[Robotics] iTaSC branch available for test at graphicall.org

[Ruben Smits]

I get an error with scons at the same point and about the same problem
as i did using the makefiles:

Linking program ==> 'blender'
/home/rsmits/workspace/build/linux2/lib/libbf_blenkernel.a(action.o):
In function `free_pose':
action.c:(.text+0x762): undefined reference to `BIK_clear_data'
/home/rsmits/workspace/build/linux2/lib/libbf_blenkernel.a(armature.o):
In function `where_is_pose':
armature.c:(.text+0x149b): undefined reference to `BIK_initialize_tree'
armature.c:(.text+0x14e3): undefined reference to `BIK_execute_tree'
armature.c:(.text+0x150c): undefined reference to `BIK_release_tree'
/home/rsmits/workspace/build/linux2/lib/libbf_blenkernel.a(pointcache.o):
In function `BKE_ptcache_object_reset':
pointcache.c:(.text+0x25d9): undefined reference to `BIK_clear_cache'
/home/rsmits/workspace/build/linux2/lib/libbf_rna.a(rna_pose_gen.o):
In function `rna_Itasc_update':
rna_pose_gen.c:(.text+0xf00): undefined reference to `BIK_update_param'
/home/rsmits/workspace/build/linux2/lib/libbf_rna.a(rna_pose_gen.o):
In function `Pose_ik_solver_set':
rna_pose_gen.c:(.text+0x176d): undefined reference to `BIK_clear_data'
/home/rsmits/workspace/build/linux2/lib/libbf_rna.a(rna_pose_gen.o):
In function `rna_Pose_IK_update':
rna_pose_gen.c:(.text+0xc12): undefined reference to `BIK_clear_data'
/home/rsmits/workspace/build/linux2/lib/libbf_converter.a(BL_ArmatureObject.o):
In function `game_free_pose(bPose*)':
BL_ArmatureObject.cpp:(.text+0x702): undefined reference to `BIK_clear_data'
collect2: ld returned 1 exit status
scons: *** [/home/rsmits/workspace/build/linux2/bin/blender] Error 1
scons: building terminated because of errors.

