#Attempt to clarify environment marker ev...

1 messages · Page 1 of 1 (latest)

mighty vault
#

(i had reviewed henry's writeup)

spiral dirge
#

Generally just "at least 1 of the editors is willing to sign off on merging it". This has the quirk that when one of us writes a PR, it needs two editors to approve it (since the repo permissions are set up to disallow merging our own PRs).

mighty vault
spiral dirge
#

Yeah, I'm with @vague trout on that one - link rot hitting every PR is a pain, but expecting us to check links manually instead isn't a good answer

mighty vault
#

i mean the link check still runs, it's just not blocking anymore

#

the problem is that the link check currently blocks any change anytime an unrelated page goes down

#

i'd like to file fixes to specs, but you currently run into an unclear approval process and technical hurdles like linkchecker rejecting valid PRs

vague trout
spiral dirge
#

Bumping this, since the link check situation has been improved (thanks @mighty vault , @vague trout )

mighty vault
#

looks like pradyun tried to merge it, but it needs an approval for the merge to go through?

#

(currently reading)

spiral dirge
#

Thanks for that review @mighty vault - the unknown markers and extra == ... clarifications were two important issues to sort out.

I agree in spirit with the Graal Python change, but only went as far as dropping the Jython call out, as I'm not sure what platform.python_implementation() actually emits there (if they haven't patched the stdlib, it will likely be nonsense)