기기마다 인증코드 발급 > 타 유저 신청 ( 기본 관리자 : 1명 )
- 채택 : 2안
2. 커뮤니티 추가
- 자신이 관리하지 않는 쉼터 정보를 알 수 없음 ( 캣맘 대상 범죄 예방 )
- 구인글 작성 : 쉼터 관리자를 구하는 글
- 지원글 작성 : 쉼터 관리자를 지원 신청하는 글
- 인증 코드 배포
3. 단체 연계
- 기업/단체 연계 시 인증코드 단체 배포
2. ##### 하이브리드 어플
- 방식 변경 : 웹 앱 > 하이브리드
- 시간이 없을 경우 Android 대상 제작
[Truncated]
- 현재 상황 : 작성 완료
- 개별 검토
- 제출 : 2021. 03. 23. (화) / 양다영
#### 2. 수행 과제
- 세부 사항 조사/정리 : input/output/data 처리
#### 3. 다음 회의
1. 팀내 회의 : 2021. 03. 23. (화) 14:00
- 안건 : 세부사항 정리/자료화
2. 멘토 회의 : 2021. 03. 24. (수) 19:00
- 미리 일정 확인 필요 |
flickerleap/phpcs | 352498247 | Title: Add env as a forbidden function
Question:
username_0: [Laravel config cache](https://laravel.com/docs/5.6/configuration#configuration-caching) is great, but when it's enabled `env` values are all null. We tend not to enable it on our dev environment, but on production. The use of the env() helper could easily slip into our code. A sniffer for it would be a good idea then.
Answers:
username_1: ```xml
<!-- Ban some functions -->
<rule ref="Generic.PHP.ForbiddenFunctions">
<properties>
<property name="forbiddenFunctions" type="array">
<!-- https://github.com/flickerleap/phpcs/issues/1#issue-352498247 -->
<!-- DO NOT USE env() or getenv() because it will break laravel env/config caching -->
<element key="env" value="config"/>
<element key="getenv" value="config"/>
</property>
</properties>
</rule>
<rule ref="Generic.PHP.ForbiddenFunctions">
<exclude-pattern>config/*.php</exclude-pattern>
</rule>
``` |
emberjs/ember.js | 39151453 | Title: [REGRESSION 1.6.x] {{#with}} + controller + {{else}}
Question:
username_0: There seems to be a regression in 1.6.x when using a falsy value in `{{#with}}` with a controller:
```handlebars
{{#with somethingFalsy controller="someName"}}
Showing main block.
{{else}}
Showing else block.
{{/with}}
```
In 1.5.x, this [shows the else block](http://emberjs.jsbin.com/gehis/2/edit), in 1.6.x, this [shows the main block](http://emberjs.jsbin.com/zapil/1/edit).<issue_closed>
Status: Issue closed |
wbhob/civic-social-app | 179667971 | Title: Geocoding service confirmation + API token
Question:
username_0: For initial proof of concept for geocoding, I used Mapbox's geocoding API + my own API token. We should confirm whether this is the best one to use, as opposed to other options like Google. It appeared to me that Mapbox's rate limits were better than Google's. Also, need to use a different API token than mine as I already use it for other projects. |
GoogleChrome/web.dev | 430522065 | Title: Redirect all the things!
Question:
username_0: This will be tricky but we need to redirect all articles and codelabs to use the new flat URL space.
We'll need to come up with a plan for how this affects breadcrumbs.
Answers:
username_0: Proposal:
We create two tags: `post` and `collectionItem`.
Anything that starts its life in the blog will have the tag `post`.
If an element only has the `post` tag then...
- the breadcrumbs will lead back to the blog.
- the next/prev buttons at the bottom will not display, instead we'll show cards for related content.
If an element gets upgraded to a collection, or if it starts its life in a collection and never appears in the blog, then it will have the tag `collectionItem`.
If an element **only** has the `collectionItem` tag then...
- the breadcrumbs will lead back to the collection.
- the next/prev buttons at the bottom will display, and take the user back to the collection landing page or to the next `collectionItem` in the collection.
If an element has **BOTH** the `collectionItem` and `post` tags then...
- the breadcrumbs will offer to take the user back to the blog OR recommend they keep reading in the collection.
- the next/prev buttons at the bottom will display, let the user know this post is part of a collection, and take the user to the next `collectionItem` in the collection.
- cards for related content will also be displayed.
Status: Issue closed
username_0: Soooo we did this. But the bottom breadcrumbs remain a point of discussion. Going to close this issue because we're already working on the breadcrumb stuff with Chrome UX folks. |
miguelgrinberg/Flask-SocketIO | 602812312 | Title: Maximum room support
Question:
username_0: I want to implement Flask SocketIO on one of my projects, which provides a dashboard to customers upon login. From what I see, Flask SocketIO would work well in pushing frequent data from the server to the clients for this purpose.
I've also noticed that some people suggest you maintain an internal array and add new clients to this array when they register/connect to the websocket. You can then iterate through this array and send data to each client's dashboard according to predefined criteria (which would typically be some JSON object).
Can I use rooms for this purpose? I think the solution might be similar to using the `request.sid` as mentioned as the solution in [https://stackoverflow.com/questions/39423646/flask-socketio-emit-to-specific-user](url). Essentially, I'll be using each room as a messaging mechanism to push data from the server to the client, where each room (client) is isolated from the rest.
Also, is there a limit to the amount of rooms I can utilise, assuming that one client will use one room?
Hope my question is clear. Any advice would be appreciated.
Answers:
username_1: Sure, you can use rooms. There is no limit, and in fact, each client gets a room of its own by default when they connect. The room name is the same as `request.sid`.
Status: Issue closed
|
zutrinken/bleak | 656652755 | Title: lightbox
Question:
username_0: Hello
Is there any way to add a lightbox to this them? I tried 3-4 different lightboxes,
none of them works.
Answers:
username_1: The issue is, that light boxes usually expect to have the image wrapped by a link to the main image. So if you integrate one, you should handle the extra link in addition for each image via JS, since Ghost doesn't do this at the moment. Nothing theme wise.
Status: Issue closed
username_0: No problem about linking, I do that with a js, problem was reloading the js in the theme.
username_0: Thanks
username_1: Ah okay, than you have to add your function to the [reload function](https://github.com/username_1/bleak/blob/master/src/js/script.js#L148-L155). |
ValveSoftware/steam-for-linux | 305066701 | Title: 64bit ONLY client pls
Question:
username_0: #### Your system information
* Steam client version (build number or date): none at moment
* Distribution (e.g. Ubuntu): Debian 9.4
* Opted into Steam client beta?: [Yes/No]
* Have you checked for system updates?: [Yes/No]
#### Please describe your issue in as much detail as possible:
I have read #5324 and #179, but this issue is different...
I understand that there are many 32bit games you want to keep supporting BUT:
There are people like me who want to stay pure 64bit.
So please provide a 64bit ONLY steam, with a filter that will only show 64bit games in it's listing.
I refuse to install 32bit crap in my 64bit system thank you...
Answers:
username_1: Hello @username_0, a feature request for a purely 64 bit client is already being tracked at #3518.
As for duplicates, those are inevitable because all of the search terms for this are very common and overlap with basic information requested for troubleshooting.
Closing in favor of the existing feature request.
Status: Issue closed
username_0: I'm not using Centos, but **Debian**, if you make it work on debian it will work on **many** other distributions that are derived from debian....
Make this thread a tracker for DEBIAN...
username_2: The Valve has announced "does not support 32bit platfrom" around 2 years ago, but the Steam framwork runs only with 32bit components?! WTF? If you does not support 32bit, I accept, but you dont ask me, we support your 32bit client.
The Steam (32bit) support is paused on blackPanther OS for native 64bit Steam client releasing by Valve. I advise for any distro developers too.
Existing any other prider as well, like GOG, etc.
username_0: https://github.com/ValveSoftware/steam-for-linux/issues/5342#issuecomment-358160158
Thats said please provide 2 distributions then, one that is able to run **32-Bit games ONLY** and one that is able to run **64-Bit games ONLY**.