apparently the libikplugin.a is not getting linked in the blender executable

Ruben

On Thu, Jul 30, 2009 at 4:54 PM, Benoit Bolsee<benoit.bolsee at online.be> wrote:
> Hi Ruben,
>
> Thanks for the file. Makefile is not the best way to compile Blender.
> Can you try with scons?
>
> Move to the directory where you checked out itasc (directory with
> Sconstruct file). You may want to tune first some scons parameter by
> copying ./config/linux2-config.py to ./user-config.py and change some
> parameters there. I usually disable the gameplayer:
> WITH_BF_PLAYER = False
>
> Then run 'python scons/scons.py'
> This will build blender in ../install directory
>
> I didn't try to compile in Linux since I ported the code to 2.5 but I
> did compile for Linux when the code was still in the 2.4 branch. I will
> try on my side to compile in Ubuntu.
>
> Regards,
> Benoit
>
>> -----Original Message-----
>> From: robotics-bounces at blender.org
>> [mailto:robotics-bounces at blender.org] On Behalf Of Ruben Smits
>> Sent: jeudi 30 juillet 2009 15:57
>> To: Blender and Robotics
>> Subject: Re: [Robotics] iTaSC branch available for test at
>> graphicall.org
>>
>>
>> On Thu, Jul 30, 2009 at 2:20 PM, Ruben
>> Smits<ruben.smits at mech.kuleuven.be> wrote:
>> > On Mon, Jul 27, 2009 at 9:32 PM, Benoit
>> > Bolsee<benoit.bolsee at online.be> wrote:
>> >> Hi Florian,
>> >>
>> >> I saw the group, very good idea thanks.
>> >> Yes building the iTaSC branch in Ubuntu should be straightforward
>> >> although I never did it. The scons building system is up
>> the date and
>> >> should work in Linux. Here is the address of the branch:
>> >>
>> >> https://svn.blender.org/svnroot/bf-blender/branches/itasc
>> >
>> > I get the following error while building the itasc branch
>> on my ubuntu
>> > jaunty system:
>> >
>> > ====> make all in intern/itasc
>> > make[2]: sdl-config: Command not found
>> > make[2]: sdl-config: Command not found
>> > make[2]: sdl-config: Command not found
>> > mkdir: cannot create directory
>> >
>> `/home/rsmits/workspace/blender-itasc/obj/linux-glibc2.9-x86_64/intern
>> > /intern/itasc':
>> > No such file or directory
>> >
>> > I guess the problem is the double intern listing in the
>> path, probably
>> > something wrong in some makefile, i looked around a bit but did not
>> > found out yet where the error comes from
>>
>> I already got a little further, but i had to apply the
>> changes from the attached patch file.
>>
>> Now i'm stuck at the following;
>> make all in source/gameengine/GamePlayer/ghost
>> make[1]: *** No rule to make target
>> `/home/rsmits/workspace/blender-itasc/obj/linux-glibc2.9-x86_6
> 4/gameengine/blphys/fuzzics/libfuzzics.a',
>> needed by
>> `/home/rsmits/workspace/blender-itasc/obj/linux-glibc2.9-x86_6
>> 4/bin/blender'.
>>  Stop.
>> make: *** [all] Error 1
>>
>> I don't find anything in the source tree that comes even
>> close to something being called fuzzics :(
>>
>> Ruben
>>
>> >> It is parallel to the blender directory of the official blender2.5
>> >> branch.
>> >>
>> >> If you can make a Ubuntu branch and test it with the demo
>> file, I can
>> >> upload it to graphicall.
>> >>
>> >> Regards,
>> >> Benoit
>> >>
>> >>
>> >> -----Original Message-----
>> >> From: robotics-bounces at blender.org
>> >> [mailto:robotics-bounces at blender.org] On Behalf Of Florian Lier
>> >> Sent: lundi 27 juillet 2009 20:36
>> >> To: Blender and Robotics
>> >> Subject: Re: [Robotics] iTaSC branch available for test at
>> >> graphicall.org
>> >>
>> >> Hey,
>> >>
>> >> ok - this excludes 2.5 for me - FOR NOW.
>> >> I was just thinking about switching to the iTaSC
>> >> branch and using the BGE for my master thesis, but
>> >> there are, as u said, too many limitations. So I will continue my
>> >> work on 2.49, which is still thrilling :)
>> >>
>> >> My 2 cent:
>> >> I think your work will extremly push the process of
>> >> using blender in robotics, because building models,
>> environments and
>> >> interactivly control those entities will be extremly "easy" in
>> >> comparison to other solutions.
>> >>
>> >> I created a Blender and Robotics group on Vimeo, so adding
>> videos in
>> >> this group will aggregate the knowledge...
>> >>
>> >> Here's the link: http://www.vimeo.com/groups/blenderandrobotics
>> >>
>> >>
>> >> Cheers, Florian
>> >>
>> >> P.S.:
>> >> If u don't have the time for a build u can send me the url of the
>> >> sources. I guess building the iTaSC branch isn't extremly
>> different
>> >> from building the "normal" 2.5...
>> >>
>> >>
>> >> Benoit Bolsee wrote:
>> >>
>> >> OK, I'll make a Ubuntu build as soon as I can.
>> >> The BGE has a different Python API than Blender and unfortunately,
>> >> there is no API yet to get the pose matrix or the joint angle. The
>> >> 'old' blender API is available, however it would be of any use in
>> >> this case because the pose data is stored on the BGE side
>> and would
>> >> need a special API to get it. This is something I will
>> work on in the
>> >> near future.
>> >>
>> >> 2.5 is indeed strongly limited but improving everyday day. It is
>> >> quite possible to build a model in 2.49 and just open it
>> in 2.5 and
>> >> use the new features. This is how I did the robot model in
>> this demo.
>> >> Editing in 2.5 is possible but very limited and still
>> buggy. Linking
>> >> data between files is not yet possible unfortunately but the ogic
>> >> editor and the BGE is functional in 2.5.
>> >>
>> >> Regards,
>> >> Benoit
>> >>
>> >>
>> >>
>> >> -----Original Message-----
>> >> From: robotics-bounces at blender.org
>> >> [mailto:robotics-bounces at blender.org] On Behalf Of Florian Lier
>> >> Sent: lundi 27 juillet 2009 19:04
>> >> To: Blender and Robotics
>> >> Subject: Re: [Robotics] iTaSC branch available for test at
>> >> graphicall.org
>> >>
>> >>
>> >> Wow, this outstanding !!
>> >>
>> >> I really want to test the build on Ubuntu!
>> >>
>> >> A few questions:
>> >>
>> >> 1) 2.5 is strongly limited at the moment, is it possible to create
>> >> own models?
>> >> 2) Is the python API the same? Is is possible to use "old" python
>> >> scripts in the BGE?
>> >>
>> >>
>> >> Cheers, and again - nice work!
>> >>
>> >> Florian
>> >>
>> >>
>> >> _______________________________________________
>> >> Robotics mailing list
>> >> Robotics at blender.org
>> >> http://lists.blender.org/mailman/listinfo/robotics
>> >>
>> >>
>> >>
>> >
>>
>
> _______________________________________________
> Robotics mailing list
> Robotics at blender.org
> http://lists.blender.org/mailman/listinfo/robotics
>