Re: [Rpm-maint] [rpm-software-management/rpm] 4.16: no longer possible to package invalid symlink (#1159)

Le mer. 1 avr. 2020 à 21:54, Ed Santiago a écrit : > We have a package, buildah , > that includes a symlink loop > > as part of its -tests subpackage: > > l

[Rpm-maint] [rpm-software-management/rpm] Fix regression causing segfault on database autodetection (#1160)

If configuration points to non-existent backend, tryBackend() will segfault on the first call. Duh. Regression introduced in commit 3eb0eed3806b41efdf86f0433d0b5d7d6c953561. You can view, comment on, or merge this pull request online at: https://github.com/rpm-software-management/rpm/pull/1160

Re: [Rpm-maint] [rpm-software-management/rpm] 4.16: no longer possible to package invalid symlink (#1159)

I'm sure we could find a workaround. It will be challenging because the same tests will need to work in CI as well as rpm-packaged, root and non-root. I will of course do that work if necessary. I filed this issue in hopes that the rpmbuild change was inadvertent/unintentional and could be rever

Re: [Rpm-maint] [rpm-software-management/rpm] 4.16: no longer possible to package invalid symlink (#1159)

> This is, however, a regression and places us in a difficult situation wrt > packaging our tests. You can just ship a tarball of this stuff instead right? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/

[Rpm-maint] [rpm-software-management/rpm] 4.16: no longer possible to package invalid symlink (#1159)

We have a package, [buildah](https://src.fedoraproject.org/rpms/buildah), that includes a [symlink loop](https://github.com/containers/buildah/blob/master/tests/bud/dockerignore2/subdir/circular-link) as part of its `-tests` subpackage: ``` lrwxrwxrwx. 1 esm esm 13 Oct 21 05:53 circular-link -

Re: [Rpm-maint] [rpm-software-management/rpm] Export CC and CXX in __build_pre (#1154)

> This seems like a fairly obvious thing on the outset, but doing this would > break anybody setting CC/CXX externally in the environment. Which is a > long-standing "feature" that somebody somewhere is almost surely using. > Do you mean that this would prevent someone from setting these environ

Re: [Rpm-maint] [rpm-software-management/rpm] Document all of rpmdeps options (#1157)

Patch updated (semi automatic description of options is not always a good idea :-( ) -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/pull/1157#issuecomment-607339902

[Rpm-maint] [rpm-software-management/rpm] RFE: %_files_listed_twice_terminate_build (#1158)

Would it be possible to have an option to make files listed twice a hard error? Similarly how `%_unpackaged_files_terminate_build` works, I imagine `%_files_listed_twice_terminate_build`. Example: ``` %files /a /a ... warning: File listed twice: /a ``` ``` %global _files_listed_twice_termin

Re: [Rpm-maint] [rpm-software-management/rpm] Document all of rpmdeps options (#1157)

@soig pushed 2 commits. 0f559437a8892f81d29e345e43b1d1ccd62e2a99 describe all rpmdeps options e67395de9df1c4ac449ec251a70a6a3d37f53a2d update -- You are receiving this because you are subscribed to this thread. View it on GitHub: https://github.com/rpm-software-management/rpm/pull/1157/files/

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: optional one-shot/cached macro expansion (#1155)

First seeing spec file and then trying to move some macros into macros file, it was very confusing, I can tell you. From packager POV, the macro file is the same as spec file, but you don't write `%global` everywhere for some unknown reason. Also the difference between `%import` and `%load` (whe

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

Fedora PR: https://src.fedoraproject.org/rpms/python-rpm-generators/pull-request/9 -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/pull/1153#issuecomment-607270301__

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUILD

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUILD

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUILD

Re: [Rpm-maint] [rpm-software-management/rpm] Document all of rpmdeps options (#1157)

@pmatilai commented on this pull request. > +Print the suggests +.TP +\fB--supplements\fR +Print the supplements +.TP +\fB--enhances\fR +Print the enhances +.TP +\fB--conflicts\fR +Print the conflicts +.TP +\fB--obsoletes\fR +Print the obsoletes +.TP +\fB--alldeps\fR +Print the alldeps There's