What would you think people would say if you added 64-bit support, and force them install needed 64-bit stuff, in a 32-bit client because it needs to be able to run both type of games? o.O
Yes im still kicking the ones in charge untill people are not forced to multi-architecture... |
umbraco/Umbraco-CMS | 595601647 | Title: Log Viewer in backoffice breaks when changing log location
Question:
username_0: ### Bug summary
When changing the location of the logs in serilog.config, the Log Viewer in the backoffice does not work anymore.
### Steps to reproduce
* Start with clean Umbraco 8 site
* Change `serilog:write-to:File.path` in config/serilog.config to something else than the default
* Make sure no old logs file remain (i.e. delete App_Data/Logs/)
* Open Log Viewer in the backoffice
### Expected result
Log Viewer works
### Actual result
Log Viewer doesn't work: No logs are displayed and the following error is shown: "Failed to retrieve state if logs can be viewed".
In the logs the following error is logged: "System.IO.DirectoryNotFoundException: Could not find a part of the path 'D:\home\site\wwwroot\App_Data\Logs'" (at `Umbraco.Core.Logging.Viewer.JsonLogViewer.CheckCanOpenLogs(LogTimePeriod logTimePeriod)`)
### Specifics
Also (kind of) reported before [on the forums](https://our.umbraco.com/forum/using-umbraco-and-getting-started/99251-change-log-files-output-directory).
I know we can [replace the build-in LogViewer with our own](https://our.umbraco.com/documentation/getting-started/Backoffice/LogViewer/#implementing-your-own-log-viewer), but I feel this functionality is basic enough to work 'out-of-the-box' or with minimal extra effort.
Looking at the [JsonLogViewer](https://github.com/umbraco/Umbraco-CMS/blob/v8/dev/src/Umbraco.Core/Logging/Viewer/JsonLogViewer.cs) code, it looks like the constructor supports adding in a different `logsPath` than the default. However, this parameter is not used in [its only usage](https://github.com/umbraco/Umbraco-CMS/blob/v8/dev/src/Umbraco.Core/Logging/Viewer/LogViewerComposer.cs). We also cannot change this in our own code, as the `JsonLogViewer` class is internal, meaning we would have to copy the entire code to change this parameter.
Additionally, [line 65](https://github.com/umbraco/Umbraco-CMS/blob/v8/dev/src/Umbraco.Core/Logging/Viewer/JsonLogViewer.cs#L65) does not use the `_logsPath` field for some reason.
### Possible solutions
The preferred solution would be that the Core `LogViewerComposer` would build the `JsonLogViewer` in a way that passes the `logsPath`. I do not see an obvious way to do so at this time.
Another solution would be to allow users to use the `JsonLogViewer` to set a different `logsPath`. I think the steps would be:
* Make class `JsonLogViewer` public.
* Change line 65 to use the private field instead of the hardcoded path
* Update docs to reflect changes needed when changing Serilog location
The steps users would have to make would be:
* Add new Composer that calls `composition.SetLogViewer(_ => new JsonLogViewer(composition.Logger, logsPath: yourNewLogsPath));`
I'm not sure if any of the above would work, but if that is the preferred direction to take, I can look into creating a PR for this.
## Umbraco version
I am seeing this issue on Umbraco version: 8.6
Answers:
username_1: Yep, that last option with a Composer sounds like a good option that could work! 👍
username_2: I've proposed a solution to this issue here: https://github.com/umbraco/Umbraco-CMS/issues/10143#issuecomment-848847021
Status: Issue closed
|
samuelgozi/firebase-auth-lite | 589842614 | Title: At leat 90% test coverage before adding or modfing any features.
Question:
username_0: This library is currently the only one from the "firebase-lite" SDK that doesn't have substantial test coverage. And it is also the library that is getting the most traction, therefore, before making any changes or refactors, It is the highest priority for the code coverage to reach 90% before moving to anything else(unless there are big bugs).
This shouldn't take long, but opened an issue so that anyone can know that I am indeed working on it. Hopefully, I'll get it done this week.<issue_closed>
Status: Issue closed |
davidhalter/jedi | 227134987 | Title: Weird `Params` for function completion
Question:
username_0: ```
In [11]: Script('bi').completions()[0].params
Out[11]: [<_Param param 2796202>]
````
function is `bin` which signature is `bin(number, /)` the `/` being end of positional-only arguments that can be done only using C. the `2796202` is still a bit weird :-)
Answers:
username_1: We're currently not using `inspect.signature(bin)`, but a weird way of `__doc__` parsing. We should just use the signature call whenever it's available.
username_1: I fixed this in d1c85191a0931bfa5dd0773d6d2f754199ee9699 by using `inspect.signature`.
Status: Issue closed
|
pimutils/vdirsyncer | 347701503 | Title: Syncing account on dav.openxchange.eu
Question:
username_0: Hello, did anyone ever sync an account on openxchange.eu successfully? I ran into the following issue:
```
error: The server returned something vdirsyncer doesn't understand. Error message: InvalidXMLResponse('Invalid XML encountered: mismatched tag: line 1, column 2777\nDouble-check the URLs in your config.',)
error: While this is most likely a serverside problem, the vdirsyncer devs are generally interested in such bugs. Please report it in the issue tracker at https://github.com/pimutils/vdirsyncer/issues
error: 1 out of 1 tasks failed.
```
I have similar problems with other Cal/CardDAV clients as well. Each of these not managing to find the avilable address books and/or calendars. The only working client I found is OX Sync [1] that offers a distinct selection of providers. This selection includes EuroDNS which is behind openxchange.eu.
OX Sync uses dmfs [2] which manages to connect to `https://dav.openxchange.eu` successfully. So I started to reverse engineer it a bit, trying to find out which URLs it attempts to connect. Unfortunately, I could not find out.
Before spending more time on this (taking into account my lack of knowledge about Cal/CardDAV): Has anyone ever synced an account on dav.openxchange.eu successfully?
[1]: https://play.google.com/store/apps/details?id=com.openexchange.mobile.syncapp.enterprise
[2]: https://github.com/dmfs/jdav
Answers:
username_1: There's some instructions [here](http://oxpedia.org/wiki/index.php?title=AppSuite:CalDAVClients#Discover_the_CalDAV_URL_of_your_Calendar_Folders) on how to obtain the exact URL of your calendar. Have you tried using that URL?
username_0: Yes, Ive also came across that article and tried that URL without any success.
I will provide logs of vdirsyncer and the official sync app for Android on
the weekend. In the logs of the app several endpoints are polled
unsuccessfully too when finally one of them works. I guess that should hint
toward the right direction for getting vdirsyncer work with this particular
provider.
Am 5. August 2018 8:16:07 nachm. schrieb <NAME> |
lucatume/wp-browser | 600200282 | Title: Support unix socks in initialization
Question:
username_0: Some local development environments, like the most recent version of Local, will not provide a database connect by means of a Unix sock.
The `codecept init wpbrowser` command should spot answers like those and change the host type, in the WPDb and WPLoader configuration, to work correctly.
This will include an update to the docs.<issue_closed>
Status: Issue closed |
DestinyItemManager/DIM | 720479859 | Title: Suggested Perks
Question:
username_0: **What version of DIM are you using?**
Find the version in "About DIM" from the menu (the button with three bars on the upper left of the screen). Is the DIM icon orange, or blue?
6.34.0
**What Browser and OS are you using?**
It helps to know what type of browser and what version you're using. Have you installed DIM as an app through Chrome or Safari? Chrome on computer, not home to check version, using the app on my phone. Issue is on PC and app.
**Describe the bug and how to make it happen**
Tell us what's wrong. A great format for bug reports is "I did X, and I expected Y, but instead Z happened." If it helps illustrate the issue, please add screenshots to help explain your problem.
