Dermont kindly advised me to ask on this mailing list whether there are any binaries available for 1.6.4 on Debian unstable 64-bit. So, if anybody has something like that, I'd love to hear about it.
In the meantime, I've started building from source. I see some people have trouble with 64-bit at runtime, but I figure if I can get it built, then we're that much closer to fixing it.
> Dermont kindly advised me to ask on this mailing list whether there are > any binaries available for 1.6.4 on Debian unstable 64-bit. So, if > anybody has something like that, I'd love to hear about it.
> In the meantime, I've started building from source. I see some people > have trouble with 64-bit at runtime, but I figure if I can get it built, > then we're that much closer to fixing it.
Thanks! I googled the error but didn't come up with that. That solved the problem and I compiled the wrappers.
Almost there. I've tried running a few of the demos, but it dies here:
Added resource location '../media/fonts' of type 'FileSystem' to resource group 'General' <snip> OGRE EXCEPTION(5:ItemIdentityException): Could not find font BlueHighway in TextAreaOverlayElement::setFontName at OgreTextAreaOverlayElement.cpp (line 360)
Note that ../media/fonts does contain bluehigh.ttf.
> > Dermont kindly advised me to ask on this mailing list whether there are > > any binaries available for 1.6.4 on Debian unstable 64-bit. So, if > > anybody has something like that, I'd love to hear about it.
> > In the meantime, I've started building from source. I see some people > > have trouble with 64-bit at runtime, but I figure if I can get it built, > > then we're that much closer to fixing it.
> > Ogre itself builds fine according to the instructions on > > http://wiki.python-ogre.org/index.php/LinuxBuildV2 and I've run the > > samples to verify everything works. The problem comes on Step 5:
Am Mittwoch 17 Februar 2010 11:24:22 schrieb Steve Freitas:
> Building the first part (ogre) dies with this error: > /usr/bin/ld: cannot find -lboost_python-gcc44-mt
Make a symlink from /usr/lib64/libboost_python-mt.so to /usr/lib64/libboost_python-gcc44-mt.so and rerun again.
The problem with the 64 bit is not building. It is running most of the demos. Some demos at least start and show something. But most die with an error.
Okay, good to hear. Andy, Dermont, do you have any suggestions for what I should do next to chase this down? I'm happy to do a debug build, apply patches, whatever, I just need a bit of guidance.
On Wed, 2010-02-17 at 12:20 +0100, Enrico Zschemisch wrote: > Hi again,
> Am Mittwoch 17 Februar 2010 12:13:25 schrieb Steve Freitas: > > Almost there. I've tried running a few of the demos, but it dies here: > this is exactly the 64 Bit problem everybody talks about ;-)
I'm doing further clean up and testing and should have an environment for you to test with later today (It will be Ogre 1.7RC1, latest boost etc)..
Once you get updated then we can work together on the debugging..
In the mean time do you have any of the demos that work? If so which ones? What are the errors you are seeing? Does an ogre.log file get produced and if so can you send it through (really only interested in the last few lines, try to see what the last thing Ogre did)..
Then my normal way for debugging this would be to add log statements throughout the ogre source, rebuild/install ogre (don't need to change Python-Ogre initially) in an attempt to lock down where the failure is occurring -- of course the 'right' way to do this is to run it under a debugger and trace back from the break point and if someone has time to post instructions on how to do this that would be very helpful (I'll have to refresh myself on debugging under Linux...)
Andy
On 18 February 2010 04:41, Steve Freitas <sfl...@ihonk.com> wrote:
> Okay, good to hear. Andy, Dermont, do you have any suggestions for what I > should do next to chase this down? I'm happy to do a debug build, apply > patches, whatever, I just need a bit of guidance.
> Thanks!
> Steve
> On Wed, 2010-02-17 at 12:20 +0100, Enrico Zschemisch wrote:
> Hi again,
> Am Mittwoch 17 Februar 2010 12:13:25 schrieb Steve Freitas: > > Almost there. I've tried running a few of the demos, but it dies here: > this is exactly the 64 Bit problem everybody talks about ;-)
> -- > You received this message because you are subscribed to the Google Groups > "Python Ogre Developers" group. > To post to this group, send email to > python-ogre-developers@googlegroups.com. > To unsubscribe from this group, send email to > python-ogre-developers+unsubscribe@googlegroups.com<python-ogre-developers% 2Bunsubscribe@googlegroups.com> > . > For more options, visit this group at > http://groups.google.com/group/python-ogre-developers?hl=en.
