repo_name
stringlengths 4
136
| issue_id
stringlengths 5
10
| text
stringlengths 37
4.84M
|
|---|---|---|
mozilla/galaxy.js-mobile-gamepad | 44667547 | Title: Use `Promise.resolve`/`Promise.reject` instead of `Promise` constructor when returning values
Question:
username_0: We can't do it in most places. But where we can, we should turn a function that's written like this
``` js
function methodThatReturnsAPromise() {
return new Promise(function (resolve, reject) {
resolve({type: 'state', data: data});
...
reject(err);
});
}
```
to this:
``` js
function methodThatReturnsAPromise() {
return Promise.resolve({type: 'state', data: data});
...
return Promise.reject(err);
}
```<issue_closed>
Status: Issue closed |
defold/editor2-issues | 317535314 | Title: Programmatically generating tilemaps
Question:
username_0: Yet an other feature request. It was requested on forum but can't find it on git issues.
I know it is not easy but something like this:
tilemap.new_tilemap(size, position, tileset, name, id...)
tilemap.add_layer(id, name)
tilemap.add_collision_layer(layer_name)
tilemap.remove(id)
Answers:
username_1: DEF-3430 |
abstiles/deluminate | 241944693 | Title: some pictures appearance were reverse like a sepia effect
Question:
username_0: Version 59.0.3071.115 (Official Build) (64-bit)
Deluminate 0.7.4
Win10 64bit home edition single language
some pictures appearance were reverse like a sepia effect in google news including my profile picture.
https://news.google.com/news/?ned=en_ph&hl=en
### Which Deluminate
Answers:
username_0: i think it's just the profile picture which was inverted my bad... |
rossfuhrman/_why_the_lucky_markov | 397495101 | Title: He empasized together with a very dim hallway, but as his eyes again, bit his lip, thinking and flicking his eyes transfixed. <NAME>.
Question:
username_0: Toot: He empasized together with a very dim hallway, but as his eyes again, bit his lip, thinking and flicking his eyes transfixed. <NAME>.
One comment = 1 upvote. Sometime after this gets 2 upvotes, it will be posted to the main account at https://mastodon.xyz/@_why_toots |
hierynomus/smbj | 244184421 | Title: Question: Listing available DiskShares?
Question:
username_0: I'm migrating a product from Samba JCIFS, where a listing of available SMB shares is available by connecting without specifying a path. Is there a similar capability in SMBJ?
I've tried various "generic" arguments to session.connectShare(), e.g., null, "", "\\", but to not avail.
And while I'm here, thank you got your excellent work on this project! It's available just in time.
Answers:
username_1: Listing shares is done through the [srvsvc](https://msdn.microsoft.com/en-us/library/cc247080.aspx?f=255&MSPPError=-2147217396) RPC interface. Support for MS-RPC over SMB hasn't been implemented yet. See #82.
username_0: Thanks for pointing me toward #82 @username_1. I just tried out the smbj-rpc extension but regrettably I'm not able to get past an access denied error, at least against the Win7 test box that's otherwise working perfectly with SMBJ. Looks like exactly what's needed though.
username_2: I think we should add some basic rpc calls to the library, but make it so that it is extensible enough to use custom rpcs.
_Sent from my Samsung SM-G950F using [FastHub](https://play.google.com/store/apps/details?id=com.fastaccess.github)_
username_3: @username_0 We've been using smbj-rpc to get all shares successfully, please open an issue with smbj-rpc if there were problems so that we can investigate. |
composer/composer.github.io | 176190039 | Title: PGP key cannot be imported by GPG
Question:
username_0: I think you do not use the correct form at (which e.g. includes an empty line after the "BEGIN PUBLIC KEY"), so one cannot import this key when using GPG.
Status: Issue closed
Answers:
username_1: It is not a PGP key that's why it can't be imported :) It's an openssl public key.
username_0: Ah oh... |
joomcode/redispipe | 499471610 | Title: Error debugging in GoLand
Question:
username_0: Hi,
I'm using this library and i found that its really delivers what it promised to !
its fast and lightweight much more than all other Go redis lib's out there so well done for that !
Another issue i'm facing is that i'm using GoLand IDE by jetbrains, everything works well beside one thing.. when i'm running on debug and i try to debug the result i got from the redis i get an error:
redispipe.io.error
But when i'm running without debug, the same lines of code works perfectly.
I Tried moving the context and the whole connection to global var and many other options but was unable to Debug the result...
Plus i tried using SyncCtx and normal Sync and yet same behaviour :(
Any idea ? how to overcome it ?
Because its pretty basic for me to be able to debug what comes back from redis "get".
Thanks a lot and keep up with the awesome work !
Answers:
username_0: Really? nothing?
username_1: Sorry, I've missed this issue.
I think you hit network timeout. By default it is not zero (ie default zero is converted to 5 seconds).
Try to set IOTimeout = -1 and DialTimeout = -1 in Connext options.
username_0: Worked like a charm ! Well done :D Thanks a lot you are a life saver!
Status: Issue closed
|
iterative/dvc | 860511308 | Title: Basic operations performance on Win
Question:
username_0: CProfile: https://drive.google.com/file/d/1grD3dHgDatjarXbS6Z-Cye6DaeQKvdgT/view?usp=sharing
DVC version:
```
DVC version: 2.0.17 (conda)
---------------------------------
Platform: Python 3.8.3 on Windows-10-10.0.19041-SP0
Supports: gdrive, http, https
```
Regular (not network drive).
<img width="1413" alt="Screen Shot 2021-04-17 at 1 38 26 PM" src="https://user-images.githubusercontent.com/3659196/115126245-7d332200-9f82-11eb-80f6-29ef0640cdd4.png">
It looks like we read itself takes 10s out of 450s. A lot of time goes into doing path operations, SQLlite, etc. Even if we could improve specific `nt.stat` and other calls, question still valid - why do we do `12x * number of files` of those? (+ a lot of other path, string manipulations).
Steps I would suggest to do:
- create Win bench for this - to start tracking basic Win operations. Can we answer the question how - is it a regression or not?
- consolidate this with other similar Win performance tickets
- try to reproduce this - make sure that we are not missing anything
- research what can we do.
Answers:
username_1: Related to #2360
username_2: I am facing the similar problem, and my add.prof file is here.
https://github.com/username_2/fyp/blob/main/add.prof
Also my dvc doctor:
DVC version: 2.1.0 (exe)
---------------------------------
Platform: Python 3.8.9 on Windows-10-10.0.14393-SP0
Supports: All remotes
Cache types: hardlink, symlink
Cache directory: NTFS on D:\
Caches: local
Remotes: None
Workspace directory: NTFS on D:\
Repo: dvc, git
username_3: We could get back to this subject, it will probably require setting up (another runner)[https://github.com/iterative/dvc-bench/issues/196#issuecomment-669877825]. We have been considering upgrading our linux one, so maybe we could join those tasks. |
zhongkaifu/CRFSharp | 144902077 | Title: Some feedback about trainfile size and decode inputfile form
Question:
username_0: 1. Trainfile size is depended on your computer memory. More large memory can process more large corpus size.
And, if your log show “Open and check training corpus and templates...“ means the corpus was still being loaded.
If your trainfile is too large, even the memory would be out of usage.
2. Decode inputfile, if you use Chinese corpus, or others language need to use word segmentation, please check the sentence size.
If the sentence is too long, maybe cmd would be stopped.
HOPE these can help you
Thanks! |
akhodakivskiy/VimFx | 123876115 | Title: [question] Compatability with `speed dial`-like extentions
Question:
username_0: Hi!
I'm currently using [Speed Dial extention](https://addons.mozilla.org/en-US/firefox/addon/speed-dial/) which unfortunately doesn't work with VimFX `follow link` command. I found that it must be fixed in the develop branch ( #475), but it still don't work for me in freshly builded .xpi from Git.
I want to know if it would be fixed or is there any good alternatives for the speed dial extention that are works with VimFX.
Thanks!<issue_closed>
Status: Issue closed |
elmsln/issues | 959439563 | Title: [elmsln] set the timeout lower for non-blocking network calls
Question:
username_0: This could just be in the xAPI calls but I think our default timeout is 30 seconds which comes from like 2012 settings / defaults. There's no reason for this to be higher then say 5 seconds, honestly anything after 2 is probably a failure. In instances where a network call might get stalled, having a lower timeout can ensure at least busted content loads as opposed to waiting, literally 30 seconds for busted content.
PSU has had a lot of odd networking issues lately, so these have popped up from time to time unfortunately ;)<issue_closed>
Status: Issue closed |
sulu/sulu | 557381328 | Title: Preview iframe disappearing
Question:
username_0: But what I would expect is that the frame is never hidden, no matter if you click "new window" or not. If I would want the preview iframe to be hidden, I'd click a "hide preview" button (which doesn't seem to exist right now 😄)
# Steps to Reproduce
When I'm clicking "Open in a new window", the iframe disappears. When clicking through the first 4 tabs (Details, SEO, Excerpt & Taxonomies, Settings) it stays hidden. When I click "Permissions", and then go back to one of the other 4 tabs, the iframe reappears again.
Answers:
username_1: That the iframe is disappearing is actually intentional. What would be the use case for opening it in a new window, and still waste valuable space, which can be taken by the form otherwise?
But what is definitely a bug is that it behaves differently for the permission tab. IMO the iFrame should not appear after returning from it, if the preview is already loaded in a separate window.
username_0: @username_1 I dunno, without really paying attention to it I regularly just open it in a new window to see how it looks full screen, when I then go back to the admin I have to reload to get the preview back. To me it seems more logical to be able to choose whether you want do display a preview or not, and also have to ability preview it in a new window.
username_1: That would mean more clicks, so I don't really see that as an advantage... If you close the tab with only the preview in it, the iframe should appear again as well. Have you known that?
I would say the primary use case is if you have two screens, so that you can open the entire preview on a separate screen. If you consider this use case, what you are suggesting would be a little bit cumbersome.
username_0: @username_1 oh I didn't know that, I also see the new window also automatically refreshes. In that case, only the permissions tab part is indeed a bug 😄 |
StansAssets/com.stansassets.android-native | 597387371 | Title: Unity Jar Resolver
Question:
username_0: 
<img width="440" alt="UnityJarResolver2" src="https://user-images.githubusercontent.com/63166821/78915490-ad550480-7a94-11ea-8e3f-c4764dd411d7.png">
<img width="452" alt="UnityJarResolver" src="https://user-images.githubusercontent.com/63166821/78915493-aded9b00-7a94-11ea-9c4f-f814d18f436d.png"><issue_closed>
Status: Issue closed |
dart-lang/args | 228121774 | Title: addSubCommand should have a "hide" option
Question:
username_0: I wish you could have hidden subcommands that didn't appear in the help, the same way you can have hidden arguments and options.
Status: Issue closed
Answers:
username_1: You can do this today, actually. `Command` has a `hidden` getter that you can override in your custom commands. |
Jermolene/TiddlyWiki5 | 167206827 | Title: SyntaxError: Unexpected end of input after loading TW5 html file
Question:
username_0: So I just installed TW on node.js using `tiddlywiki cosmos --init server` and then I did `tiddlywiki --load cosmos.html` where `cosmos.html` is a TW5 html file I had with lots of tiddlers. It created `.tid` files in the `tiddlers` folder as expected. However, then I run `tiddlywiki --server` on the root folder, and I am getting this:
```
undefined:1
SyntaxError: Unexpected end of input
at Object.parse (native)
at $tw.modules.define.application/json (/usr/lib/node_modules/tiddlywiki/boot/boot.js:1296:23)
at $tw.Wiki.deserializeTiddlers (/usr/lib/node_modules/tiddlywiki/boot/boot.js:1218:23)
at $tw.loadTiddlersFromFile (/usr/lib/node_modules/tiddlywiki/boot/boot.js:1449:23)
at $tw.loadTiddlersFromPath (/usr/lib/node_modules/tiddlywiki/boot/boot.js:1527:22)
at /usr/lib/node_modules/tiddlywiki/boot/boot.js:1522:40
at $tw.utils.each (/usr/lib/node_modules/tiddlywiki/boot/boot.js:69:12)
at $tw.loadTiddlersFromPath (/usr/lib/node_modules/tiddlywiki/boot/boot.js:1520:15)
at $tw.loadWikiTiddlers (/usr/lib/node_modules/tiddlywiki/boot/boot.js:1686:21)
at $tw.loadTiddlersNode (/usr/lib/node_modules/tiddlywiki/boot/boot.js:1754:27)
```
I checked on `boot.js` but can't see anything wrong. The server worked before adding the new tiddlers (of which there are about 1000)
Answers:
username_0: After removing tiddlers that begun with `$__` the server is starting now. I guess those shadow tiddlers can't be imported from the html TW5?
username_0: After stopping the server and starting it again, the same error comes up
username_0: I found that the error was due to plugin tiddlers. I think the instructions on TW on node.js should be more clear about how to install plugins. Installing a plugin from TW itself creater a tiddler that created the error. I manually copied the `tiddlywiki/katex` folder from github to a `plugins` directory, after adding the plugin to `tiddlywiki.info` too. Even then, it was complaining about `/plugins/tiddlywiki/plugin.info'` not existing.. so I just copied the `plugin.info` from the katex folder to `/plugins/tiddlywiki/`. Doesn't seem this is working very well.. or am I missing something?
username_1: @username_0
- If you remove all the tiddlers in the `tiddlers` directory does the problem persist?
- Would you be willing to post your tiddlywiki.info on gist.github.com and post a link?
username_0: If I remove all the tiddlers the problem goes away; but as I said above, it suffices to remove any plugin tiddlers.
This is my `tiddlywiki.info`: https://gist.github.com/username_0/b39e6613fa6a421139d817b7813d6b1a
I just think that adding plugins in the node.js doesn't appear straightforward.
username_1: So the problem is in the tiddlers. Do you have any custom plugins in the tiddlers directory? Delete each one and add them back one at a time till the problem shows up. That will be the one that is malformed. Open it in an editor and see what went wrong.
Also the Node environment for TiddlyWiki splits out all the data into organized files. Unlike the single file version which packages everything into the HTML. This means each tiddler is it's own file. It also means that types of tiddlers are organized. Regular tiddlers go in `tiddlers` folder while plugins are organized in a `plugins` directory. There is also meta data (`plugins/my-plugin/plugin.info`) for the plugin because plugins are usually more then just one file. Finally, both the node server and the single page version use the same boot code. Which means plugins need to be runnable in _both_ environments. This is why most plugins will have a `if ($tw.node)` or `if ($tw.browser)` check in them.
The reasoning is that in the single page version all the interaction is focused in the browser where organization is managed though the normal TiddlyWiki means. But on the server side most interaction is in the source code (like saving to a version control system or editing in your favorite text editor). And so it leave all the heavy lifting to the user and that is what a server side user would most likely want (i.e. control).
The `plugins` section of the `tiddlywiki.info` is only for importing official plugins from the TiddlyWiki repository. Custom ones go into your own `plugins` directory and are not referenced in the `tiddlywiki.info`.
I have an example of custom plugins on my blog: https://github.com/username_1/dev-tritarget-org
username_0: Thanks. I removed the custom plugin from the `tiddlywiki.info` file, and things are working as expected. However, I think that installation of plugins from TW itself should be made to work if possible :P
username_2: Hi @username_0 I updated the docs for plugin installation under Node.js:
http://tiddlywiki.com/#Installing%20a%20plugin%20from%20the%20plugin%20library:%5B%5BInstalling%20a%20plugin%20from%20the%20plugin%20library%5D%5D%20%5B%5BUninstalling%20a%20plugin%5D%5D
username_0: Thanks, I get it now!
username_1: Can this issue be closed now?
Status: Issue closed
|
test-repo-billy/sdk-release-request | 653704338 | Title: [resource-manager] Python: 'test issue template 4'
Question:
username_0: ## Release request for <i>test issue template 4</i>
**Link**: [test issue template 4](test issue template 4)
**Readme Tag**: test issue template 4
**Target release date**: 2020-07-23
**Contacts**:
<EMAIL> `<NAME>` @username_0
**Comments**: test issue template 4
This checklist is used to make sure that common guidelines for a sdk release are followed.
---
**Checklist**:
This checklist is used to make sure that common guidelines for a sdk release are followed.
- [ ] SDK generated, changelog and version number are ready?
- Breaking Change:
- [ ] breaking in single version of swagger with preview tag?
- [ ] breaking in single version of swagger with stable tag?
- [ ] breaking across swagger api-version?
- [ ] breaking in codegen?
- [ ] breaking only in SDK?
- Test and Validatation:
- [ ] run non-live test successfully?
- [ ] run live test successfully?
- [ ] requester validated the package?
- [ ] The PR is merged and package public released?
---
**Requester comments for breaking reason**
For example: we think this is a necessary change.
---
**Breaking swagger links**
For example: https://azure-rest-api-spec/xxxx
---
**Released Package&Version**
For example: https://pypi.org/project/azure-mgmt-compute/12.0.0/
---
**Changlog for this release**
For example:
12.0.0 (2020-06-19)
Features |
theQRL/QRL | 303670196 | Title: "OS Error" when trying to generate slaves.json
Question:
username_0: ubuntu@ip:~/.qrl$ qrl -r --host 172.16.17.32 slave_tx_generate
Src []: 0
Addr from (Leave blank in case same as source) []:
Number of slaves [0]: 4
Access type [0]: 1
Fee [0.0]: 0
Generating Slave #1
Successfully Generated Slave 1/4
Generating Slave #2
Successfully Generated Slave 2/4
Generating Slave #3
Successfully Generated Slave 3/4
Generating Slave #4
Successfully Generated Slave 4/4
**OS Error**
ubuntu@ip:~/.qrl$ ls -alh
total 12K
drwxrwxr-x 2 ubuntu ubuntu 4.0K Mar 8 23:18 .
drwxr-xr-x 6 ubuntu ubuntu 4.0K Mar 8 23:18 ..
-rw-rw-r-- 1 ubuntu ubuntu 298 Mar 8 23:16 wallet.qrl
Answers:
username_1: It appears as though the wallet already existed and an attempt to overwrite the index wallet 0 happened.