All my suggested Perks are gone. I must have hit something cause suddenly I had like 22k rolls in my wishlist! So I cleared that, however now no thumbs up on anything. Last night it was only that way on my PC. Now it's like that on my app too, so I can't see any of them.
**Include devtools console logs**
If you can, open the devtools console (ctrl+shift+J on PC and cmd+option+J on Mac) and paste what you see in the window that pops up:
Sorry I'm not at my PC right now, just the app. Is there a way to see on the app?
```
Paste console logs here
```
Answers:
username_1: If you cleared your wish list, it'll be cleared everywhere. You can reapply it from Settings.
username_0: But I never created a wishlist. It used to be suggested Perks from y'all I
thought? Like when I logged on yesterday suddenly everything had thumbs up
perk with a white circle around them, which was way different than the day
before. drop down box with fav, junk, keep etc.. was gray, and the button
you used to hit to compare (looked kind of like 2 piece of paper laid over
one another kinda) is now a balance deal.
I did go to settings to try this wishlist but it goes to my files to find
it. I for sure have never saved a file I've not been using this long. I'm
sorry I'm probably just being totally dumb here.
When I joined DIM it automatically showed me if my weapons had suggested
Perks, with a thumbs up icon. Sometimes even a quick tip from players on
their thoughts and recommended masterwork. Typically it was pretty similar
to what is on light gg, but was nice I didn't have to use that site anymore
and could see right on the card. Now not so much. Any chance this makes
more sense to the type of issue I'm having?
I appreciate any direction as I'd like to see your suggested Perks, not a
wishlist I create on the side. I don't know enough to wishlist anything 😂
username_0: Omg nevermind, I finally found the premade link to put in. Now I'm just
confused why the appearance is different suddenly, but I'll shut up and
move on. 😂😂 Thanks for the help 👍
On Tue, Oct 13, 2020, 5:28 PM <NAME> <<EMAIL>> wrote:
> But I never created a wishlist. It used to be suggested Perks from y'all I
> thought? Like when I logged on yesterday suddenly everything had thumbs up
> perk with a white circle around them, which was way different than the day
> before. drop down box with fav, junk, keep etc.. was gray, and the button
> you used to hit to compare (looked kind of like 2 piece of paper laid over
> one another kinda) is now a balance deal.
>
> I did go to settings to try this wishlist but it goes to my files to find
> it. I for sure have never saved a file I've not been using this long. I'm
> sorry I'm probably just being totally dumb here.
>
> When I joined DIM it automatically showed me if my weapons had suggested
> Perks, with a thumbs up icon. Sometimes even a quick tip from players on
> their thoughts and recommended masterwork. Typically it was pretty similar
> to what is on light gg, but was nice I didn't have to use that site anymore
> and could see right on the card. Now not so much. Any chance this makes
> more sense to the type of issue I'm having?
>
> I appreciate any direction as I'd like to see your suggested Perks, not a
> wishlist I create on the side. I don't know enough to wishlist anything 😂
>
>
username_2: We'll make it easier in the future to 'reset' to the default and automatically load:
https://raw.githubusercontent.com/48klocs/dim-wish-list-sources/master/voltron.txt
in the box if you clear it- we'll track that in #5337
Glad you got it working!
Status: Issue closed
username_2: We moved from the green thumbs up to the new blue with circle as we're currently redesigning the item inspect sheet to be more clear. There are some additional changes on the way. It's all going to make sense really soon! Existing users should find it easier to use, and use users should be able to quickly jump in.
username_0: Can't wait, I absolutely love this site! I'm a data person and I love being
able to dig in and manage all the things. Thanks again!! |
foltysM/foodCheck-Android-Java | 718661450 | Title: Fix java.lang.NullPointerException: Attempt to invoke virtual method 'java.lang.Boolean net.foltys.foodcheck.data.FavProdViewModel.getOneFav(java.lang.String)' on a null object reference in net.foltys.foodcheck.AfterScanActivity.onCreate (AfterScanActivity.java:87)
Question:
username_0: ### Version 0.6.0(2) ###
### Stacktrace ###
net.foltys.foodcheck.AfterScanActivity.onCreate (AfterScanActivity.java:87);
### Reason ###
java.lang.NullPointerException: Attempt to invoke virtual method 'java.lang.Boolean net.foltys.foodcheck.data.FavProdViewModel.getOneFav(java.lang.String)' on a null object reference
### Link to App Center ###
* [https://appcenter.ms/users/MFoltys/apps/FoodCheck/crashes/errors/1686805242u](https://appcenter.ms/users/MFoltys/apps/FoodCheck/crashes/errors/1686805242u)<issue_closed>
Status: Issue closed |
bazelbuild/bazel | 205412582 | Title: Action cache is not working in some cases
Question:
username_0: After I build a new Bazel binary, I get almost no action cache hits, even if the binaries are identical (I think). This happens on MacOS at least. I'm investigating.
Answers:
username_0: It happens even after a shutdown, for things as simple as bazel build //third_party/ijar.
username_0: It looks like the round-trip through the on-disk representation is buggy. I get the exact same digest on subsequent runs, but it's always not matching the digest read from the on-disk action cache.
username_0: Previous analysis was incorrect - it seems to only apply to some C++ compile actions, and seems likely to be related to input discovery / subsetting.
username_0: Here's the action cache entry we write:
12e745723ce3d938d4f9c946e8754342 <KEY>
actionKey = a9f1d0ed60bfa5141202b858a610cd25
digestKey = <KEY>
bazel-out/local-fastbuild/bin/foo/_objs/foo/foo/foo.pic.d
bazel-out/local-fastbuild/bin/foo/_objs/foo/foo/foo.pic.o
bazel-out/local-fastbuild/internal/_middlemen/_S_Sfoo_Cfoo-null
external/local_config_cc/cc_wrapper.sh
foo/foo.cc
foo/foo.h
Here's the (partial) entry we compare it to:
<snip>
bazel-out/local-fastbuild/bin/foo/_objs/foo/foo/foo.pic.o
bazel-out/local-fastbuild/bin/foo/_objs/foo/foo/foo.pic.d
external/local_config_cc/cc_wrapper.sh
foo/foo.cc
bazel-out/local-fastbuild/internal/_middlemen/_S_Sfoo_Cfoo-null
Order of files does not matter, but note that we're missing foo/foo.h.
username_0: It seems like this is causing a lot of recompiling for C++ applications after a shutdown.
username_0: This is draining my battery because every experiment I run rebuilds a lot of C++ code. However, I think ActionExecutionFunction.collectInputs is misbehaving.
username_0: Well, that diagnosis was incorrect as well. New theory: in Bazel, CppCompileAction returns true for inputsKnown(), because include scanning is disabled (and not implemented). However, it still updates the action inputs after execution. At the same time ActionExecutionFunction is only using the files resolved from the cache for the action cache lookup if inputsKnown() returns false.
username_0: I still think my last explanation is correct, but it doesn't seem to be sufficient - even with the inputsKnown() method fixed, I still see unnecessary action execution. My battery is now down to <30%. :-(
username_0: So, looks better, but I get a null md5 digest for external/local_config_cc/cc_wrapper.sh.
username_0: We have two artifacts for cc_wrapper.sh, and one of them doesn't actually exists, and so does not have a md5:
Artifact:[/Users/username_0/Google/bazel[source]]external/local_config_cc/cc_wrapper.sh=singleton marker artifact value (1638529830)
Artifact:[/private/var/tmp/_bazel_username_0/ea721f29c7ff430b1d606476a9c89991[source]]external/local_config_cc/cc_wrapper.sh=RegularFileArtifactValue{digest=[44, 94, -17, 119, -89, -55, 22, -35, -31, -123, -125, 122, 4, -67, 63, 25], size=2730}
What I don't understand is that we somehow seem end up with the first artifact before running the action and with the second artifact after. In any case, if it only affects cc_wrapper.sh, then it may work on Linux, where we don't use a wrapper.
username_0: Apparently, the code that's resolving artifacts from exec paths is ending up with the wrong artifact, while the action has the correct artifact as input, and we use the resolved artifacts to find an action cache hit, where we ignore that one of the inputs doesn't actually exist.
