On 2019-06-12 17:23:31 +0200, Ivo De Decker wrote: > Hi, > > On Thu, Jun 13, 2019 at 12:13:30AM +0900, Vincent Lefevre wrote: > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=930430 contains: > > > > "Package: libasan5; Maintainer for libasan5 is Debian GCC Maintainers > > <[email protected]>; Source for libasan5 is src:gcc-9 (PTS, > > buildd, popcon). > > [...] > > Found in version gcc-8/8.3.0-7" > > > > The "src:gcc-9" is incorrect, as the source for libasan5 is gcc-8 > > (at least for the reported version), not gcc-9. The bug should > > appear on > > > > https://bugs.debian.org/cgi-bin/pkgreport.cgi?src=gcc-8 > > > > not on > > > > https://bugs.debian.org/cgi-bin/pkgreport.cgi?src=gcc-9 > > The BTS looks at the newest know version to decide the source <-> binary > mapping. If the mapping is different in unstable and experimental, the one > from experimental will be used (because the version is higher). That's what's > happening here (libasan5 is built by gcc-9 in experimental). > > It might be better to prefer the mapping from unstable over the one > from experimental, even though that will be wrong in some other > cases. I don't know if it's realistic to have the mapping depend on > the version information in the bug.
Perhaps it should use the latest version *and* add any source from the "found" field. This should affect the page of the bug and the one for any such source, e.g. https://bugs.debian.org/cgi-bin/pkgreport.cgi?src=gcc-8 This will be useful even when unstable gets the new version. -- Vincent Lefèvre <[email protected]> - Web: <https://www.vinc17.net/> 100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/> Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)