Status: Issue closed
username_2: slaves are not necessary anymore. closing |
nebulab/renderful | 550226372 | Title: Add support for Prismic
Question:
username_0: Currently, Renderful is focused on [Contentful](https://www.contentful.com/), but nothing prevents us from creating a generic API that can work with multiple headless CMSs. A good addition would be [Prismic](https://prismic.io/).
Mostly, this can be accomplished with a generic "provider" class that acts as a bridge between the API Renderful expects on content entries and the API exposed by the CMSs (Contentful, Prismic).
I'm still not sure how we could handle features that are only offered by one provider but not another. Probably that's just a problem to think about when it presents itself.<issue_closed>
Status: Issue closed |
realfastvla/rfgpu | 365092670 | Title: suggestions for multigpu use case in rfpipe
Question:
username_0: Currently, rfpipe uses a single function to set up rfgpu on a single GPU. For realfast, that requires a one-to-one mapping of data read and data searched on a GPU. That will limit GPU utilization if large amounts of memory need to be read at once.
As an example, reading L-band with 10 ms sampling and an FRB search requires reading ~10 GB of memory. For a server with 8 GPUs, we will either need to limit the memory usage per read or share the search work for a single read over multiple GPUs.
I can see one way to use multiple GPUs in rfpipe, but it is pretty invasive. I wanted to ask for suggestions before implementing it. The basic flow is:
```
rfgpu.cudaSetDevice(devicenum)
[set up grid, etc.]
for i in resampling_list:
for dm in dm_list:
grid.operate()
image.operate()
[threshold]
return peaks
```
Is there overhead to setting`cudaSetDevice`? Could we use it within the loops to run two GPUs concurrently?
Answers:
username_1: Yeah, being able to use multiple GPUs from a single host thread may be useful. According to the CUDA doca, `cudaSetDevice` "should be considered a very low overhead call." It might be cleaner to handle this inside rfgpu though; for example each `Grid`, `Image`, etc instance will be associated with a certain device, and rfgpu will automatically switch as appropriate when each is called. That way the calling program doesn't have to manage all this. Sound good?
username_0: Yes. I was hoping it could be worked into rfgpu for the sake of the overall design. Do you think it would also be possible to make the `vis_raw.h2d()` command transfer to a set of devicenums?
Is there a clever way to transfer concurrently? If not, the `h2d` transfer time will scale up linearly with number of devices, while the search time will scale down in the same way. When multiplexing over many GPUs, we'll eventually be spending more time moving data than searching.
username_1: Yes we should be able to do concurrent transfers no problem. Even if we didn't, the way the `Array` class is currently structured, there is a one-to-one mapping between GPU and host memory, so this needs changing to address the overall memory usage issue mentioned earlier.
username_1: Not checked in yet but I've done some work on this. Here is a bit of info on the `Array` interface changes. I've tried to keep it all backwards-compatible. Let me know if you have any suggestions:
`a = GPUArrayComplex((1024,1024))` allocates a 1024-by-1024 array on whatever device is currently selected (via `cudaSetDevice`). This is the same as it always was.
`a = GPUArrayComplex((1024,1024),(1,2,3))` allocates the array on devices 1, 2 and 3.
`a.h2d()` sends the data from CPU memory to all devices associated with this array. Transfer is done asynchronous, so should go in parallel to all the devices. Also the function returns before the transfer is finished. The GPUs will handle this automatically (ie not start other stuff til it's done) but there could possibly be issues if you modfiy the CPU array while data transfer is still happening.
`a.d2h()` will transfer the data from the default device (first one in the list) to the CPU memory. `a.d2h(n)` will transfer from device `n`.
username_1: Yes, that's right. Everything I described in the previous comment has to do with moving chunks of raw data between CPU and GPU memory. In short the change allows for a single chunk of CPU memory to map into several GPUs, rather than a strictly one-to-one mapping as before.
How I plan `Grid` to work is that when an instance is declared, it will have an extra argument specifying which GPU it uses, ie `grid = rfgpu.Grid(nbl, nchan, ntime, upix, vpix, device)`. Then any time you call a `Grid` method such as `operate()` it will automatically use that GPU. Will raise an error if you try to pass it an array that is not defined on that device, etc. Since each `Grid` instance is tied to a specific GPU, you'll need to have one per each GPU in use.
Does that all sound OK?
username_0: I'm sure I could work with that.
I suppose another approach would be to make the methods of `Grid` (e.g., downsample, operate), take a `device` as an argument. Perhaps that would mean there'd only be a need to make a single `Grid` object? Either way, multiple `Array` objects would be needed to manage the resampled data and images.
username_0: Ok, I imagined there could be structural issues. I'm excited to play with this!
username_0: Do I understand that you've only implemented the GPU memory for the `GPUArrayComplex` class? I tried to set it up for GPUArrayReal` and got this error:
```
TypeError: __init__(): incompatible constructor arguments. The following argument types are supported:
1. rfgpu.GPUArrayReal()
2. rfgpu.GPUArrayReal(arg0: int)
3. rfgpu.GPUArrayReal(arg0: List[int])
4. rfgpu.GPUArrayReal(arg0: List[int], arg1: List[int])
Invoked with: (1944, 2304), 0
```
username_0: Sorry, ignore that last question. I see the syntax in the code.
username_1: OK, hope that makes sense. In the "two lists" version the first is the array dimensions and the second is the list of devices. I should add some better arg names and docstrings at some point.
username_1: Just trying this out a bit more today. While it seems to work (in the sense of not crashing or otherwise producing bad results), the GPUs are not being used in parallel in the way I thought they would when calling multiple devices from a single thread. So it probably needs a bit more work before it should be integrated "for real" into the pipeline.
Also, I'm adding some docstrings and argument names to the python interface, should help with code readability, for example you can do `a = rfgpu.GPUArrayComplex(dims=(1024,512),devices=(1,2,3))`. Some of this has been committed already, some still in progress.
username_0: I have set up some concurrent execution code around the grid/image portion of the rfgpu code. It works pretty well for 1 or 2 GPUs, but does not scale linearly beyond that.
username_1: Do you have a simple example of this? Or is it in rfpipe somewhere?
username_0: Yes, in the development branch. The concurrent part is done at:
https://github.com/realfastvla/rfpipe/blob/development/rfpipe/search.py#L137
I've restructured this a few times to try to get it to scale well. It could be simpler than this version and still scale the same (i.e. sublinear for >2 gpus).
username_1: That might be the explanation. I'd like to reproduce the rfpipe usage in a simple example and run it through nvprof to really see what is going on though. I started this but haven't finished yet.
I think putting an iteration over multiple images into rfgpu is probably a good idea in any case though, and is something I've thought about before. This would also allow things like using batched FFTs and should improve efficiency for small images sizes where it looked like the current implementation is dominated by kernel launch overheads.
username_1: OK, I figured out part of the problem, I accidentally committed a version of the Makefile that had the GPU timing code enabled. This will in some circumstances slow down multi-GPU usage because it makes many of the routines wait for the GPU to finish before returning, preventing some parallel operations from happening. That may not completely explain what you were seeing, but you can pull the latest (with those two line re-commented) and try again. Good idea to `make clean` first to make sure everything gets re-built.
username_1: I have a test script set up and am able to reproduce your results; run time improves by ~50% going from 1 to 2 GPUs, and not much after that. I'll play around with it and let you know if I find out anything useful.
username_0: I've pulled the latest on the the multi_gpu branch and rebuilt, but importing to Python fails:
```
In [1]: import rfgpu
---------------------------------------------------------------------------
ImportError Traceback (most recent call last)
<ipython-input-1-37fd4ba54c94> in <module>
----> 1 import rfgpu
ImportError: /home/cbe-master/realfast/anaconda/envs/development3/lib/python3.6/site-packages/rfgpu.so: undefined symbol: _ZN5rfgpu5Timer4stopEv
```
Going back one commit (to 4c94821e8b980a12887155003a333aa484714020) fixes it.
username_1: I added `Makefile` as a dependency so everything will get rebuilt if it's changed, sorry for the confusion.
username_1: Just noting here that as discussed in #13 multi_gpu branch has been merged into master.
username_1: I took another look at this today. Reducing the amount of `cudaSetDevice` and `cudaGetDevice` calls did not have any effect on the multi-GPU performance so I think I will leave that code as-is.
I think the bad scaling is as you say due to python. I made a simple change to have the `Image.stats()` function release the GIL and this seemed to have helped a lot. I now get linear runtime scaling for up to 4 GPUs with 1k-by-1k images (this probably depends on image size and other params). Could you test it out sometime and see if this works in your code? Also all this GIL stuff is always semi-mysterious to me.. I think this change is safe but if you think otherwise please let me know ;)
For additional improvement I will still think about processing batches of N images at a time as we discussed last week. But maybe hold of on this for now in favor of working on phasing / dynamic spectra?
username_0: I built the new version (with nice new build scripts!), but can't get the scaling to improve beyond what it did before. I see 2x improvement for 2 GPUs, but none beyond that.
I have some ideas, so I'll play more on my end. |
doobleweb/sublime-dooble | 123225574 | Title: select text issue
Question:
username_0: when i try to select a word, by pressing double click or <kbd>ctrl</kbd>+<kbd>shift</kbd>+<kbd>arrow</kbd> in HTML++ syntax the text that is select take more than the specific word
Status: Issue closed
Answers:
username_1: caused because of the removel of / and \ from the word separators. they have been added back. |
pacas00/ModdersGearbox | 377898149 | Title: Better Loading Screen Handling
Question:
username_0: Need to handle loading better,
Currently, we block everything including UI updates.
Proposed new handling,
Postfix on someones .ctor
- Setup vars we need
- Make sure we are ready to run our own update method
Prefix on someones UnityUpdate()
- If we are not running, start up and return
- If we are running return until we are ready
- If we are ready, stop returning.
Our Update Method
- Needs to update the UI
- Or at the least, not block
Answers:
username_0: New Plan,
Prefix ModManager.HandleActivation
if (WorldLoadState.State != WorldLoadStateEnum.LoadModUnityData) leave prefix
if (WorldLoadState.State == WorldLoadStateEnum.LoadModUnityData) then
Check if we are ready, if not, call our own update method for loading and return
If we are ready, then leave prefix allowing normal execution
username_0: Prefix LoadingControl.Update
if (WorldLoadState.State != WorldLoadStateEnum.LoadModUnityData) leave prefix
if (WorldLoadState.State == WorldLoadStateEnum.LoadModUnityData) then
- if we are ready, leave prefix
- Set our loading text string as needed
`this.LoadingLabel.text = PersistentSettings.GetString("Game_Loading") + "[" + text + "]";`
username_0: Thinking about this, We cannot patch a method we are running.
Not happening any time soon.
Status: Issue closed
username_0: Closing this until i think of another way to do this.
username_1: Can't set it up as one of those; "Must restart program to apply setting change," options you see in plenty of other programs, games, etc? |
orange451/LWJGUI | 598309107 | Title: Graphic issues with SplitPane
Question:
username_0: Hi,
While redoing my GUI with your lib I've stumbled into two graphical issues regarding the usage of SplitPane.
First, when I do the following hierarchy: **SplitPane <- ScrollPane <- TreeView.** (Read "<-" as "contains")
If I set the scroll pane to fill it's parent width and height (which I belive is the correct thing to do when a component must take the whole space it has in a SplitPane) the content of the tree takes the minimal amount of space it can (which is ok since we didn't set it to fill space) but if I set it to fill the space from it's parent I get the following glitch.
And when this happens I can also scoll to the right, making the text disappear even tought the text is smaller than the space inside the divider.
My second issue is when I put a StackPane in the middle of a SplitPane with two divisors, it seems that I can't get it to fill it's space (actually I wanted to put an OpenGL pane but to avoid errors comming from my use of the OpenGLPane I replaced it).
If I set to fill width / height to true I get the following result.
Whereas if I just set it's pref size to a big enought number I get what I want
The code the reproduce this is available here:
https://github.com/username_0/ProtectedEngine/pull/19/files
The first issue mentionned is in HierarchyPane.java
The second one in at EditorLayout.java
(On a side note, is there a way to make the divisors of the splitpane thinner ?)
Hope my description is clear enough, feel free to ping me if more infos are needed :)
Answers:
username_1: Thanks for all this info! I’ll play around with your examples and see if I can fix anything :)
username_1: I assume you want some functionality similar to this?
In order to make this, I removed the setFillToParentHeight() and setFillToParentWidth() calls in HierarchyPane.java
Then I overrode the position method like this:
`@Override
public void position(Node parent) {
super.position(parent);
sceneHierarchy.forceSize(this.getViewportWidth(), this.getViewportHeight());
}`
The reason being, the parent of sceneHierarchy is the CONTENT of the scrollpane, which is infinite size. So viewport must be used to make sure you size it to the correct element.
username_1: Also to answer your other question, you can use ScrollPane#setScrollBarThickness()
Also in the latest version, I added a scrollbar-width css property.
username_0: Yep that's all good now, thanks a lot for your speedy response !
Status: Issue closed
username_0: Actually, there's just the remaining problem of the stackpane in the middle, if I set it to fill it's parent the whole layout gets weird, for now I just set it to have a big pref size but I don't know if that can cause issues in the future
username_0: Hi,
While redoing my GUI with your lib I've stumbled into two graphical issues regarding the usage of SplitPane.
First, when I do the following hierarchy: **SplitPane <- ScrollPane <- TreeView.** (Read "<-" as "contains")
If I set the scroll pane to fill it's parent width and height (which I belive is the correct thing to do when a component must take the whole space it has in a SplitPane) the content of the tree takes the minimal amount of space it can (which is ok since we didn't set it to fill space) but if I set it to fill the space from it's parent I get the following glitch.
And when this happens I can also scoll to the right, making the text disappear even tought the text is smaller than the space inside the divider.
My second issue is when I put a StackPane in the middle of a SplitPane with two divisors, it seems that I can't get it to fill it's space (actually I wanted to put an OpenGL pane but to avoid errors comming from my use of the OpenGLPane I replaced it).
If I set to fill width / height to true I get the following result.
Whereas if I just set it's pref size to a big enought number I get what I want
The code the reproduce this is available here:
https://github.com/username_0/ProtectedEngine/pull/19/files
The first issue mentionned is in HierarchyPane.java
The second one in at EditorLayout.java
(On a side note, is there a way to make the divisors of the splitpane thinner ?)
Hope my description is clear enough, feel free to ping me if more infos are needed :)
username_0: I've tired to use
`@Override
protected void position(Node parent) {
super.position(parent);
forceSize(parent.getWidth(), parent.getHeight());
}`
In the OpenGL pane but while the parent width and height seems to be correct doing that messes up the layout
username_0: Update: Actually not setting the size of the opengl pane seems to work in terms of layout but still, the clear happens everywhere but on the openglpane
Status: Issue closed
|
VeeamHub/vbo365-rest-self-service | 552900650 | Title: Cannot install composer
Question:
username_0: I have version 3 working OK but need to move to version 4.
I have created a VM from Ubuntu 18.04 and installed a LAMP package
I get an error when downloading and installing composer with the command
curl -sS https://getcomposer.org/installer | /usr/bin/php && /bin/mv -f composer.phar /usr/local/bin/composer
/bin/mv: cannot move 'composer.phar' to '/usr/local/bin/composer': Permission denied
I can manually move this using sudo but has anything else been left incomplete?
How can I successfully install composer?
I can clone the repository
git clone https://github.com/username_2/vbo365-rest.git
Should I be in any particular directory to do this?
I have created /portal for all downloads and run the command from here.
The instruction "Place these files under the web service root (/var/www/)" needs clarification
is it the directory vbo365-rest I should be moving or just the files and directories in it?
When running Composer install I get an error
No composer.json in current directory, do you want to use the one at /var/www? [Y,n]?
[RuntimeException]
/var/www/vendor does not exist and could not be created.
Is vendor a variable or should I just manually create it as a directory?
How do I overcome these errors?
After running the composer routines as above I can no longer access the apache default page.
Thanks
Answers:
username_1: @username_0 manually moving the composer.phar file should be enough when you do not have root permissions on your environment. If that won't work, I would suggest to login as root via `sudo su -` (assuming your account has that permission) and installing composer again.
According to the [documentation](https://github.com/VeeamHub/vbo365-rest-self-service#2-clone-this-repository) the files should be moved to the correct folder. In my environment we have cloned the environment to `/var/www/vbo365-rest-self-service` and pointed our Apache configuration to that folder.
you should run the `composer` command inside the cloned repository folder. The user running that command should have enough permissions to create files / folders in that repository folder.
Good luck!
Maurice
username_0: Maurice
Many thanks for your response.
I will give it a go.
Thanks
username_0: I have moved composer
mv composer.phar /usr/local/bin/composer
I have cloned the repository to /var/html
I have tried to run composer from this directory but I still get the error
[RuntimeException]
/var/www/vendor does not exist and could not be created.
Trying to run as sudo gives the following error
Do not run Composer as root/super user! See https://getcomposer.org/root for details
username_1: @username_0 The error message you get when trying to run composer, please make sure the `composer.json` file is in that folder and the user that you use has permissions to create and write files / folder in that folder.
username_2: Could you do a 'ls' in the /var/html folder? I presume you still have the vbo365-rest folder in which you should enter and run the composer command.
username_0: vboadmin@vbo365rest:/var/www$ ls
html vbo365-rest
Should I move "vbo365-rest" inside folder "html"?
username_2: You need to move all the content from inside the vbo365-rest folder into the html folder.
Normally these should be enough:
mv /var/www/vbo365-rest/* /var/www/html
mv /var/www/vbo365-rest/.* /var/www/html
Afterward, go to the /var/www/html folder and try to run composer again.
username_0: Thanks Niels
so I need the folder "/var/www/vbo365-rest" but the content needs to be in "/var/www/html"
Regards
username_2: Correct, the content from in the vbo365-rest folder needs to be in the html folder.
username_0: Thanks Niels
This has now worked
Regards
Clive
username_0: [Veeam for O365 portal .docx](https://github.com/VeeamHub/vbo365-rest-self-service/files/4097562/Veeam.for.O365.portal.docx)
Hi Niels
I have created an install from scratch doc, which maybe of use.
Regards
Status: Issue closed
username_2: Thanks for sharing. I will look into it and if needed update the readme file. |
uber-go/ratelimit | 537821056 | Title: Evaluate atomic and mutex based Take implementations.