username_0: It looks like ArtifactFactory cannot correctly handle artifacts under external/.
username_0: Ok, I found a workaround, which is to treat files under external/ as output files in ActionCacheChecker.getCachedInputs:
@@ -319,6 +340,11 @@ public class ActionCacheChecker {
allowedDerivedInputsMap.put(derivedInput.getExecPath(), derivedInput);
}
}
+ for (Artifact input : action.getMandatoryInputs()) {
+ if (input.getExecPath().startsWith(new PathFragment("external"))) {
+ allowedDerivedInputsMap.put(input.getExecPath(), input);
+ }
+ }
List<Artifact> inputArtifacts = new ArrayList<>();
List<PathFragment> unresolvedPaths = new ArrayList<>();
username_0: The other part of the fix is this:
+++ b/src/main/java/com/google/devtools/build/lib/rules/cpp/CppCompileAction.java
@@ -334,7 +334,7 @@ public class CppCompileAction extends AbstractAction
this.shouldPruneModules = shouldPruneModules;
this.usePic = usePic;
this.useHeaderModules = useHeaderModules;
- this.inputsKnown = !shouldScanIncludes;
+ this.inputsKnown = !shouldScanIncludes && !shouldPruneModules && !cppSemantics.needsDotdInputPruning();
this.cppCompileCommandLine =
new CppCompileCommandLine(
sourceFile, dotdFile, copts, coptsFilter, features, variables, actionName);
username_0: For completeness: for actions whose inputs can change over their lifetime, the action cache behaves differently than for actions who don't. The action cache uses inputsKnown() to determine whether the inputs can change or not.
If the inputs can change, the action cache reads the list of input files from the action cache and uses them to check if the action can be cached. However, the action cache only stores relative paths, so the action cache needs to resolve them to artifacts.
This is tricky - there are three kinds of artifacts. For output files, it considers the list of declared inputs of the current action to find the corresponding artifacts. For source files, it needs to find the correct --package_path entry and package that's responsible for that source file. That uses Skyframe to look up the corresponding locations. This generally all works.
Unfortunately, the code treats files in external/ as if they were source files, even though they are actually generated files (more or less - at least they are not in the source tree). Thus, it ends up with the wrong artifact when it checks the action cache before execution - the corresponding source file does not exist, and so the action cache check silently fails. Silent failure is necessary because a user might have legitimately deleted a file.
The code snippet above makes the action cache checker treat files in external/ as if they were generated files, i.e., it requires that they are declared inputs to the current action, and uses the relative path to look them up.
username_0: I wonder if 840acb10fb91b04e87d45f5627f888464899e577 is related.
username_1: The files in <output_base>/external/ should be treated as source files, I think.
username_0: For the purpose of the C++ compile action, the external/ paths are special, because they are not in the source tree, and they are also not under any of the output roots, so - right now - neither code path is correct. However, it so happens that I can put the external/ paths into the same code path as generated files and that works. Also note that `external/local_config_cc/cc_wrapper.sh` in particular happens to be a generated file, even if generated by the workspace.
username_0: Isn't it?
username_2: Hah! I have spent three days diagnosing this same issue and, now that I had found a cause, knew what to look for -- which led me to this bug. Let me paste some of the details from my investigation, which include a much simpler test case to iterate on this.
Consider these files:
-- foo/BUILD --
cc_library(name = "foo", srcs = ["foo.cc", "foo.h"])
-- foo/foo.cc --
#include "foo/foo.h"
void foo(void) { }
-- foo/foo.h --
// Nothing.
Then, the following operations:
bazel build --explain log --verbose_explanations --check_visibility //foo:foo
bazel build --explain log --verbose_explanations --check_visibility //foo:foo
bazel build --explain log --verbose_explanations --nocheck_visibility //foo:foo
The first command should build things as usual. The second command will yield an empty log file (disregarding the details for `stable-status.txt`). The third command, however, will report that `foo.cc` has changed and needs to be recompiled.
Removing the `#include "foo/foo.h"` from `foo.cc` makes the problem go away.
Focusing on `ActionCacheChecker`, in `validateArtifacts` during the third command above, the `Entry` object properly tracks `foo.h` as a path... but the `actionInputs` *do not* contain `foo.h`. Thus the newly computed digest does not match what we recorded in the action cache and miss, triggering the rebuild. Tricking the code into consider the header file causes the hashes to match and a cache hit.
username_3: @username_2 : @username_0's issue and yours seem to be orthogonal. I don't think his fix will fix your issue (although conceivably a fix to your issue might fix his).
username_0: Hmm? The --nocheck_visibility flag triggers re-analysis, which in turn triggers an action cache check for the new actions even if the actions are identical. Fixing the action cache check for C++ actions seems like it should improve things significantly (compared to actually running the compiler).
username_2: I'm happy to move the issue I found to a separate bug if there is more than one fix to be made (and it seems like that's the case). The reason I added the details here is because @username_0 troubleshooting steps looked very similar to mine and I wondered if they were obscured by what I found.
username_3: I agree that both issues are related to the action cache and Bazel. I'm just saying that @username_0's issue appears to be specific to external files, and @username_2's is about all C++ compilation. Perhaps the latter will resolve the former (although I doubt it), but not vice versa.
username_0: I actually found two bugs, one in CppCompileAction, and another related to our handling of //external. I had to fix both to get it to work on my Laptop.
username_0: So, assuming that Kristina is right and that these files should be treated as source files, I dug even deeper. The symptom is that ArtifactFactory doesn't return the correct Artifact. What's likely happening is that the PackageRootResolver is returning the wrong package root. PackageRootResolver is implemented by SkyframeExecutor.getArtifactRoots, which contains this code:
PathFragment parent = Preconditions.checkNotNull(
execPath.getParentDirectory(), "Must pass in files, not root directory");
packageKeys.add(ContainingPackageLookupValue.key(
PackageIdentifier. *createInMainRepo* (parent)));
That's incorrect. It tries to look for the corresponding path in the main repo, even though it's actual in an external repo. Maybe we should check for an "external" prefix and then create the right repo identifier instead:
if (parent.startsWith(new PathFragment(Label.EXTERNAL_PATH_PREFIX)) && parent.segmentCount() >= 2) {
RepositoryName repository = RepositoryName.create("@" + parent.getSegment(1));
packageKeys.add(ContainingPackageLookupValue.key(
PackageIdentifier.create(repository, parent.subFragment(2, parent.segmentCount()))));
} else {
packageKeys.add(ContainingPackageLookupValue.key(
PackageIdentifier.createInMainRepo(parent)));
}
username_0: I accidentally tried to reproduce this with an old bazel binary - pre 4ec02263fa630e97ce9a749a02efe9cf64838690 - and couldn't reproduce, so this problem must have been introduced in the last ~25 days.
username_0: It fails at 54223b019817c329d7bff1c174d497ef3d64695a, at least, so somewhere between 16 and 25 days ago.
username_0: No, my bad. It works at 54223b0.
username_0: It fails at 840acb1, so that's where this bug got introduced. @lberki
username_4: @Ulf- binary search? ;)
username_0: Meh. We have two Skyframe implementations of PackageRootResolver, and I was looking at the wrong one. The correct one is ActionExecutionFunction.PackageRootResolverWithEnvironment.
username_0: Well, this isn't working. I get the wrong source root back from the root resolution, except I don't know why. Here's the source root I get: /private/var/tmp/_bazel_username_0/fe875d5ca7c5e5732f6aae397d7cdf99/external/local_config_cc[source]
Here's the source root of the artifact that's written to the cache:
/private/var/tmp/_bazel_username_0/fe875d5ca7c5e5732f6aae397d7cdf99[source]
username_0: Ok, I have a fix, but I'm not sure this is correct. :-/
username_1: Please cc me on the fix, /private/var/tmp/_bazel_username_0/fe875d5ca7c5e5732f6aae397d7cdf99/external/local_config_cc will be the correct source root once the execroot change is in (just waiting on TF to merge https://github.com/tensorflow/tensorflow/pull/7165!).
username_0: I'd be more than happy for you to fix this, @username_1. :-P
username_0: @username_1 how do you plan to switch the source root behavior? Is there a flag or something?
username_0: Is the plan to roll forward 82d43279f93d95e4c41b4bc598a3cc05ddd1ae1a? Unfortunately, I can't easily patch it in right now as it doesn't apply cleanly anymore.
username_1: Basically yes. The internal # is 143777249, if you want to check it out.
