https://bugs.freedesktop.org/show_bug.cgi?id=64360
--- Comment #5 from David Henningsson <[email protected]> --- (In reply to comment #4) > (In reply to comment #3) > > (In reply to comment #1) > > > Even if pa_format_info_free2() was never part of public API as exposed in > > > headers? > > > > Yes. > > > > That's the rules of libtool. > > I don't think so. pa_format_info_free2() was never part of the ABI, except > in a very technical sense that isn't relevant here, in my opinion. > > > And Debian/Ubuntu nowadays automatically checks > > this and complains loudly (i e build failure) if a symbol is removed. > > Is there no way to skip the check? Having read up a little, it seems there are a few different ways of working around this check, yes. > > We need to put it back. > > > > Btw, I did the same thing for FluidSynth once (removed a symbol not exposed > > in the headers), and got loud complaints from more than one distro > > maintainer. I ended up putting it back. Looking back, I misremembered, it was only one distro maintainer. The other complained about something else. (Reference - http://sourceforge.net/apps/trac/fluidsynth/ticket/98 ) > Why do the distro maintainers demand adding cruft back? I understand if they > complain and call us stupid and whatever, because we cause extra work for > them, but there's absolutely no long term need to add pa_format_info_free2() > back. It's only the distro tools that will cause trouble, no application > code will ever break because of this. It would be good to have more opinions on this. Sjoerd, what do you think? -- You are receiving this mail because: You are the QA Contact for the bug. You are the assignee for the bug.
_______________________________________________ pulseaudio-bugs mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/pulseaudio-bugs