Question:
username_0: #15 brought in a new implementation of Take to avoid starving out the older mutex implementation (a major performance win). Before we cut a new release, let's dive a little deeper on the new implementation's behaviors and aim for a single implementation.
Answers:
username_0: The atomic implementation has two warts that make me weary. I don't like reaching for unsafe if it's at all avoidable. The fixed padding is also architecture specific. I'm not sure yet, but it may be possible to avoid the unsafe pointer swap. I'm not too sure what to do about the padding.
The older mutex implementation held its lock during the blocking sleep call. Holding the lock doesn't appear to be necessary (compute and set last & sleepFor, unlock, sleep). Should this hold true, let's use the new benchmark to compare the three implementations. |
spring-io/initializr | 774346247 | Title: @ConditionalOnBuildSystem fails to locate bean ProjectDescription
Question:
username_0: ## Problem
Given the following configuration class:
```java
@ProjectGenerationConfiguration
@ConditionalOnBuildSystem(CasOverlayBuildSystem.ID)
public class CasOverlayProjectGenerationConfiguration {
}
```
...and the following spring.factories file:
```
io.spring.initializr.generator.project.ProjectGenerationConfiguration=\
org.apereo.cas.config.CasOverlayProjectGenerationConfiguration
io.spring.initializr.generator.buildsystem.BuildSystemFactory=\
org.apereo.cas.overlay.buildsystem.CasOverlayBuildSystemFactory
```
Spring Initializr `0.9.2` produces the following error on startup:
```
***************************
APPLICATION FAILED TO START
***************************
Description:
A component required a bean of type 'io.spring.initializr.generator.project.ProjectDescription' that could not be found.
Action:
Consider defining a bean of type 'io.spring.initializr.generator.project.ProjectDescription' in your configuration.
Process finished with exit code 1
```
It seems as though the configuration class is initialized eagerly on startup, rather than on demand when the project is requested.
## Reproduce
```
git clone https://github.com/username_0/initializr-sample .
./gradlew bootRun
```
Status: Issue closed
Answers:
username_0: Thanks for the note, @username_1. |
defuse/juggler-pow | 109753823 | Title: Use uint32_t when possible.
Question:
username_0: Make a preprocessor statement check if all of the stuff will fit in a `uint32_t`, and if so, use that type instead of `uint64_t`. This will make the code more efficient and will make the proofs smaller.
Status: Issue closed
Answers:
username_0: Done. |
lbryio/lbry-android | 809813254 | Title: Fix mass tip unlocking
Question:
username_0: Many times if doesn't work as expected...we may not be passing all claim ids/txids.
Don't pass claim / txids, just run with --type=support and --is_not_my_input --blocking
Answers:
username_1: @username_0 we think this is fixed. If you think it's not, can you re-open and provide context?
Status: Issue closed
|
ApiGen/ApiGen | 184348580 | Title: [4.1 Phar Release] ErrorException in GroupSorter, strpos(): needle is not a string or an integer
Question:
username_0: `Found 7711 classes, 1 constants and 1541 functions
Destination is not empty. Do you want to erase it? [yes] yes
Generating API documentation
exception 'ErrorException' with message 'strpos(): needle is not a string or an integer' in phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/GroupSorter.php:141
Stack trace:
#0 [internal function]: Tracy\Debugger::errorHandler(2, 'strpos(): needl...', 'phar://D:/WebPr...', 141, Array)
#1 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/GroupSorter.php(141): strpos('Piwik Tracker V...', Array)
#2 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/GroupSorter.php(66): ApiGen\Parser\Elements\GroupSorter->compareGroups('Piwik\\Tracker\\V...', 'ApiGen', Array)
#3 [internal function]: ApiGen\Parser\Elements\GroupSorter->ApiGen\Parser\Elements\{closure}('Piwik\\Tracker\\V...', 'ApiGen')
#4 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/GroupSorter.php(67): uksort(Array, Object(Closure))
#5 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/ElementStorage.php(277): ApiGen\Parser\Elements\GroupSorter->sort(Array)
#6 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/ElementStorage.php(246): ApiGen\Parser\Elements\ElementStorage->sortNamespacesAndPackages()
#7 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/ElementStorage.php(329): ApiGen\Parser\Elements\ElementStorage->categorizeParsedElements()
#8 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Parser/Elements/ElementStorage.php(118): ApiGen\Parser\Elements\ElementStorage->ensureCategorization()
#9 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Generator/TemplateGenerators/NamespaceGenerator.php(83): ApiGen\Parser\Elements\ElementStorage->getNamespaces()
#10 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Generator/GeneratorQueue.php(66): ApiGen\Generator\TemplateGenerators\NamespaceGenerator->isAllowed()
#11 [internal function]: ApiGen\Generator\GeneratorQueue->ApiGen\Generator\{closure}(Object(ApiGen\Generator\TemplateGenerators\NamespaceGenerator))
#12 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Generator/GeneratorQueue.php(71): array_filter(Array, Object(Closure))
#13 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Generator/GeneratorQueue.php(81): ApiGen\Generator\GeneratorQueue->getAllowedQueue()
#14 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Generator/GeneratorQueue.php(37): ApiGen\Generator\GeneratorQueue->getStepCount()
#15 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Command/GenerateCommand.php(184): ApiGen\Generator\GeneratorQueue->run()
#16 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Command/GenerateCommand.php(151): ApiGen\Command\GenerateCommand->generate(Array)
#17 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/vendor/symfony/console/Symfony/Component/Console/Command/Command.php(257): ApiGen\Command\GenerateCommand->execute(Object(ApiGen\Console\Input\LiberalFormatAr
gvInput), Object(Symfony\Component\Console\Output\ConsoleOutput))
#18 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/vendor/symfony/console/Symfony/Component/Console/Application.php(874): Symfony\Component\Console\Command\Command->run(Object(ApiGen\Console\Input\LiberalForma
tArgvInput), Object(Symfony\Component\Console\Output\ConsoleOutput))
#19 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/vendor/symfony/console/Symfony/Component/Console/Application.php(195): Symfony\Component\Console\Application->doRunCommand(Object(ApiGen\Command\GenerateComma
nd), Object(ApiGen\Console\Input\LiberalFormatArgvInput), Object(Symfony\Component\Console\Output\ConsoleOutput))
#20 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Console/Application.php(55): Symfony\Component\Console\Application->doRun(Object(ApiGen\Console\Input\LiberalFormatArgvInput), Object(Symfony\Component\Co
nsole\Output\ConsoleOutput))
#21 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/vendor/symfony/console/Symfony/Component/Console/Application.php(126): ApiGen\Console\Application->doRun(Object(ApiGen\Console\Input\LiberalFormatArgvInput),
Object(Symfony\Component\Console\Output\ConsoleOutput))
#22 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/src/Console/Application.php(64): Symfony\Component\Console\Application->run(Object(ApiGen\Console\Input\LiberalFormatArgvInput), NULL)
#23 phar://D:/WebProject/SPHERE-Framework/Library/MOC-V/Component/Documentation/Vendor/4.1/apigen.phar/bin/apigen(35): ApiGen\Console\Application->run()
#24 D:\WebProject\SPHERE-Framework\Library\MOC-V\Component\Documentation\Vendor\4.1\apigen.phar(10): require('phar://D:/WebPr...')
#25 {main}
Unable to log error: Directory is not specified.
`
Answers:
username_1: Can you clarify what you mean by `main` option?
username_0: [see usage instructions](https://github.com/ApiGen/ApiGen/tree/4.1#1-as-a-phar-recommended)
There is an option available called `main`, which is now referred to be an array and is (was?) used to specify the main project name prefix
username_1: I don't think that option is supported in 4.2. It was only available in pre 4.1 version. Your link also links to `4.1` branch.
Suggesting to close the issue.
username_0: on unsupported config options but a more "meaningful" message instead perhaps? ;-)
---
did no test on 4.2 with this .. possibly already fixed?
Status: Issue closed
username_2: Closed. PHAR not supported since #788. |
bitcoin/bitcoin | 993647425 | Title: ORO COIN
Question:
username_0: **Is your feature request related to a problem? Please describe.**
<!-- A clear and concise description of what the problem is. Ex. I'm always frustrated when [...] -->
**Describe the solution you'd like**
<!-- A clear and concise description of what you want to happen. -->
**Describe alternatives you've considered**
<!-- A clear and concise description of any alternative solutions or features you've considered. -->
**Additional context**
<!-- Add any other context or screenshots about the feature request here. -->
ORO COIN
Answers:
username_0: ![Uploading FB_IMG_16311421746539698.jpg…]()
username_0: Burqaa coin
Status: Issue closed
|
SpectraLogic/ds3_net_sdk | 169654345 | Title: Fully Qualified Filename Length Limitation
Question:
username_0: Due to .Net's limit of 248 character paths, despite NTFS supporting 32000, I am getting exceptions when I call Ds3.Helpers.Ds3ClientHelpers.StartWriteJob.
In my own code, I have addressed this issue by using Delimon.Win32.IO; instead of System.IO.
Is it possible to resolve this here instead of me maintaining a patch set for ds3_net_sdk?
Answers:
username_1: Are you getting the exception when using the StartWriteJob function or when using job.Transfer?
Can you paste here a snippet of your failing code?
username_0: It's actually in the FileHelpers class that I was getting exceptions.
Ds3/Helpers/FileHelpers.cs
It takes a while to get the exception, so this screenshot is the best I can provide at the moment.
I'm currently testing the attached patched version of FileHelpers.cs. Sorry for being lazy and not forking, branching, and pull-requesting.
[FileHelpers.cs.txt](https://github.com/SpectraLogic/ds3_net_sdk/files/404638/FileHelpers.cs.txt)
username_1: Our file helpers has the .Net limitation concerning the number of chars, and it looks like your new file helpers could be your solution in this case.
Let me know how it goes
username_0: So, yes moving to the other library worked just fine for me. I've just broken that code out into a new helper class to avoid having to maintain patches. Thanks for confirming the .Net limitation within this library.
Status: Issue closed
username_1: Sure :+1: |
elischutze/grrrrl | 193321977 | Title: List of features and wireframes
Question:
username_0: Hi Eli, I really like your idea and love that you've invited others to collaborate. Do you think there should be an issue to suggest features? After that, some wireframes could be submitted?
Answers:
username_1: sounds good to me!
username_0: Okay! So going by the README, the idea is to have two pages:
Search page:
- search box
- search results
Discover page:
- tech talks by topic (users select a topic and are presented with related talks)
Suggestion (contributors, feel free to suggest otherwise!):
Have 4 pages:
Home page/Landing page:
- Search box
- Short description of what Grrrrl is
Discover page:
- tech talks by topic (this could be designed in a fun and engaging way. I don't have amazing design skills though 😂 )
Results page:
- Display of results based on search string or selected topic
Submit a talk page:
- A form to submit a talk. Input fields could include: Talk name, URL of talk, description, topic
I can submit a pull request to add initial wireframes
username_1: Yeah, that sounds great. Definitely important to have a submission page!
username_0: Ah, just realised some of the other issues already mentioned submission page. @username_1 please could you add me as a contributor so that I can submit a pr? |
Glitchfiend/ToughAsNails | 329335926 | Title: Thirst status effect is pointless
Question:
username_0: The thirst status effect seems completely pointless. Water being infinite, you can just keep drinking until your bar is filled and the effect will go away leaving you with plenty of water left. One option to fix this is to have the thirst effect prevent you from drinking and make sure the thirst effect takes off more water than you restore from drinking the bad water, or replace it with other status effects like long duration nausea, slowness, etc. representing that you are getting sick off of the water.
Status: Issue closed
Answers:
username_1: It works the same as the vanilla Hunger effect that you get from eating raw chicken/rotten flesh. There are issues with it on older versions (Before 1.12.2) though, where the effect didn't work properly. |
CodeForBoulder/CfB-Website-New | 153154321 | Title: look into meetup widget for 'join us' section
Answers:
username_1: FWIW, I have started looking into this. Meetup has a [Widget Foundry](http://www.meetup.com/meetup_api/foundry/) that will provide the desired functionality. I have something working locally, but getting it to work without an API key has been the current hurdle. |
wmulroy/Kotlin_Fun | 640663808 | Title: Screenshots
Question:
username_0: 
 |
antlr/antlr4 | 183228633 | Title: XPathLexer.java missing from src?
Question:
username_0: When compiling:
antlr4-master\runtime\Java\src
I get the following compilation error:
antlr4-master\runtime\Java\src\org\antlr\v4\runtime\tree\xpath\XPath.java:
Error:(87, 17) java: cannot find symbol
symbol: class XPathLexer
location: class org.antlr.v4.runtime.tree.xpath.XPath
Answers:
username_1: Hi,
you are missing the maven build step which generates these files.
Eric
username_2: Yeah, see https://github.com/antlr/antlr4/blob/master/doc/building-antlr.md and run `mvn -DskipTests install`
Status: Issue closed
|
ritsrivastava01/ngxFlip | 357218009 | Title: Make module ready for AOT
Question:
username_0: First of all, thanks for the module :) I had a custom made library doing the same, but this one is better tested and less glitchy than mine, so I'm swapping mine by yours!
Anyways, I've been trying to add it in my project and it works flawless. However, when building i production mode with AOT enabled, the build crashes.
The reason seems to be that this module is not AOT ready and it only works in JIT mode. I've been doing a bit of research and it seems it only changing some build configuration:
https://blog.mgechev.com/2017/01/21/distributing-an-angular-library-aot-ngc-types/ (check AOT section).
At the moment I'm using this module for development, but I'd be willing (when I get/if I can get some time) to help to do this change in the build!
Answers:
username_1: Nice to hear that everything is working good, it's also thankful thought that you want to improve this lib.
Thanks in advance for your efforts and finding.
By this way we can improve this lib day by day.
Thanks a lot.
--Ritz
username_0: Hi @username_1 ,
Small updates (note down that I'm a bit confused).
I tried cloning your library and compiling it localy. It builds fine though reported version is 6.0.0 and package name is `@ritsrivastava/ngxflip` instead of ngx-flip. It should be ok and still work fine.
I tried installing it from my local build (`npm install ../ngxFlip/dist/ngxflip/ritsrivastava-ngxflip-6.0.0.tgz`), tried building my project (changed the library to be loaded from the new package name) and.... voila! Works perfect with AOT enabled.
I tried again uninstalling locally made version and using the online one, and I get again the same error.
I think (and it's my best guess), that there is something wrong with the naming inside of the library. if you check the error I get (you can reproduce just creating an empty project and using the library that is server from NPM) you read the following:
```
ERROR in ./src/app/app.module.ngfactory.js
Module not found: Error: Can't resolve 'ngxflip' in '/Users/sturgelose/git/frontend/apps/dash/src/app'
ERROR in /Users/sturgelose/git/frontend/libs/cards/src/lib/card/card.component.ngfactory.js
Module not found: Error: Can't resolve 'ngxflip' in '/Users/sturgelose/git/frontend/libs/cards/src/lib/card'
ERROR in /Users/sturgelose/git/frontend/node_modules/ngx-flip/ngxflip.ngfactory.js
Module not found: Error: Can't resolve 'ngxflip' in '/Users/sturgelose/git/frontend/node_modules/ngx-flip'
```
It seems it's trying to resolve `ngxflip` when the package name is `ngx-flip` (with the dash). May this be a small missconfiguration in the namings inside the package?
Mostly because the package name has the dash and all the files inside of it are missing it (maybe some magic from the compiler, though...)
So in sum, it isn't an issue with AOT precompiling, it's more with a naming issue somewhere that when using the --prod option and runs AOT it fails. I tried to attach the tgz build I created myself just so you can check, but seems this format is not allowed here....
Any ideas what could happen? Thanks again!
username_1: Thanks, @username_0,
I'll try to create a sample project with online lib and then try to build--prod.
let see how I can resolve it. :)
username_2: I am also facing the same issue on prod build. It seems to be a mismatch in the naming that causes the error.
username_1: Dear @username_0 and @username_2,
Thanks for your input, I got some time this weekend to work on this issue and I changed the package name (previously it was '@username_1/ngx-flip', now it's only 'ngx-flip').
Pls let me know if it working for you OR not.
Hope its working for you.
--Ritz
Status: Issue closed
username_0: @username_1 I can confirm it's fixed :) 👍 Thanks a lot! |
eggheadio/gatsby-starter-egghead-blog | 400335868 | Title: Example site?
Question:
username_0: Do you perhaps have an example of this starter hosted on Netlify or something? I'm curious what it looks like and being able to see it before setting it up locally is great!
Answers:
username_1: would be nice to see in the sample how at least thing should be setup. it does not show even after cloning.
username_2: I just forked and deployed it on Netlify. You can check it out there: [https://nostalgic-swartz-ab79c4.netlify.com/demo-01](https://nostalgic-swartz-ab79c4.netlify.com/demo-01)
username_3: Will add this to the list and get an example up and running on Netlify.
username_3: https://egghead-gatsby-starter.netlify.com/
Status: Issue closed
|
sqlalchemy/sqlalchemy | 631628183 | Title: proposal: I don't think we really need sqlalchemy.future as a public import space
Question:
username_0: As I've gone through and implemented the biggest parts of the 2.0 architecture in 1.4, it turns out there is almost nothing in sqlalchemy.future. I would propose that some kind of "future.engine" package remain present as a place for basiclally the new engine API to go, but that "future" style operation only need be calling create_engine() with a "future=True" flag. the rest of the API seems to be functional without any "future" need, and at most perhaps the ORM Session would enable/disable "future" mode with a similar future=True flag sent to sessionmaker().
The reason it would be nice to not have users import from "sqlalchemy.future" would be that we don't have to worry about copying the entire import space into these packages, and it would continue to give us room to adjust the import namespaces between now, 1.4 and 2.0 without an abrupt change. addtionally, I think it's going to be really confusing to have two different select() functions that are 99% the same except for subtle behavioral differences.
