Bug#702882: Status of igraph packaging for Debian
Hi Tamás, thanks for your work on this package. I guess it is OK for you if I added you to the Uploaders field and injected you as the owner as the latest changelog entry. On Fri, Aug 29, 2014 at 11:54:30PM +0200, Tamás Nepusz wrote: ... Anyway, I think that this bug can safely be closed once the updated package (0.7.1-2) is uploaded. Most of the patches that I added to 0.7.1-2 now can also be removed once upstream releases the next version because these patches are also merged into upstream. You can close bugs inside d/changelog and so I did. Thanks again Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#702882: Status of igraph packaging for Debian
Hi Andreas, I realised that there remains an open bug to version 0.6.5. Could you please check whether this exists or not. I have recently committed a patch that seems to have fixed this bug. In particular, we are not linking to f2c any more because it is needed only if we also use the internal BLAS, LAPACK or ARPACK implementations (which we don't use). One of my other patches also ensures that dh_auto_test runs all the unit tests, and they all compile just fine. FWIW, the patch I committed is temporary; the configure script in the upstream repo already does a better job at deciding whether linking to f2c is needed or not, but this version is likely to be included in the next release only. The patch I committed here only ensures that we are not linking to f2c at all, it does not attempt to fix the detection of libf2c or the logic that decides whether linking to f2c is necessary or not. Anyway, I think that this bug can safely be closed once the updated package (0.7.1-2) is uploaded. Most of the patches that I added to 0.7.1-2 now can also be removed once upstream releases the next version because these patches are also merged into upstream. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#702882: Status of igraph packaging for Debian
On Thu, Aug 21, 2014 at 8:33 AM, Andreas Tille andr...@fam-tille.de wrote: Hi Tamás, I realised that there remains an open bug to version 0.6.5. Could you please check whether this exists or not. Mathieu, what do you think? I am not interested in igraph anymore: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=729516 the remaining bug is still the same as back with our previous discussion: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=702882#77 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#702882: Status of igraph packaging for Debian
Hi Tamás, I realised that there remains an open bug to version 0.6.5. Could you please check whether this exists or not. Mathieu, what do you think? Kind regards Andreas. On Mon, Aug 18, 2014 at 12:09:51PM +0200, Tamás Nepusz wrote: Hi Andreas, OK, I have updated debian/watch file (please checkout svn[1]). I'd suggest to always use this for packaging issues and keep us (= the Debian Med team) informed about needed changes. The new version was just uploaded to unstable. Thanks, looks great! Also, I'm happy to maintain the offician debian/ dir of igraph at svn.debian.org in the future. I'll send a request to join the Debian-med team on Alioth soon. Also, I was wondering whether I should also maintain the debian/ dir of python-igraph, but it looks like that belongs to a different team so maybe it's better if I ask there. What I would be really wondering is the Role of R igraph. I realised that it contains a code copy of (c-)igraph which is somehow against Debian policy. I'm not that familiar with R packaging (on our side it is Gabor who is responsible for developing and maintaining the R interface of igraph) but I have a vague recollection that he once said that R packages that depend on an external C library usually also contain a copy of the source code of that library and compile the whole thing (i.e. the dependent library and the actual glue code between R and the library) in one batch when someone runs install.packages(...) from R. This is probably the reason why the R interface of igraph also contains a copy of the C library. But again, I never created R packages so I might be wrong here and the best is probably ask someone who is more familiar with R packages to see how it is done in practice (as I'm pretty sure that igraph is not the only R package that depends on an external C library). Best regards, Tamas -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
