Description of problem: New CAD package Name of program: FreeCAD Home page of project: http://sourceforge.net/apps/mediawiki/free-cad/index.php?title=Main_Page Making: http://sourceforge.net/apps/mediawiki/free-cad/index.php?title=Compiling comment: in mandriva 2010.2 I using open-cascade, python, qt4lib,.... from rpm´s but soqt, coin and pivy from base source. Coin Mercurial Repositories http://hg.sim.no/ - base source for COIN,.... Stable version: is 0.11 (source from trunk I do not know compiling)
You may use spec file from http://ftp.blogdrake.net/mandriva/SPECS/, but I do not know if is working.
Summary: New CAD program => add: Free-cad, New CAD programCC: (none) => manuel
CC: (none) => anaselli
CC: (none) => john.gibbe
CC: (none) => marja11Summary: add: Free-cad, New CAD program => free-cad, FreeCAD is a general purpose Open Source 3D CAD/MCAD/CAx/CAE/PLM modelerSource RPM: (none) => free-cadSeverity: normal => enhancement
Blocks: (none) => 531Whiteboard: (none) => X
I'm working on it - it won't be a quick job though as it involves many packages we don't yet have.
Status: NEW => ASSIGNEDCC: (none) => zen25000Assignee: bugsquad => zen25000
Hello Barry, Some packages for mageia exist on http://ftp.blogdrake.net/mageia/mageia1/free/x86_64/ the freecad one is Ok but doesn't check for prerequisites that you have to install manually. Anyway messages when you try to execute freecad are explicite and after several attempts (execute freecad, read message, install the missing package) it works. At the end, the only really missing package is python-pivy. But freecad launch and you can draw. Hope it will help Best regards
Hello all, well just like to share some more information. Sources of pivy-0.5.0 can be found on from the address http://hg.sim.no/Pivy/default. It requires swig to be build. With swig-1.3.40 (mageia 1) it works but with swig-2.0.4 (cauldron) it fails. Regards
Hallo all, thank you for working on this package request. I'm very interested in availability of this program in the repos, because I'm not able to compile from source code. If you need testers for the RPMs, I would offer my help. Kind regards Florian
CC: (none) => alfaflo
I also would like this packet. It is a nice development
CC: (none) => thierry.rouillon
CC: (none) => adrien_d
I am re-assigning this back to bugsquad as I no longer have time to look further into this. I spent a lot of time on it but hit too many brick walls on dependencies. Maybe someone else will have more time to figure it out. I imported some deps, and I did see someone working on opencascade recently so maybe now it is closer to being possible. Sorry, Barry
Assignee: zen25000 => bugsquad
Never thought about seeing if its been requested, Put a post on forums about it. Anyway, Set to and had a go, finally managed it , new to the packaging game so could do with been tested/checked. Have figured out all deps, and packaged also. Most deps are rebuilds of packages of mga2 that Ole Lukoie had done. Took some working out all the deps as Barry pointed out. Getting it to install and work including the modules has been the main problem. Done 64bit packages only at the moment.(need 32 build doing) There not signed, just uploading to my webserver. will post on forum when done Hope i've helped in some way. Pete.
CC: (none) => worzel910
Pete, Well done. May I suggest that you join the packaging team as an apprentice so that you can take this project further. For this to be useful to Mageia all the dependencies and build requires need to be packaged correctly with care taken about licencing etc. and also ultimately it would require a maintainer to continue to look after the package(s). If you are keen to do this then here is the place to start: https://wiki.mageia.org/en/Becoming_a_Mageia_Packager Have fun, Barry
(In reply to Peter Wallace from comment #8) > Most deps are rebuilds of packages of mga2 that Ole Lukoie had done. My packages for FreeCAD are simple ports of Fedora ones and thus rely on OCE instead of original OpenCASCADE. And there are newer versions already that I had not ported yet. And also people say FreeCad 0.12 has some severe memory leaks...
CC: (none) => olegbosis
Hope you guessed that "oleukoie"/"Ole Lukoie" is my nickame ;)
CC: manuel.mageia => (none)
hehe, no I didn't Oleg, The OCE thing was one of the hiccups that I had. Just rebuilding freecad again due to the fact I missed the Partdesign.so on install. Must have been lurking from previous attempts. With regards the apprentice (Not been one of those in a lot of years) that's some food for thought... Thanks for the suggestion Barry. The licensing for Freecad I would have thought would need to be none-free repo I think, The licensing side of software is just plain confusing to me. Pete.