The major change that has led to this is that the select() construct has a "plugin" model now whereby it can call into orm-specific functions once it has the so-called "orm plugin". I originally wasn't sure if this model was going to actually end up being used but it's turned out to be the best way to have "select()" with a few ORM-specific behaviors. the end result here is that the "select()" and "future_select()" are nearly the same, future_select() just adds some new methods, and the new constructor which we want to unify in #5284 . the only method that changes is join() / outerjoin(), which is a method that has always done something useless (creates a subquery and joins to it) and at the moment in 1.4 the base form of join() just raises an error. so this would be one "backwards incomptible" bump for a function that nobody should be using anyway.
the "Removed in 2.0" warnings are now turned on/off by a single environment-style variable in https://github.com/sqlalchemy/sqlalchemy/blob/master/lib/sqlalchemy/util/__init__.py#L152 which we just need to hook into os.environ().
for the engine, the newer API is in sqlalchemy.future.engine, and it is essentially the rollback() and commit() and new execute() methods, which are one-liners, and the Engine which sets up as no-ops all the old methods. then there's a bunch of nicer docstrings. So it is nice to have a place for the new Connection / Engine in terms of documentation but there's nothing really to "import" except for create_engine().
the ORM Session has a big behavioral change which is that session.execute() will do ORM things if you pass it select/update/delete objects with ORM elements in them. this is probably fine for everyone, but in case it's not we can also do a future=True mode for Session / sessionmaker().
in ORM there is otherwise no new API in the plan. The one element was how do to query(User).from_statement(statement), but thanks to the ORM plugin model I think I'm going to put it right on select(), select(User).from_statement(statement).
Answers:
username_0: there's also the new declarative base, but I was planning to make this API shift in 1.4 in any case that declarative_base() can from from sqlalchemy.orm.
username_1: **<NAME>** has proposed a fix for this issue in the **master** branch:
**Add future=True to create_engine/Session; unify select()** https://gerrit.sqlalchemy.org/c/sqlalchemy/sqlalchemy/+/2058 |
polarsignals/issues | 844984106 | Title: Same series shortcut when selecting profiles to compare
Question:
username_0: When I select one profile and I want to compare it with another profile from the same series, I have to enter the series matchers manually.
It would be good to have a button with text like `compare with the same series` so that the input can be filled with the same matchers by clicking it.
Answers:
username_1: Yeah, I 100% agree. I have even thought of just defaulting directly to the same label set as the first selection, or the same query as what was used for the first selection. Would you have a preference?
username_0: I prefer the latter. But either way is good to me!
username_1: Ok awesome! I hope to find some time in the next weeks to do this. Thank you for your feedback!
Status: Issue closed
username_0: I think this is done. Thanks @metalmatze |
jonbhanson/flutter_native_splash | 1169843370 | Title: SVG/PNG asset support
Question:
username_0: can you guys add svg/png format support
i mean we can use a logo with SVG file instead of PNG
Status: Issue closed
Answers:
username_1: Please add your request to [this thread](https://github.com/username_1/flutter_native_splash/issues/245) |
quasarframework/quasar | 615180333 | Title: QStepper Label Slot
Question:
username_0: Add a slot to the QStepper for the Label (Title and Caption).
Similar to Vuetify: https://vuetifyjs.com/en/components/steppers/
Answers:
username_1: May I ask why a slot for title and caption are necessary? What is your use-case?
Status: Issue closed
username_1: Quasar's stepper is a bit more powerful than the one indicated by you because it supports both vertical and horizontal modes **with same html template**. This however hinders the ability to use a "title" and "caption" slots. However, since these two props can be dynamically binded, there is little reason to add slots for them. This is why I was asking about your use-case. Sorry for any inconveniences. If I get some details on your use case then I can point you in the right direction on how to achieve what you want.
username_0: I'd like to add buttons and other things into the Step label
username_0: @username_1
username_1: Hi,
Due to the reasons I mentioned above, QStepper can't support the slots that you are mentioning. However, why not use QTabs + QTabPanels for this? |
abcdefgs0324/issue-label-bot-test | 490695546 | Title: Presubmit failure: 404 when fetching workflow
Question:
username_0: example:
https://prow.k8s.io/view/gcs/kubernetes-jenkins/pr-logs/pull/kubeflow_kubeflow/4013/kubeflow-presubmit/1166833694397173760/
#4013
All the tests seem passed, but it's trying to wait for a non-existing workflow deployapp,
#4013 (comment) |
zalando/postgres-operator | 801218290 | Title: pod_environment_secret is not passed along to backup cronjob
Question:
username_0: Hi,
we are currently using the backup functionality of the operator, which creates a cronjob within kubernetes that regularly executes logical backups as described [here ](https://postgres-operator.readthedocs.io/en/latest/administrator/#logical-backups)
For this we can pass along _logical_backup_s3_secret_access_key_ in order to provide access to the storage. However this is currently passed in a configmap which is insecure. We therefore tried to use the _pod_environment_secret_ mechanism in order to mount a custom secret which holds the credentials necessary upload the backup to storage.
It turned out that the secrets configured with _pod_environment_secret_ are only considered / mounted on the operator itself. They are not present on the cronjob created by the operator to do the backups. Which leaves the whole backup mechanism insecure in terms of credential management.
I would suggest to either consider the _pod_environment_secret_ also for creating the cronjob or to introduce a new variable specifically used for the logical backup.
- **Which image of the operator are you using?**
---> registry.opensource.zalan.do/acid/postgres-operator:v1.6.0
- **Type of issue?** [Bug report, question, feature request, etc.]
--> BugReport / feature request
Answers:
username_1: Same. I was wondering if it is possible to pass `pod_environment_configmap` to logical backup cron jobs so that I can need to hardcode the S3 credentials to the `configLogicalBackup` section. Is it supported? |
utexas-bwi/segbot | 157284998 | Title: segbot_sensors: SegbotVelodyneOutlierRemovalConfig.h missing dependency
Question:
username_0: This is the printout error:
/home/users/szhang/catkin_ws_scavenger/src/segbot/segbot_sensors/include/segbot_sensors/segbot_velodyne_outlier_removal.h:60:63:
fatal error: segbot_sensors/SegbotVelodyneOutlierRemovalConfig.h: No such file or directory
#include <segbot_sensors/SegbotVelodyneOutlierRemovalConfig.h>
Answers:
username_0: @piyushk @maxsvetlik
username_1: Probably a missing dependency on the gencfg output (or something like that).
Things like that might work on one machine and fail on another due to build parallelism and the number of processors available.
username_0: @username_1 Do you mean the issue comes from our use of dynamic reconfiguration or you are pointing to a solution? Sorry that I am new to dynamic reconfiguration.
username_1: I was gone Sunday and Monday. I'll take a look at it today.
I am just guessing what the problem may be.
username_1: I tried to reproduce this problem in nibbler, but it built without error. I'll try another lab machine.
username_1: Still working for me. I'll try to find the problem tomorrow via code review.
username_0: It builds well on the robots, but not on the lab machines. I can try again tomorrow on the lab machines and provide more detailed info. Thanks! @username_1
username_1: I found the bug, fixing it now.
Status: Issue closed
username_1: I am pretty sure c95eff2 fixes the problem.
Please re-open if I am wrong and it happens again. :smile:
username_0: @username_1 Thanks! It compiles on the lab machines now. |
http4s/http4s | 213222909 | Title: Host contributors' guide on http4s.org
Question:
username_0: We currently link to GitHub for `CONTRIBUTING.md`. Look and feel would be more consistent if it was copied to Hugo site similar to `CHANGELOG.md` [in build.sbt](https://github.com/http4s/http4s/blob/v0.15.6/build.sbt#L304).
Answers:
username_1: I've also considered slimming down contributing.md and making a developers' guide. I'm open to either idea.
username_1: We [have this](https://http4s.org/contributing/) now.
Status: Issue closed
|
ME3Explorer/ME3Explorer | 156649014 | Title: Packaged Editor exports start at 0 instead of 1... issue?
Question:
username_0: The exports list in package editor start at index 0. However, values in interpreter are indexed starting at 1. This leads to a ton of confusion and I'm not sure if it's correct behavior or an oversight we've had for some time.
Answers:
username_1: UE3 indexes the exports starting at 1, and the imports starting at -1, with 0 meaning Class. So one signed int can represent any export or import. These are the values displayed in the interpreter.
When ME3Explorer reads the PCC, it puts all the exports in one array and the imports in another. So the number shown in the export list is just the index in that array. I'm pretty sure this is a prime example of an implementation detail that should not be exposed to the user. Fixing it at this point would be a huge task though, as it's like that all over the toolset, not just in the PackEd. It would also probably be very confusing to all the users who are used to it by now.
username_0: Only reason I brought it up is when I am looking up a value to put in interpreter (in this case I am making geth prime use pulse rifle instead of his gethprimerifle), the indexes are all off by 1 in the exports list which really irritating. When I look in interpreter of the gun I want to use to find values, its the right values, but its sooooooo easy to mix them up and I end up wasting a lot of time fixing them.
username_2: I was going to bring this up during the re-write, @username_1.
I don't think it's something we want to address yet, but when it comes to overhauling the interface and functionality of Package Editor with the new GUI implementation, I think we're going to want to address it.
Over the past 6 months we've added in a bunch of new editable/informational fields in Package Editor, and it has indeed gotten very confusing. These types of changes are the types that I'd like to see us discuss more on the forums in those new dev threads, as while more information seems like it should be a no-brainer, it isn't always.
This problem is why I recently got very confused during a exchange about NPC editing and archetypes. I think You and Kinko were in on this conversion. There's now _multiple_ areas in Package Editor where sometimes it's +1 and sometimes it isn't. Trying to keep them all straight is getting harder and it's very, very easy to forget and make a mistake. Before all the new fields were added we didn't have the problem as much. Generally it was all just +1.
If we can start *thinking* right now about how we could deal with this, that would be a good thing. But, I don't think anything needs to happen until we overhaul the Package Editor GUI.
Status: Issue closed
|
potree/PotreeConverter | 245297122 | Title: Build on OS X Fails
Question:
username_0: attempting to build 1.31 on OS X.
have installed CMAKE (command line utility), have installed and built LASTOOLS.
fails due to not finding 'experimental/filesystem'
any help would be greatly appreciated. thanks,
`Scanning dependencies of target PotreeConverter
[ 8%] Building CXX object PotreeConverter/CMakeFiles/PotreeConverter.dir/src/BINPointReader.cpp.o
clang: warning: -Z-reserved-lib-stdc++: 'linker' input unused
clang: warning: -lm: 'linker' input unused
/Users/bettercallsaul/dev/workspaces/username_1converter/master/PotreeConverter/src/BINPointReader.cpp:6:10: fatal error:
'experimental/filesystem' file not found
#include <experimental/filesystem>
^
1 error generated.
make[2]: *** [PotreeConverter/CMakeFiles/PotreeConverter.dir/src/BINPointReader.cpp.o] Error 1
make[1]: *** [PotreeConverter/CMakeFiles/PotreeConverter.dir/all] Error 2
make: *** [all] Error 2`
Answers:
username_1: What compiler are you using? Should be working with the
username_0: I have not specified the compiler. Just followed the instructions from the home screen https://github.com/username_1/PotreeConverter
I am a bit of a noob. I have installed homebrew and cmake. perhaps this is the issue, the compiler is not specified.
username_0: I am using apple clang 8
-- The C compiler identification is AppleClang 8.0.0.8000042
-- The CXX compiler identification is AppleClang 8.0.0.8000042
-- Check for working C compiler: /Library/Developer/CommandLineTools/usr/bin/cc
-- Check for working C compiler: /Library/Developer/CommandLineTools/usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /Library/Developer/CommandLineTools/usr/bin/c++
-- Check for working CXX compiler: /Library/Developer/CommandLineTools/usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Configuring done
-- Generating done
-- Build files have been written to: /Users/bettercallsaul/dev/workspaces/PotreeConverter/master/build
Status: Issue closed
username_2: sorry to bother you.
how is this issue fixed?
I'm having the same error ...
username_0: no worries, I was using a super old Macbook pro. 2009 I believe.
anyways our older mac's don't support it (OSX version). so the fix was not using a super old OSX. |
rocketeers/rocketeer | 63255543 | Title: Wrong tree on explainer lines before a connection is made
Question:
username_0: This is easiest to reproduce by having a multi-server connection and running the same command twice, with a before hook on it. On the second run, the explainers first line will include the handle of the last connection on the previous run. For example, if you have `server a` and `server b`, it will show `server b => task (description)` as the first explainer'd line of output, even though it hasn't made a connection yet.
Answers:
username_1: Hm can't seem to reproduce this one, did a small self-contained test:
```php
$this->swapConnections([
'production' => [
'servers' => [
['host' => 'foo.com'],
['host' => 'bar.com'],
]
],
]);
$this->tasks->before('Update', function (AbstractTask $task) {
$task->explainer->line('BEFORE');
});
$this->queue->run('Update');
```
And output was correct:
```
[anahkiasen@production/foo.com] BEFORE
[anahkiasen@production/foo.com] Pulling changes
[anahkiasen@production/foo.com] [...]
[anahkiasen@production/bar.com] BEFORE
[anahkiasen@production/bar.com] Pulling changes
[anahkiasen@production/bar.com] [...]
```
Is this still occuring?
username_0: Add a hook before the deploy *command*, see if that works
username_1: With a command it seems correct too, with the only difference that the before hook is only fired once (but that's normal):
```
|-- Closure fired by rocketeer.commands.update.before
|===> BEFORE
@production/foo.com | Update (Update the remote server without doing a new release)
@production/foo.com |-- Primer (Run local checks to ensure deploy can proceed)
@production/foo.com |-- Deploy/Clone (Clones a fresh instance of the repository by SCM)
@production/foo.com |=> Pulling changes
@production/foo.com |=> Sharing file /Users/anahkiasen/Sites/rocketeers/rocketeer/tests/TestCases/../_server/foobar/releases/20000000000000/tests/Elements
@production/foo.com |-- Dependencies (Installs or update the dependencies on server)
@production/foo.com |---- Dependencies/Polyglot (Runs all of the above package managers if necessary)
@production/foo.com |=> Setting permissions for /Users/anahkiasen/Sites/rocketeers/rocketeer/tests/TestCases/../_server/foobar/releases/20000000000000/tests
@production/foo.com |=> Successfully updated application
@production/bar.com | Update (Update the remote server without doing a new release)
@production/bar.com |-- Primer (Run local checks to ensure deploy can proceed)
@production/bar.com |-- Deploy/Clone (Clones a fresh instance of the repository by SCM)
@production/bar.com |=> Pulling changes
@production/bar.com |=> Sharing file /Users/anahkiasen/Sites/rocketeers/rocketeer/tests/TestCases/../_server/foobar/releases/20000000000000/tests/Elements
@production/bar.com |-- Dependencies (Installs or update the dependencies on server)
@production/bar.com |---- Dependencies/Polyglot (Runs all of the above package managers if necessary)
@production/bar.com |=> Setting permissions for /Users/anahkiasen/Sites/rocketeers/rocketeer/tests/TestCases/../_server/foobar/releases/20000000000000/tests
@production/bar.com |=> Successfully updated application
```
username_0: Hmm. I've got a tree on the first line. Suppose this is an issue in my fork. Future issues will include a reproducible test case :)
Status: Issue closed
|
phetsims/axon | 51487030 | Title: rename DerivedProperty.detach to dispose?
Question:
username_0: I've been following the scenery/scenery-phet/sun convention of using 'dispose' as the name of the function that does cleanup at the end of an object's lifetime. Should we rename DerivedProperty.detach to dispose, so that it's consistent with this naming convention?
Assigning to @username_1 as the author of DerivedProperty.detach, but other developers please chime in.
Answers:
username_0: @jbphet @username_1 @username_2 What do you think of changing AXON to use `dispose` rather than `detach`?
username_1: I like that the same name gives a sense of symmetry to the Axon/Sun APIs, but practically it also means that searching for all usages of axon.dispose will be more difficult because sun.dispose will show up in text searches as well. Not sure what is best here. I'd like to hear from others.
username_0: Yes, I'd like to hear from others too. @jbphet and @username_2 again, your feedback is requested.
username_2: From the words alone, I would have the intuition "I can't use the derived property after dispose()", but "I could use the derived property after detach(), but it just won't update from its dependencies".
If it's truly the "end of lifetime" for the property (i.e. don't call property.value afterwards), then I slightly prefer "dispose". Otherwise "detach" seems to work better.
username_0: We discussed either leaving as is, or changing to dispose and removing all of its listeners. Consensus was to leave as is.
Status: Issue closed
username_1: I've been following the scenery/scenery-phet/sun convention of using `dispose` as the name of the function that does cleanup at the end of an object's lifetime. Should we rename `DerivedProperty.detach` to `dispose`, so that it's consistent with this naming convention? (And rename similarly for other AXON APIs.)
Assigning to @username_1 as the author of DerivedProperty.detach, but other developers please chime in.
username_1: I should have been a bit more vocal at the meeting--though I do not feel very strongly about it, I do think it would be a good move to switch to dispose() but that this is not a very high priority issue. Thoughts?
username_2: I don't mind either way.
username_1: @username_0 are you ok to leave this as a "to do" item at low priority as described above?
username_0: Yes, ok leaving low priority.
username_1: I'm about to add Property.dispose and PropertySet.dispose. Note that renaming DerivedProperty.detach=> dispose will have a conflict after I add those.
username_1: I added PropertySet.dispose and Property.dispose above.
Wondering if this could cause any problems in existing PropertySet subclasses that also declare dispose. (Other than the fact that they would probably want to call super.dispose()). @username_0 do you think this will be a problem?
username_0: In an uninstrumented sim, pairing of link/unlink and new/dispose is required to prevent memory leaks. So when instrumenting code that has properly paired link/unlink and new/dispose, Property.dispose should have to remove 0 observers. If it removes any observers, then it's masking a memory leak. So I think that calling dispose should throw an assertion if there are any observers registered for the Property. Similarly for PropertySet.
username_1: @username_3 are you available to familiarize yourself with this thread, see if asserts make sense there, and test a few sims out to see if they are throwing assertion errors for this case?
username_3: I can take a look. What's the priority of this? Next week is the earliest I could get to it.
username_1: Anything within 2 weeks would be fine. If you don't get time before then, just let me know. Thanks!
username_4: Looks like this is an old issue. @username_1 can you take care of, close, or reassign (not to @username_3 ) if further action is needed.