Re: [Rpm-maint] [rpm-software-management/rpm] Unset SOURCE_DATE_EPOCH for the test-suite (#1148)

Merged #1148 into master. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/pull/1148#event-3187921348___ Rpm-maint mailing list Rpm-mai

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

Unfortunately there's no meaningful way to track build-side rpm capabilities (just an RFE ticket at #717), so you'll just need to use a versioned dependency on rpm. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://g

[Rpm-maint] [rpm-software-management/rpm] Document all of rpmdeps options (#1157)

This will update the man page in order to describe all rpmdeps options. You can view, comment on, or merge this pull request online at: https://github.com/rpm-software-management/rpm/pull/1157 -- Commit Summary -- * describe all rpmdeps options * update -- File Changes -- M doc/rpmde

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: Add variable arrays to rpm (#1149)

That would be awfully nice. The more native elements (macros variable tags) can be manipulated/set/unset from lua, the less we need to worry about the legacy rpm tangle. I don’t think anyone wants to write and maintain complex logic using the native rpm macro language, it’s just that pushing dat

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: add variable named scoping to rpm (#1150)

Even you you do not touch the existing spec tangle, we need to produce new spec tangles all the time, so what can rpm do to make the new spec tangles more sane? State of upstream projects is not frozen, tooling needs to grow like the projects it processes. -- You are receiving this because you

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: Add variable arrays to rpm (#1149)

That said, I'm thinking more about further blurring the line between %{lua:...} and the native rpm macro subsystem. For example, could we expose Lua tables (and/or other variables) as native macro elements. -- You are receiving this because you are subscribed to this thread. Reply to this email

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: Add variable arrays to rpm (#1149)

There are no *variables* in rpm. There are only macros, and %{lua:...} is nothing more, nothing less than a means to generate macro content programmatically. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.co

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

Is there a way to do something like: ``` Requires: rpmlib(ParametricMacroGenerators) ``` Or do I have to do: ``` Requires: rpm > 4.15.90-0 ``` -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: add variable named scoping to rpm (#1150)

The point is, this resembles #573 more than just a little, and is in the ponies department for the same reason: it's not possible to retrofit sanity and order into the existing spec tangle. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view

Re: [Rpm-maint] [rpm-software-management/rpm] Export CC and CXX in __build_pre (#1154)

This seems like a fairly obvious thing on the outset, but doing this would break anybody setting CC/CXX externally in the environment. Which is a long-standing "feature" that somebody somewhere is almost surely using. If we add CC/CXX then why not CPP too? And the whole other bunch of related en

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

Oh, right. Totally forgot about that. I certainly wont object if you want to give it a spin in rawhide before we hit the merge button here, but certainly not required either. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub:

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

In Fedora, we ship this from a different package. So it is not tight on when rpm will be updated but on when python-rpm-generators will. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-managem

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUILD

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

I think we can just merge and deal with possible fallout when it hits rawhide, that shouldn't be too far from now. Just letting this sit here for a day or two to give other Python folks a chance to comment. -- You are receiving this because you are subscribed to this thread. Reply to this email

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@pmatilai commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUIL

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

> The shell version ran for 2m 13s, this version in 0m 18s. Of that 18s, the > part of actual dependency generation is half at most. Nice job! Thanks. Do you want to test this in rawhide before we merge this, or merge and fix later if it appears broken somehow? -- You are receiving this becau

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUILD

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok pushed 1 commit. 1b074aa288082c539e99cf25e6d434818274eeac Drop tabs from python.attr -- You are receiving this because you are subscribed to this thread. View it on GitHub: https://github.com/rpm-software-management/rpm/pull/1153/files/3af2111524b7194f6d7bed70c91d21dcce98556d..1b074aa

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@pmatilai commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUIL

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

> Should I convert the indentation to tabs as well, commit another whitespace > only tab-removal, or leave as is? Up to you. In an attempt to get some timings for this thing: - moved pythondist.attr out of the way (that thing is *expensive*) - ran `for p in $(rpm -qa 'name=python*'); do rpm -ql

