Re: fasttree: hard-coded limit on branch length precision leads to erroneous results
Hi Charles, On Fri, Mar 27, 2015 at 08:55:14AM +0900, Charles Plessy wrote: > > I'll tag #781259 as serious. > > Hi Roland, > > to get the fix in Jessie at this point of the Freeze, it has to be limited the > strict necessary changes. > > I recommend to temporarly revert the following commits: I'll checkout the currently uploaded source and fix the bare minimum. > In any case, many thanks for spotting this and preparing a correction so > quickly ! +1 Kind regards Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327071438.ga10...@an3as.eu
Re: fasttree: hard-coded limit on branch length precision leads to erroneous results
> "C" == Charles Plessy writes: Hi Charles, C> Le Thu, Mar 26, 2015 at 05:15:36PM +0100, Andreas Tille a écrit : >> On Thu, Mar 26, 2015 at 05:14:24PM +0100, Andreas Tille wrote: >> > > The above article describes the patch needed to prevent the >> > > flaw. Upstream incorporated the fix in version 2.1.8. In my >> > > opinion this bug is release critical and should be fixed >> > > before the release of jessie. >> > >> > Please file an RC bug! >> >> I'll tag #781259 as serious. C> to get the fix in Jessie at this point of the Freeze, it has to C> be limited the strict necessary changes. C> I recommend to temporarly revert the following commits: C> --- C> commit 5788cecbb05a4394c3fed722c47bdba5c20432ef Author: C> tbooth-guest Date: Tue Feb 25 13:43:34 C> 2014 + C> Fixed package not cleaning 100% after build. this is a perfect example why it's so important to tag package releases. Unfortunately, fasttree doesn't have any so far. So for someone unfamiliar with the package history, it's guesswork or tedious detective research to find out what went into a release version of the package. Good job fasttree is a tiny package making things easier. Looking at the date of the last commit in fasttree, I assumed it must have been included in the jessie version. On the other hand, the biolinux1 in the version number should have made me more suspicious ... :) C> I also think the upstream changelog belongs in the package, C> according to policy, so I added it. But maybe I'm mistaken? C> git-svn-id: C> svn://svn.debian.org/debian-med/trunk/packages/fasttree/trunk@16316 C> d8681a01-af0d-0410-a158-b4166a59cfaa C> commit dcef62b4c80a2c43b3c17af428e977b6535c3dc3 Author: jamessan C> Date: Sun Feb 23 04:46:19 2014 + C> Move debian/upstream to debian/upstream/metadata C> git-svn-id: C> svn://svn.debian.org/debian-med/trunk/packages/fasttree/trunk@16250 C> d8681a01-af0d-0410-a158-b4166a59cfaa C> --- C> I think that the commits about VCS URL and maintainer are C> acceptable, but if the release team prefers the package without, C> the best will be to revert the commit on VCS URLs The change of the VCS URLs was in a commit that is not part of the tag 2.1.7-2. C> and upload the changes as a "Team Upload", without change of C> maintainers (this can be done when uploading version 2.1.8 after C> the release). You guys know best what to do in such cases. Just go ahead with whatever you think is right. C> In any case, many thanks for spotting this and preparing a C> correction so quickly ! It was a simple fix and given the findings described in the blog post pretty important. C> Do you think it would be possible to backport the change for C> Wheezy as well ? I guess so. We have already built the package for Qlustar/Wheezy, so no principal problem. We'd have to rebuild it with a different version number suitable for the backports repo, but that's done in no time. Just a question how the upload process to the official backport repo works. I'm not familiar with that. -- Roland --- http://www.q-leap.com / http://qlustar.com --- HPC / Storage / Cloud Linux Cluster OS --- -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/21781.10873.301925.713...@gargle.gargle.howl
Re: [MoM] dwv
Hi Yves, ping, I neither noticed any activity nor any question of yours. Please let us know if the hints below are not helpful enough. Kind regards Andreas. On Wed, Mar 18, 2015 at 11:05:37AM +0100, Andreas Tille wrote: > Hi Yves, > > On Tue, Mar 17, 2015 at 10:08:18PM +0100, Yves wrote: > > Yes, this plugin depends on Orthanc 0.8.6 (the latest). So either > > liborthancclient-dev or the new orthanc-dev. > > Please pull my latest commit with the versioned Build-Depends which is > expressing this requirement. > > > For javascript, it seems that I either need to include debian packaged ones > > (like jquery) or provide the sources. From what I get that was done in the > > orthanc-webviewer, by javascript sources, it is meant non minified code? Is > > that correct? > > Sébastien has given a sensible answer to this question. > > Other issues are: The resulting *.deb package is basically empty just > containing the broken doc-base template. If you intend to install > documentation files you need to add these to debian/docs (see > man dh_installdocs) or delete this file. > > Moreover you need to install your real install target into the package. > > Kind regards > > Andreas. > > -- > http://fam-tille.de > > > -- > To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org > with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org > Archive: https://lists.debian.org/20150318100537.ga...@an3as.eu > > -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327114423.gc10...@an3as.eu