Great news, Andy. Thanks. I'm happy to give it all a go.
Just ran through them all and none of the demos work. Here are the results of my checks. Log file for each demo is attached.
Demos which opened the setup dialog fine then crashed after that due to the following error:
OGRE EXCEPTION(5:ItemIdentityException): Could not find font BlueHighway in TextAreaOverlayElement::setFontName at OgreTextAreaOverlayElement.cpp (line 360)
Demo_Basic.py: Would not start, it just shuts down after complaining about not finding "plugins.cfg.linux". Demo_Compositor.py: Would not start due to missing libCEGUIBase.so.1 Demo_NetworkClient.py: Didn't try Demo_NetworkServer.py: Didn't try Demo_OgreCollision02.py: Didn't try, haven't compiled ode Demo_Shadows.py: Would not start due to missing libCEGUIBase.so.1 Demo_Video.py: Does nothing and exits normally Demo_Water.py: Does nothing and exits normally Demo_WX.py: Didn't try
Demo_SSAO_standalone.py: Would not start, error is:
23:59:34: OGRE EXCEPTION(7:InternalErrorException): Could not load dynamic library /home/steve/dev/sim/src/python-ogre/plugins/RenderSystem_GL. System Error: /home/steve/dev/sim/src/python-ogre/plugins/RenderSystem_GL.so: cannot open shared object file: o such file or directory in DynLib::load at OgreDynLib.cpp (line 81)
Demo_Spinner.py: Would not start, error is:
OGRE EXCEPTION(7:InternalErrorException): Could not load dynamic library ../../plugins/RenderSystem_GL.dll. System Error: ../../plugins/RenderSystem_GL.dll.so: cannot open shared object file: No such file or directory in DynLib::load at OgreDyn ib.cpp (line 81)
On Thu, 2010-02-18 at 10:59 +0800, Andy Miller wrote: > Steve
> I'm doing further clean up and testing and should have an environment > for you to test with later today (It will be Ogre 1.7RC1, latest boost > etc)..
> Once you get updated then we can work together on the debugging..
> In the mean time do you have any of the demos that work? If so which > ones? What are the errors you are seeing? Does an ogre.log file get > produced and if so can you send it through (really only interested in > the last few lines, try to see what the last thing Ogre did)..
> Then my normal way for debugging this would be to add log statements > throughout the ogre source, rebuild/install ogre (don't need to change > Python-Ogre initially) in an attempt to lock down where the failure > is occurring -- of course the 'right' way to do this is to run it > under a debugger and trace back from the break point and if someone > has time to post instructions on how to do this that would be very > helpful (I'll have to refresh myself on debugging under Linux...)
> Andy
> On 18 February 2010 04:41, Steve Freitas <sfl...@ihonk.com> wrote:
> Okay, good to hear. Andy, Dermont, do you have any suggestions > for what I should do next to chase this down? I'm happy to do > a debug build, apply patches, whatever, I just need a bit of > guidance.
> Thanks!
> Steve
> On Wed, 2010-02-17 at 12:20 +0100, Enrico Zschemisch wrote:
> > Hi again,
> > Am Mittwoch 17 Februar 2010 12:13:25 schrieb Steve Freitas: > > > Almost there. I've tried running a few of the demos, but it dies here: > > this is exactly the 64 Bit problem everybody talks about ;-)
> -- > You received this message because you are subscribed to the > Google Groups "Python Ogre Developers" group. > To post to this group, send email to > python-ogre-developers@googlegroups.com. > To unsubscribe from this group, send email to > python-ogre-developers+unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/python-ogre-developers?hl=en.
> -- > You received this message because you are subscribed to the Google > Groups "Python Ogre Developers" group. > To post to this group, send email to > python-ogre-developers@googlegroups.com. > To unsubscribe from this group, send email to python-ogre-developers > +unsubscribe@googlegroups.com. > For more options, visit this group at > http://groups.google.com/group/python-ogre-developers?hl=en.
Probably best to create and run a minimal example that crashes, something like the following. I would also try extracting OgreCore.zip to a directory such as test64bit and load resources from that to rule out a zip/loading/parsing problem.
import sys import os sys.path.insert(0,'..') import PythonOgreConfig import ogre.renderer.OGRE as ogre
> I'm doing further clean up and testing and should have an environment for > you to test with later today (It will be Ogre 1.7RC1, latest boost etc)..
> Once you get updated then we can work together on the debugging..
> In the mean time do you have any of the demos that work? If so which ones? > What are the errors you are seeing? Does an ogre.log file get produced and > if so can you send it through (really only interested in the last few lines, > try to see what the last thing Ogre did)..