(In reply to Peter Wallace from comment #12) > The licensing for Freecad I would have thought would need to be none-free > repo I think, The licensing side of software is just plain confusing to me. AFAIK FreeCAD is a GPLv2/LGPLv2-licensed software and thus can go into core repository. There could be some patent problems in some countries with some data formats or algorithms but I hadn't heard of them.
CC: (none) => hhielscher
Hello, i need it to recover some files created for a 3D printer, so, i'll try to package it for Mageia 4 ( at worst, i'll find help to do that. ).
CC: (none) => yves
(In reply to Yves Gaël Chény from comment #14) > Hello, i need it to recover some files created for a 3D printer, so, i'll > try to package it for Mageia 4 ( at worst, i'll find help to do that. ). All ready done it ;) https://forums.mageia.org/en/viewtopic.php?f=10&t=5661#p38456
Yeah ! good news ! Why it's not into core repository ?
Seems that the license issue is fixed, and Fedora have it now in their repositories http://pkgs.fedoraproject.org/cgit/freecad.git
CC: (none) => makowski.mageia
I read here : http://www.freecadweb.org/wiki/index.php?title=Release_notes_014#General that freecad change his license for the LGPL. "Move to pyside, FreeCAD is now fully LGPL With the many complications caused by the double-license model of FreeCAD (LGPL & GPL), some of the components of FreeCAD (namely the OpenCasCade kernel) being incompatible with GPL code, we decided to switch all the remaining bits of GPL code of FreeCAD to LGPL" So it seems there are no more problèmes with the license.
CC: (none) => marc
CC: (none) => rick
Depends on: (none) => 14862
CC: (none) => 231036448
I just took ownership of this one. Using some of Pete's work as a base, I was able to complete PoC packages for 0.14.3256 on cauldron. If we can get SoQt back in the distro (still researching the removal), this should just be a matter of bringing 7 packages up to standards and getting the change set committed. That's more work that in sounds, though. Given my current workload, I predict a February completion date.
Source RPM: free-cad => freecadAssignee: bugsquad => rick
(In reply to Richard Houser from comment #19) > If we can get SoQt back in the distro (still researching the removal), this > should just be a matter of bringing 7 packages up to standards and getting > the change set committed. > soqt was removed because it was not required by anything. iirc I imported it when I was looking at freecad a few years ago but gave up due to un-buildable/obsolete dependencies. I hope you have more success. ;)
SoQt is back in cauldron, but I don't think it's pushed yet (will do so when I get pivy pushed). This will get you a copy for now: $ mgarepo co soqt $ cd soqt $ bm -l
All dependencies I know of are now present in cauldron. My build is now failing during configure, however. This worked for me a couple days ago, so I think it might be related to some other update that went recently. Error: .... Use local sources of PyCXX & zipios++... yes checking for xerces-c... checking for /usr/include/xercesc... yes checking xercesc/framework/XMLBuffer.hpp usability... yes checking xercesc/framework/XMLBuffer.hpp presence... yes checking for xercesc/framework/XMLBuffer.hpp... yes checking whether xerces lib is available... yes checking for eigen3... checking for /usr/include/eigen3/Eigen... yes checking Eigen/Eigenvalues usability... yes checking Eigen/Eigenvalues presence... yes checking for Eigen/Eigenvalues... yes checking for boost... configure: error: failed error: Bad exit status from /var/tmp/rpm-tmp.tHJTWo (%build) Work in progress: https://drive.google.com/file/d/0Bx7kRA46qjpFYU9WUmRwdTFlOE0/view?usp=sharing
(In reply to Richard Houser from comment #22) > Work in progress: > https://drive.google.com/file/d/0Bx7kRA46qjpFYU9WUmRwdTFlOE0/view?usp=sharing I tried the buildprocess with your spec-file. I used the actual source freecad-0.14.3702. My system is mageia 5 RC, 64 Bit. I saved the complete build messages in a textfile. Shall I give a link to this file? The build process worked complete. The problem ist the missing manpage. Apparently we do not have a freecad-manpage. Is that right? Who knows more?
CC: (none) => neue_chance
Man page is not mandatory for Mageia. Will you commit your work in Mageia svn?
CC: (none) => loginov_alex
FreeCAD 0.15 packaged in cauldron, using fedora's SPEC and some tweaking thanks to Richard's work.
Status: ASSIGNED => RESOLVEDCC: (none) => yann.cantinResolution: (none) => FIXED
Thanks a lot for this excellent news ! Should it be possible to backport it on mga 5 ? The next mga6 is so far ;) Thanks a lot for you job !
freecad 0.15 in mga5 core/backport_testing. Marc : It should be really helpful if you could test this freecad package functionalities, i don't know CAD software enough to look further than some examples.
