Well then, I don't know what your draft looked like before you cleaned it up, but it should be easier to plug in another method now, I think . :)

Is the current support for PAM sufficient? https://git.sr.ht/~sircmpwn/meta.sr.ht/commit/4d11e6017625ef2eb34f09c0522c4486d70bd8b9

Followup on #156

It's now possible to visit e.g.

https://git.sr.ht/~sircmpwn/git.sr.ht/log/master/setup.py

Instead of providing links to each commit such as

https://git.sr.ht/~sircmpwn/git.sr.ht/commit/c1f95e2a03b5b3264ac3d029cc4efb4d0b152dd0

the link should be

https://git.sr.ht/~sircmpwn/git.sr.ht/commit/c1f95e2a03b5b3264ac3d029cc4efb4d0b152dd0#setup.py

This is helpful in commits with many changed files.

pkgconf contains a very strange regression from pkg-config

I've never seen pkgconf print anything other than 'virtual:world' and it makes the error message unnecessarily-confusing.

Your description of error message wording implementations as a "regression" makes your bugtracker ticket submission unnecessarily confusing.

Other list software sometimes implements the same feature.

majordomo@domain.com can accept the mailing list command "index" to get a list of files which can be retrieved using the mailing list command "get"; these files may be generated archives of the list but apparently can be anything, or nothing as archives don't seem to be an automatic feature.

ezmlm explicitly supports past message retrieval; it can be emailed at {list}-get.{startnum}-{endnum}@domain.com and returns one attachment per email

The UX for this is a bit on the bad side, but you can already click "Details" -> Download raw message, and open the result in e.g. Thunderbird via File -> Open -> Saved Message.

You could manually terminate stdin e.g. run commands with </dev/null. Depending on what the goal is in allocating a tty, this might be possible to do automatically? Though note that not all tools actually check stdin, some of them check /dev/tty...

Duplicate of:

• ~sircmpwn/git.sr.ht#122
• ~sircmpwn/git.sr.ht#264
• ~sircmpwn/git.sr.ht#290

Will be solved by ~sircmpwn/git.sr.ht#205

So it turns out that this is generally an issue that also causes https://github.com/swaywm/sway/issues/4603

Meh, looks like someone already sent the exact same patch in: https://lists.sr.ht/~sircmpwn/sr.ht-dev/patches/8336