> Then my normal way for debugging this would be to add log statements > throughout the ogre source, rebuild/install ogre (don't need to change > Python-Ogre initially) in an attempt to lock down where the failure > is occurring -- of course the 'right' way to do this is to run it under a > debugger and trace back from the break point and if someone has time to post > instructions on how to do this that would be very helpful (I'll have to > refresh myself on debugging under Linux...)
> Andy
> On 18 February 2010 04:41, Steve Freitas <sfl...@ihonk.com> wrote:
> > Okay, good to hear. Andy, Dermont, do you have any suggestions for what I > > should do next to chase this down? I'm happy to do a debug build, apply > > patches, whatever, I just need a bit of guidance.
> > Thanks!
> > Steve
> > On Wed, 2010-02-17 at 12:20 +0100, Enrico Zschemisch wrote:
> > Hi again,
> > Am Mittwoch 17 Februar 2010 12:13:25 schrieb Steve Freitas: > > > Almost there. I've tried running a few of the demos, but it dies here: > > this is exactly the 64 Bit problem everybody talks about ;-)
> > -- > > You received this message because you are subscribed to the Google Groups > > "Python Ogre Developers" group. > > To post to this group, send email to > > python-ogre-developers@googlegroups.com. > > To unsubscribe from this group, send email to > > python-ogre-developers+unsubscribe@googlegroups.com<python-ogre-developers% 2Bunsubscribe@googlegroups.com> > > . > > For more options, visit this group at > > http://groups.google.com/group/python-ogre-developers?hl=en.
@demont - thanks for the debugging info -- agree on the extracting resources etc as steve is getting a strange error..
Steve -- You can take a look here<http://wiki.python-ogre.org/index.php/LinuxBuildV4>for the latest build instructions (I'm still testing however ogre and ois code generation is OK -- testing compile at the moment - slow Linux machine)..
As you can see from the instructions we are using as many standard system libraries as possible with the only exceptions being Nvidia's CG and GCCXML (the standard one possibly works however we know it's buggy) -- and we are installing everything into standard system locations.
None of the C++ libraries are patched (including Ogre) so if you already have Ogre/OIS/CEGUI installed then they can be used -- however note that I'm using OGRE 1.7.0RC1 and would want to use this for any debugging so please do retrieve and build it.
There is also a 'new' library to build which is boost_python (instead of the older complete 'boost) -- I've included a zip with just the required source in the SVN and it extracts and builds as part of the '... -b boost_python ...' step.
Things are in flux so please feel free to update the wiki page and let me know if you have any issues.. -- might also make sense to start a new thread - perhaps 'Working on the 64bit issues" :)
Andy
On 18 February 2010 16:52, dermont <dermontg...@gmail.com> wrote:
> Probably best to create and run a minimal example that crashes, > something like the following. I would also try extracting OgreCore.zip > to a directory such as test64bit and load resources from that to rule > out a zip/loading/parsing problem.
> import sys > import os > sys.path.insert(0,'..') > import PythonOgreConfig > import ogre.renderer.OGRE as ogre
> ogre.ResourceGroupManager.getSingleton().initialiseResourceGroup(resGroup) > del root
> Andy Miller wrote: > > Steve
> > I'm doing further clean up and testing and should have an environment for > > you to test with later today (It will be Ogre 1.7RC1, latest boost etc)..
> > Once you get updated then we can work together on the debugging..
> > In the mean time do you have any of the demos that work? If so which > ones? > > What are the errors you are seeing? Does an ogre.log file get produced > and > > if so can you send it through (really only interested in the last few > lines, > > try to see what the last thing Ogre did)..
> > Then my normal way for debugging this would be to add log statements > > throughout the ogre source, rebuild/install ogre (don't need to change > > Python-Ogre initially) in an attempt to lock down where the failure > > is occurring -- of course the 'right' way to do this is to run it under a > > debugger and trace back from the break point and if someone has time to > post > > instructions on how to do this that would be very helpful (I'll have to > > refresh myself on debugging under Linux...)
> > Andy
> > On 18 February 2010 04:41, Steve Freitas <sfl...@ihonk.com> wrote:
> > > Okay, good to hear. Andy, Dermont, do you have any suggestions for > what I > > > should do next to chase this down? I'm happy to do a debug build, apply > > > patches, whatever, I just need a bit of guidance.
> > > Thanks!