(Now just waiting on https://github.com/tensorflow/serving/pull/321...)
username_1: ...and I accidentally reverted the CL ($*%&$) so now it's 147374155.
username_5: We got around this issue of the on-disk-roundtrip-exclude-invalidation by adding a symlink to the external execroot in the workspace root. We can post a diff if that's an acceptable workaround (my guess is not).
username_0: I'm still on it, even if it takes a bit longer. It looks like there's a third place in the code that's doing artifact discovery and that's also getting incorrect results. :-/
username_0: And that one got rolled back as well. I have a complete fix, but the change I committed depends on Kristina's change, which got rolled back. I think I can make it so it works both with and without Kristina's change... next week.
username_5: With my workaround, and changing inputsKnown to not be dependent upon shouldScanIncludes (the negation thereof) and just always false for CppCompileAction, the original patch works fine for me.
username_6: @username_0: will you have time to work on the fix, should we ask @username_7 to take over?
username_0: I've sent a fix for review to Kristina. Hopefully it'll be merged later today.
username_5: I was just looking at the test from the prune fix (`src/test/shell/integration/cpp_test.sh`), and realized that the test is relying on `Compiling a/a.cc` appearing in the output, which, unless I'm mistaken, is a timing artifact, and may not appear when the compile actually takes place - it should probably be using `-s` and looking for something indicating the compile activity
username_0: I thought I had a fix, but I was wrong. I've spent a couple more hours tracing the code and I think I've got it now. I'm running all of our tests again.
username_5: I have put together a test case for some unnecessary recompilations that are taking place that fails under the #2372 patch, but succeeds when inputsKnown is decoupled from shouldScanIncludes in CppCompileAction, see this [diff](https://gist.github.com/username_5/cf53b5201a8b0be19e462b3566f1cdfd)
username_7: cc @lberki
username_5: After testing further with a daemon shutdown between builds, it was only the inputsKnown initially false for compiles that kept recompilations from occurring after a build, for any cc_test *or* cc_binary compilation action.
Status: Issue closed
username_0: We also need to add a regression test, maybe based on @username_5's patch.
username_5: Regression test based on shutdown to incur on disk roundtrip.
https://github.com/bazelbuild/bazel/pull/2586
I'm investigating the action reevalulation prevention that is occurring in the odd corner case with cc_test compilations + crosstool + interleaved query. Will post another PR when I get there.
username_2: Is this fully fixed? I tried the repro in https://github.com/bazelbuild/bazel/issues/2490#issuecomment-278660955 with a head-built Bazel and still hit the problem. Should I file a separate bug for that issue? |
apotdevin/thunderhub | 1016248856 | Title: Create account link on startup is outdated
Question:
username_0: The link here goes to the GitHub repo, and not the docs site.
<img width="1119" alt="bilde" src="https://user-images.githubusercontent.com/16610775/136019740-30c371eb-3e63-420b-b898-08fe785ce96b.png"> |
uikit/uikit | 271015929 | Title: Lightbox not firing events.
Question:
username_0: <!--
Got a question?
===============
The issue list of this repo is exclusively for bug reports and feature requests. For simple questions, please use the following resources:
- Read the docs: https://getuikit.com/docs
- Ask in the Gitter chat room: https://gitter.im/uikit/uikit
- Look for/ask questions on stack overflow: https://stackoverflow.com/questions/ask?tags=getuikit
-->
<!-- BUG REPORT TEMPLATE -->
### UIkit version
<!-- Check if the issue is reproducible with the latest stable version. -->
3.0.0-beta.32
### Browser
All
### Reproduction Link
<!-- A minimal Codepen that can reproduce the bug. -->
<!-- You could start with this template: http://codepen.io/anon/pen/XMpryM -->
https://codepen.io/username_0/pen/xPZvqR
### Steps to reproduce
Open the Lightbox
### What is Expected?
Events fire
### What is actually happening?
They doesn´t fire.
Answers:
username_1: The events are triggered on the Lightbox Panel, not the Lightbox _trigger_. That is not optimal to be honest.
Also:
```
UIkit.util.on('$element', 'show', function () {
alert("Show");
});
```
That will not work, instead try:
```
UIkit.util.on('#box', 'show', function () {
alert("Show");
});
```
Status: Issue closed
username_0: Perfect, thanks for the explanation! 👍
username_0: @username_1 And how can i reference to the Lightbox Panel in order to use events?
https://codepen.io/username_0/pen/YEqaRW
Sorry, i didn´t found an example on the docs.
Thanks.
username_2: Hi,
I would like to use the 'showitem' event, but the example below, return with 'not a function'
UIkit.util.on('element', 'itemshow', function () {
//do something
});
Could you please reveal, which object has the event listeners for the Lighbox component, and how the UIKit can access it? Very thanks.
username_3: @username_0 did you manage to figure this one out? I've seen another example of
```
UIkit.util.on(document, '.uk-lightbox', 'beforeshow show shown itemshow hide hidden', listenerFunction)
```
Which doesn't work either
username_0: @username_3 Hi! Try this:
```
UIkit.util.on(document, 'shown', '.uk-lightbox.uk-open', function () {
console.log('Lightbox is displayed!');
});
```
username_3: @username_0
I can not for the life of me get these to work.
```
UIkit.util.on(document, 'beforeshow', '.uk-lightbox', function(e) {
var items = e.target.querySelector('.uk-lightbox-items');
UIkit.util.on(items, 'click', 'li', function() {
UIkit.lightboxPanel(e.target).show('next');
});
});
```
Any ideas? |
ajs/perl6-Math-Sequences | 464862296 | Title: A005588 / binary-rooted-trees
Question:
username_0: [A005588 / Number of free binary trees admitting height n.](https://oeis.org/A005588)
The sequence is not implemented yet.
<code>
2 7 52 2133 2590407 3374951541062 5695183504479116640376509 16217557574922386301420514191523784895639577710480 131504586847961235687181874578063117114329409897550318273792033024340388219235081096658023517076950
</code> |
WheelsOnRoll/Extensions | 248177487 | Title: Include more functionality in both Chrome and Mozilla extensions
Question:
username_0: * Allow user to see WoR coins in Mozilla and Chrome extensions.
* Also allow the user with a link to the main Wheels On Roll website page!
* Allow user to access the Play Store link of the app at the same time as well! |
groove-x/trio-util | 851383232 | Title: compose and transform without context managers?
Question:
username_0: Currently compose and transform functionality take the form of context managers. This is the orderly way, ensuring that derived values don't outlive the scope of their parents, and that callbacks (e.g. the transform function) won't be called after the corresponding AsyncValue has been finalized.
```python3
with compose_values(x=async_x, y=async_y) as async_xy, \
async_xy.open_transform(lambda val: val.x * val.y) as x_mul_y \
async_xy.open_transform(lambda val: val.x / val.y) as x_div_y:
...
```
This form is a bit uncomfortable at first, and requires creative use of `ExitStack()`-- but it works.
## How would things be without context managers?
```python3
async_xy = compose_values(x=async_x, y=async_y)
x_mul_y = async_xy.transform(lambda val: val.x * val.y)
x_div_y = async_xy.transform(lambda val: val.x / val.y)
```
The concern is about lifetimes, and avoiding cycles that might keep objects around forever. The parent AsyncValue would need a weak collection of transform functions, and the child AsyncValue would hold the main ref to the function. The transform callback may be called for a while even after the child expires, until the GC cleans up the weak ref. (Users may be surprised by this, especially if the transform function has side effects.)