Status: Issue closed
username_1: The main problem in this issue has been addressed, the reason the issue is still open is described in https://github.com/phetsims/axon/issues/38#issuecomment-89396284 . I'll create a separate issue for that and close this one. |
rit-sse/rapdev2015-api | 53530410 | Title: Event Tagging
Question:
username_0: This issue involves implementing tagging for events. To complete this issue the following should be accomplished:
* Implement the /api/event/{id}/tags endpoint
GET - Get all of the tags for a event item
PUT - Add a tag to a event item
DEL - Remove a tag from a event item |
ultimachine/Mini-Rambo | 401371562 | Title: Failure to detect open-circuit thermistor
Question:
username_0: I don't know if this is the right place to report this but I have found an issue with the design of mini-Rambo board supplied with original Prusa i3 Mk2s where the board would not detect an open-circuit heat-block thermistor, thus risking overheating or worse.
The problem is caused by the voltage on zener diode D12 being too low due to insufficient zener current and/or device tolerances. On my board the voltage was about 4.4V rather than 4.7.
A new diode did not help but increasing the zener current to just over 3mA by reducing the value of the 10K dropper gives the correct voltage. The thermistor failure is now detected.
I started a thread in the Prusa Research support forum here:
https://shop.prusa3d.com/forum/hardware-firmware-and-software-help-f44/danger-extruder-thermistor-failsafe-didn-t-work-t27201.html |
CliMT/climt | 332105248 | Title: Use base class docstrings
Question:
username_0: Currently there are many component docstrings on __init__, __call__ or array_call methods which don't really say much of anything useful that the base class docstrings don't already say. For those components, we should remove the docstring entirely so that Sphinx will use the base class docstrings (which are written to be generic enough to apply to any component). The important part of the documentation is present in the *_properties dictionaries of the components.
We also could think about a way to include those property dictionaries in the documentation, but that's a separate issue.
Answers:
username_1: In most components `__init__` always has quite detailed docstring. I agree that we should eliminate docstrings for `__call__` which don't really do much.
username_0: Yes, there are just a couple `__init__` methods with no kwargs where I'll remove the docstring. |
rodikno/Automation_course | 208167611 | Title: You have somthing
Question:
username_0: Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Shd';sadl a';dlsa';lsa';dl';sld';sad';asl';saas;as'';ALD;'SA'SA'sa';\d;as
Answers:
username_1: тебе ж на ужин надо идти
что ты тут делаешь?
username_0: @username_1 это я пытался доказать товарищу Родиону, что форма не лочится при сабмите )
username_1: @username_0
так а как же ужин? :)
username_0: ужин - done :)
ишью можно закрывать)
username_2: Verified on one glass of IPA, one glass of Whisky, some meal and dessert.
Status: Issue closed
|
coreos/go-oidc | 798469620 | Title: Support: Example docs, next steps...
Question:
username_0: I am trying to use this go-oidc v3 in gin-gonic. I started to create "middleware" based on the [examples](https://github.com/coreos/go-oidc/tree/v3/example) included here. This was what I came up with initially: https://github.com/username_0/gin-gonic-oidcauth/tree/basic_insecure.
As the branch name states, that is a basic_insecure example, but it is really not useful for authentication, so...
I then did a bit of research and determined that "nonce" and "state" should definitely not be static strings. It seemed like "state" should be something specific to the client (user's browser), and should thus be stored in the session. The nonce *could* be stored in the session as well, but I found https://pkg.go.dev/github.com/letsencrypt/boulder/nonce which seems to implement "nonce as a service" :p ... and it seems to be working as desired. I am actually just using a nonce for both the nonce and the state. I am not really sure whether thats good or bad :) ... Anyhow, now I am here: https://github.com/username_0/gin-gonic-oidcauth/tree/main ... which I think is a "better" (more secure at least) example, but it is still a contrived example, that dumps a bunch of information into the session, and is able to change from "Hello, world" to "Hello, NAME" ...
Now, I feel like the initial authentication is handled (for the callback), but I am not sure how to handle subsequent requests from the user.
1. What "token" should I store for the user (in the session). The `idToken.AccessToken()` is what I am currently storing, but I am not sure how to validate it. (I guess I could just "assume" that the encrypted/authenticated session has not been tampered with, but that seems bad?)
2. Should I do any token validation on subsequent requests, or just store the "iat" and "exp" (currently storing all claims), and validate that they are still valid.
3. Are there any examples about "refresh" tokens? (renewing the token from the app while the user is still active)
Thanks in advance,
Tommy
Answers:
username_1: Hey Tommy,
These seem like general questions about OAuth 2.0 security and session management, not necessarily a bug in this library.
Unfortunately, the maintainers of this package generally don't have bandwidth to answer support questions like this. Below are some links that might help getting started:
* https://auth0.com/docs/protocols/state-parameters
* https://cheatsheetseries.owasp.org/cheatsheets/Session_Management_Cheat_Sheet.html
* https://openid.net/specs/openid-connect-core-1_0.html#NonceNotes
Hope this helps,
Eric
Status: Issue closed
username_0: Thanks for responding @username_1! As per the title of my issue I was asking for support, and that is what you offered. Given the information contained in the documents that you linked (by the way the first and third ones were referenced over the weekend), you have answered several questions that I already solved (state/nonce) that were not part of the example code. The owasp information on sessions is probably the information I will need to solve my issue with validating tokens later in the user's session. I am going with the idea that I have validated the user's token one time, and the user's possession of the session ID is as they say "equivalent" to authentication (temporarily).
I do propose that the examples should be a little more "secure" by design. The example code says to "not" do this in production, but it doesn't really give any direction or hints about what we "should" do. :) Perhaps this ticket will help someone in the future.
Thanks again,
Tommy |
adazzle/react-data-grid | 158927291 | Title: NODE_ENV not replaced in minified version
Question:
username_0: The file I am including is `dist/react-data-grid.min.js`.
Would it be possible for you to deliver the minified versions with NODE_ENV properly replaced? Or can I as a consumer of your library do something about it in my build process?
Thanks in advance!
Answers:
username_1: This happened when we added the code for context menu. Rather than bundling all of the redux library, I think we will remove this from the main dist file and have the code for this in another `.js` file. I'll also set the NODE_ENV variable correctly
username_0: @username_1 awesome, thanks! Looking forward to it 👍
Status: Issue closed
username_0: Just updated to version 1.0.4 which seems to include the fix!
username_0: The file I am including is `dist/react-data-grid.min.js`.
Would it be possible for you to deliver the minified versions with NODE_ENV properly replaced? Or can I as a consumer of your library do something about it in my build process?
Thanks in advance!
username_0: Ok, I was wrong: `react-data-grid.ui-plugins.min.js` still contains NODE_ENV variable in version 1.0.4.
@username_1 Any updates on this?
username_0: https://npmcdn.com/[email protected]/dist/react-data-grid.ui-plugins.min.js
Status: Issue closed
|
ArtemChubatyi/home-dsgn | 376787372 | Title: Studio and News blocks consume a lot of space on tablet
Question:
username_0: Try to make colorful blocks horizontal as a solution:
Answers:
username_1: @username_0 fixed
Status: Issue closed
|
ljlm0402/typescript-express-starter | 987395005 | Title: Custom api error
Question:
username_0: ### Summary (요약)
I wanna add errorType in response when api error.
### Additional Details (추가적인 세부 사항)
- I have added this field in HTTPException class
- But the response have only message.
<img width="1106" alt="Screen Shot 2021-09-03 at 11 50 25 AM" src="https://user-images.githubusercontent.com/82807964/131952067-4b430033-89d1-4a7a-8fb5-1906b3a0cc4c.png"><issue_closed>
Status: Issue closed |
odknt/bspwmbar | 689381697 | Title: bspwmbar exiting when system is put to sleep
Question:
username_0: Hi there!
bspwmbar seems to exit when the system goes to sleep, it is exiting cleanly, no error messages or bad return codes.
It does that regardless if it was launched from bspwmrc or manually.
Is this intended behavior?
Greetings,
Jan<issue_closed>
Status: Issue closed |
EA-Pods-Team/pdf-bullets | 560614227 | Title: case insensitive..
Question:
username_0: This is probably related to #7
But case shouldn't matter..
Answers:
username_1: For me it ended up being a non issue, because all words except proper nouns are required to be lower case unless they are an acronym.
Except when the word is the first word of the bullet, but my leadership has a policy of not abbreviating the first word as well.
Regardless, I am working on adding an option to enable case sensitive matching.
username_0: Good Point, I guess my example was very generic. "Weapons Officers" is a better example which breaks that rule (not sure why I chose to use "Military Officer" maybe because it showed 2 abbreviations..
But your fix action should account for that. |
color-book/web_server | 365164890 | Title: Optimize AddLineItem function
Question:
username_0: Currently we're looping through the line items in the api handler and calling `dataStore.Store.AddLineItem()` for each item. I feel like this might be a little inefficient? It just depends on how many line items are usually added.
It would be better if we could send the whole line item slice to the datastore and some how insert them all at once. |
snordgren/fjord | 434470844 | Title: Add support for importing other modules
Question:
username_0: See https://github.com/purescript/purescript/issues/2437 for an interesting perspective on modules.
import Prelude from "prelude-fj"
The `from`-part establishes that this is an *external dependency*, meaning it is not part of this project, and the `"prelude-fj"` part tells the compiler to look in the `prelude-fj/` directory for the modules.
import Prelude (..) from "prelude-fj"
I want two kinds of syntax: the first example includes everything in the Prelude module in the "prelude-fj" package as `Prelude.<name>`, while the second example includes everything without a prefix.
Status: Issue closed
Answers:
username_0: See https://github.com/purescript/purescript/issues/2437 for an interesting perspective on modules.
import Prelude from "prelude-fj"
The `from`-part establishes that this is an *external dependency*, meaning it is not part of this project, and the `"prelude-fj"` part tells the compiler to look in the `prelude-fj/` directory for the modules.
import Prelude (..) from "prelude-fj"
I want two kinds of syntax: the first example includes everything in the Prelude module in the "prelude-fj" package as `Prelude.<name>`, while the second example includes everything without a prefix.
Status: Issue closed
username_0: Basic imports are now supported. Closing to create more fine-grained issues. |
sodiumoxide/sodiumoxide | 379812061 | Title: libsodium-sys does not compiled on win10 for static library
Question:
username_0: ```shell
Compiling libsodium-sys v0.1.0
Running `c:\projects\tz\target\debug\build\libsodium-sys-85ac9c20faea7bd2\build-script-build`
Running `rustc --crate-name libsodium_sys C:\Users\rafael\.cargo\registry\src\github.com-1ecc6299db9ec823\libsodium-sys-0.1.0\src\lib.rs --crate-type lib --emit=dep-info,link -C debuginfo=2 -C metadata=54726e2c99f8b2f3 -C extra-filename=-54726e2c99f8b2f3 --out-dir c:\projects\tz\target\debug\deps -L dependency=c:\projects\tz\target\debug\deps --extern libc=c:\projects\tz\target\debug\deps\liblibc-bc70762130ffb1e9.rlib --cap-lints allow -Ctarget-feature=+crt-static -L native=C:\projects\sodium\x64\Release\v141\static -l static=sodium`
error: could not find native static library `sodium`, perhaps an -L flag is missing?
error: aborting due to previous error
error: Could not compile `libsodium-sys`.
Caused by:
process didn't exit successfully: `rustc --crate-name libsodium_sys C:\Users\rafael\.cargo\registry\src\github.com-1ecc6299db9ec823\libsodium-sys-0.1.0\src\lib.rs --crate-type lib --emit=dep-info,link -C debuginfo=2 -C metadata=54726e2c99f8b2f3 -C extra-filename=-54726e2c99f8b2f3 --out-dir c:\projects\tz\target\debug\deps -L dependency=c:\projects\tz\target\debug\deps --extern libc=c:\projects\tz\target\debug\deps\liblibc-bc70762130ffb1e9.rlib --cap-lints allow -Ctarget-feature=+crt-static -L native=C:\projects\sodium\x64\Release\v141\static -l static=sodium` (exit code: 101)
c:\projects\tz>rustc --version --verbose
rustc 1.26.0 (a77568041 2018-05-07)
binary: rustc
commit-hash: a7756804103447ea4e68a71ccf071e7ad8f7a03e
commit-date: 2018-05-07
host: x86_64-pc-windows-msvc
release: 1.26.0
LLVM version: 6.0
c:\projects\tz>
```
And sodium library is located like below.
The environment variables are following.
SODIUM_INC_DIR=C:\projects\sodium\include
SODIUM_LIB_DIR=C:\projects\sodium\x64\Release\v141\static
SODIUM_STATIC=1
Answers:
username_1: To build sodiumoxide using libsodium from vcpkg, you have to add either VCPKGRS_DYNAMIC=1 (for dynamic linking) or RUSTFLAGS=-Ctarget-feature=+crt-static (for static linking) to environment variables.
Please report if it does not help.
username_1: Huh. It worked fine for me even if I don't set `SODIUM_STATIC` or `RUSTFLAGS`. Maybe you should update rustc?
```
C:\Users\humbu\dev\sodiumoxide>cargo run --example simple
Finished dev [unoptimized + debuginfo] target(s) in 0.17s
Running `target\debug\examples\simple.exe`
nonce = Nonce([66, 73, 58, 215, 160, 166, 136, 146, 241, 14, 67, 229, 126, 104, 156, 248, 99, 142, 117, 255, 70, 217, 5, 171])
C:\Users\humbu\dev\sodiumoxide>echo %SODIUM_LIB_DIR%
C:\Users\humbu\dev\libsodium\x64\Release\v141\static
C:\Users\humbu\dev\sodiumoxide>echo %SODIUM_INC_DIR%
C:\Users\humbu\dev\libsodium\include
C:\Users\humbu\dev\sodiumoxide>echo %SODIUM_STATIC%
%SODIUM_STATIC%
C:\Users\humbu\dev\sodiumoxide>rustc --version --verbose
rustc 1.30.1 (1433507eb 2018-11-07)
binary: rustc
commit-hash: 1433507eba7d1a114e4c6f27ae0e1a74f60f20de
commit-date: 2018-11-07
host: x86_64-pc-windows-msvc
release: 1.30.1
LLVM version: 8.0
```
examples/simple.rs:
```
extern crate sodiumoxide;
use sodiumoxide::crypto::box_::gen_nonce;
fn main() {
let nonce = gen_nonce();
println!("nonce = {:?}", nonce);
}
```
username_2: Interestingly enough, from error in the build script:
```
-L native=C:\projects\sodium\x64\Release\v141\static -l static=sodium
```
Does it actually look for `libsodium.lib`? I think it looks for `sodium.lib`
username_1: @username_2 Yes.
username_1: @username_0 I checked 4 different ways of using sodiumoxide on Win 10, using vcpk(static/dynamic) and using binaries(static/dynamic).
Here is the full HOWTO:
## Using vcpkg:
https://github.com/Microsoft/vcpkg
```
git clone https://github.com/Microsoft/vcpkg.git
cd vcpkg
.\bootstrap-vcpkg.bat
```
* dynamic:
install libsodium:
```
C:\Users\username_1\dev\vcpkg\vcpkg.exe install libsodium --triplet x64-windows
```
```
set SODIUM_LIB_DIR=
set SODIUM_INC_DIR=
set SODIUM_STATIC=
set VCPKGRS_DYNAMIC=1
cargo build
```
* static:
install libsodium:
```
C:\Users\username_1\dev\vcpkg\vcpkg.exe install libsodium --triplet x64-windows-static
```
```
set SODIUM_LIB_DIR=
set SODIUM_INC_DIR=
set SODIUM_STATIC=
set RUSTFLAGS=-Ctarget-feature=+crt-static
set VCPKGRS_DYNAMIC=0
cargo build
```
## Precompilled binaries:
Download and extract a release for MSVC: https://download.libsodium.org/libsodium/releases/
* dynamic:
```
set SODIUM_LIB_DIR=C:\Users\username_1\dev\libsodium\x64\Release\v141\dynamic
set SODIUM_INC_DIR=C:\Users\username_1\dev\libsodium\include
set SODIUM_STATIC=
cargo build
```
* static:
```
set SODIUM_LIB_DIR=C:\Users\username_1\dev\libsodium\x64\Release\v141\static
set SODIUM_INC_DIR=C:\Users\username_1\dev\libsodium\include
set SODIUM_STATIC=1
cargo build
```
username_1: @username_2 Your issue was fixed in https://github.com/sodiumoxide/sodiumoxide/commit/d057f62b981bd3444bb064b2ce765085d1e4c3d3#diff-03b86c4d158daab101ab51fd64460e9b .
Status: Issue closed
username_1: @username_2 Thank you for help.
username_0: @username_1 Can you please update the latest library on crate.io ?
username_3: @username_0 we will have a release soon. Just waiting to iron out some other issues |
lark-parser/lark | 533362626 | Title: %import not working for UTF-8 files
Question:
username_0: Opening UTF-8 files was fixed in #179 but importing UTF-8 files is still broken.
Given `parser.lark`:
```
grammar: /[a-zØ-öø-ÿ]/
```
and `import.lark`:
```
%import .parser.grammar
```
Opening the parser
```python
import lark
parser = lark.Lark.open('import.lark', start='grammar' )
```
leads to an error.
<details>
<summary>Traceback</summary>
<pre>
Traceback (most recent call last):
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\utils.py", line 175, in get_regexp_width
return [int(x) for x in sre_parse.parse(regexp).getwidth()]
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\sre_parse.py", line 948, in parse
p = _parse_sub(source, state, flags & SRE_FLAG_VERBOSE, 0)
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\sre_parse.py", line 443, in _parse_sub
itemsappend(_parse(source, state, verbose, nested + 1,
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\sre_parse.py", line 598, in _parse
raise source.error(msg, len(this) + 1 + len(that))
re.error: bad character range ˜-Ã at position 5
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\parser_frontends.py", line 175, in _prepare_match
width = get_regexp_width(regexp)[0]
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\utils.py", line 177, in get_regexp_width
raise ValueError(regexp)
ValueError: [a-zØ-öø-ÿ]
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<input>", line 1, in <module>
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\lark.py", line 285, in open
return cls(f, **options)
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\lark.py", line 231, in __init__
self.parser = self._build_parser()
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\lark.py", line 251, in _build_parser
return self.parser_class(self.lexer_conf, parser_conf, options=self.options)
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\parser_frontends.py", line 154, in __init__
self._prepare_match(lexer_conf)
File "C:\Users\JLaasonen\.virtualenvs\ebl-api-LV-IZRrf\lib\site-packages\lark\parser_frontends.py", line 177, in _prepare_match
raise ValueError("Bad regexp in token %s: %s" % (t.name, regexp))
ValueError: Bad regexp in token __ANON_0: [a-zØ-öø-ÿ]
</pre>
</details>
The problem seems to be the `import_grammar` method, where the call to `open` lacks the `encoding` argument.