> > > Steve
> > > On Wed, 2010-02-17 at 12:20 +0100, Enrico Zschemisch wrote:
> > > Hi again,
> > > Am Mittwoch 17 Februar 2010 12:13:25 schrieb Steve Freitas: > > > > Almost there. I've tried running a few of the demos, but it dies > here: > > > this is exactly the 64 Bit problem everybody talks about ;-)
> > > -- > > > You received this message because you are subscribed to the Google > Groups > > > "Python Ogre Developers" group. > > > To post to this group, send email to > > > python-ogre-developers@googlegroups.com. > > > To unsubscribe from this group, send email to > > > python-ogre-developers+unsubscribe@googlegroups.com<python-ogre-developers% 2Bunsubscribe@googlegroups.com> > <python-ogre-developers%2Bunsubscribe@googlegroups.com<python-ogre-develope rs%252Bunsubscribe@googlegroups.com>
> -- > You received this message because you are subscribed to the Google Groups > "Python Ogre Developers" group. > To post to this group, send email to > python-ogre-developers@googlegroups.com. > To unsubscribe from this group, send email to > python-ogre-developers+unsubscribe@googlegroups.com<python-ogre-developers% 2Bunsubscribe@googlegroups.com> > . > For more options, visit this group at > http://groups.google.com/group/python-ogre-developers?hl=en.
let me share my findings as I am somewhat lost :-(
Problem: Font can not be found. The font can not be found because it is not loaded. The font is not loaded because in OgreResourceGroupManager.cpp around line 920 "StringVector &patterns" is of unreasonable size and does not contain anything at all. The resource manager the loop is currently working at is OgreFontManager. When bypassing this place in OgreResourceGroupManager.cpp and manually adding the font to the list of resources to load, the font is loaded (correctly?). When bypassing and letting the demo (I am using Demo_Grass.py) run I get this segmentation fault (see stacktrace at end of email).
The stacktrace confirms my suspicion that something might be wrong with the bindings and/or boost at this place.
@Andy: gccxml is included in most distributions, so no need to install it additionally. So you can take it out from the wiki.
Hints? Ideas?
I will try with Ogre 1.7 now.
Regards, Enrico
Stacktrace:
Program received signal SIGSEGV, Segmentation fault. 0x00007f2105f6a053 in std::basic_ostream<char, std::char_traits<char> >& std::operator<< <char, std::char_traits<char>, std::allocator<char>
Hehe, and again. Any ideas, Andy? Do I have to specifiy certain options when building code for Ogre 1.7? I can not generate code for 1.7:
python python-ogre/BuildModule.py -g ogre --usesystem PythonOgre.BuildModule: INFO Building Source code for ogre PythonOgre.BuildModule: WARNING Task Failed 11:10:35 - /mnt/data/projekte/po : cat log.out 02-18 11:10 PythonOgre.BuildModule INFO Building Source code for ogre 02-18 11:10 PythonOgre.BuildModule DEBUG Spawning 'python generate_code.py --usesystem' in '/mnt/data/projekte/po/python-ogre/code_generators/ogre' 02-18 11:10 PythonOgre.BuildModule WARNING Task Failed 02-18 11:10 PythonOgre.BuildModule DEBUG 02-18 11:10 PythonOgre.BuildModule DEBUG INFO Creating xml file "/mnt/data/projekte/po/python-ogre/code_generators/cache/ogre_1.6.4_cache.x ml" from source file "/mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre.h" ... INFO gccxml cmd: /usr/bin/gccxml -I"/mnt/data/projekte/po/python-ogre" - I"/mnt/data/projekte/po/python-ogre/code_generators/ogre" - I"/usr/include/boost" -I"/usr/include/OGRE" -I"/usr/include/python2.5/" - D"OGRE_NONCLIENT_BUILD" -D"OGRE_GCC_VISIBILITY" -D"__PYTHONOGRE_BUILD_CODE" - D"__GCCXML__" -D"OGRE_VERSION_1.6.4" "/mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre.h" -fxml="/mnt/data/projekte/po/python- ogre/code_generators/cache/ogre_1.6.4_cache.xml" Traceback (most recent call last): File "generate_code.py", line 1696, in <module> generate_code() File "generate_code.py", line 1431, in generate_code , cflags=environment.ogre.cflags File "/usr//lib/python2.5/site- packages/pyplusplus/module_builder/boost_python_builder.py", line 95, in __init__ , indexing_suite_version) File "/usr//lib/python2.5/site- packages/pyplusplus/module_builder/boost_python_builder.py", line 138, in __parse_declarations decls = reader.read_files( files, compilation_mode ) File "/usr//lib/python2.5/site-packages/pygccxml/parser/project_reader.py", line 217, in read_files return self.