Re: EDFlib 1.12 (new version available)
Hi, Because of a bad feature check, the build failed on hurd. I tried to properly fix it using cmake, which I have no previous experience with. Please review and dput again. ☺ Fabian
Re: CI for parsinsert [Outreachy]
Dear Nadiya, On Mon, Aug 21, 2017 at 01:49:50PM -0700, Nadiya Sitdykova wrote: > I pushed tests for parsinsert package during DebConf but forgot to write > you about it. You could have even told me. :-P > I used some test data from upstream version > https://sourceforge.net/projects/parsinsert/files/ and I'm not sure if it's > worth mentioning in README.source and additions to debian/copyright. What > do you think? Please add it to README.source. Its just helpful documentation for those who might want to reproduce or enhance anything. As always thanks for your work on this Andreas. -- http://fam-tille.de
Re: CI for parsinsert [Outreachy]
Ahhh, please wait for my migration SVN -> Git before you push anything. I'll keep you updated once I'm finishing. On Tue, Aug 22, 2017 at 11:28:45AM +0200, Andreas Tille wrote: > Dear Nadiya, > > On Mon, Aug 21, 2017 at 01:49:50PM -0700, Nadiya Sitdykova wrote: > > I pushed tests for parsinsert package during DebConf but forgot to write > > you about it. > > You could have even told me. :-P > > > I used some test data from upstream version > > https://sourceforge.net/projects/parsinsert/files/ and I'm not sure if it's > > worth mentioning in README.source and additions to debian/copyright. What > > do you think? > > Please add it to README.source. Its just helpful documentation for > those who might want to reproduce or enhance anything. > > As always thanks for your work on this > >Andreas. > > -- > http://fam-tille.de > > -- http://fam-tille.de
Sponsored upload: odil 0.8.0-3
Dear all, I believe I've fixed #871301 in the Git repo of Odil: * the build dep is fixed * I don't generate a symbols file, but the required symbols are both present in the .so and I've added the dh_makeshlibs override I've also fixed the detection of Python version and updated the standards version. It builds correctly and Lintian does not complain, so could I get an upload to unstable? Cheers, -- Julien signature.asc Description: OpenPGP digital signature
Re: CI for grinder [Outreachy]
Dear Nadiya, On Mon, Aug 21, 2017 at 03:08:04PM -0700, Nadiya Sitdykova wrote: > There was a problem with rebuilding of this package, MYMETA.yml file was > missing, so after some time I realized how to fix it. This file was copied > into the folder that didn't exist, so I created this folder and package > rebuilds just fine. > > I used `autopkgtest-pkg-perl` as a default test-suite here, but it runs > only the test which is loads the program. The thing is in debian/rules I > find those lines: > > # Because of Grinder's random output, some tests are brittle and may > fail...# Only test that the program loads successfully > > > and when I run all tests some of them really fail and it changes from run > to run. So I decided to run only 00-load.t as autopkgtest and wrote > information to the README.test how to run all tests. I think that's OK and thus I uploaded. Thanks for your work on this Andreas. -- http://fam-tille.de
Re: CI for parsinsert [Outreachy]
On Tue, Aug 22, 2017 at 11:35:23AM +0200, Andreas Tille wrote: > Ahhh, please wait for my migration SVN -> Git before you push > anything. I'll keep you updated once I'm finishing. Package moved to Git now. Please ping me if you have added the information about data sets (in case I might not realise your commit). Kind regards Andreas. > On Tue, Aug 22, 2017 at 11:28:45AM +0200, Andreas Tille wrote: > > Dear Nadiya, > > > > On Mon, Aug 21, 2017 at 01:49:50PM -0700, Nadiya Sitdykova wrote: > > > I pushed tests for parsinsert package during DebConf but forgot to write > > > you about it. > > > > You could have even told me. :-P > > > > > I used some test data from upstream version > > > https://sourceforge.net/projects/parsinsert/files/ and I'm not sure if > > > it's > > > worth mentioning in README.source and additions to debian/copyright. What > > > do you think? > > > > Please add it to README.source. Its just helpful documentation for > > those who might want to reproduce or enhance anything. > > > > As always thanks for your work on this > > > >Andreas. > > > > -- > > http://fam-tille.de > > > > > > -- > http://fam-tille.de > > -- http://fam-tille.de