Status: RESOLVED => REOPENEDVersion: Cauldron => 5Resolution: FIXED => (none)Summary: free-cad, FreeCAD is a general purpose Open Source 3D CAD/MCAD/CAx/CAE/PLM modeler => Backport Request: free-cad, FreeCAD is a general purpose Open Source 3D CAD/MCAD/CAx/CAE/PLM modelerWhiteboard: X => backport
I am glad I... I mean we, can use FreeCAD on Mageia 5. Me, too, says "Thanks!". It must have been a lot of work. About the testing, the program installs fine, and is usable (create, edit, save, etc). The only issue, so far, is with the Ship workbench. But it might be a FreeCAD bug. When you choose "Ship" from the dropdown menu, the program stalls for a while, and a pop-up appears (see attachment). The workbench doesn't seem to load. My system is Mageia 5, 64bit.
Created attachment 7175 [details] "Ship" workbench not loading
Hi. I install it successfully on mageia 5 64. My first (quick) test work fine. I continue to test it. But for now, all is ok.
Working fine here too, including the "Ship workbench" Done some basic stuff with it without issues.
I also installed it in Mga5 and it seems OK, however with backports_testing still enabled, urpmi --auto-update now wants to remove it: The following packages have to be removed for others to be upgraded: freecad-0.15-1.mga5.x86_64 (due to missing python-pivy, due to unsatisfied freecad-data == 0.15-1.mga5) freecad-data-0.15-1.mga5.noarch (due to unsatisfied freecad == 0.15-1.mga5) python-pivy-0.5.0-2.mga5.x86_64 (due to missing libSoQt.so.20()(64bit)) I have checked that libSoQt.so.20()(64bit)) is correctly provided by lib64soqt20-1.5.0-4.mga5 which is of course installed: rpm -qa|grep $(urpmq --whatprovides "libSoQt.so.20()(64bit)") lib64soqt20-1.5.0-4.mga5 So I do not as yet understand the cause of the problem.
The integration of backports* with regular repos is not polished. Can you disable all backports repos and retry the urpmi --auto-update ?
CC: loginov_alex => (none)
With only fully updated regular media enabled: medium "Core Release (zmrepo1)" is up-to-date medium "Core Updates (zmrepo3)" is up-to-date medium "Nonfree Release (zmrepo11)" is up-to-date medium "Nonfree Updates (zmrepo13)" is up-to-date medium "Tainted Release (zmrepo21)" is up-to-date medium "Tainted Updates (zmrepo23)" is up-to-date medium "Core 32bit Release (zmrepo31)" is up-to-date medium "Core 32bit Updates (zmrepo32)" is up-to-date medium "Nonfree 32bit Release (zmrepo36)" is up-to-date medium "Nonfree 32bit Updates (zmrepo37)" is up-to-date medium "Tainted 32bit Release (zmrepo41)" is up-to-date medium "Tainted 32bit Updates (zmrepo42)" is up-to-date medium "srpms" is up-to-date medium "Core Release Debug (distrib2)" is up-to-date tail of urpmi --auto update --debug --test selecting task-obsolete-5-41.mga5.tainted.noarch set_rejected: lib64soqt20-1.5.0-4.mga5.x86_64 installed python-pivy-0.5.0-2.mga5.x86_64 is conflicting because of unsatisfied libSoQt.so.20()(64bit) set_rejected: python-pivy-0.5.0-2.mga5.x86_64 installed freecad-0.15-1.mga5.x86_64 is conflicting because of unsatisfied python-pivy promoting python-pivy-0.5.0-2.mga5.i586 because of conflict above no packages match python-pivy (it is either in skip.list or already rejected) no packages match python-pivy (it is either in skip.list or already rejected) installed freecad-0.15-1.mga5.x86_64 is conflicting because of unsatisfied python-pivy set_rejected: freecad-0.15-1.mga5.x86_64 installed freecad-data-0.15-1.mga5.noarch is conflicting because of unsatisfied freecad[== 0.15-1.mga5] set_rejected: freecad-data-0.15-1.mga5.noarch installed freecad-0.15-1.mga5.x86_64 is conflicting because of unsatisfied freecad-data[== 0.15-1.mga5] opening rpmdb (root=, write=) The following packages have to be removed for others to be upgraded: freecad-0.15-1.mga5.x86_64 (due to missing python-pivy, due to unsatisfied freecad-data == 0.15-1.mga5) freecad-data-0.15-1.mga5.noarch (due to unsatisfied freecad == 0.15-1.mga5) python-pivy-0.5.0-2.mga5.x86_64 (due to missing libSoQt.so.20()(64bit)) (test only, removal will not be actually done) (y/N) n unlocking rpm database unlocking urpmi database EXITING (pid=10668) Just a thought... Was soqt removed from task-obsolete when it was re-introduced I wonder?