Re: [MoM] dwv
Ping received! I got stuck in other stuff but should have some time this week end... About the empty package, how do I "install your real install target into the package"? On 27 March 2015 at 12:44, Andreas Tille wrote: > Hi Yves, > > ping, I neither noticed any activity nor any question of yours. Please > let us know if the hints below are not helpful enough. > > Kind regards > > Andreas. > > On Wed, Mar 18, 2015 at 11:05:37AM +0100, Andreas Tille wrote: > > Hi Yves, > > > > On Tue, Mar 17, 2015 at 10:08:18PM +0100, Yves wrote: > > > Yes, this plugin depends on Orthanc 0.8.6 (the latest). So either > > > liborthancclient-dev or the new orthanc-dev. > > > > Please pull my latest commit with the versioned Build-Depends which is > > expressing this requirement. > > > > > For javascript, it seems that I either need to include debian packaged > ones > > > (like jquery) or provide the sources. From what I get that was done in > the > > > orthanc-webviewer, by javascript sources, it is meant non minified > code? Is > > > that correct? > > > > Sébastien has given a sensible answer to this question. > > > > Other issues are: The resulting *.deb package is basically empty just > > containing the broken doc-base template. If you intend to install > > documentation files you need to add these to debian/docs (see > > man dh_installdocs) or delete this file. > > > > Moreover you need to install your real install target into the package. > > > > Kind regards > > > > Andreas. > > > > -- > > http://fam-tille.de > > > > > > -- > > To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org > > with a subject of "unsubscribe". Trouble? Contact > listmas...@lists.debian.org > > Archive: https://lists.debian.org/20150318100537.ga...@an3as.eu > > > > > > -- > http://fam-tille.de > > > -- > To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org > with a subject of "unsubscribe". Trouble? Contact > listmas...@lists.debian.org > Archive: https://lists.debian.org/20150327114423.gc10...@an3as.eu > >
gatk: Any hint for Java (Maven) packaging
Hi, I just commited Vcs-Git: git://anonscm.debian.org/debian-med/gatk.git I had no real luck with mh_make which only created debian/maven.* files but no d/contol and d/rules file due to some error in the last step. I also noticed the following output from mh_make process: In pom.xml: This dependency cannot be found in the Debian Maven repository. Ignore this dependency? samtools:htsjdk:jar:null In pom.xml: This dependency cannot be found in the Debian Maven repository. Ignore this dependency? picard:picard:jar:null In pom.xml: This dependency cannot be found in the Debian Maven repository. Ignore this dependency? colt:colt:jar:null In pom.xml: This dependency cannot be found in the Debian Maven repository. Ignore this dependency? it.unimi.dsi:fastutil:jar:null In pom.xml: This dependency cannot be found in the Debian Maven repository. Ignore this dependency? com.google.code.cofoja:cofoja:jar:null In pom.xml: This dependency cannot be found in the Debian Maven repository. Ignore this dependency? net.sf.jgrapht:jgrapht:jar:null So there are definitely some missing Build-Dependencies in the current state. However, my problem is even before these Build-Depends are needed since the Java build is not started at all. Any idea how to change d/rules to get some build process triggered? Thanks in advance Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327122754.gd10...@an3as.eu
Re: [MoM] dwv
Hi Yves, On Fri, Mar 27, 2015 at 12:52:23PM +0100, Yves wrote: > About the empty package, how do I "install your real install target into > the package"? Usually dh_auto_install calls the install target of the provided upstream build system. This obviously does nothing since the package remains empty. So you need to use dh_install The best way to use it is via a debian/install file which contains line entries source target The source files in your build tree will be installed to the target directory. Just imagine that a "cp " will be prepended. See also `man dh_install`. Hope this clarifies things - if not feel free to keep on asking Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327125745.ga17...@an3as.eu
