Re: Proposed MBF: Removal of libfreetype6-dev (causing FTBFS)
On Saturday, 19 August 2023 11:14:02 CEST Andreas Metzler wrote: > > What is the recommended/appropriate way to deal with such issues? > > The agreed reached was not "let's ignore it, lintian has been warning > about it". Instead a way forward that /should/ have avoided any breakage > (versioned provides) was proposed and chosen. Ah, the versioned provides is what makes it *not* FTBFS! I missed that as had been added on 2021-12-28 already (bug #1002049) and I didn't look back in the history far enough. (I may also not have made the connection though) Thanks :-) signature.asc Description: This is a digitally signed message part.
Re: Proposed MBF: Removal of libfreetype6-dev (causing FTBFS)
On Saturday, 19 August 2023 10:54:20 CEST Sven Joachim wrote: > On 2023-08-19 10:03 +0200, Diederik de Haas wrote: > > [please CC me as I'm not subscribed to debian-devel] > > > > On Mon, 17 Jul 2023 at 21:45:13 +1000, Hugh McMaster wrote: > >> On Mon, 17 Jul 2023 at 00:07, Simon McVittie wrote: > >> > On Sun, 16 Jul 2023 at 22:38:20 +1000, Hugh McMaster wrote: > >> > > Currently, there are 219 build-dependencies and 29 (direct) > >> > > dependencies on libfreetype6-dev, which has been released with > >> > > bullseye and bookworm. > >> > > >> > Lintian diagnoses this as "[build-]depends-on-obsolete-package" since > >> > 2.116.0 (MR at [1], instances of the relevant tags listed at [2] and > >> > [3]) which will hopefully help progress towards dropping the > >> > transitional > >> > package. > >> > >> Thanks for pointing this out. I wasn't aware Lintian had started > >> flagging dependencies on obsolete packages some 10 months ago. > >> > >> Having Lintian issue a warning or error instead of bug filing is > >> preferable.> > > While it's true that lintian did issue an error, now that src:freetype has > > been updated and libfreetype6-dev has been dropped, there are a number of > > packages which hadn't been updated and now FTBFS. > > Could you please name an example? Hmm. It appears there is something wrong in my reasoning. I first looked at https://tracker.debian.org/pkg/xft, so that would be my example. It Build-Depends on libfreetype6-dev (and libfontconfig1-dev), so I assumed that when that B-D no longer exists, it would thus FTBFS. So I made https://salsa.debian.org/xorg-team/lib/xft/-/merge_requests/3 to fix that, but while it did make the CI pipeline succeed, it was only after your message that I realized that the 'before' pipeline *did* succeed in the 'build' job ... which indicates it does NOT FTBFS. But I still don't understand why. Can you point out where my reasoning is incorrect and that a 'disappearing' B-D does not (automatically) cause a FTBFS? > At the time I recommended just removing the libfreetype6-dev package[2], > based on my experience with the transitional -dev packages in ncurses, > where this approach worked without a hitch. What is different in > freetype? I did see your message, but as described above I didn't/don't understand why that apparently does work. Cheers, Diederik signature.asc Description: This is a digitally signed message part.