Seems not: from 5/task-obsolete # (wally) 2014-03-09 # moved to obsolete in SVN Obsoletes: libsimvoleon40 < 2.0.2 Obsoletes: lib64simvoleon40 < 2.0.2 Obsoletes: libsimvoleon-devel < 2.0.2 Obsoletes: lib64simvoleon-devel < 2.0.2 Obsoletes: libsoqt20 < 1.5.1 Obsoletes: lib64soqt20 < 1.5.1 Obsoletes: libsoqt-devel < 1.5.1 Obsoletes: lib64soqt-devel < 1.5.1
(In reply to Barry Jackson from comment #34) > With only fully updated regular media enabled: > > medium "Core Release (zmrepo1)" is up-to-date > medium "Core Updates (zmrepo3)" is up-to-date > medium "Nonfree Release (zmrepo11)" is up-to-date > medium "Nonfree Updates (zmrepo13)" is up-to-date > medium "Tainted Release (zmrepo21)" is up-to-date > medium "Tainted Updates (zmrepo23)" is up-to-date > medium "Core 32bit Release (zmrepo31)" is up-to-date > medium "Core 32bit Updates (zmrepo32)" is up-to-date > medium "Nonfree 32bit Release (zmrepo36)" is up-to-date > medium "Nonfree 32bit Updates (zmrepo37)" is up-to-date > medium "Tainted 32bit Release (zmrepo41)" is up-to-date > medium "Tainted 32bit Updates (zmrepo42)" is up-to-date > medium "srpms" is up-to-date > medium "Core Release Debug (distrib2)" is up-to-date > > tail of urpmi --auto update --debug --test > > selecting task-obsolete-5-41.mga5.tainted.noarch > set_rejected: lib64soqt20-1.5.0-4.mga5.x86_64 > installed python-pivy-0.5.0-2.mga5.x86_64 is conflicting because of > unsatisfied libSoQt.so.20()(64bit) > set_rejected: python-pivy-0.5.0-2.mga5.x86_64 > installed freecad-0.15-1.mga5.x86_64 is conflicting because of unsatisfied > python-pivy > promoting python-pivy-0.5.0-2.mga5.i586 because of conflict above > no packages match python-pivy (it is either in skip.list or already rejected) > no packages match python-pivy (it is either in skip.list or already rejected) > installed freecad-0.15-1.mga5.x86_64 is conflicting because of unsatisfied > python-pivy > set_rejected: freecad-0.15-1.mga5.x86_64 > installed freecad-data-0.15-1.mga5.noarch is conflicting because of > unsatisfied freecad[== 0.15-1.mga5] > set_rejected: freecad-data-0.15-1.mga5.noarch > installed freecad-0.15-1.mga5.x86_64 is conflicting because of unsatisfied > freecad-data[== 0.15-1.mga5] > opening rpmdb (root=, write=) > The following packages have to be removed for others to be upgraded: > freecad-0.15-1.mga5.x86_64 > (due to missing python-pivy, > due to unsatisfied freecad-data == 0.15-1.mga5) > freecad-data-0.15-1.mga5.noarch > (due to unsatisfied freecad == 0.15-1.mga5) > python-pivy-0.5.0-2.mga5.x86_64 > (due to missing libSoQt.so.20()(64bit)) > (test only, removal will not be actually done) (y/N) n > unlocking rpm database > unlocking urpmi database > EXITING (pid=10668) > > Just a thought... > Was soqt removed from task-obsolete when it was re-introduced I wonder? I had soqt removed from task-obsolete early this year (before MGA5 was released), as that was part of the dependency chain needed to compile freecad 0.14 (0.15 was not yet released). The entire dependency chain for 0.14, including python-pivy made it into MGA5. * Fri Jan 09 2015 wally <wally> 5-84.mga5 + Revision: 809551 - don't force removal of soqt and simvoleon Would you please confirm that your system is fully patched? Specifically, which version of task-obsolete do you have? I show 5-109 was released with MGA5. If you have not fully upgraded to MGA5, the MGA4 (or < 2015-01-09 Cauldron) task-obsolete would explain these symptoms. I encountered this same dependency problem building python-pivy on MGA5 (it worked on MGA3), hence the update to task-obsolete.