Re: fasttree: hard-coded limit on branch length precision leads to erroneous results
Hi Roland, On Fri, Mar 27, 2015 at 11:01:29AM +0100, Roland Fehrenbacher wrote: > C> commit 5788cecbb05a4394c3fed722c47bdba5c20432ef Author: > C> tbooth-guest Date: Tue Feb 25 13:43:34 > C> 2014 + > > C> Fixed package not cleaning 100% after build. > > this is a perfect example why it's so important to tag package > releases. Unfortunately, fasttree doesn't have any so far. So for > someone unfamiliar with the package history, it's > guesswork or tedious detective research to find out what went into a > release version of the package. You are right. Some maintainers in our team are a bit sloppy with tagging. The rationale of them is that we have snapshot.debian.org. While I do not subscribe to this opinion personally I understand their point. Apropos tagging: There are people in the team who said they will never tag any of their uploads to avoid that all these tags will consume more and more space on their harddisks. I also do not fully subscribe to this opinion but to find some compromise I deleted "historical" tags quite systematicly only leaving tags of major versions, somehow "important" tags (for whatever reason) and the last three tags. To my (possibly poor) understanding this was the best compromise to invite people to do some tagging at all without filling up to much disk space. So please be prepared for a mostly incomplete tagging in SVN (feel free to criticise this but its hard to change the past.) > Good job fasttree is a tiny package > making things easier. Looking at the date of the last commit in > fasttree, I assumed it must have been included in the jessie version. > On the other hand, the biolinux1 in the version number should have made > me more suspicious ... :) I did an upload based upon apt-get source fasttree and afterwards importing your patches. I stored this in the Jessie branch of the newly created Git repository. Cases like this are not covered by our policy document (but should). I hope I found a solution that is easy to understand and helpful for this case. > C> and upload the changes as a "Team Upload", without change of > C> maintainers (this can be done when uploading version 2.1.8 after > C> the release). > > You guys know best what to do in such cases. Just go ahead with whatever > you think is right. So I did at least for the user visible part in the Debian pool. Feel free to find a better solution for the Git repository - I do not consider myself sensibly experienced with Git. > C> In any case, many thanks for spotting this and preparing a > C> correction so quickly ! > > It was a simple fix and given the findings described in the blog post > pretty important. +1 > C> Do you think it would be possible to backport the change for > C> Wheezy as well ? > > I guess so. We have already built the package for Qlustar/Wheezy, so no > principal problem. We'd have to rebuild it with a different version > number suitable for the backports repo, but that's done in no time. > Just a question how the upload process to the official backport repo > works. I'm not familiar with that. I admit I never cared about backports and I was hoping that somebody from the team would care for it. Unfortunately this was not (yet) the case and I'm crossing thumbs that this will become better in the future. Kind regards Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327131038.gb17...@an3as.eu
Re: fasttree: hard-coded limit on branch length precision leads to erroneous results
Le Fri, Mar 27, 2015 at 02:10:38PM +0100, Andreas Tille a écrit : > On Fri, Mar 27, 2015 at 11:01:29AM +0100, Roland Fehrenbacher wrote: > > > C> Do you think it would be possible to backport the change for > > C> Wheezy as well ? > > > > I guess so. We have already built the package for Qlustar/Wheezy, so no > > principal problem. We'd have to rebuild it with a different version > > number suitable for the backports repo, but that's done in no time. > > Just a question how the upload process to the official backport repo > > works. I'm not familiar with that. > > I admit I never cared about backports and I was hoping that somebody > from the team would care for it. Unfortunately this was not (yet) the > case and I'm crossing thumbs that this will become better in the future. Actually, I have been a bit confusing: I did not mean uploading version 2.1.7-2 to Wheezy, but to apply the patches to 2.1.4-1 as a Stable update if it makes sense. Have a nice day, Charles -- Charles Plessy Debian Med packaging team, http://www.debian.org/devel/debian-med Tsurumi, Kanagawa, Japan -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327131736.gb8...@falafel.plessy.net
Re: Intent to package seg3d-- need mentoring/advice?