__parse_file_by_file(files) File "/usr//lib/python2.5/site-packages/pygccxml/parser/project_reader.py", line 254, in __parse_file_by_file reader.create_xml_file( header, prj_file.cached_source_file ) File "/usr//lib/python2.5/site-packages/pygccxml/parser/source_reader.py", line 179, in create_xml_file raise error pygccxml.parser.source_reader.gccxml_runtime_error_t: Error occured while running GCC-XML: In file included from /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre.h:23: /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:453: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:453: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:453: error: '_Tmap_traits' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:453: error: wrong number of template arguments (6, should be 3) /usr/include/OGRE/OgreIteratorWrapper.h:46: error: provided for 'template<class T, class IteratorType, class ValType> class Ogre::IteratorWrapper' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:453: error: expected unqualified-id before '>' token /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:454: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:454: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:454: error: '_Tmap_traits' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:454: error: wrong number of template arguments (6, should be 3) /usr/include/OGRE/OgreIteratorWrapper.h:46: error: provided for 'template<class T, class IteratorType, class ValType> class Ogre::IteratorWrapper' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:454: error: expected unqualified-id before '>' token /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:455: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:455: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:455: error: '_Tmap_traits' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:455: error: wrong number of template arguments (6, should be 2) /usr/include/OGRE/OgreIteratorWrapper.h:255: error: provided for 'template<class T, class IteratorType> class Ogre::MapIteratorWrapper' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:455: error: expected unqualified-id before '>' token /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:456: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:456: error: '_Tree' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:456: error: '_Tmap_traits' is not a member of 'std' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:456: error: wrong number of template arguments (6, should be 2) /usr/include/OGRE/OgreIteratorWrapper.h:255: error: provided for 'template<class T, class IteratorType> class Ogre::MapIteratorWrapper' /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_aliases.h:456: error: expected unqualified-id before '>' token In file included from /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre.h:31: /mnt/data/projekte/po/python-ogre/code_generators/ogre/python_ogre_sizeof.h : In function 'void python_ogre::details::instantiate()': /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_sizeof.h:29: error: 'stdext' was not declared in this scope /mnt/data/projekte/po/python- ogre/code_generators/ogre/python_ogre_sizeof.h:29: error: template argument 2 is invalid
Did you grab the latest svn ?? Can you check what rev you are using..
Also from a test perspective (for 64 bit) I put Test_01.py in the demos/ogre directory which is the basic program from Dermont. You might want to try it and then extract the ogrecore.zip , edit the test program (you can see where it is commented out) and try again, just in case the 64 bit issue is .zip file related..
> Hehe, and again. Any ideas, Andy? Do I have to specifiy certain options > when > building code for Ogre 1.7? > I can not generate code for 1.7:
> python python-ogre/BuildModule.py -g ogre --usesystem > PythonOgre.BuildModule: INFO Building Source code for ogre > PythonOgre.BuildModule: WARNING Task Failed > 11:10:35 - /mnt/data/projekte/po : cat log.out > 02-18 11:10 PythonOgre.BuildModule INFO Building Source code for ogre > 02-18 11:10 PythonOgre.BuildModule DEBUG Spawning 'python > generate_code.py > --usesystem' in '/mnt/data/projekte/po/python-ogre/code_generators/ogre' > 02-18 11:10 PythonOgre.BuildModule WARNING Task Failed > 02-18 11:10 PythonOgre.BuildModule DEBUG > 02-18 11:10 PythonOgre.BuildModule DEBUG INFO Creating xml file
> "/mnt/data/projekte/po/python-ogre/code_generators/cache/ogre_1.6.4_cache.x ml" > from source file "/mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre.h" ... > INFO gccxml cmd: /usr/bin/gccxml -I"/mnt/data/projekte/po/python-ogre" - > I"/mnt/data/projekte/po/python-ogre/code_generators/ogre" - > I"/usr/include/boost" -I"/usr/include/OGRE" -I"/usr/include/python2.5/" - > D"OGRE_NONCLIENT_BUILD" -D"OGRE_GCC_VISIBILITY" -D"__PYTHONOGRE_BUILD_CODE" > - > D"__GCCXML__" -D"OGRE_VERSION_1.6.4" "/mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre.h" > -fxml="/mnt/data/projekte/po/python- > ogre/code_generators/cache/ogre_1.6.4_cache.xml" > Traceback (most recent call last): > File "generate_code.py", line 1696, in <module> > generate_code() > File "generate_code.py", line 1431, in generate_code > , cflags=environment.ogre.cflags > File "/usr//lib/python2.5/site- > packages/pyplusplus/module_builder/boost_python_builder.py", line 95, in > __init__ > , indexing_suite_version) > File "/usr//lib/python2.5/site- > packages/pyplusplus/module_builder/boost_python_builder.py", line 138, in > __parse_declarations > decls = reader.read_files( files, compilation_mode ) > File > "/usr//lib/python2.5/site-packages/pygccxml/parser/project_reader.py", > line 217, in read_files > return self.