I have tainted enabled and the latest task-obsolete was never pushed to tainted. The version in tainted seems to be taking precedence over the new version in core. > selecting task-obsolete-5-41.mga5.tainted.noarch :\
[baz@localhost ~]$ urpmq -f task-obsolete task-obsolete-5-109.mga5.noarch|task-obsolete-5-41.mga5.tainted.noarch|task-obsolete-5-109.mga5.noarch|task-obsolete-5-41.mga5.tainted.noarch Yet looking at... http://svnweb.mageia.org/packages/updates/5/task-obsolete/current/SPECS/task-obsolete.spec?revision=824915&view=markup ...I cannot see where task-obsolete-5-109 comes from. The most recent I see is the same version as tainted - what am I missing? Removing all tainted media does stop the attempted removals on update.
In fact, after testing an : urpmi --auto-update, I see I have exactly the same problème : The following packages have to be removed for others to be upgraded: freecad-0.15-1.mga5.x86_64 (due to missing python-pivy, due to unsatisfied freecad-data == 0.15-1.mga5) freecad-data-0.15-1.mga5.noarch (due to unsatisfied freecad == 0.15-1.mga5) python-pivy-0.5.0-2.mga5.x86_64 (due to missing libSoQt.so.20()(64bit))
(In reply to Marc Mascré from comment #39) > In fact, after testing an : urpmi --auto-update, I see I have exactly the > same problème : It's been fixed - https://bugs.mageia.org/show_bug.cgi?id=17064 but it will not take effect until qa pushes the updated task-obsolete packages.
Hi I saw the update for the task-obsolete package but when I try urpmi --auto-update I continue to have the same error message : he following packages have to be removed for others to be upgraded: freecad-0.15-1.mga5.x86_64 (due to missing python-pivy, due to unsatisfied freecad-data == 0.15-1.mga5) freecad-data-0.15-1.mga5.noarch (due to unsatisfied freecad == 0.15-1.mga5) python-pivy-0.5.0-2.mga5.x86_64 (due to missing libSoQt.so.20()(64bit)) Thanks a lot
Indeed :\ selecting task-obsolete-5-41.mga5.tainted.noarch set_rejected: lib64soqt20-1.5.0-4.mga5.x86_64 [baz@localhost ~]$ urpmq -i task-obsolete|grep Release Release : 109.mga5 Release : 109.1.mga5 Release : 41.mga5.tainted Release : 109.1.mga5.tainted ...and 41.mga5.tainted is being chosen over 109.1.mga5.tainted.
OK the workaround is to just: urpmi task-obsolete this will install the new one. Then you should be OK to use: urpmi --auto-update
yes it work fine ! No more problem with this package.
CC: makowski.mageia => (none)
Hi everybody ! It seems to works fine. What is need to push the freecad package from backport-testing to backport ?
Is this one ready for QA?
Hi something new about it ?
Assignee: rick => yann.cantin
Nothing new, ready for QA.
Assignee: yann.cantin => qa-bugsWhiteboard: backport => backport MGA5-64-OK
What is?
Priority: Normal => LowComponent: New RPM package request => BackportsWhiteboard: backport MGA5-64-OK => MGA5-64-OK
Whiteboard: MGA5-64-OK => MGA5-64-OK feedback
Packages : freecad-0.15-1.mga5.i586.rpm freecad-0.15-1.mga5.x86_64.rpm freecad-data-0.15-1.mga5.noarch.rpm from SRPM : freecad-0.15-1.mga5.src.rpm To be moved from core-backports_testing to core-backports
Thankyou.
Whiteboard: MGA5-64-OK feedback => MGA5-64-OK
Keywords: (none) => validated_updateCC: (none) => davidwhodgins, sysadmin-bugs
Keywords: validated_update => (none)
Ready for push. Adding new validated_backport keyword. Installed with # urpmi --searchmedia "Core Backports Testing" freecad Checked freecad-data also installed. Binary is /usr/bin/FreeCAD. Starts, does stuff & saves a file OK. $ FreeCAD FreeCAD 0.15, Libs: 0.15RUnknown © Juergen Riegel, Werner Mayer, Yorik van Havre 2001-2015 ##### #### ### #### # # # # # # # ## #### #### # # # # # #### # # # # # # # ##### # # # # #### #### # # # # # # # # # # # # # # ## ## ## # # #### #### ### # # #### ## ## ## Arch workbench activated Arch workbench deactivated closing Sketch editArch workbench activated
Keywords: (none) => validated_backport
pushed to backports
Status: REOPENED => RESOLVEDCC: (none) => tmbResolution: (none) => FIXED