Hi Robert, thanks for your interest in Debian Med. On Thu, Mar 26, 2015 at 05:02:58PM -0400, Robert Blake wrote: > To whom it may concern: > > I would like to package seg3d > (http://www.sci.utah.edu/cibc-software/seg3d.html) for the > debian-med PPA and eventually debian unstable. I'd consider it more valuable if we target directly to Debian (unstable). > I have a package > called "seg3d" in progress here: > > https://launchpad.net/~rob.c.blake.3/+archive/ubuntu/trayanova-cep Without checking the content you seem to have some packaging skills which allow you to create some working package. > I'm running into some difficulties: > > 1.) Upstream has bundled 14 libraries into their tarball and is > resistant to change this. See > > https://lists.debian.org/debian-med/2013/05/msg00052.html > > I have written a patch that allows a debian packager to ignore all > these libraries and substitute in debian package equivalents. That's pretty cool! > I have tested the patch extensively for the past 3+ months with no > stability problems. Good. > Given the extent of the software included (i.e. all of boost, > python, itk, etc) I have no idea how to write a proper > debian/copyright file. I admit writing the copyright file is usually the most boring part in packaging but it needs to be done anyway. Otherwise it will not pass ftpmaster who will upload to the Debian packaging pool. Since you said you have found a patch to enable using the packaged libraries it would be the best idea to strip all unused code from the upstream tarball using the Files-Excluded field in debian/copyright (see `man uscan`). This makes you really sure that this code is not used and saves you from the work to mention all the files in your d/copyright file. Moreover I'd recommend the command licensecheck to parse upstream code for copyright and license statements. Just let us know here on the list if these hints might not be sufficient. > 2.) After adding a seg3d.desktop to the build, I can't get the icon > to appear properly for Gnome or KDE. Hmmm, I admit I'm no expert in this. If nobody else here on this list has any clue asking on debian-mentors list might help. However, before you target this list it would be probably helpful to move you work to a Git archive of the Debian Med team on alioth. How this can be approached is explained in our team policy[1]. > 3.) This is my first time packaging anything, so I am blind to many > debian best practices. I've never written a manpage, etc. I would > love an extra set of eyes on what I'm doing so that I can file an > ITP with unstable. > > I read the MoM guideline linked previously this month. Is there any > way I could sign up? Usually you sign up by adding a line to the table on the MoM page. However, I need to check whether Malihe Asemani who is in the MoM queue and wanted to take the April slot wants to really start in April (as she told me in private mail recently). For sure we will not leave you alone if you ask questions here on this list and perhaps the hints given above might bring you on track easily but I'd like to make sure that we somehow keep the proper sequence in the MoM effort. Kind regards Andreas. [1] http://debian-med.alioth.debian.org/docs/policy.html -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327135422.gc17...@an3as.eu
fw4spl's bug reported
Hi everybody I have just been notified of two "Release Critical Bug" concerning fw4spl. As it is the first time that one of my packages is concerned by a report, I would like some clarifications before to do something wrong, About the first bug [1], I presume that it comes from this line of my FindHdf5.cmake : /usr/lib/*x86_64-linux-gnu*/hdf5/serial I replaced it with : /usr/lib/*/hdf5/serial I can test it on my computer. Does that seem right to you ? The second bug [2] is about files from libqt4-qt3support copied during build. I followed this link [3], added rdfind and symlinks in d/control and these lines in d/rules (in override_dh_auto_install) : rdfind -outputname /dev/null -makesymlinks true debian/fw4spl/ symlinks -r -s -c debian/fw4spl If all what I've done is correct, what is the next step ? Update the d/changelog in mentionning the closes numbers 781294 and 781298 ? Thank you for your help, Best regards, Corentin [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=781294 [2] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=781298 [3] https://wiki.debian.org/dedup.debian.net