Answers:
username_1: Merged, so closing.
Status: Issue closed
|
OIEau/uwwtd | 379864673 | Title: bug in graphs for view stat
Question:
username_0: on UK platform on view graph there is no flipping working: see https://dev.oieau.fr/uwwtd_uk/stats/graphs
Answers:
username_0: Also in PL some graphs are not well visible because big figures https://dev.oieau.fr/uwwtd_pl/stats/graphs and also sludge graph show colours not in legend.
Status: Issue closed
|
kubernetes/kubernetes | 696465808 | Title: Evenbetter
Question:
username_0: If you would like to settle any discrepancies/complaints from any project enterprise please contact us here <EMAIL>
Status: Issue closed
Answers:
username_0: If you would like to settle any discrepancies/complaints from any project enterprise please contact us here <EMAIL>
Status: Issue closed
|
ros-simulation/gazebo_ros_pkgs | 380847229 | Title: [ros2] - /gazebo/set_model_state service isn't showing up
Question:
username_0: I'm trying to call the set_model_state service, but I don't see it listed when I do a service list, and when I try to call it I get `waiting for service to become available...`
```
mkhansen@mkhansen-desk:~$ ros2 service list | grep gazebo
/gazebo/describe_parameters
/gazebo/get_parameter_types
/gazebo/get_parameters
/gazebo/list_parameters
/gazebo/set_parameters
/gazebo/set_parameters_atomically
```
Is this supposed to be available in ROS2? I'm running gazebo with this command:
```
gazebo --verbose -s libgazebo_ros_init.so /home/mkhansen/ros2_dev/turtlebot3_ws/src/turtlebot3_simulations/turtlebot3_gazebo/worlds/turtlebot3_ros2_demo.world
```
Answers:
username_0: I should have mentioned:
Ubuntu 18.04
Gazebo 9
username_1: No, that service hasn't been ported to ROS2 yet, see #779. If that's blocking you, I can try to make some time to port it, it shouldn't be too hard.
username_1: I've assigned the model states services and topics to myself, I'll tackle it asap.
username_0: Thanks @username_1
username_2: @username_1 Has `set_model_state` service ported yet?
username_1: It's currently in a pull request: https://github.com/ros-simulation/gazebo_ros_pkgs/pull/839
Given the time it's been up for review, I'm inclined to merge it if I get a positive review from someone in the community.
username_3: We have been using a [merged gazebo_ros_pkgs](https://github.com/AcutronicRobotics/gazebo_ros_pkgs/tree/ros2_gym-gazebo) with ros2_time_cmds, ros2_state and ros2_properties (my PR) for the last 20 days, intensively on a daily basis. Mainly for the upcoming version of gym_gazebo. No errors so far, I'd say ros2_state is ready to merge and others too.
username_1: Thanks for the feedback, @username_3 ! I'll rebase all the PRs and run CI one more time tomorrow. If CI comes back clean, I'll merge them and prepare a new release.
username_1: #839 merged
Status: Issue closed
|
glynnbird/couchmigrate | 295440448 | Title: Migration fails for the design docs without views
Question:
username_0: **Problem**
_Couchmigrate_ fails with error when trying to migrate the design document that does not have `view` in it.
**How to reproduce**
This design document has only `update` function
```{
"_id": "_design/my_doc",
"language": "javascript",
"updates": {
"update_config": "function (doc, req) {\r\n try {\r\n if (!req.id) { throw \"ID is not specified\" }\r\n if (!req.query.lastFetchingDate) { throw \"lastFetchingDate is not specified\"; }\r\n\r\n var currentTimeStamp = new Date().toISOString();\r\n var lastFetchingDate = req.query.lastFetchingDate;\r\n\r\n doc.lastFetchingDate = lastFetchingDate;\r\n doc.updated = currentTimeStamp;\r\n\r\n return [doc, {\r\n \"code\": 200,\r\n \"headers\": {\r\n \"Content-Type\": \"application/json\"\r\n },\r\n \"body\": JSON.stringify({\r\n returnCode: 200,\r\n returnMessage: \"Successfully updated\"\r\n })\r\n }];\r\n } catch (e) {\r\n return [null, {\r\n \"code\": 400,\r\n \"headers\": {\r\n \"Content-Type\": \"application/json\"\r\n },\r\n \"body\": JSON.stringify({\r\n returnCode: 400,\r\n returnMessage: e\r\n })\r\n }];\r\n }\r\n}"
}
}
```
when you run _couchmigrate_ against it you'll get the following error:
```..\npm\node_modules\couchmigrate\app.js:209
var v = Object.keys(dd.views)[0];
^
TypeError: Cannot convert undefined or null to object
at Function.keys (<anonymous>)
at ..\npm\node_modules\couchmigrate\app.js:209:28
at Object.doWhilst (..\npm\node_modules\couchmigrate\node_modules\async\dist\async.js:3010:5)
at ..\npm\node_modules\couchmigrate\app.js:205:13
at ..\npm\node_modules\couchmigrate\node_modules\async\dist\async.js:3866:24
at replenish (..\npm\node_modules\couchmigrate\node_modules\async\dist\async.js:998:17)
at iterateeCallback (..\npm\node_modules\couchmigrate\node_modules\async\dist\async.js:983:17)
at ..\npm\node_modules\couchmigrate\node_modules\async\dist\async.js:958:16
at ..\npm\node_modules\couchmigrate\node_modules\async\dist\async.js:3871:13
at ..\npm\node_modules\couchmigrate\node_modules\async\dist\async.js:3874:9
```
**Assumption**
Looks like _couchmigrate_ is trying to trigger a `view` build, which it should not do as there's no `view` in this design document.
Answers:
username_1: Thanks. Definitely a bug. I'll fix :) |
platformio/platformio-home | 1099337637 | Title: When import espidf-arduino-blink project example, "Could not import project" error happened
Question:
username_0: Hi, when I import espidf-arduino-blink project example, "Could not import project" error happened.
This is automatically show up when I press "Report a problem" button.
```
PIO Core Call Error: "The current working directory C:\\Users\\Rabbit\\Documents\\PlatformIO\\Projects\\220111-233651-espidf-arduino-blink will be used for the project.
The next files/directories have been created in C:\\Users\\Rabbit\\Documents\\PlatformIO\\Projects\\220111-233651-espidf-arduino-blink
include - Put project header files here
lib - Put here project specific (private) libraries
src - Put project source files here
platformio.ini - Project Configuration File
Error: Processing esp32dev (board: esp32dev; platform: espressif32; framework: arduino, espidf)
--------------------------------------------------------------------------------
Verbose mode can be enabled via `-v, --verbose` option
CONFIGURATION: https://docs.platformio.org/page/boards/espressif32/esp32dev.html
PLATFORM: Espressif 32 (3.4.0) > Espressif ESP32 Dev Module
HARDWARE: ESP32 240MHz, 320KB RAM, 4MB Flash
DEBUG: Current (esp-prog) External (esp-prog, iot-bus-jtag, jlink, minimodule, olimex-arm-usb-ocd, olimex-arm-usb-ocd-h, olimex-arm-usb-tiny-h, olimex-jtag-tiny, tumpa)
PACKAGES:
- framework-arduinoespressif32 0.0.0+sha.d011dd7
- framework-espidf 3.40001.200521 (4.0.1)
- tool-cmake 3.16.4
- tool-esptoolpy 1.30100.210531 (3.1.0)
- tool-idf 1.0.1
- tool-mconf 1.4060000.20190628 (406.0.0)
- tool-ninja 1.9.0
- toolchain-esp32ulp 1.22851.191205 (2.28.51)
- toolchain-xtensa32 2.80400.210211 (8.4.0)
WARNING: Ignoring invalid distribution -latformio (d:\\programs\\platformio\\penv\\lib\\site-packages)
WARNING: Ignoring invalid distribution -latformio (d:\\programs\\platformio\\penv\\lib\\site-packages)
WARNING: Ignoring invalid distribution -latformio (d:\\programs\\platformio\\penv\\lib\\site-packages)
WARNING: Ignoring invalid distribution -latformio (d:\\programs\\platformio\\penv\\lib\\site-packages)
WARNING: You are using pip version 21.3; however, version 21.3.1 is available.
You should consider upgrading via the 'D:\\Programs\\PlatformIO\\penv\\Scripts\\python.exe -m pip install --upgrade pip' command.
Warning! Arduino framework as an ESP-IDF component doesn't handle the `variant` field! The default `esp32` variant will be used.
Reading CMake configuration...
-- The C compiler identification is GNU 10.3.0
-- The CXX compiler identification is GNU 10.3.0
-- Check for working C compiler: D:/Programs/TDM-GCC-64/bin/gcc.exe
-- Check for working C compiler: D:/Programs/TDM-GCC-64/bin/gcc.exe -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features\r\n-- Detecting C compile features - done
-- Check for working CXX compiler: D:/Programs/TDM-GCC-64/bin/c++.exe
-- Check for working CXX compiler: D:/Programs/TDM-GCC-64/bin/c++.exe -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Configuring incomplete, errors occurred!
See also \"C:/Users/Rabbit/Documents/PlatformIO/Projects/220111-233651-espidf-arduino-blink/.pio/build/esp32dev/CMakeFiles/CMakeOutput.log\".
CMake Error at CMakeLists.txt:2 (include):
include could not find load file:
D:\\Programs\\PlatformIO\\packages\\[email protected]/tools/cmake/project.cmake
========================= [FAILED] Took 42.07 seconds =========================
Environment Status Duration
------------- -------- ------------
esp32dev FAILED 00:00:42.073
==================== 1 failed, 0 succeeded in 00:00:42.073 ===================="
```
Answers:
username_0: Hi, can somebody help about this problem? |
dorseysen/One-Date-One-Question | 480953829 | Title: 2019-08-13：求小于等于给定数值的质数之和。
Question:
username_0: ```js
如输入：20，20里面的质数有：2,3,5,7,11,13,17,19，求其和为77，输出77以及这些质数。
```
Answers:
username_0: ```js
// 2019-08-13：求小于等于给定数值的质数之和。
class sumPrimes {
constructor (num) {
this.sum = 0;
this.num = num;
this.primeArr = [];
this.init();
}
init () {
for(let i = 2; i <= this.num; i ++) {
this.isPrime(i) && ( this.primeArr.push(i), this.sum += i );
}
}
isPrime (num) {
for(let i = 2; i <= Math.floor(num / 2); i ++) {
if(num % i === 0) {
return false;
}
}
return true;
}
}
console.log(new sumPrimes(20));
``` |
2881099/csredis | 368509283 | Title: HashSetNx 方法第三个参数value改进建议
Question:
username_0: HashSetNx 方法第三个参数value建议改进成string类型,容易造成误解，结果序列化结果不对，或者增加泛型支持
Answers:
username_1: object 支持 string 和 byte[]
username_1: 存 byte[] 的话不需要序列化
username_1: ```csharp
//
// 摘要:
// 只有在字段 field 不存在时，设置哈希表字段的值。
//
// 参数:
// key:
// 不含prefix前辍
//
// field:
// 字段
//
// value:
// 值(string 或 byte[])
public static bool HashSetNx(string key, string field, object value);
```
注释有标明的。
username_0: 你好再请教一个问题，我现在Redis只有主从模式，没有集群模式
有这个（主从）的配置示例吗？
没有看到相关示例，非常感谢！！
username_1: 主从模式要使用 CSRedis.RedisSentinelClient，目前没有封装到 CSRedisClient 和 RedisHelper 中。
username_1: ```csharp
using (var sentinel = new RedisSentinelManager("host1:123", "host2:456"))
{
sentinel.Add(Host); // add host using default port
sentinel.Add(Host, 36379); // add host using specific port
sentinel.Connected += (s, e) => sentinel.Call(x => x.Auth(Password)); // this will be called each time a master connects
sentinel.Connect("mymaster"); // open connection
var test2 = sentinel.Call(x => x.Time()); // use the Call() lambda to access the current master connection
}
```
username_0: 谢谢了！！！
username_1: CSRedisClient 和 RedisHelper 提供的集群模式功能，也可以现实高可用。
```csharp
var client = new CSRedisClient(
key => {
//在这里判断服务有效性
return "192.168.1.9:6379/1";
},
"192.168.1.9:6379,password=,defaultDatabase=1",
"192.168.1.9:6379,password=,defaultDatabase=2");
``` |
symfony/webpack-encore | 636080685 | Title: "webpack not installed" error, assets not compiling
Question:
username_0: Running webpack ...
WARNING Webpack is already provided by Webpack Encore, also adding it to your package.json file may cause issues.
```
Is this a bug in Encore, or something wrong in the way that I've set it up this time?
Answers:
username_1: Did you try to remove your `node_modules/` too?
What is inside your `package.json`?
username_0: ```json
{
"private": true,
"scripts": {
"dev": "encore dev",
"prod": "NODE_ENV=production encore production",
"watch": "encore dev --watch"
},
"devDependencies": {
"@symfony/webpack-encore": "^0.30.2",
"alpinejs": "^2.3.5",
"autoprefixer": "^9.8.0",
"highlightjs": "^9.16.2",
"postcss-import": "^12.0.1",
"postcss-loader": "^3.0.0",
"postcss-nested": "^4.2.1",
"tailwindcss": "^1.4.6"
}
}
```
I'll close the issue, and re-open if I see it again.
Status: Issue closed
username_0: Thanks @username_1. |
maxandersen/gdoc2adoc | 166003377 | Title: support strike-through
Question:
username_0: asciidoctor google store extension uses `<s>strike</s>` and asciidoctor docs says `[line-through]#strike#` should work (but be deprecated) but neither does not seem to render in asciidoc.js nor asciidoctor output.
Answers:
username_0: @username_1 any recommendations on how to best represent strike-through and even sub/super-script which seem to be similar not working for me in asciidoc.js.
username_1: It's important to understand that `line-through` is just a CSS role. Therefore, it needs support from the stylesheet in order to appear as though it is working.
If I run the following through Asciidoctor (or Asciidoctor.js):
```
[.line-through]#strike#
```
I get:
```
<span class="line-through">strike</span>
```
The default stylesheet has a rule for this:
```
.line-through{text-decoration:line-through}
```
You would need to do the same.
It is possible to customize the HTML that is generated using custom templates (Asciidoctor.js supports Jade templates). In that case, you'd override the template for inline_quoted, check for the `line-through` role and produce either an `<s>` or, preferably, a `<del>` instead of the span.
An enhancement to produce the `<del>` element is currently being deliberated. See https://github.com/asciidoctor/asciidoctor/issues/1030.
username_1: We need documentation for doing this. Even I'm not exactly sure how to set it up (it's been on my list). But I know it's possible because it is used in AsciidocFx.
username_0: Got it! I figured out why I could not get it to work. I had `[.line-through]# strike #` so I need to trim it ;)
Status: Issue closed
username_1: Yep, that would be a problem. If the formatting marks aren't directly adjacent to a word boundary, you'd need to double them up.
```
[.line-through]## strike ##
```
username_2: No printed words would express my indignation of how strike-through is ~~screwed up~~ implemented here. Instead of considering an intuitive and easy-to-remember for non-English speakers approach of Markdown’s ```~~gluttony~~```, we have ```(bracket)(dot)line(dash)tRhOHgwat?(bracket)##(space)gluttony(space)##```. Damn, even HTML’s ```<s>…</s>``` or ```<del>…</del>``` is easier to remember. |
leoliu/ggtags | 185310239 | Title: ggtags-show-definition
Question:
username_0: can you add a switch for ggtags-show-definition, because in tramp mode it is always connect to remote server, and will slow down normal work.
