There are almost 3k issues open in Julia’s GitHub repository:
Taking a glance at those, it got me wondering. Has there been a sustained effort to review old issues and close the ones that are no longer relevant to the supported versions of Julia?
Look at the issues labeled “doc” for example. Many of those were opened even before Julia 1.0 and it seems many of those are related to technical issues. Not knowing the internals of Julia before that, I can’t tell if those still apply, but I can guess many of them do not.
imho - better links with
If you want to do this, I imagine that
- fixes to some old issues where there are no open questions/decisions (“just” work that someone needs to do) are welcome,
- asking that clearly outdated issues (because the language moved on so the question is no longer relevant, the bug got fixed, etc) are closed is also helpful,
- but many old issues are sitting around waiting for some kind of a major/minor feature to be implemented, which may not even be decided yet. Reviewing these is best left to core devs, but of course people could revive these issues if the want to (especially if they feel like contributing, and just want some guidance). However, pinging these issues en masse for no good reason may not be ideal.
If you want to contribute, the “good first issue” label is the best place to strart.
People occasionally go through and try to clear out stale issues, but it’s kind of tedious work, so it’s not done all that often. Might be good to have a “spring cleaning” period at some point where people collaborate on the cleaning.
I do a pass of old issues approximately bi-annually. Many can be closed but many are still relevant.
Tangentially, I tried to figure out a way to use the github API to see which user closed an issue (so I could count them) but didn’t figure out a way.
Check here: https://developer.github.com/v3/issues/#response-4
It is in
Here is a Stack Overflow question regarding
merged_by.login for PRs (very similar) and a Shell script written for tldr-pages/tldr repository.