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: add variable named scoping to rpm (#1150)

Well, if it existed I would not request it :) It is part of adapting rpm to today’s software projects, that require more subpackaging and more subpackaging context than traditional `make install` things. Dev tools get more parametric, handling the result requires heavier variable passing. --

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: optional one-shot/cached macro expansion (#1155)

Um. And use what instead? The spec file format? Whose parser is the one of the most compilicated and loathed piece of code in rpm and would be difficult to lift out of librpmbuild to put it mildly, and slow every rpm startup in the process? In contrast, the macro file format is a nice, declarat

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: add variable named scoping to rpm (#1150)

Oh. I'm afraid this falls into the ponies department, at least in the realm of the existing spec machinery. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/1150#issuecomm

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: optional one-shot/cached macro expansion (#1155)

Is there any benefit in having macro files? Wouldn't be better to deprecate them instead? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/issues/1155#issuecomment-607124518_

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: Add variable arrays to rpm (#1149)

It’s not a lua vs non-lua thing. I’m an heavy lua user myself. I expect that this feature would unlock more lua use, not less. While very cool and useful, lua is ultimately limited by the stage at which lua expressions are evaluated. So, no matter how much lua you add to a spec, you need, at so

[Rpm-maint] [rpm-software-management/rpm] fix testsuite on Mageia (#1156)

it consistently fails b/c readelf returns more stuff: $ readelf --debug-dump=info ./usr/lib/debug/usr/local/bin/hello2*.debug | grep comp_dir <24> DW_AT_comp_dir: (indirect string, offset: 0x52a): /home/iurt/rpmbuild/BUILD/glibc-2.31/csu <43> DW_AT_comp_dir: (indirect string,

Re: [Rpm-maint] [rpm-software-management/rpm] RFE: add variable named scoping to rpm (#1150)

Thanks for the feedback! It makes me realise, that I forgot to emphasise things that were obvious to me (as a packager). A scope needs an explicit name/anchor to be useful packaging side. It’s not a local vs global thing, it’s a “I do build things for context X now” and an later “I do checks fo

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

Testing has shown this seems to work for the cases where it is supposed to work. I haven't really done any "this package looks like it is almost supposed to provide `python(abi)`, but not really" testing, because I don't know such cases. -- You are receiving this because you are subscribed to th

[Rpm-maint] [rpm-software-management/rpm] RFE: optional one-shot/cached macro expansion (#1155)

Some macros are really expensive to expand, notorious examples include python/perl/etc related macros that invoke the interpreter to fish out a configuration path or such. When these are expanded repeatedly, it really adds up. In a spec one can use %global to achieve this, but that's not possib

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok commented on this pull request. > @@ -1,4 +1,28 @@ -%__python_provides %{_rpmconfigdir}/pythondeps.sh --provides -%__python_requires %{_rpmconfigdir}/pythondeps.sh --requires +%__python_provides() %{lua: +-- Match buildroot/payload paths of the form +--/PATH/OF/BUILD

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

Some testing: ``` $ mock -r fedora-rawhide-x86_64 --enablerepo=local --init $ mock -r fedora-rawhide-x86_64 --enablerepo=local --update $ rpm -q --root ~/rpmbuild/mock/fedora-rawhide-x86_64/root/ rpm rpm-4.15.90-0.git14971.1.fc33.x86_64 $ mock -r fedora-rawhide-x86_64 --copyin python.attr $(rpm -

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

Done. Should I convert the indentation to tabs as well, commit another whitespace only tab-removal, or leave as is? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/pull/1153#iss

Re: [Rpm-maint] [rpm-software-management/rpm] Reimplement pythondeps.sh as parametric macro generators (#1153)

@hroncok pushed 1 commit. 3af2111524b7194f6d7bed70c91d21dcce98556d Reimplement pythondeps.sh as parametric macro generators -- You are receiving this because you are subscribed to this thread. View it on GitHub: https://github.com/rpm-software-management/rpm/pull/1153/files/a140a5b0d6d2da9fc9