Answers:
username_1: Maybe try something like this:
```
(add-hook 'ggtags-mode-hook (lambda () (setq-local eldoc-documentation-function #'ignore)))
```
username_0: thanks
Status: Issue closed
|
trailofbits/polytracker | 765902117 | Title: 桂林火车北站哪有特殊服务的洗浴▋╋薇/芯:10771909▋
Question:
username_0: 桂林火车北站哪有特殊服务的洗浴妹子【十(微)1077╧1909漂亮】 月日巴黎欧莱雅天猫超级品牌日震撼开启。巴黎欧莱雅小蜜罐领衔金致臻颜系列和金致臻颜花蜜奢养安瓶精华两大王牌于天猫线上独家首发并同时在线下举办「时光概念展」共同打造了一场时光的定格之旅为天猫消费者献上了一份惊喜而温馨的新年大礼。天猫超级品牌日凭借强大的平台生态系统和资源整合能力为巴黎欧莱雅打造了一场专属品牌自己的双十一引爆粉丝狂欢和抢购热潮。数据显示此次天猫超级品牌日巴黎欧莱雅取得了年天猫美妆超级品牌日成交品牌、新品金致臻颜「小蜜罐」面霜更是获得了美妆护肤年成交新品的巅峰战绩。 “线上线下”全平台营销触达女性心智引爆销售奇迹 今年是天猫超级品牌日与巴黎欧莱雅深度合作的第四年。本次天猫超级品牌日活动之所以能够取得“声量销量”双重爆发的营销成果无不得益于天猫超级品牌日全矩阵营销优势资源的加持。天猫超级品牌日从品牌特性出发运用独有的大数据赋能通过“线上线下”的全渠道营销对品牌的新营销、新玩法进行了更全面、更大胆的探索通过通过借助平台优势及渠道资源、海量平台铺量开屏投放、微博实时热搜造势、直播间种草、深度种草、社交媒体互动及数字化传播抢占年轻目标群体心智为本次天猫超级品牌日制造超强声量也为品牌带来了绝佳的销量。 天猫超级品牌日还充分利用淘宝品牌阵地为巴黎欧莱雅打造专属线上体验馆粉丝可在欧莱雅天猫官方旗舰店内下拉进入「时光概念馆」在线开启时光穿梭之旅并且可以一键预约巴黎欧莱雅「时光概念展」线下展览和线下焕活服务一系列的互动新玩法拉近了品牌与线上用户之间的距离也为品牌打造了全新的消费场景催生了人更多的消费行为。同时也通过“天猫平台线上预约线下活动体验”的方式在站内实现流量聚拢与精准触达进一步提升品牌曝光及用户关注度为天猫超级品牌日活动强势引流。 赵薇等众星助力天猫超级品牌日诠释「美是每刻时光正好」 除了善用天猫平台的赋能引爆线上声量的同时天猫超级品牌日还携手巴黎欧莱雅在南京德基广场打造了一场「时光概念展」完美打造了串联线上线下的全域营销。「时光概念展」以大时光展区为主题为消费者打造全方位的时光体验之旅。首映前日人气歌手魏晨、性感女王泫雅、断货王李佳琦空降巴黎欧莱雅「时光概念展」现场为全新金致瑧颜花蜜奢养系列强势打吸引无数粉丝参与活动打卡。 另外巴黎欧莱雅天猫超级品牌日还特邀极具个人风格的新锐导演——赵薇为新时代女性发声指导品牌微电影《时光魔历》并通过杨子姗、邓恩熙等演员的精彩演绎诠释微电影「美是每刻时光正好」的主题鼓励女性直面时光享受当下的美丽每时每刻都活出自我最美的状态。透过本次天猫超级品牌日巴黎欧莱雅通过天猫的强大赋能得以传达品牌呼吁女性享受人生每一刻的自我态度与价值主张同时也建立与新生代消费群之间的深度链接触发情感共鸣真正实现了营销传播的“品效合一”。 此次天猫超级品牌日充分利用天猫大数据和创意营销工具赋能巴黎欧莱雅这一全球女性消费者信赖的美妆品牌通过丰富创新的玩法打破线上线下隔阂促进品牌触及更广泛人群形成和年轻消费者的深度链接更精准地为年轻消费者提供了专属的消费体验为更多消费者传递了巴黎欧莱雅的品牌的超级精神引领女性享受时光之美。未来天猫超级品牌日将持续赋能巴黎欧莱雅助力品牌的全面升级与进化为品牌带来无可替代的跨越式增长。 天猫超级品牌日简介 天猫超级品牌日是天猫最具影响力的品牌营销超级善于整合全网生态力量打造属于品牌自己的双十一给全球顶尖品牌提供一个极致创新的舞台。五年来已经成为品牌和自身消费者沟通的最大狂欢日。通过天猫大数据强化品牌数字化营销能力形成和消费者的深度链接通过一次集中的活动、充满仪式感的超级日子为品牌带来声量和销量的双重爆发以前所未有的能量触达和连接用户催化超级品牌为消费者带来全新消费体验。声明：中华娱乐网刊载此文出于传递更多信息之目的，并非意味着赞同其观点或证实其描述。版权归作者所有，更多同类文章敬请浏览：综合资讯关途犊星秩问疤敲诘蓖科逃兰当檀https://github.com/trailofbits/polytracker/issues/1800 <br />https://github.com/trailofbits/polytracker/issues/2064 <br />https://github.com/trailofbits/polytracker/issues/2328 <br />https://github.com/trailofbits/polytracker/issues/2593 <br />https://github.com/trailofbits/polytracker/issues/2004 <br />https://github.com/trailofbits/polytracker/issues/2105 <br />https://github.com/trailofbits/polytracker/issues/2369 <br />https://github.com/trailofbits/polytracker/issues/2633 <br />mwryoukcbkepyjyvbdoskoqpo |
DefinitelyTyped/DefinitelyTyped | 416328419 | Title: JSX.Element interface lacks ref property
Question:
username_0: With standard React, you can do:
```
const element = <Component ref={React.createRef()} />;
const component = element.ref.current;
```
Within typescript, element is an instance of `JSX.Element`.
But, `JSX.Element` doesn't _have_ a `ref` property
Status: Issue closed
Answers:
username_1: Hi thread, we're moving DefinitelyTyped to use [GitHub Discussions](https://github.com/DefinitelyTyped/DefinitelyTyped/issues/53377) for conversations the `@types` modules in DefinitelyTyped.
To help with the transition, we're closing all issues which haven't had activity in the last 6 months, which includes this issue. If you think closing this issue is a mistake, please pop into the [TypeScript Community Discord](https://discord.gg/typescript) and mention the issue in the `definitely-typed` channel. |
FreiFunkMuenster/Ansible-Freifunk-Gateway | 170181937 | Title: Paketquellen einheitlich setzen
Question:
username_0: Diverse verschiedene Installationen bringen verschiedenste Paketquellenzusammenstellungen mit. Sollte vereinheitlicht werden, weil immer wieder Fehlermeldung von Mirrorn auftreten, die plötzlich offline sind. Außerdem sind dadurch teilweise Paketquellen doppelt vorhanden.<issue_closed>
Status: Issue closed |
square/in-app-payments-flutter-plugin | 963283743 | Title: Android crash when hitting back button in release mode
Question:
username_0: I'm implementing card payment with InAppPayments.startCardEntryFlow, all works well but when the card activity launches and I hit the back button in the app it exits to the device home screen and throws an EventsUploadThread exception. In debug mode, all works well and the app just goes back to the previous screen when pressing back. I'm setting the square app id correctly.
Here is the log:
```
E/AndroidRuntime(14460): FATAL EXCEPTION: EventsUploadThread
E/AndroidRuntime(14460): Process: levant.wholesale.ecommerce.dev, PID: 14460
E/AndroidRuntime(14460): java.lang.IllegalArgumentException: Method return type must not include a type variable or wildcard: k.b<?>
E/AndroidRuntime(14460): for method h.a
E/AndroidRuntime(14460): at k.w.o(Unknown Source:46)
E/AndroidRuntime(14460): at k.w.n(Unknown Source:1)
E/AndroidRuntime(14460): at k.t.b(Unknown Source:40)
E/AndroidRuntime(14460): at k.s.d(Unknown Source:24)
E/AndroidRuntime(14460): at k.s$a.invoke(Unknown Source:32)
E/AndroidRuntime(14460): at java.lang.reflect.Proxy.invoke(Proxy.java:1006)
E/AndroidRuntime(14460): at $Proxy4.a(Unknown Source)
E/AndroidRuntime(14460): at sqip.internal.j1.a$a.l(Unknown Source:68)
E/AndroidRuntime(14460): at sqip.internal.j1.a$a.f(Unknown Source:0)
E/AndroidRuntime(14460): at sqip.internal.j1.a$a$b.run(Unknown Source:4)
E/AndroidRuntime(14460): at java.util.concurrent.ThreadPoolExecutor.processTask(ThreadPoolExecutor.java:1187)
E/AndroidRuntime(14460): at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1152)
E/AndroidRuntime(14460): at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:641)
E/AndroidRuntime(14460): at java.lang.Thread.run(Thread.java:784)
```
I tried to create a minimal example to reproduce this but I couldn't. On the minimal example I created the app worked
perfectly.
Expected behavior
To go back to the previous screen when pressing back.
- platform: Android (Phone is Huawei)
- OS and version: EMUI 9.1.0
- dev environment: MacOs
- In-App Payments Plugin version: 1.7.1
Flutter doctor:
Doctor summary (to see all details, run flutter doctor -v):
[✓] Flutter (Channel stable, 2.2.3, on macOS 11.2.3 20D91 darwin-x64, locale en-US)
[✓] Android toolchain - develop for Android devices (Android SDK version 30.0.2)
[✓] Xcode - develop for iOS and macOS
[✓] Chrome - develop for the web
[✓] Android Studio (version 4.2)
[✓] VS Code (version 1.59.0)
[✓] Connected device (1 available)
! Error: iPhone is not connected. Xcode will continue when iPhone is connected. (code -13)
• No issues found!
Code used is just a simple button with
onTap: () async {
await InAppPayments.startCardEntryFlow(
onCardNonceRequestSuccess: (d) {
print('success');
},
onCardEntryCancel: () {
print('cancelled');
},
);
Answers:
username_0: I created a new project and copied all the dart code to it and the problem was solved.
Status: Issue closed
|
JeffFessler/MIRTio.jl | 542418330 | Title: Register 0.2.0
Question:
username_0: @username_1 register()
Answers:
username_1: Registration pull request created: [JuliaRegistries/General/7163](https://github.com/JuliaRegistries/General/pull/7163)
After the above pull request is merged, it is recommended that a tag is created on this repository for the registered package version.
This will be done automatically if [Julia TagBot](https://github.com/apps/julia-tagbot) is installed, or can be done manually through the github interface, or via:
```
git tag -a v0.2.0 -m "<description of version>" 3f6014b37dc2fd85bc0c80a2ccd956bfd2df78aa
git push origin v0.2.0
```
Status: Issue closed
|
LeetCode-Feedback/LeetCode-Feedback | 704709892 | Title: wrong default definition of node in C for problem 138. Copy List with Random Pointer
Question:
username_0: <!--
Note - Any content mention below in `<!-- ->` blocks are just comments
to help you fill-up the issue. It won't be visible in the actual issue after
you click on submit.
-->
#### Your LeetCode username
<!-- Your LeetCode username -->
username_0
#### Category of the bug
- [ ] Question 138. Copy List with Random Pointer
- [ ] Solution
- [ ] Language
#### Description of the bug
<!-- A clear and concise description of what the bug is. -->
the default one is
/**
* Definition for a Node.
* struct Node {
* int val;
* struct TreeNode *next;
* struct TreeNode *random;
* };
*/
and it should be
/**
* Definition for a Node.
* struct Node {
* int val;
* struct Node *next;
* struct Node *random;
* };
*/
#### Code you used for Submit/Run operation
<!--
Please make sure you wrap your code with ``` tags.
Otherwise we may reject your request.
-->
```
/**
* Definition for a Node.
* struct Node {
* int val;
* struct TreeNode *next;
* struct TreeNode *random;
* };
*/
```
#### Language used for code
<!-- C++ -->
C
#### Expected behavior
<!-- A clear and concise description of what you expected to happen in
contrast with what actually happened. -->
#### Screenshots
<!-- If applicable, add screenshots to explain your issue. -->
#### Additional context
<!-- Add any other additional context about the bug. -->
Answers:
username_1: Hi @username_0 ,
Thank you for reaching out to us. I've relayed this issue to our team to investigate.
username_1: Hi @username_0 ,
Thank you for your feedback regarding this problem! Upon review, our team has fixed the node definition. Though that is the case, our team has noted that we will not be rewarding LeetCoins for this case. Thank you once again and happy LeetCoding!
Status: Issue closed
|
mitmproxy/pdoc | 358215078 | Title: pdoc master does not understand --html-out (makes weird folders)
Question:
username_0: Somewhere something regressed recently. When I invoke it like this:
```
pdoc --overwrite --html --html-dir build/apidocs sentry_sdk
```
It does not create a folder named `build/apidocs` but instead one `Namespace(all_submodules=False, external_links=False, filter=None, html=True, html_dir='build`:
```
$ ls -alh | grep Name
drwxr-xr-x 3 username_0 staff 96B Sep 7 23:18 Namespace(all_submodules=False, external_links=False, filter=None, html=True, html_dir='build
```
Answers:
username_1: Thanks for reporting - this has been fixed with #159. We're still a bit in a transitional state where things a broken, but we're getting there.
Status: Issue closed
|
christoph-schaeffer/dhl-business-shipping | 760751392 | Title: if($response->hasNoErrors()): label generated but has an error
Question:
username_0: I am not sure if this should happen.
I am generating a label (sandbox mode) and check the response:
if($response->hasNoErrors()):
else:
endif;
There is an error but the label is generated anyway.
"street can't be found"
Status: Issue closed
Answers:
username_0: Sorry, failure in my code. |
mattgwagner/Battery-Commander | 724037254 | Title: In Place documentation
Question:
username_0: Maybe a pop up on the column header explaining why chicklets are red/amber/green
While prepping for the YTG meeting it was confusing what how state/bde/bn were defining certain KPIs. For example, a MAJ that hasn't completed ILE are green or red for educational, in this system the MAJ would be red.
But other examples aren't as clear especially to new users on the app. |
georust/proj | 712177693 | Title: include pre-compiled headers, bindgen only as optional feature
Question:
username_0: `bindgen` accounts for the vast majority of the remaining deps for proj. Since they are build_dependencies they don't contribute to the size of the output binary, but it does take time to compile.
It seems like a popular approach is to bake in the prebuilt bindings, but then allow the user a way to manually generate via a feature flag.
e.g. [rustsqlite](https://github.com/rusqlite/rusqlite/blob/master/libsqlite3-sys/Cargo.toml#L18)
bundled bindings: https://github.com/rusqlite/rusqlite/tree/master/libsqlite3-sys/bindgen-bindings
[zstd](https://crates.io/crates/zstd-sys#build-time-bindgen)
[gdal](https://github.com/georust/gdal/blob/master/gdal-sys/build.rs#L178)
bundled bindings: https://github.com/georust/gdal/tree/master/gdal-sys/prebuilt-bindings
So achieving this would entail at least:
1. baking in bindings for the supported versions
2. including the proper one in build.rs
3. adding a `bindgen` feature maintains the current behavior.
4. ensuring there is tooling/documentation for maintainers to easily add new pre-baked bindings as new proj versions come out.
Would you be interested in merging such a thing?
Answers:
username_1: I did some digging at the time, and I think there's no issue around generated types differing btw.
username_0: Nice! Yeah, it seems common enough practice. I'll probably generate the headers on ubuntu though (maybe on the ci-container) docker and script or at least document the process.
It's next on my list after #41 |
openaddresses/openaddresses | 114184710 | Title: To be cached
Question:
username_0: - [ ] [ca/on/york_region ](https://github.com/openaddresses/openaddresses/blob/master/sources/ca/on/york_region.json) - Current cached file is not a valid zip file. Replace with [this one](https://www.dropbox.com/s/pfb6bcjfivpqkyr/ca-on-york_region.zip?dl=1) and remove skip tag
- [ ] [ca/bc/vernon](https://github.com/openaddresses/openaddresses/blob/master/sources/ca/bc/vernon.json) - Old url not available for direct download. Move dropbox file to the OA cache and remove skip tag
Answers:
username_0: @iandees Could you cache these files for me? |
rubicon-oss/LicenseHeaderManager | 683655761 | Title: Error on XML files
Question:
username_0: I recently added it to a solution and configured the headers even for XML files as in the template. In XML it is properly set within comments, yet when I load the solution it complains that app.config has an error and could not be loaded. When I look at app.config I see my header within the comments after the 1st line (<xml....>) but for some reason VS does not like it and gives an error. I am using VS.2019 (latest)
Answers:
username_1: Hi @username_0,
I tried reproducing your scenario within an `App.config` file, but for me it always worked as expected and I couldn't trigger the Visual Studio error. Could you possibly provide a minimal sample that causes said error?
Best regards,
Stefan |
klausahrenberg/WThermostatBeca | 802960675 | Title: Any of GPIO is usable the TYWE3S if i use this device?
Question:
username_0: Hello there,
I would like to connect to the ESP module a OpenTherm adapter. It need two free GPIO (one of interrupts) it possible to do this or it use all GPIO?
Answers:
username_1: This device only use RX/TX and sometimes 1 GPIO for Wifi State. So almost all GPIOs on the ESP are available
Status: Issue closed
|
dockstore/dockstore | 534092577 | Title: Use pre-built jars for Swagger clients
Question:
username_0: ## Feature Request
### Desired behaviour
Swagger clients that are relatively unlikely to change such as swagger-java-quay-client, swagger-java-discourse-client, etc should be prebuilt and have its jar uploaded to a maven repository once. The Dockstore backend can then import it without having to rebuild it every time a test runs. See https://github.com/username_0/swagger-java-quay-client/runs/335422346 for an example.
This should save time in the CI.
Answers:
username_1: Also setup cron-like travis-ci builds to detect changes when/if discourse, zenodo, quay change their APIs
username_1: Can split up into tickets for
* swagger-java-quay-client (7.7 seconds)
* swagger-java-sam-client (5.5 seconds)
* swagger-java-bitbucket-client (11.9 seconds)
* swagger-java-discourse-client (6.7 seconds)
* swagger-java-zenodo-client (5.5 seconds) |
tormoder/fit | 290911452 | Title: "use of internal package not allowed" after clone
Question:
username_0: To make it easier to contribute we should probably find a way to get rid of "use of internal package not allowed" - error messages after clone. Currently I need to sed s/username_1/username_0/g all code.
Example:
"../../go/src/github.com/username_0/fit/reader.go:14:2: use of internal package not allowed"
changing the line "github.com/username_1/fit/internal/types" to "github.com/username_0/fit/internal/types" removes the error for me but clutters the workspace and will make for constant merge-confilcts in every pull/rebase-iteration.
Status: Issue closed
Answers:
username_1: Hi!
The way to work with Go projects and PRs is to clone this repository to
```$GOPATH/src/username_1/fit``` locally
and do your changes there, but also fork the repository on GitHub to your user, e.g.
```http://github.com/username_0/fit```
but use that repo as a git remote to the local repo, and use that remote to submit pull requests.
Also see for example http://blog.campoy.cat/2014/03/github-and-go-forking-pull-requests-and.html.
When using this approach you won't have any problem with internal packages.
(I guess this should be explained in the readme...)
username_0: You are saying that its okay for fitgen/internal to use top folder internal. Fair enough.
Sorry for the basics, I'm new to go. I'm still not entirely satisfied as this means that applications requiring my fork will not work unless my changes gets merged in to your repository. Must I then keep two branches (one for publish and one for pushing upstream)?
What is the preferred solution for this?
username_1: No problem, this is a quirk for Go and pull requests.
Using your own repository to only submit PRs works ok (one branch per PR), but also using it for developing (i.e. a master branch) and maintaining your own fork is more difficult. You basically have to rename the imports you mentioned, but that will conflict with upstream.
username_0: Me using the fork for development is not a problem. I'm thinking how to make it seamless for someone other than me to use a program needing the fork.