Once `transform()` is built this way, `compose_values()` could be implemented with it by taking advantage of callback side effects. |
conan-io/conan | 207795702 | Title: Locating installed package from command line
Question:
username_0: It would be convenient to be able to locate the path to an installed package from the command line
eg something like
```
$ conan path my-package/0.5.0@username_0/stable
```
could give output
```
/Users/username_0/.conan/data/my-package/0.5.0/username_0/stable/package/ed52a5841768e616e14f89e1f52ae18ee5d1456a
```
Answers:
username_1: Such path doesn't exist. You get a different path for every posible binary you have for that package. You would need to add all the arguments you have to provide to the install command. If anything it could make sense as a ``conan info`` item.
What would be the problem you are trying to solve, what do you need that information for? Please note that you also have the path in the conan model, so you can access if from your conan recipes (even for project, using a conanfile.py instead of a conanfile.txt) very easily.
Status: Issue closed
username_0: Ok, I was considering using conan for distributing a command line tool to a docker image I am maintaining.
But I see conan is probably not the right tool for this type of setup, this is not really a dev-environment, it is a runtime environment where I run the installed binaries. So I should probably be using some .deb package or similar instead.
username_1: yes, you are right, conan is for developers, totally dev-oriented. It will probably include features to help create different deploys in the future, but this is not in the roadmap yet. |
D34DPlayer/Projet-Dev-Web | 891105628 | Title: Redirection www.boucherie.tk vers boucherie.tk
Question:
username_0: Le sous-domaine www.boucherie.tk a la même adresse ip que boucherie.tk. Mais aucune redirection n'est mise en place pour l'instant, ce qui fait une erreur SSL car le sous-domaine n'est pas inclus dans le certificat. |
kberov/Mojolicious-Command-generate-resources | 560002674 | Title: Test suite started to fail (with Mojolicious 8.27?)
Question:
username_0: A sample fail report: http://www.cpantesters.org/cpan/report/d93a4d48-3e13-11ea-9df8-533a5b1f184e
Statistical analysis suggests that the breakage started with Mojolicious 8.27:
```
****************************************************************
Regression 'mod:Mojolicious'
****************************************************************
Name Theta StdErr T-stat
[0='const'] 1.0000 0.0000 55661404870717376.00
[1='eq_8.02'] 0.0000 0.0000 0.00
[2='eq_8.05'] 0.0000 0.0000 2.05
[3='eq_8.07'] 0.0000 0.0000 2.39
[4='eq_8.08'] 0.0000 0.0000 1.22
[5='eq_8.09'] 0.0000 0.0000 0.24
[6='eq_8.12'] -0.0000 0.0000 -0.34
[7='eq_8.13'] -0.0000 0.0000 -0.34
[8='eq_8.14'] -0.0000 0.0000 -0.83
[9='eq_8.15'] 0.0000 0.0000 1.36
[10='eq_8.16'] -0.0000 0.0000 -2.44
[11='eq_8.17'] 0.0000 0.0000 2.73
[12='eq_8.26'] -0.0000 0.0000 -1.47
[13='eq_8.27'] -1.0000 0.0000 -18763455754174240.00
[14='eq_8.30'] -1.0000 0.0000 -12293547766686428.00
[15='eq_8.31'] -1.0000 0.0000 -16976569330977352.00
[16='eq_8.32'] -1.0000 0.0000 -31857886981972700.00
R^2= 1.000, N= 104, K= 17
****************************************************************
``` |
Shopify/mobile-buy-sdk-ios | 174816953 | Title: issue with adding new lineItem
Question:
username_0: I'm having issues with adding a new line item to cart. I'm following the example given in the integration guide/checkout.
cart = client.modelManager.insertCartWithJSONDictionary(nil)
//Button to add item
addItemButtum.addTarget(self, action: #selector(self.addToCart(_:)), for controlEvents: . TouchUpInside)
// My function to add item to cart
func addToCart(button: UIButton) if let variant = product?.variants.firstObject as? BuyProductVariant { cart.addVariant(variant) } print("Number of items: \(cart.lineItems.count)")
The addToCart is overriding the previously added item instead of creating a new line. How do I create a new lineItem
Status: Issue closed
Answers:
username_1: Please ignore my comment on the other issue.
If you create a new cart every time, yes, you will lose the contents of the previous cart.
And that's what your first line of code is doing:
`Swift
cart = client.modelManager.insertCartWithJSONDictionary(nil)
`
that creates a new cart with no line items. You have to keep a reference to the original cart you created.
username_1: Please use https://ecommerce.shopify.com/c/shopify-apis-and-technology for questions. Only create issues on github if you can prove there is a bug, or you want to request a feature. Thanks. |
ooni/explorer | 573932940 | Title: Write copy for new Tor test results
Question:
username_0: @sarathms I have written the copy for the new Tor test here: https://docs.google.com/spreadsheets/d/1QAeDVF5wYj7bnTE6rJ0XX3C8sBF0o6A0TxeLw9fpTvM/edit#gid=0
Please note that I have **created 2 new strings (rows 160-161)** and that this needs to be reflected in the codebase.
Status: Issue closed
Answers:
username_0: @sarathms I have written the copy for the new Tor test here: https://docs.google.com/spreadsheets/d/1QAeDVF5wYj7bnTE6rJ0XX3C8sBF0o6A0TxeLw9fpTvM/edit#gid=0
Please note that I have **created 2 new strings (rows 160-161)** and that this needs to be reflected in the codebase.
Status: Issue closed
|
terraform-google-modules/terraform-example-foundation | 863396120 | Title: 3-networks deployment feedback
Question:
username_0: Feedback from the deploy of step `3-networks`:
- There is no information regarding the `enable_hub_and_spoke` and `enable_hub_and_spoke_transitivity` configuration in the main README.md.
- `enable_hub_and_spoke` and `enable_hub_and_spoke_transitivity` should be included commented in the https://github.com/terraform-google-modules/terraform-example-foundation/blob/master/3-networks/common.auto.example.tfvars file, otherwise the user will not know where to add them.
- The main README should have a section with information regarding the two possible network models, at least a reference to the relevant section in the guide.
- the configuration of the variable `target_name_server_addresses` in the file `shared.auto.example.tfvars` lacks context for their usage in the foundation
- there is no context for the variable `domain` in the `common.auto.example.tfvars` file. It is not clear that it is the DNS name of a peering managed zone unless you go to one of the READMEs inside one of the environment in the `envs` folder. It is not evidente that its value **must** end with a period.
- The example for the terraform service account should be more helpful, instead of
https://github.com/terraform-google-modules/terraform-example-foundation/blob/1d5b05bf35e98d62b4145a349d490df872449dee/3-networks/common.auto.example.tfvars#L19 it should be:
```terraform
terraform_service_account = "[email protected]"
```
- The terraform service account created in step `0-bootstrap` expects to be the owner of the project as a side effect of creating the projects, since not all necessary permissions were added to the terraform account when it was created. for example the service account is missing the necessary permission to create firewall rules. This was found out when using the workaround for the quota limit issue from https://github.com/terraform-google-modules/terraform-example-foundation/issues/417
- In the deploy instructions with Cloud build, the user should be instructed to wait for the cloud build for the `development` branch to finish before starting the `non-production` build to avoid concurrent peering changes
```
Step #4 - "tf apply": Error: Error adding network peering: googleapi: Error 400: There is a peering operation in progress on the local or peer network. Try again later., badRequest
Step #4 - "tf apply":
Step #4 - "tf apply": on .terraform/modules/restricted_shared_vpc.peering/modules/network-peering/main.tf line 22, in resource "google_compute_network_peering" "local_network_peering":
Step #4 - "tf apply": 22: resource "google_compute_network_peering" "local_network_peering" {
``` |
domluna/JuliaFormatter.jl | 862894393 | Title: chaining with `|>` inside a function gives weird behavior
Question:
username_0: I have a function that is set up to chain a bunch of stuff along. For simplicity's sake, I've made one up with no packages. When I format it, it looks off to me.
