| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
Sorry, something went wrong.
|
I'd clarify that this isn't for reporting bugs in Python, but that it's for bugs in documents. But yeah, I think this would be an improvement. |
Sorry, something went wrong.
|
Would some headings help? (This is for code, but for example.) On that .github/ISSUE_TEMPLATE.md example, I see GitHub now tells me:
So shall we use the new .github/ISSUE_TEMPLATE/config.yml version? |
Sorry, something went wrong.
@hugovk To my mind, it seems overkill for bugs in documentation. I assume that people reporting bugs here are experienced developers (if they follow the PEP1 process, i.e. they report only bugs), and they don't need such issue forms. What's your opinion on that? But if there's a consensus, I can update the template to use the yaml file. |
Sorry, something went wrong.
| @@ -0,0 +1,23 @@ | |||
| Issue tracker is **ONLY** used for reporting bugs on documents (PEPs), not | |||
There was a problem hiding this comment.
| Issue tracker is **ONLY** used for reporting bugs on documents (PEPs), not | |
| This issue tracker is **ONLY** used for reporting bugs on documents (PEPs), not |
Sorry, something went wrong.
| @@ -0,0 +1,23 @@ | |||
| Issue tracker is **ONLY** used for reporting bugs on documents (PEPs), not | |||
| bugs in Python. | |||
There was a problem hiding this comment.
We could link to https://devguide.python.org/tracker/ as the place to report bugs in Python (and let's call it CPython).
Sorry, something went wrong.
| New features should be discussed according to the following process: | ||
| [PEP 1](https://peps.python.org/pep-0001/#start-with-an-idea-for-python) |
There was a problem hiding this comment.
| New features should be discussed according to the following process: | |
| [PEP 1](https://peps.python.org/pep-0001/#start-with-an-idea-for-python) | |
| New features should be discussed according to | |
| [PEP 1](https://peps.python.org/pep-0001/#start-with-an-idea-for-python). |
Sorry, something went wrong.
There was a problem hiding this comment.
@hugovk To my mind, it seems overkill for bugs in documentation. I assume that people reporting bugs here are experienced developers (if they follow the PEP1 process, i.e. they report only bugs), and they don't need such issue forms. What's your opinion on that?
I've not used the new format yet. If it's only for forms, sure, this one is simple enough :)
Sorry, something went wrong.
| ### What did you spot, and where? | ||
|
|
||
| ### What did you expect to be changed? | ||
|
|
||
| ### Why? |
There was a problem hiding this comment.
Add a bit of space for text entry:
| ### What did you spot, and where? | |
| ### What did you expect to be changed? | |
| ### Why? | |
| ### What did you spot, and where? | |
| ### What did you expect to be changed? | |
| ### Why? | |
Sorry, something went wrong.
A lot of them aren't, and many are experienced developers and still might gloss over the issue template. The main goal isn't to give contributors more boilerplate text to read, edit and delete (or ignore), which the Contributing Guide and PEP 1 (especially with #2378 ) already cover. Rather, it should be to provide clear, hard to miss, easy to use and actionable pre-triage, so they can understand at a glance whether their issue is appropriate here, and if not, where to take it—which is what the template chooser does.
Actually, no, the config.yaml isn't the configuration for the issue creation forms, which I don't think we need here, its for the far more useful (at least in this case) template chooser. See, for example, Numpy's; we wouldn't need that many options, but it gives you an example of what it looks like and what can be done with it. In summary, it gives users a set of short, simple and actionable options (that we configure) when clicking the New Issue button, which can be to one or more issue templates (or even just a blank issue), or to external sites (Python-Ideas, Discourse, BPO, etc), rather than dumping them straight to a raw issue with a bunch of pre-filled plain-text boilerplate. This is way friendlier and more effective, especially for what we're trying to do (redirect users with issues that don't belong here to better places) versus just a generic issue template, because it actually makes sure users read them and affirmatively choose one, while keeping them short and very quick to navigate. |
Sorry, something went wrong.
|
I'd also suggest waiting just a bit until #2378 is merged (which maybe we can go ahead and do now, since every active PEP editor has now approved it and the last comments were addressed), since that changes and improves the existing text and provides new/updated sections that we can link to directly in PEP 1 and the contributing guide. |
Sorry, something went wrong.
|
I am not enthusiastic about this change. Inappropriate issues are rare and easily dealt with, and the proposed templates add more overhead for us maintainers to deal with and more boilerplate for new contributors to read. |
Sorry, something went wrong.
|
I'd suggest just ditching the template and adding the recommended an issue chooser instead, with one button to just open a blank issue with a line of text explaining what issues here are for (or perhaps one for PEP text issues and one for website issues, with perhaps a link of text each), and maybe a couple other options with links to other places, e.g. "Report a bug in Python" -> BPO/GH, "Propose a new idea" -> Python-Ideas/Discourse Ideas, etc. This is fairly simple for us to maintain, minimizes the text new users need to read to the most important few lines while ensuring they read it, and allows experienced users to go right to an issue with one click and no boilerplate. |
Sorry, something went wrong.
|
@vpoulailleau Would you like to make a fresh PR following CAM's suggestion? For reference, the CPython repo just added something a bit like this Thanks! |
Sorry, something went wrong.
|
@hugovk @CAM-Gerlach OK, I'll do this this week. |
Sorry, something went wrong.
| Back | FazBrowse Home | New Git URL |
Many issues are useless and can be avoided with an issue template. This template would have avoided (among others):
#2491 #2486 #2469 #2100 #2076 #1898 #1867 #1747 #1273