Re: CI for minia [Outreachy]
Dear Nadiya, On Mon, Aug 21, 2017 at 02:09:43PM -0700, Nadiya Sitdykova wrote: > I've noticed that you uploaded new version of minia and I wanted to check > if my tests still working. I did not really *uploaded* a new version of minia I just commited a packaging attempt. > The file which I used as test data is missing in > new package version so I was concerned what would happen with my test and I > thought maybe I need to add this file. I cloned current version of minia > and tried to built it, but it failed with this message in verbose mode: > > dh_auto_configure: cmake .. -DCMAKE_INSTALL_PREFIX=/usr > -DCMAKE_VERBOSE_MAKEFILE=ON -DCMAKE_BUILD_TYPE=None > -DCMAKE_INSTALL_SYSCONFDIR=/etc -DCMAKE_INSTALL_LOCALSTATEDIR=/var > -DCMAKE_EXPORT_NO_PACKAGE_REGISTRY=ON > -DCMAKE_FIND_PACKAGE_NO_PACKAGE_REGISTRY=ON returned exit code 1 > cd /home/nadiya/minia > debian/rules:9: recipe for target 'build' failed > make: *** [build] Error 2 > > Is this happening because the package is still "work in progress" and I > shouldn't worry about it, or there is really something wrong? Your assumption is perfectly correct. I was also asking here on this list why we are packaging legacy minia version. I think I answered this question in my last changelog entry: TODO: package https://github.com/GATB/gatb-core If you want to try the packaging that would be perfectly welcome. I think you have seen a sufficient amount of packages to know how to do the packaging. Feel free to decide whether you want to add more tests to any package of your choice or do the packaging of gatb-core. Both is equally valuable for the project and feel free to take the challenge which fits you more. Please always remember: Just ask here before sitting for hours on a problem that could be answered here easily. Have fun Andreas. -- http://fam-tille.de
Re: Last commit did not help (Was: [libbpp-phyl] branch master updated (85dac66 -> c4bc05c))
Hi Julien, On Tue, Aug 22, 2017 at 06:50:02AM +0200, Julien Yann Dutheil wrote: > Dear Andreas, > > Ok, this is as far as I get and need a bit of help I'm afraid. > Using pdebuilder shows no issue (apart from a few lintian warnings). > Using gbp buildpackage --git-ignore-new however leads to a long list of > errors starting by: Hmmm, I have configured gbp buildpackage in a way that pdebuild is actually used to build the package. You can do this in ~/.gbp.conf (see Debian Med policy for an example). For me it looks like a broken clean target in debian/rules when trying to build the package twice in a row. > I: using cowbuilder as pbuilder > dh clean >dh_auto_clean >dh_autoreconf_clean >dh_clean > dpkg-source: info: using source format '3.0 (quilt)' > dpkg-source: info: building libbpp-core using existing > ./libbpp-core_2.3.1.orig.tar.gz > dpkg-source: error: cannot represent change to build/src/libbpp-core.so: > dpkg-source: error: new version is symlink to libbpp-core.so.3 > dpkg-source: error: old version is nonexistent > dpkg-source: error: cannot represent change to build/src/libbpp-core.so.3: > dpkg-source: error: new version is symlink to libbpp-core.so.3.0.0 > dpkg-source: error: old version is nonexistent > dpkg-source: error: cannot represent change to > build/CMakeFiles/3.7.2/CMakeDetermineCompilerABI_CXX.bin: binary file > contents changed > dpkg-source: error: add > build/CMakeFiles/3.7.2/CMakeDetermineCompilerABI_CXX.bin in > debian/source/include-binaries if you want to store the modified binary i > n the debian tarball > dpkg-source: warning: executable mode 0755 of > 'build/CMakeFiles/3.7.2/CMakeDetermineCompilerABI_CXX.bin' will not be > represented in diff > dpkg-source: error: cannot represent change to > build/CMakeFiles/3.7.2/CompilerIdCXX/a.out: binary file contents changed > dpkg-source: error: add build/CMakeFiles/3.7.2/CompilerIdCXX/a.out in > debian/source/include-binaries if you want to store the modified binary in > the debian tarball > > > Not really sure what I'm missing. I admit I can not reproduce this at my side. The directory build is not really used by the Debian packaging. Its possibly some remaining from a manual build you were doing. What about manually removing the dir build/ and trying again? Kind regards Andreas. -- http://fam-tille.de
