Re: [Rpm-maint] [rpm-software-management/rpm] Add rpm.spawn() Lua API (PR #3241)

2024-08-16 Thread Panu Matilainen
@pmatilai pushed 1 commit. 52938a9d4825a659710d545482baf160480dcd99 Add rpm.spawn() Lua API -- View it on GitHub: https://github.com/rpm-software-management/rpm/pull/3241/files/6e8d0ed17dabd2a287818dfba70cf0ba6798bbfa..52938a9d4825a659710d545482baf160480dcd99 You are receiving this because you

[Rpm-maint] [rpm-software-management/rpm] Refactor ordering code to use STL containers (PR #3243)

2024-08-16 Thread Panu Matilainen
This is all a rather crude, mechanical conversion, I bet there's tonnes of further cleanup possible. The main point here is to get rid of C-side allocations. You can view, comment on, or merge this pull request online at: https://github.com/rpm-software-management/rpm/pull/3243 -- Commit Summa

Re: [Rpm-maint] [rpm-software-management/rpm] Consider making %buildsystem_name_conf optional (Discussion #3191)

2024-08-16 Thread Panu Matilainen
Specs utilizing a known buildsystem are a very different animal to the rest of the wild west. The idea was basically to push people towards taking advantage of the %conf section in the Buildsystem, because "surely all major buildsystem has a configure-like step", right? Seems it's been a while

[Rpm-maint] [rpm-software-management/rpm] Consider making %buildsystem_name_conf optional (Issue #3244)

2024-08-16 Thread Panu Matilainen
### Discussed in https://github.com/rpm-software-management/rpm/discussions/3191 Originally posted by **hroncok** July 3, 2024 `%conf` is an optional section. Hence, I believe mandating the presence of `%buildsystem_name_conf` for a declarative buildsystem is suboptimal. I was able to "solve"

Re: [Rpm-maint] [rpm-software-management/rpm] Consider making %buildsystem_name_conf optional (Issue #3244)

2024-08-16 Thread Panu Matilainen
Specs utilizing a known buildsystem are a very different animal to the rest of the wild west. The idea was basically to push people towards taking advantage of the %conf section in the Buildsystem, because "surely all major buildsystem has a configure-like step", right? Seems it's been a while

Re: [Rpm-maint] [rpm-software-management/rpm] Consider making %buildsystem_name_conf optional (Discussion #3191)

2024-08-16 Thread Panu Matilainen
Opened https://github.com/rpm-software-management/rpm/issues/3244 from this, along with a more elaborate answer why it's that way. Thanks for the suggestion. -- Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/discussions/3191#discussioncomment-10

Re: [Rpm-maint] [rpm-software-management/rpm] Consider making %buildsystem_name_conf optional (Discussion #3191)

2024-08-16 Thread Panu Matilainen
Closed #3191 as resolved. -- Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/discussions/3191 You are receiving this because you are subscribed to this thread. Message ID: ___ Rpm-maint mailing list Rp

Re: [Rpm-maint] [rpm-software-management/rpm] posix.redirect2null deprecated without replacement (Issue #3192)

2024-08-16 Thread Panu Matilainen
On a related note, I just realized that we can't really *remove* these even in v6, because as rare as the use of these functions are, those uses are on some of the most central packages of distros and removing them would suddenly make all current Fedora/RHEL versions uninstallable. Which would b