Re: fw4spl's bug reported
Hi Corentin, On Fri, Mar 27, 2015 at 04:02:02PM +0100, Corentin Desfarges wrote: > I have just been notified of two "Release Critical Bug" concerning fw4spl. > As it is the first time that one of my packages is concerned by a report, I > would like some clarifications before to do something wrong, > > About the first bug [1], I presume that it comes from this line of my > FindHdf5.cmake : >/usr/lib/*x86_64-linux-gnu*/hdf5/serial > > I replaced it with : >/usr/lib/*/hdf5/serial > > I can test it on my computer. Does that seem right to you ? I'm no cmake expert but this sound plausible somehow. > The second bug [2] is about files from libqt4-qt3support copied during > build. I followed this link [3], added rdfind and symlinks in d/control and > these lines in d/rules (in override_dh_auto_install) : > >rdfind -outputname /dev/null -makesymlinks true debian/fw4spl/ >symlinks -r -s -c debian/fw4spl > > If all what I've done is correct, what is the next step ? Update the > d/changelog in mentionning the closes numbers 781294 and 781298 ? Yes. Do some dch -i which creates a new changelog entry with a Debian revision "-2". Than describe what you did to fix the bug and add a "Closes: #781294" or "Closes: #781298" respectively to the entries. In any case to be fully sure you can perfectly ask the bug reporter by replying to the bug mail address whether the proposed fix is OK. This might safe another upload cycle where the reporter needs to reopen the bug if it is not fixed. Thanks for your work on this Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150327153511.gh17...@an3as.eu
Re: [MoM] dwv
Ok, I can create this install file. The source will be where ever the built shared library is and the target is the plugin folder of orthanc. Or do you mean something else by 'source'? And a newbie question, how do I switch to using sid packages for my test builds? On 27 March 2015 at 13:57, Andreas Tille wrote: > Hi Yves, > > On Fri, Mar 27, 2015 at 12:52:23PM +0100, Yves wrote: > > About the empty package, how do I "install your real install target into > > the package"? > > Usually dh_auto_install calls the install target of the provided > upstream build system. This obviously does nothing since the package > remains empty. So you need to use > > dh_install > > The best way to use it is via a debian/install file which contains > line entries > > source target > > The source files in your build tree will be installed to the target > directory. Just imagine that a "cp " will be prepended. See also > `man dh_install`. > > Hope this clarifies things - if not feel free to keep on asking > > Andreas. > > > -- > http://fam-tille.de > > > -- > To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org > with a subject of "unsubscribe". Trouble? Contact > listmas...@lists.debian.org > Archive: https://lists.debian.org/20150327125745.ga17...@an3as.eu > >
Re: [MoM] dwv
On Sat, Mar 28, 2015 at 12:29:27AM +0100, Yves wrote: > Ok, I can create this install file. The source will be where ever the built > shared library is and the target is the plugin folder of orthanc. Yes. > And a newbie question, how do I switch to using sid packages for my test > builds? Seek for pbuilder in Debian Med policy. :-) Kind regards Andreas. -- http://fam-tille.de -- To UNSUBSCRIBE, email to debian-med-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150328062056.gc23...@an3as.eu