Re: Sponsored upload: odil 0.8.0-3
Hi Julien, On Tue, Aug 22, 2017 at 11:51:28AM +0200, Julien Lamy wrote: > Dear all, > I believe I've fixed #871301 in the Git repo of Odil: > * the build dep is fixed > * I don't generate a symbols file, but the required symbols are both > present in the .so and I've added the dh_makeshlibs override > > I've also fixed the detection of Python version and updated the > standards version. It builds correctly and Lintian does not complain, so > could I get an upload to unstable? I've uploaded to _unstable_ as you requested here and as I think is correct (despite changelog was pointing to experimental). Thanks for your contribution Andreas. -- http://fam-tille.de
Re: Sponsored upload: odil 0.8.0-3
Hi Andreas, Le 22/08/2017 à 16:09, Andreas Tille a écrit : > Hi Julien, > > On Tue, Aug 22, 2017 at 11:51:28AM +0200, Julien Lamy wrote: >> Dear all, >> I believe I've fixed #871301 in the Git repo of Odil: >> * the build dep is fixed >> * I don't generate a symbols file, but the required symbols are both >> present in the .so and I've added the dh_makeshlibs override >> >> I've also fixed the detection of Python version and updated the >> standards version. It builds correctly and Lintian does not complain, so >> could I get an upload to unstable? > > I've uploaded to _unstable_ as you requested here and as I think is > correct (despite changelog was pointing to experimental). changelog was indeed wrong: thanks for the correction and for the upload. -- Julien signature.asc Description: OpenPGP digital signature
Re: CI for parsinsert [Outreachy]
Just let you know that I added info about source of test data. Regards, Nadiya On Tue, Aug 22, 2017 at 4:23 AM, Andreas Tille wrote: > On Tue, Aug 22, 2017 at 11:35:23AM +0200, Andreas Tille wrote: > > Ahhh, please wait for my migration SVN -> Git before you push > > anything. I'll keep you updated once I'm finishing. > > Package moved to Git now. Please ping me if you have added the > information about data sets (in case I might not realise your commit). > > Kind regards > > Andreas. > > > On Tue, Aug 22, 2017 at 11:28:45AM +0200, Andreas Tille wrote: > > > Dear Nadiya, > > > > > > On Mon, Aug 21, 2017 at 01:49:50PM -0700, Nadiya Sitdykova wrote: > > > > I pushed tests for parsinsert package during DebConf but forgot to > write > > > > you about it. > > > > > > You could have even told me. :-P > > > > > > > I used some test data from upstream version > > > > https://sourceforge.net/projects/parsinsert/files/ and I'm not sure > if it's > > > > worth mentioning in README.source and additions to debian/copyright. > What > > > > do you think? > > > > > > Please add it to README.source. Its just helpful documentation for > > > those who might want to reproduce or enhance anything. > > > > > > As always thanks for your work on this > > > > > >Andreas. > > > > > > -- > > > http://fam-tille.de > > > > > > > > > > -- > > http://fam-tille.de > > > > > > -- > http://fam-tille.de > >
Re: CI for parsinsert [Outreachy]
Hi Nadiya, On Tue, Aug 22, 2017 at 06:04:44PM -0700, Nadiya Sitdykova wrote: > Just let you know that I added info about source of test data. Thanks for the ping. I have uploaded with the change that testdata are no separated into an extra binary package with architecture all. This saves bandwidth and space on mirrors and does not force users to install the data if they do not want to test. The extra binary package will move parsinsert into the new queue so there might be some delay until it will hit unstable. Thanks for your work on this Andreas. -- http://fam-tille.de