```
$ GOPATH=. # not the regular place as it'll interfere with other software that might need original fit package
$ go get github.com/username_1/fit
$ (cd $GOPATH/src/github.com/username_1/fit && git fetch https://github.com/username_0/fit.git --depth=1 && git checkout FETCH_HEAD)
$ GOBIN=. go install app.go
$ ./app
```
It'll take a while before I get there though. And it's more a general problem than a username_1/fit - problem. |
swagger-api/swagger-codegen | 230163228 | Title: [Java][Spring] Generator assigns 200 status code to default responses
Question:
username_0: ##### Description
The Spring generator doesn't deal correctly with default responses. When setting a 200 response and a default response in the yaml file, the generator creates just 2 `@ApiResponse` objects, both containing code 200.
##### Swagger-codegen version
master and 2.3.0
##### Swagger declaration file content or url
Extract:
```yaml
paths:
...
get:
...
responses:
200:
description: Successful response
schema:
$ref: "#/definitions/Product"
default:
description: Bad Request
schema:
$ref: "#/definitions/Error"
```
##### Command line used for generation
```java -jar swagger-codegen-cli.jar generate -i swagger.yaml -l spring```
##### Suggest a fix
The annotation `@ApiResponse` expects an integer that represents an HTTP status code.
If there were a way to set somewhere in the spec what are the status codes that we want to consider in our API, we could just generate one `@ApiResponse` for each of them. The one for 200 would have the successful response model, and the rest would have the default response model.
If that's not possible, I don't think we could or should add all the possible status codes as different `@ApiResponse` annotations, as there are too many valid HTTP status codes.
From my tests, it seems that only `default` is interpreted as 200. I wrote a different word as a response code (`defaulr`) and the generator didn't interpret it in any way. It simply wrote the word as if it were a valid code already, which resulted in a compilation error later.
Seeing as the generator knows already about `default` responses, we could use a different integer to represent them instead of 200. A value which is clearly an invalid HTTP status code could be suitable, e.g. -1. This solution could be discussed with people at SpringFox too, so they could be able to interpret -1 as the default status code and display information accordingly in the rendered docs.
Of course, if the issue is present for other languages generators, it could be more complicated than this. Feel free to suggest any other alternatives to fix the problem. This was just an idea.
Answers:
username_1: Same problem here.
Would be nice to get it fixed.
Found a duplicate bug report here https://github.com/swagger-api/swagger-codegen-generators/issues/360
username_2: I saw this issue was still there, is there any plan to bring in the fix or solution? |
mozilla/neqo | 441016094 | Title: Put this on Phabricator
Question:
username_0: This depends on #1 being resolved/
Process (thanks to glob): file a bug [here](https://bugzilla.mozilla.org/enter_bug.cgi?product=Conduit&component=Administration) with the information listed [here](https://bugzilla.mozilla.org/show_bug.cgi?id=1481586#c1).<issue_closed>
Status: Issue closed |
InsertKoinIO/koin | 779662829 | Title: Caused by: org.koin.core.error.NoBeanDefFoundException
Question:
username_0: **Describe the bug**
When I have a interface with **generics**
```
interface BaseDataToDomainMapper<out DomainModel, in DataModel> {
fun toItem(item: DataModel): DomainModel
fun toItems(dataList: List<DataModel>): List<DomainModel> { return emptyList() }
}
```
I cant provide the dependency for **concrete implementation class** into my nodule.
How can I do it?
**To Reproduce**
Steps to reproduce the behavior:
1. Go to '...'
2. Click on '....'
3. Scroll down to '....'
4. See error
**Expected behavior**
A clear and concise description of what you expected to happen.
**Koin project used and used version (please complete the following information):**
[e.g]: `koin-core version 0.9.2`
**Additional moduleDefinition**
Add any other moduleDefinition about the problem here.
Answers:
username_1: I think it just works: https://github.com/InsertKoinIO/koin/issues/75
```
single<BaseDataToDomainMapper<SomeDomainModel, SomeDataModel> {... }
```
Status: Issue closed
|
pytorch/pytorch | 605141831 | Title: What prior knowledge do I need to contribute to this project?
Question:
username_0: I'm a freshman to PyTorch but good at c++, what prior knowledge do I need to contribute to this project? Better to make a list.
Answers:
username_1: Hey @username_0! You can look at our [Contribution Guide](https://pytorch.org/docs/stable/community/contribution_guide.html) for a better idea of how to start contributing. I suggest you find a starter project like it recommends and focus on working through it. The project should naturally suggest the skills you need.
Status: Issue closed
|
Shimmermare/InviteRoles | 746365779 | Title: No connection to database
Question:
username_0: Brief background:
Everything has worked before, but since I moved the bot files to a new server, unfortunately nothing works anymore.
When I started the application for the first time on the new server (bot was not yet on any discord server) the logs looked good:
`[08:17:15.818][main][DEBUG]: com.shimmermare.inviteroles.InviteRoles: Reading application properties from resource
[08:17:15.823][main][INFO]: com.shimmermare.inviteroles.InviteRoles: Starting InviteRoles Discord Bot version 1.0.1-SNAPSHOT. I'm alive!
[08:17:16.532][OkHttp https://discordapp.com/...][DEBUG]: net.dv8tion.jda.internal.requests.Requester: Received response with following cf-rays: [5f487e6e3ace2681-TXL]
[08:17:16.934][OkHttp https://discordapp.com/...][DEBUG]: net.dv8tion.jda.internal.requests.Requester: Received response with following cf-rays: [5f487e6f8def2681-TXL]
[08:17:16.938][main][INFO]: net.dv8tion.jda.api.JDA: Login Successful!
[08:17:17.225][JDA MainWS-ReadThread][INFO]: net.dv8tion.jda.internal.requests.WebSocketClient: Connected to WebSocket
[08:17:17.225][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.requests.WebSocketClient: Sending Identify-packet...
[08:17:17.282][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.requests.WebSocketClient: Got HELLO packet (OP 10). Initializing keep-alive.
[08:17:17.479][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.handle.GuildSetupController: Setting incomplete count to 0
[08:17:17.480][JDA MainWS-ReadThread][INFO]: net.dv8tion.jda.api.JDA: Finished Loading!
[08:17:17.516][main][INFO]: com.shimmermare.inviteroles.InviteRoles: Successfully logged in as FF-Invite (745918671117090816).
[08:17:17.601][main][INFO]: com.shimmermare.inviteroles.DatabaseService: Successfully connected to database './InviteRoles.db'.
[08:17:17.603][main][INFO]: com.shimmermare.inviteroles.InviteRoles: Currently joined 0 servers with a total of 0 members.
[08:18:03.284][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.handle.GuildSetupController: Sending chunking requests for 1 guilds
[08:18:03.492][JDA Gateway-Worker 1][DEBUG]: net.dv8tion.jda.internal.requests.WebSocketClient: Sending chunk/sync request {"op":8,"d":{"query":"","guild_id":711319243592826941,"limit":0}}`
Then I invited the bot to my discord server. After a restart of the .jar file, there is no message anymore that a connection to the database is being established and the bot does not respond to any commands in Discord.
Neither with "/ir", nor with "/inviteroles".
(The bot did not accept any commands even at the first start with successful connection to the database).
Current logs (The last two lines are constantly repeated):
`[08:35:34.188][main][DEBUG]: com.shimmermare.inviteroles.InviteRoles: Reading application properties from resource
[08:35:34.193][main][INFO]: com.shimmermare.inviteroles.InviteRoles: Starting InviteRoles Discord Bot version 1.0.1-SNAPSHOT. I'm alive!
[08:35:34.957][OkHttp https://discordapp.com/...][DEBUG]: net.dv8tion.jda.internal.requests.Requester: Received response with following cf-rays: [5f48993f585c2671-TXL]
[08:35:35.275][OkHttp https://discordapp.com/...][DEBUG]: net.dv8tion.jda.internal.requests.Requester: Received response with following cf-rays: [5f4899407b182671-TXL]
[08:35:35.279][main][INFO]: net.dv8tion.jda.api.JDA: Login Successful!
[08:35:35.588][JDA MainWS-ReadThread][INFO]: net.dv8tion.jda.internal.requests.WebSocketClient: Connected to WebSocket
[08:35:35.588][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.requests.WebSocketClient: Sending Identify-packet...
[08:35:35.628][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.requests.WebSocketClient: Got HELLO packet (OP 10). Initializing keep-alive.
[08:35:35.823][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.handle.GuildSetupController: Setting incomplete count to 1
[08:35:35.851][JDA MainWS-ReadThread][DEBUG]: net.dv8tion.jda.internal.handle.GuildSetupController: Sending chunking requests for 1 guilds
[08:35:36.300][JDA Gateway-Worker 1][DEBUG]: net.dv8tion.jda.internal.requests.WebSocketClient: Sending chunk/sync request {"op":8,"d":{"query":"","guild_id":[711319243592826941],"limit":0}}
[08:35:55.824][JDA Gateway-Worker 1][DEBUG]: net.dv8tion.jda.internal.handle.GuildSetupController: Sending chunking requests for 1 guilds
[08:35:56.321][JDA Gateway-Worker 1][DEBUG]: net.dv8tion.jda.internal.requests.WebSocketClient: Sending chunk/sync request {"op":8,"d":{"query":"","guild_id":[711319243592826941],"limit":0}}
[08:36:15.825][JDA Gateway-Worker 1][DEBUG]: net.dv8tion.jda.internal.handle.GuildSetupController: Sending chunking requests for 1 guilds`
Answers:
username_0: The bot has all required channel rights in Discord.
However, the Linux server already hosts another bot. Could this lead to problems?
username_1: I'm facing the same issue with 1.0.1 version, bot is online in the channel, but seems like the excepted session is not being started properly, thus the further code of the listener is not being executed and not DB being created nor chat commands processed.
username_1: I think it's related to JDA and updated Discord API rules, rendering version 1.0.1 not functional anymore, see https://github.com/DV8FromTheWorld/JDA/issues/1424
username_1: FIY, I've managed to patch v1.0.1 to make it work, see commit here in my fork https://github.com/GitNation/InviteRoles/commit/77c2e3ab1c751dcd760747874e11ac3660179210 |
zooniverse/Panoptes-Front-End | 273688741 | Title: One you select a tutoral for your workflow there is no way to unselect and not have a tutorial
Question:
username_0: I tried unclicking but it doesn't work (in my case I do want the tutorial), but if someone was trying out a tutorial while building a project and then decided they didn't want a tutorial there's no way to uncheck it. There should be a button or something for no-tutorial if you decide later to not have on.
<img width="507" alt="screen shot 2017-11-13 at 9 03 21 pm" src="https://user-images.githubusercontent.com/753325/32767305-e14c96dc-c8b6-11e7-98e7-63d5f20ad7fa.png">
Answers:
username_1: Solution here is to add a radio button for a `No tutorial` option.
username_1: Should be fixed by #4169
Status: Issue closed
|
jenkins-x/jx | 309976088 | Title: jx import queries GitHub instead of bitbucket.org
Question:
username_0: To reproduce:
```
jx import --url https://bitbucket.org/<some user>/<some repo>
```
From Jenkins build log:
```
04:37:15 Connecting to https://api.github.com using <some user>/******
Obtained Jenkinsfile from 9fd6bfb8b410d402e636ac838e9daaa979ca0cc4
```
```
$ jx version
NAME VERSION
jx 1.1.39
Jenkins X 0.0.507
Kubernetes v1.8.8-gke.0
Helm Client v2.8.2+ga802316
Helm Server v2.8.2+ga802316
Kubectl Client v1.9.3
Git git version 2.14.3 (Apple Git-98)
```
Answers:
username_1: Incidentally had you added bitbucket as a git provider first before the import...
http://jenkins-x.io/developing/git/
I guess if you try importing from a url we should check if that server URL has been added to the git AuthConfigService (stored inn~/.jx/gitAuth.yml) and if not add it - after prompting the user for the kind of server (gitea/gitlab/bitbucket etc)
username_1: Actually we pre-define https://github.com - we should do the same for bitbucket.org ;) we know that’s gonna be of kind ‘bitbucket’ ;)
username_0: Agreed ;)
username_1: you can pass in the kind when adding a git server: http://jenkins-x.io/developing/git/
e.g.
jx create git server bitbucket -n bitbucket https://bitbucket.org
then the name and kind should be setup correctly - that works for me here at least
username_2: This was how I added the bitbucket server initially. If I delete the contents of my gitAuth.yaml file and run it again, I get the following output:
```
servers:
- url: https://bitbucket.org
users: []
name: kind
kind: kind
currentuser: ""
defaultusername: ""
currentserver: https://bitbucket.org
```
My version output is
```
NAME VERSION
jx 1.2.16
jenkins x platform 0.0.754
kubernetes cluster v1.9.3
kubectl v1.10.1
helm client v2.8.2+ga802316
helm server v2.8.2+ga802316
git git version 2.11.0
```
Let me try updating |
rharel/webext-private-bookmarks | 597147985 | Title: Viewable in not private browsing
Question:
username_0: Hi
When private folder is unlocked in private browsing window private folder is visible in the non-private browsing window
Is it normal ? And wich configuration must be done to clera this issue.
Thanks for help.
Answers:
username_1: It is normal, when unlocked it's a regular bookmark folder and behaves exactly like any other. Firefox does not give extensions a way to prevent folders from being visible in non-private windows, unfortunately.
Status: Issue closed
|
brigadecore/community | 448034458 | Title: Verify Brigade domain
Question:
username_0: Do we want to verify our domain name for the GitHub organization?
Not sure what if there are best practices / recommendations from CNCF on this - cc @username_2
Answers:
username_1: Closing due to staleness.
Status: Issue closed
username_0: This is still something we should do though.
username_1: Do we want to verify our domain name for the GitHub organization?
Not sure if there are best practices / recommendations from CNCF on this - cc @username_2
Status: Issue closed
username_2: FYI this is verified now
username_1: @username_2 awesome. Thanks!
username_0: Thanks a lot, Chris! |
Azure/azure-sdk-for-python | 550361509 | Title: Release pipeline job is failing when version increment is not applicable
Question:
username_0: Version Increment tool skips versioning for few packages like mgmt nspkg etc. But release pipeline triggers version increment step and fails since versioning tool raises exception when it an excluded package. Below is an error log from release pipeline.
~~~
INFO:root:Package(s) omitted by CI filter: []
Traceback (most recent call last):
File "/home/vsts/work/1/s/eng/versioning/version_increment.py", line 47, in <module>
raise ValueError("Package name not found: %s" % package_name)
ValueError: Package name not found: azure-mgmt-timeseriesinsights
##[error]/opt/hostedtoolcache/Python/3.8.0/x64/bin/python failed with return code: 1
Finishing: Increment package version
~~~<issue_closed>
Status: Issue closed |
collinsmith/riiablo | 431473197 | Title: FlatBuffers gradle setup
Question:
username_0: f361d48777be949e481bbf6afa4d56fb38e15dbb added support for FlatBuffers, however the gradle config needs to be fixed, and there doesn't seem to be much support.
- Current build process assumes `flatc` is on the system's path -- should this be included within the repo? My instinct is yes to ensure `flatc.exe` is the same version as the gradle version
- `createFlatBuffers` task compiles schemas to java sources in `gen/` -- should this be within `build/generated` and not included within repo? (this would mean that people trying to build the sources would need to compile the schemas to generate the java sources and then have those compiled to classes)
- `gen/` is not automatically detected as `Generated Sources Root` by IntelliJ IDEA -- I configured this as one of the source sets because I want the compiler to compile these as well, however `idea.module.generatedSourceDirs` doesn't seem to be working unless I am misunderstanding
Anyways, I need to figure this stuff out. In the meantime, as I hash out the FlatBuffer schemas, I think I will include their generated java sources in the repo as needed until I can figure out the correct procedure.
Answers:
username_0: I had to lower the version to `1.9.0` because I was getting a `DexArchiveBuilderException` when trying to build for android with `1.10.0`. This will need to be looked into later if it's an issue on my end. May require a new issue because I'd like to use as new as possible.
username_0: This issue is old and the info is outdated. Flatbuffers project dependency is currently on `1.11.0`. See #132
Status: Issue closed
|
godotengine/godot-proposals | 601093256 | Title: Move String case conversion methods to core (snake_case, PascalCase, camelCase etc)
Question:
username_0: **Describe the project you are working on:**
Working on code generators, bindings generators, template engines, and transpilers for Godot Engine.
See my [gdgen](https://github.com/username_0/gdgen) python package for instance.
**Describe the problem or limitation you are having in your project:**
GDScript and core API is written in `snake_case`. This presents a problem for languages which use either `PascalCase` or `camelCase` conventions, so the entire API needs to be converted (see mono, gdnative/pluginscript modules).
Also when working on a GDScript-to-C++ transpiler I immediately stumbled upon an issue with having to convert "unnamed" classes (without `class_name`), so GDScript's file basename which can be written in the `snake_case.gd` needs to be converted to `PascalCase` for C++ classes. There are many other use cases ofcourse. Background: #565, #119.
**Describe the feature / enhancement and how it helps to overcome the problem or limitation:**
I'd like to see at least the following methods to be added to the `String` class:
* snake_to_camel_case
* snake_to_pascal_case
**Describe how your proposal will work, with code, pseudocode, mockups, and/or diagrams:**
Some methods are [already implemented](https://github.com/godotengine/godot/blob/27ddb27da8a78b21a83ee7ee049fe92d32c66727/modules/mono/editor/bindings_generator.cpp#L117-L183) as used by `mono` module, it's a matter of moving them to the `String` class. Preserves names prefixed with/without an underscore, useful for determining private/public access specifiers implicitly. Related: #641.
**If this enhancement will not be used often, can it be worked around with a few lines of script?:**
Not a few lines. It's a matter of copy-and-pasting existing implementations if talking about C++ specifically, but additional work is required to port this to GDScript. But mostly, these are not even necessary to expose, it would be enough that those methods can be simply moved to core without exposing them to scripting if they are too specific for most users.
**Is there a reason why this should be core and not an add-on in the asset library?:**
I personally expect this to be present on the core level, yet again it's not necessary for those methods to be exposed to scripting in the first place.
Answers:
username_1: need them too for creating plugins to convert script paths to class names
while they are not there, I've created [gdscript methods (gist)](https://gist.github.com/username_1/443b40ba79d5b589a96a16c565952419) |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.