I'm not sure if this is a bug or a feature request, and I would of course understand if you think the first example is desirable, but this seemed off to me.
```julia
function demo_function(x)
output =
x ->
x + 1 |>
x ->
x^2 |>
x ->
sqrt(x) |>
x -> maximum(x) - minimum(x) |> x -> x + 1 |> x -> sin(x) - cos(x)
return output
end
```
I don't have really strong opinions about what it *should* look like, but IMO something like this would be clearer and more readable:
```julia
function demo_function(x)
output =
x -> x + 1 |>
x -> x^2 |>
x -> sqrt(x) |>
x -> maximum(x) - minimum(x) |>
x -> x + 1 |>
x -> sin(x) - cos(x)
return output
end
```
Answers:
username_1: @username_0 so after digging around a bit it formats this way because of the semantics. If you wrapped all the anon functions calls with parenthesis, i.e. `(x -> x + 10)` then it would format how you want.
```julia> s
"a |> b -> b*10 |> c |> d |> e"
julia> s1
"a |> (b -> b*10) |> c |> d |> e"
julia> Meta.parse(s)
:(a |> (b->begin
#= none:1 =#
((b * 10 |> c) |> d) |> e
end))
julia> Meta.parse(s1)
:((((a |> (b->begin
#= none:1 =#
b * 10
end)) |> c) |> d) |> e)
```
Notice how using parenthesis completely changes the semantics since it introduces a new local scope. Taking that into consideration I would argue the default formatting reflects that and and is informative in that manner.
username_0: Interesting, thanks! That's definitely a fix for my problem. Longer term perhaps it would be nice to do this without the parentheses. However, an even better option here might be to use Pipe.jl or Chain.jl.
Thanks, |
hyperhq/www.hyper.sh | 399688339 | Title: Service is down
Question:
username_0: Hi there,
I know this isn't the right place to ask what is going on, but our containers went down more than one and a half hour ago and I cannot reach hyper.sh forum or anything relevant. I don't get any notification around the service stop. Can you please tell me what is going on?
Answers:
username_1: @username_0 Haven't you gotten the email from Hyper.sh about them shutting down? :) They have send out dozen of them :)
username_0: @username_1 I've got none of them and I don't know why, because I got emails about the invoices every time...
username_1: @username_0 Hope it works out, it was also short notice, only 25 days, https://news.ycombinator.com/item?id=18734658
username_0: @username_1 Thanks. I've just seen this article... In the meantime, I had a chance to contact the Hyper team, the problem seems to be resolved soon.
username_2: The data export issue has been solved. Thanks @username_1 for your help and thanks @username_0 for your patience.
Status: Issue closed
|
pboettch/vim-cmake-syntax | 469141058 | Title: Indent: wrongly indented if bracket comment is present in previous command-args
Question:
username_0: This
```CMake
message(STATUS "Hello" #[[Bracket Comment]] "second")
add_custom_command() # TODO this will wrongly align to ( due to bracket-comment
```
becomes
```CMake
message(STATUS "Hello" #[[Bracket Comment]] "second")
add_custom_command() # TODO this will wrongly align to ( due to bracket-comment
```
because the indent-code does not see the closing-parenthesis due to the bracket-comment.
Answers:
username_1: I have no idea how to do anything, i apologize. I was directed to etherscan from a payment sent from coinbase. im just punching buttons. I have no clue how to verify address or sign messages so i have all this stuff and its floating around, again, my apologies |
danielgindi/Charts | 162786562 | Title: Combined chart crashing when marker added
Question:
username_0: I have tested this on the demo as well as in my own app, it appears that the combined chart is crashing at `getEntryForHighlight`
It appears that `highlight.dataIndex` is -1 and causing an index out of bounds error. Not sure what the proper way to fix this is, as I have inserted a bounds check, but it then crashes inside of `ChartData`
Answers:
username_1: https://github.com/username_3/Charts/pull/1094/files
username_2: This is a known issue. be careful using @username_1's change, if using 0 has effects on your charts.
username_2: @username_0 could you please take a look at #737, if this are the same issue?
username_2: also, could you please provide steps to reproduce with ChartsDemo? I may have some time in the weekend
username_0: @username_2 Yes this appears to be the same issue, although in #737 it is irrelevant if you set the highlight to be false on data set and true on others. Even with 1 data set in a combined chart it will crash.
In order to repro, you just need to add in CombinedChartViewController.m
```
BalloonMarker *marker = [[BalloonMarker alloc] initWithColor:[UIColor colorWithWhite:180/255. alpha:1.0] font:[UIFont systemFontOfSize:12.0] insets: UIEdgeInsetsMake(8.0, 8.0, 20.0, 8.0)];
marker.minimumSize = CGSizeMake(80.f, 40.f);
_chartView.marker = marker;
```
in the view did load before `[self updateChartData];`
username_2: Alright, I finally got some time looking into this..
Yes, CombinedChartView did set self.highlightFullBarEnabled = true by default, and ChartMarker will crash for sure. Because those -1 values.
For now, please set self.highlightFullBarEnabled = false to avoid the crash. I am trying to find a fix
username_2: So it's not only combined chart, but all charts, as long as `highlightFullBarEnabled = true`
username_3: Is this still happening on Charts 3.0?
Status: Issue closed
username_3: I couldn't reproduce this on Charts 3.0, it anyone experiences this - please re-open the issue.
username_4: highlightValueWithXIndex = true ;
is still crash on ios 'Charts' ,'2.2.5'
username_2: 2.2.5 will crash for sure. try master. |
dotnet/orleans | 317416478 | Title: Exception thrown in MembershipOracle.TryToSuspectOrKill when debug level set to Debug
Question:
username_0: The following line is incorrect:
``` csharp
if (logger.IsEnabled(LogLevel.Debug)) logger.Debug("-TryToSuspectOrKill {0}: The current status of {0} in the table is {1}, its entry is {2}", entry.SiloAddress.ToLongString(), entry.Status, entry.ToFullString());
````
The parameter `format` does not follow the behavior of `String.Format`, so it will throw an exception "System.FormatException : Index (zero based) must be greater than or equal to zero and less than the size of the argument list".
Format should respect this format: https://docs.microsoft.com/en-us/aspnet/core/fundamentals/logging/?view=aspnetcore-2.1&tabs=aspnetcore2x#log-message-template
We need to do a pass in Orleans codebase to see if we did the same mistake in other places.
Answers:
username_1: Did this get past tests because the tests run with a higher minimum LogLevel? Perhaps we should use Trace for LogLevel in tests.
username_0: Yes indeed. I am trying to run them with loglovel set to trace to see if it has a big impact or not in the test time. Some of our test relies on timing, so I wouldn't be surprised if we have more test failure using this log level
username_2: Resolved via #4508.
Status: Issue closed
|
jlippold/tweakCompatible | 414010308 | Title: `AnimationsBeFast` working on iOS 12.1.1
Question:
username_0: ```
{
"packageId": "com.adamseiter.animationsbefast",
"action": "working",
"userInfo": {
"arch32": false,
"packageId": "com.adamseiter.animationsbefast",
"deviceId": "iPhone10,5",
"url": "http://cydia.saurik.com/package/com.adamseiter.animationsbefast/",
"iOSVersion": "12.1.1",
"packageVersionIndexed": true,
"packageName": "AnimationsBeFast",
"category": "Tweaks",
"repository": "eXqusic's Repo",
"name": "AnimationsBeFast",
"installed": "1.4.3",
"packageIndexed": true,
"packageStatusExplaination": "This package version has been marked as Working based on feedback from users in the community. The current positive rating is 100% with 1 working reports.",
"id": "com.adamseiter.animationsbefast",
"commercial": false,
"packageInstalled": true,
"tweakCompatVersion": "0.1.0",
"shortDescription": "Speeds up animations",
"latest": "1.4.3",
"author": "<NAME>",
"packageStatus": "Working"
},
"base64": "<KEY>
"chosenStatus": "working",
"notes": ""
}
```<issue_closed>
Status: Issue closed |
MicrosoftDocs/windows-itpro-docs | 1114818452 | Title: Detail the links between DeviceEvents AppControl actiontype events and eventviewer
Question:
username_0: Hi,
This page could use more actionable information. I am trying to understand what is being registered in DeviceEvents under actiontype AppControl*.