__parse_file_by_file(files) > File > "/usr//lib/python2.5/site-packages/pygccxml/parser/project_reader.py", > line 254, in __parse_file_by_file > reader.create_xml_file( header, prj_file.cached_source_file ) > File "/usr//lib/python2.5/site-packages/pygccxml/parser/source_reader.py", > line 179, in create_xml_file > raise error > pygccxml.parser.source_reader.gccxml_runtime_error_t: Error occured while > running GCC-XML: In file included from /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre.h:23: > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:453: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:453: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:453: error: '_Tmap_traits' > is > not a member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:453: error: wrong number of > template arguments (6, should be 3) > /usr/include/OGRE/OgreIteratorWrapper.h:46: error: provided for > 'template<class T, class IteratorType, class ValType> class > Ogre::IteratorWrapper' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:453: error: expected > unqualified-id before '>' token > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:454: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:454: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:454: error: '_Tmap_traits' > is > not a member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:454: error: wrong number of > template arguments (6, should be 3) > /usr/include/OGRE/OgreIteratorWrapper.h:46: error: provided for > 'template<class T, class IteratorType, class ValType> class > Ogre::IteratorWrapper' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:454: error: expected > unqualified-id before '>' token > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:455: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:455: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:455: error: '_Tmap_traits' > is > not a member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:455: error: wrong number of > template arguments (6, should be 2) > /usr/include/OGRE/OgreIteratorWrapper.h:255: error: provided for > 'template<class T, class IteratorType> class Ogre::MapIteratorWrapper' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:455: error: expected > unqualified-id before '>' token > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:456: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:456: error: '_Tree' is not > a > member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:456: error: '_Tmap_traits' > is > not a member of 'std' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:456: error: wrong number of > template arguments (6, should be 2) > /usr/include/OGRE/OgreIteratorWrapper.h:255: error: provided for > 'template<class T, class IteratorType> class Ogre::MapIteratorWrapper' > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_aliases.h:456: error: expected > unqualified-id before '>' token > In file included from /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre.h:31:
> /mnt/data/projekte/po/python-ogre/code_generators/ogre/python_ogre_sizeof.h : > In function 'void python_ogre::details::instantiate()': > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_sizeof.h:29: error: 'stdext' was not > declared in this scope > /mnt/data/projekte/po/python- > ogre/code_generators/ogre/python_ogre_sizeof.h:29: error: template argument > 2 > is invalid
> -- > You received this message because you are subscribed to the Google Groups > "Python Ogre Developers" group. > To post to this group, send email to > python-ogre-developers@googlegroups.com. > To unsubscribe from this group, send email to > python-ogre-developers+unsubscribe@googlegroups.com<python-ogre-developers% 2Bunsubscribe@googlegroups.com> > . > For more options, visit this group at > http://groups.google.com/group/python-ogre-developers?hl=en.
> just another small note about the new Build Instructions: Nvidia's CG is > included in Debian and Ubuntu (gccxml, too).
> -- > You received this message because you are subscribed to the Google Groups > "Python Ogre Developers" group. > To post to this group, send email to > python-ogre-developers@googlegroups.com. > To unsubscribe from this group, send email to > python-ogre-developers+unsubscribe@googlegroups.com<python-ogre-developers% 2Bunsubscribe@googlegroups.com> > . > For more options, visit this group at > http://groups.google.com/group/python-ogre-developers?hl=en.
I just installed GCCXML and CG with the commands supplied by the wiki. The wrapper code can be generated now and it is compiling...
If possible, I would like to replace the "make install" commands in the scripts with "checkinstall", if the checkinstall-program is installed. This will generate and install the required packages using the distribution's package management system. Yes/No?