Re: Proposed MBF: Removal of libfreetype6-dev (causing FTBFS)
On 2023-08-19 Diederik de Haas wrote: > [please CC me as I'm not subscribed to debian-devel] > On Mon, 17 Jul 2023 at 21:45:13 +1000, Hugh McMaster wrote: > > On Mon, 17 Jul 2023 at 00:07, Simon McVittie wrote: > > > On Sun, 16 Jul 2023 at 22:38:20 +1000, Hugh McMaster wrote: > > > > Currently, there are 219 build-dependencies and 29 (direct) > > > > dependencies on libfreetype6-dev, which has been released with > > > > bullseye and bookworm. [...] > > > Lintian diagnoses this as "[build-]depends-on-obsolete-package" since [...] > > Thanks for pointing this out. I wasn't aware Lintian had started > > flagging dependencies on obsolete packages some 10 months ago. > > Having Lintian issue a warning or error instead of bug filing is preferable. > While it's true that lintian did issue an error, now that src:freetype has > been updated and libfreetype6-dev has been dropped, there are a number of > packages which hadn't been updated and now FTBFS. [...] > As the FTBFS wrt libfreetype6-dev was predicted and announced [1], wouldn't > it > have been better if the MBF had taken place? > What is the recommended/appropriate way to deal with such issues? The agreed reached was not "let's ignore it, lintian has been warning about it". Instead a way forward that /should/ have avoided any breakage (versioned provides) was proposed and chosen. https://lists.debian.org/debian-devel/2023/07/msg00193.html cu Andreas
Re: Proposed MBF: Removal of libfreetype6-dev (causing FTBFS)
On 2023-08-19 10:03 +0200, Diederik de Haas wrote: > [please CC me as I'm not subscribed to debian-devel] > > On Mon, 17 Jul 2023 at 21:45:13 +1000, Hugh McMaster wrote: >> On Mon, 17 Jul 2023 at 00:07, Simon McVittie wrote: >> > On Sun, 16 Jul 2023 at 22:38:20 +1000, Hugh McMaster wrote: >> > > Currently, there are 219 build-dependencies and 29 (direct) >> > > dependencies on libfreetype6-dev, which has been released with >> > > bullseye and bookworm. >> > >> > Lintian diagnoses this as "[build-]depends-on-obsolete-package" since >> > 2.116.0 (MR at [1], instances of the relevant tags listed at [2] and >> > [3]) which will hopefully help progress towards dropping the transitional >> > package. >> >> Thanks for pointing this out. I wasn't aware Lintian had started >> flagging dependencies on obsolete packages some 10 months ago. >> >> Having Lintian issue a warning or error instead of bug filing is preferable. > > While it's true that lintian did issue an error, now that src:freetype has > been updated and libfreetype6-dev has been dropped, there are a number of > packages which hadn't been updated and now FTBFS. Could you please name an example? > AFAIUI there are people and/or tools which periodically rebuild packages to > see if a 'sudden' change has caused a FTBFS and that then gets followed up by > a MBF effort. > As the FTBFS wrt libfreetype6-dev was predicted and announced [1], wouldn't it > have been better if the MBF had taken place? At the time I recommended just removing the libfreetype6-dev package[2], based on my experience with the transitional -dev packages in ncurses, where this approach worked without a hitch. What is different in freetype? > [1] https://lists.debian.org/debian-devel/2023/07/msg00193.html Cheers, Sven 2. https://lists.debian.org/debian-devel/2023/07/msg00195.html
Re: Proposed MBF: Removal of libfreetype6-dev (causing FTBFS)
[please CC me as I'm not subscribed to debian-devel] On Mon, 17 Jul 2023 at 21:45:13 +1000, Hugh McMaster wrote: > On Mon, 17 Jul 2023 at 00:07, Simon McVittie wrote: > > On Sun, 16 Jul 2023 at 22:38:20 +1000, Hugh McMaster wrote: > > > Currently, there are 219 build-dependencies and 29 (direct) > > > dependencies on libfreetype6-dev, which has been released with > > > bullseye and bookworm. > > > > Lintian diagnoses this as "[build-]depends-on-obsolete-package" since > > 2.116.0 (MR at [1], instances of the relevant tags listed at [2] and > > [3]) which will hopefully help progress towards dropping the transitional > > package. > > Thanks for pointing this out. I wasn't aware Lintian had started > flagging dependencies on obsolete packages some 10 months ago. > > Having Lintian issue a warning or error instead of bug filing is preferable. While it's true that lintian did issue an error, now that src:freetype has been updated and libfreetype6-dev has been dropped, there are a number of packages which hadn't been updated and now FTBFS. AFAIUI there are people and/or tools which periodically rebuild packages to see if a 'sudden' change has caused a FTBFS and that then gets followed up by a MBF effort. As the FTBFS wrt libfreetype6-dev was predicted and announced [1], wouldn't it have been better if the MBF had taken place? What is the recommended/appropriate way to deal with such issues? [1] https://lists.debian.org/debian-devel/2023/07/msg00193.html signature.asc Description: This is a digitally signed message part.