It appears to contain items from the Microsoft-Windows-AppLocker/EXE and DLL and Microsoft-Windows-CodeIntegrity/Operational logs, but that doesn't appear to be all of it.
Secondly, I find it hard to figure out which time in the DeviceEvents table relates to the time the event was created in EventViewer.
There's several different times in the DeviceEvents table:
- TimeStamp (I believe this is the time the data was added to Defender for Endpoint)
- ProcessCreationTime
- InitiatingProcessCreationTime
- InitiatingProcessParentCreationTime
It seems the actual timestamp of the eventlog entry is not available.
I find it quite hard to work with the data around appcontrol data in Defender for Endpoint. There's equally zero information posted by anyone using this on the entire internet.
---
#### Document Details
⚠ *Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.*
* ID: 1a868903-95b1-288c-72d2-be936171044a
* Version Independent ID: 1456cb15-b4ec-2c0d-958e-340766c27788
* Content: [Query Application Control events with Advanced Hunting (Windows) - Windows security](https://docs.microsoft.com/en-us/windows/security/threat-protection/windows-defender-application-control/querying-application-control-events-centrally-using-advanced-hunting)
* Content Source: [windows/security/threat-protection/windows-defender-application-control/querying-application-control-events-centrally-using-advanced-hunting.md](https://github.com/MicrosoftDocs/windows-itpro-docs/blob/public/windows/security/threat-protection/windows-defender-application-control/querying-application-control-events-centrally-using-advanced-hunting.md)
* Product: **m365-security**
* Technology: **windows-sec**
* GitHub Login: @jsuther1974
* Microsoft Alias: **dansimp**
Answers:
username_1: Assigned internal ADO tracking item Task 37996755: MDE AH App Control topic needs more descriptive and actionable guidance to myself to update the article |
COSIMA/datetime-fortran | 550557200 | Title: Hour comparison in num2date is incorrect
Question:
username_0: In `num2date` in [mod_datetime.f90](https://github.com/COSIMA/datetime-fortran/blob/51-avoid-date2num-compiler-bug/src/lib/mod_datetime.f90), line 1245 has:
```
if(num2date % hour == 60)then
```
Surely that comparison should be to 24, not 60?
Answers:
username_1: Definitely incorrect code.
username_2: Good catch - this is in the master at wavebitscientific - maybe raise the issue there?
https://github.com/wavebitscientific/datetime-fortran/blob/master/src/lib/mod_datetime.f90#L1238
username_0: Done. https://github.com/wavebitscientific/datetime-fortran/issues/64
username_2: This mix of integer and real arithmetic makes me nervous too.
But I think I'm right in saying 1e3 has an exact representation as a float, so I guess that helps...
https://github.com/wavebitscientific/datetime-fortran/blob/master/src/lib/mod_datetime.f90#L1215-L1228 |
Azure/app-service-announcements | 637349560 | Title: Removal of older Azure Functions Runtime version: 3.0.13760
Question:
username_0: Older versions of Azure Functions are regularly removed from the production environment. Going forward we will be using issues in this repo to announce when that is happening. The following versions of Azure Functions V2 are being removed from the production environment.<br /><br />- 3.0.13760<br /><br />If your function app is configured to run on any of these versions (using the FUNCTIONS_EXTENSION_VERSION app setting) you will soon find that the app is instead loading the latest version automatically.<br /><br />Please use https://github.com/Azure/app-service-announcements-discussions/issues/165 for discussions related to the removal of these versions.<issue_closed>
Status: Issue closed |
episphere/quest | 807513281 | Title: Module 2- Intro text missing INTROWH
Question:
username_0: I think INTROWH is missing.
Did not appear for SEX=2
Answers:
username_1: @username_0 I looked at the markup and it should be showing for SEX=2 or (SEX=3 and SEX2=6 or 7). I'm assuming this is related to the testing issues you encountered last week.
username_0: @username_1 It is rendering now, but has a bar at the end of the text
username_2: @username_1 The INTROWH is still not rendering for SEX=3 and SEX2=6. Did not test with SEX2=7.
Tested with:
{"firstName":"Amelia","AGE":"55","SEX":["3"],"SEX2":["6"], "GEN":["3"],"WORK":["1"]}
{"firstName":"Amelia","AGE":"55","SEX":["3"],"SEX2":["6"], "GEN":["2"],"WORK":["1"]}
username_2: Still missing INTROWH for SEX=3 and SEX2=male sex organs. Tested with: {"firstName":"Amelia","AGE":"55","SEX":["3"],"SEX2":["6"], "GEN":["3"],"WORK":["1"]}
Moving to triage.
username_2: @hzhao392 I am still not seeing INTROWH for SEX=3 and SEX2=6 or 7 (female sex organs). It is actually skipping past the rest of the women's health section (except for INTRONONCISWOM and HORNONCISWOM) and the testosterone questions and going straight to the physical activity section. |
James-Yu/LaTeX-Workshop | 767553369 | Title: Xelatex build error with the single xe->bib->xe->xe recipe command
Question:
username_0: ## Preliminary questions [Required]
### Disable all the other extensions except for LaTeX Workshop, and check that you still see this issue. [Required]
You still see this issue?: Yes
### Make sure to visit the [wiki FAQ](https://github.com/James-Yu/LaTeX-Workshop/wiki/FAQ) before filling an issue.
You visited the wiki?: Yes
### If your issue is with compiling a document (not having to do with finding the root file of a project), check first that you can compile manually.
You can compile a TeX document manually?: Yes
I am also using Sublime Text 3 with latextool.
## Describe the bug [Required]
### To Reproduce
Steps to reproduce the behavior:
1. Go to '...'
2. Click on '....'
3. See error
### Expected behavior
A clear and concise description of what you expected to happen.
I am expecting to initiate a single recipe that would compile without any interruptions. In other format like: IEEE trans, it works out of the box using pdf->bib->pdf->pdf.
A clear and concise description of what the bug is.
Hi,
I am having problem compiling my thesis that is initially downloaded from [Here](https://github.com/BIT-thesis/LaTeX-template).
I am not sure whether it is a bug or not. Most likely it can be the latex formatting issue from the styling format that I am using. But then my question, why it works within a single command when I am running the same file under SublimeText but it does not happenning within vscode. I do not have issue compiling other format like IEEE trans, elsevier. However, I did face similar issue like the above while trying to compile acm class. The main problem lies dealing with the bib/bibtex (most probably, my guess).
I am using the following recipe to build and compile
```json
{
"name": "xelatex",
"command": "xelatex",
"args": [
"-synctex=1",
"-interaction=nonstopmode",
"-file-line-error",
"%DOCFILE%"
]
},
{
"name": "xe->bib->xe->xe",
"tools": [
"xelatex",
"bibtex",
"xelatex",
"xelatex"
]
},
[Truncated]
## Additional questions
### Are you using VSCodium?
No
### Are you using LaTeX Workshop with VS Code Remote?
No
If the answer is Yes, please write which one you are using. Write the versions of the remote extension.
- Remote Containers/Remote SSH/Remote WSL
- Remote Extension Version: [e.g. 0.48.0]
### Additional context
Add any other context about the problem here.
Answers:
username_1: The issue is not related to the extension. Please ask at https://tex.stackexchange.com/
Status: Issue closed
username_2: You have errors in your `.bib`, which make `bibtex` fails and therefore the following steps of the recipe are not run. Fix them and your recipe will work `xe->bib->xe->xe`.
See #1170 for defining a `latexmk` recipe using `xelatex`.
username_0: Thanks |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.