> Also from a test perspective (for 64 bit) I put Test_01.py in the > demos/ogre directory which is the basic program from Dermont. You might > want to try it and then extract the ogrecore.zip , edit the test program > (you can see where it is commented out) and try again, just in case the 64 > bit issue is .zip file related..
a quick look at the 1.7 RC1 build with python ogre looks great :) I tried some demos which did not work before and they are working now :-)
Some notes I took while building (the whole day :-/ ):
This file is created incorrectly: python-ogre/generated/ogre_1.7.0RC1/mapProfilerProfileHistory\>.pypp.hpp Andy already corrected that one.
Next, the class RequestHandlerHolder in OgreWorkQueue.h is protected, but needs to be public. Similar: struct Ogre::EdgeListBuilder::vectorLess in OgreEdgeListBuilder.h is protected but has to be public.
CEGUI fails to build. Does not matter if installed separately or using the distribution's package. This seems to be something in the wrapper code, which I have no idea of:
02-18 15:44 PythonOgre.BuildModule INFO Building Source code for cegui 02-18 15:44 PythonOgre.BuildModule DEBUG Spawning 'python generate_code.py' in '/mnt/data/projekte/po/python-ogre/code_generators/cegui' 02-18 15:44 PythonOgre.BuildModule WARNING Task Failed 02-18 15:44 PythonOgre.BuildModule DEBUG 02-18 15:44 PythonOgre.BuildModule DEBUG INFO Creating xml file "/mnt/data/projekte/po/python- ogre/code_generators/cache/cegui_0.7.1_cache.xml" from source file "/mnt/data/projekte/po/python-ogre/code_generators/cegui/python_CEGUI.h" ... INFO gccxml cmd: /usr/bin/gccxml -I"/mnt/data/projekte/po/python-ogre" - I"/mnt/data/projekte/po/root/usr/include/boost" - I"/mnt/data/projekte/po/root/usr/include/CEGUI" - I"/mnt/data/projekte/po/CEGUI-0.7.1" - I"/mnt/data/projekte/po/root/usr/include/CEGUI/RendererModules/Ogre" - I"/mnt/data/projekte/po/root/usr/include/OGRE" - I"/mnt/data/projekte/po/root/usr/include/OGRE" -D"OGRE_NONCLIENT_BUILD" - D"CEGUI_NONCLIENT_BUILD" -D"OGRE_GCC_VISIBILITY" -D"__PYTHONOGRE_BUILD_CODE" - D"OGRE_GUIRENDERER_EXPORTS" -D"VERSION_0_7_1" "/mnt/data/projekte/po/python- ogre/code_generators/cegui/python_CEGUI.h" - fxml="/mnt/data/projekte/po/python- ogre/code_generators/cache/cegui_0.7.1_cache.xml" Traceback (most recent call last): File "generate_code.py", line 381, in <module> generate_code() File "generate_code.py", line 312, in generate_code , cflags=environment.ogre.cflags File "/usr//lib/python2.5/site- packages/pyplusplus/module_builder/boost_python_builder.py", line 95, in __init__ , indexing_suite_version) File "/usr//lib/python2.5/site- packages/pyplusplus/module_builder/boost_python_builder.py", line 138, in __parse_declarations decls = reader.read_files( files, compilation_mode ) File "/usr//lib/python2.5/site-packages/pygccxml/parser/project_reader.py", line 217, in read_files return self.__parse_file_by_file(files) File "/usr//lib/python2.5/site-packages/pygccxml/parser/project_reader.py", line 254, in __parse_file_by_file reader.create_xml_file( header, prj_file.cached_source_file ) File "/usr//lib/python2.5/site-packages/pygccxml/parser/source_reader.py", line 179, in create_xml_file raise error pygccxml.parser.source_reader.gccxml_runtime_error_t: Error occured while running GCC-XML: In file included from /usr/include/c++/4.4/tr1/type_traits:47, from /usr/include/c++/4.4/tr1/unordered_map:43, from /mnt/data/projekte/po/root/usr/include/OGRE/OgreStdHeaders.h:40, from /mnt/data/projekte/po/root/usr/include/OGRE/OgrePrerequisites.h:315, from /mnt/data/projekte/po/root/usr/include/OGRE/OgreBlendMode.h:31, from /mnt/data/projekte/po/root/usr/include/CEGUI/RendererModules/Ogre/CEGUIOgre Renderer.h:35, from /mnt/data/projekte/po/python- ogre/code_generators/cegui/python_CEGUI.h:18: /usr/include/c++/4.4/tr1_impl/type_traits:207: error: a function call cannot appear in a constant-expression /usr/include/c++/4.4/tr1_impl/type_traits:207: error: template argument 2 is invalid /usr/include/c++/4.4/tr1_impl/type_traits:213: error: a function call cannot appear in a constant-expression /usr/include/c++/4.4/tr1_impl/type_traits:213: error: template argument 2 is invalid /usr/include/c++/4.4/tr1_impl/type_traits:219: error: a function call cannot appear in a constant-expression /usr/include/c++/4.4/tr1_impl/type_traits:219: error: template argument 2 is invalid /usr/include/c++/4.4/tr1_impl/type_traits:226: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:226: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:227: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:227: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:227: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:229: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:229: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:230: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:230: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:230: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:232: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:232: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:233: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:233: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:233: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:235: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:235: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:236: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:236: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:236: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:238: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:238: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:239: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:239: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:239: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:241: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:241: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:242: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:242: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:242: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:244: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:244: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:245: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:245: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:245: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:247: error: expected nested-name- specifier before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:247: error: expected '>' before '...' token /usr/include/c++/4.4/tr1_impl/type_traits:248: error: '_ArgTypes' was not declared in this scope /usr/include/c++/4.4/tr1_impl/type_traits:248: error: type/value mismatch at argument 1 in template parameter list for 'template<class _Tp> struct std::tr1::is_function' /usr/include/c++/4.4/tr1_impl/type_traits:248: error: expected a type, got '(_Res)(<expression error>)' /usr/include/c++/4.4/tr1_impl/type_traits:333: error: a function call cannot appear in a constant-expression /usr/include/c++/4.4/tr1_impl/type_traits:333: error: template argument 2 is invalid /usr/include/c++/4.4/tr1_impl/type_traits:339: error: a function call cannot appear in a constant-expression /usr/include/c++/4.4/tr1_impl/type_traits:339: error: template argument 2 is invalid /usr/include/c++/4.4/tr1_impl/type_traits:345: error: a function call cannot appear in a constant-expression /usr/include/c++/4.4/tr1_impl/type_traits:345: error: template argument 2 is invalid In file included from /usr/include/c++/4.4/tr1/unordered_map:43, from /mnt/data/projekte/po/root/usr/include/OGRE/OgreStdHeaders.h:40, from /mnt/data/projekte/po/root/usr/include/OGRE/OgrePrerequisites.h:315, from /mnt/data/projekte/po/root/usr/include/OGRE/OgreBlendMode.h:31, from /mnt/data/projekte/po/root/usr/include/CEGUI/RendererModules/Ogre/CEGUIOgre Renderer.h:35, from /mnt/data/projekte/po/python- ogre/code_generators/cegui/python_CEGUI.h:18: /usr/include/c++/4.4/tr1/type_traits:147: error: expected primary-expression before ',' token /usr/include/c++/4.4/tr1/type_traits:147: error: expected primary-expression before ')' token /usr/include/c++/4.4/tr1/type_traits:147: error: a function call cannot appear in a constant-expression /usr/include/c++/4.4/tr1/type_traits:241: error: requested alignment is not a constant
On Thu, 2010-02-18 at 00:52 -0800, dermont wrote: > I would also try extracting OgreCore.zip > to a directory such as test64bit and load resources from that to rule > out a zip/loading/parsing problem.
Okay, I created
src/python-ogre/demos/test64bit
Then I unzipped the contents of OgreCore.zip into it and added the following line to src/python-ogre/demos/resource.cfg under the General section (tried Bootstrap too):
FileSystem=../test64bit
The change had no effect on the font loading problem.
> ogre.ResourceGroupManager.getSingleton().initialiseResourceGroup(resGroup) > del root
Thanks for the sample code. It crashes the same way as all the others looking for the BlueHighway font. I tried uncommenting the line referring to the test64bit directory and it didn't change the result.
On Thu, 2010-02-18 at 10:51 +0100, Enrico Zschemisch wrote: > Hi,
> let me share my findings as I am somewhat lost :-(
> Problem: Font can not be found. > The font can not be found because it is not loaded. > The font is not loaded because in OgreResourceGroupManager.cpp around line 920 > "StringVector &patterns" is of unreasonable size and does not contain anything > at all. The resource manager the loop is currently working at is > OgreFontManager. > When bypassing this place in OgreResourceGroupManager.cpp and manually adding > the font to the list of resources to load, the font is loaded (correctly?). > When bypassing and letting the demo (I am using Demo_Grass.py) run I get this > segmentation fault (see stacktrace at end of email).
> The stacktrace confirms my suspicion that something might be wrong with the > bindings and/or boost at this place.
It looks like you've closed in on the error. Should we all continue to try to get 1.6.4 working? Or abandon it for 1.7?
