Control: severity -1 serious Control: reassign -1 libschroedinger-maeparser1 1.2.2-1 Control: retitle -1 libschroedinger-maeparser1: SONAME change without package name change
On Fri, Jan 10, 2020 at 08:41:47PM +1100, Stuart Prescott wrote: > Hi Julien, > > On Friday, 10 January 2020 19:44:37 AEDT Julien Cristau wrote: > > On Fri, Jan 10, 2020 at 03:57:01PM +1100, Stuart Prescott wrote: > > > Package: release.debian.org > > > Severity: normal > > > User: release.debian....@packages.debian.org > > > Usertags: binnmu > > > > > > nmu schroedinger-coordgenlibs_1.3-1 . ANY . unstable . -m "Rebuild against > > > libschroedinger-maeparser-dev >= 1.2.2-1" > > > > > > The binNMU of rdkit (#946247) did not fix #945985; it appears that it is > > > schroedinger-coordgenlibs that actually needs rebuilding to pick up the > > > new > > > filename of libmaeparser: > > > > > > $ ldd /usr/lib/x86_64-linux-gnu/libcoordgen.so.1.3 > > > > > > linux-vdso.so.1 (0x00007fff9f9bd000) > > > libmaeparser.so.1.2 => not found > > > libstdc++.so.6 => /lib/x86_64-linux-gnu/libstdc++.so.6 > > > (0x00007f5445999000) > > > libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007f5445854000) > > > libgcc_s.so.1 => /lib/x86_64-linux-gnu/libgcc_s.so.1 > > > (0x00007f544583a000) > > > libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007f544567a000) > > > /lib64/ld-linux-x86-64.so.2 (0x00007f5445c12000) > > > > > > (rebuilding the package has the linking pick up libmaeparser.so.1 instead) > > > > That sounds wrong. If libmaeparser.so.1.2 was renamed to > > libmaeparser.so.1 then its package name should have changed? > > Yes, I'd say so. I'm guessing that libschroedinger-maeparser1 was the wrong > name to begin with or more likely that the "1.2" suffix was incorrect and > then > fixed; they are right(ish) now for that soname; I assume this was an > accidental > transition. > > Since there is only source package with a dependency on libschroedinger- > maeparser1 (this one), a binNMU seems like the simplest solution; it fixes > the > current packages and then lets me get on to the RC bugs I was actually trying > to fix... > Yeah, no, let's fix this properly in schroedinger-maeparser. Cheers, Julien
