[Bf-python] a direct x importer addon for 2.6x (naming concern)

jerome jerome.le.chat at free.fr
Mon Aug 20 19:07:27 CEST 2012 

Hi,
I updated it to 2.63a at last.
regards,
Jerome

Le 23/01/2012 17:59, Campbell Barton a écrit :
> Since moving the module I've updated the imports, it should work now,
> tho the only files I have to test with run very slow so wasnt able to
> ensure this finishes an import.
>
> On Tue, Jan 24, 2012 at 3:39 AM, jerome <jerome.le.chat at free.fr> wrote:
>> I fully agree. for the house keeping need and bpy_extras I mean :)
>> important :
>> the bel module won't work as is if you let it in the direct_x_bel folder :
>> a suitable location would be, for now, for this rev, in .../2.61/scripts/modules/bel
>>
>> maybe I can mod it to fit in /bpy_extras/bel (as a step) ?
>>
>> Jerome
>>
>>
>> Le 23/01/2012 17:16, Campbell Barton a écrit :
>>> Hi Jerome,
>>>
>>> Great to see your script in SVN,
>>>
>>> I've moved bel/ into being a submodule of the  io_directx_bel since
>>> this is the only user so far.
>>>
>>> For general utility functions I prefer extend bpy_extras, noticed
>>> theres already some overlap between `bel` and `bpy_extras`,
>>> http://www.blender.org/documentation/blender_python_api_2_61_release/bpy_extras.html
>>>
>>> At some stage we can look at moving parts of bel/ into bpy_extras/.
>>>
>>> Also, noticed none of the files in bel/ have a GPL header, please add
>>> one - copy from any of the py scripts in trunk/ should do.
>>>
>>> this is annoying house keeping, but rather keep this stuff in order.
>>>
>>> - Campbell
>>>
>>> On Mon, Jan 23, 2012 at 6:44 PM, jerome<jerome.le.chat at free.fr>  wrote:
>>>> Campbell,
>>>>
>>>> I committed two folders in svn addons (2931) : bel and io_directx_bel
>>>> the bel folder should reside in .../2.61/scripts/modules/ to avoid this
>>>> message :
>>>>
>>>> Warning: addon-module 'bel' found module but without __addon_enabled__
>>>> field, possible name collision from file:
>>>> '.../2.61/scripts/addons/bel/__init__.py'
>>>>
>>>> but it's  ok if you don't move it from /addons
>>>>
>>>> no big updates, I essentially added a 'last run' feature that saves the
>>>> last importer config automatically in the operator preset path.
>>>> so one can recall it from the operator preset dropdown ('last run' set).
>>>> method should be available for any other scripts, allowing to refine an
>>>> operator setup after several tries.
>>>> (bel.fs.saveOptions() at line 214 in io_directx_bel)
>>>>
>>>>
>>>> Le 12/01/2012 17:09, Campbell Barton a écrit :
>>>>> checked the updated version, looks good so +1 to commit this to
>>>>> contrib/, after some testing can probably go in trunk for release.
>>>>>
>>>>> @jerome, if you don't have commit access mail me with your login name
>>>>> on http://projects.blender.org, otherwise, feel free to commit into:
>>>>>
>>>>> https://svn.blender.org/svnroot/bf-extensions/contrib/py/scripts/addons
>>>>>
>>>>> But only the script, we have a different place for test files:
>>>>> https://svn.blender.org/svnroot/bf-blender/trunk/lib/tests/io_tests
>>>>>
>>>>> ... but this needs access to blender/trunk svn.
>>>>>
>>>>> On Fri, Jan 13, 2012 at 1:01 AM, jerome<jerome.le.chat at free.fr>    wrote:
>>>>>> thanks for your encouragement Campbell :)
>>>>>>
>>>>>> updated according to your last mail below :
>>>>>> blender naming for img and tex, 'image exists ?' tests made by path
>>>>>> comparison module not by bpy.img names, etc... all details appended to
>>>>>> README
>>>>>> excepted for the dict suggestion, but this should work without it.
>>>>>> https://github.com/littleneo/directX_blender/wiki
>>>>>>
>>>>>> fyi, the bel module provided is a working multi files draft that acts as
>>>>>> a layer for import/export/data injection stuff, as an extra helper module.
>>>>>> in this case the x importer (should) only cares about data
>>>>>> parsing/conversion and ask bel to solve things about naming and data
>>>>>> injection.
>>>>>> I supposed most of the existing methods/functions in import/export or
>>>>>> mesh creation addons could reside in such a kind of module, since cases
>>>>>> and concerns about naming, data generations/creations are the same. so
>>>>>> its my long-term attempt to do this.
>>>>>> thanks for any c&c about it. feel free to mod.
>>>>>>
>>>>>> Jerome
>>>>>>
>>>>>>
>>>>>> Le 29/12/2011 00:43, Campbell Barton a écrit :
>>>>>>> Hi, I had a look over the script and think its mostly ok but still
>>>>>>> some things I think should be resolved.
>>>>>>>
>>>>>>> the obvious issues are still with namespace. from checking:
>>>>>>> littleneo-directX_blender-a3ed6b4
>>>>>>>
>>>>>>> these lines assume an image filename can fit into an ID name which may
>>>>>>> well fail if the names over 22 chars.
>>>>>>> if filename not in bpy.data.images :
>>>>>>> if filename not in bpy.data.textures :
>>>>>>>
>>>>>>> suggest to have a dictionary which maps names to datablocks.
>>>>>>> if its desireable to use existing datablocks this dict could
>>>>>>> optionally be initialized from existing data when the import starts.
>>>>>>>
>>>>>>> Another is...
>>>>>>>      bpy.ops.object.select_name(name=arm.name)
>>>>>>>
>>>>>>> This should be replaces with...
>>>>>>>      arm.select = True
>>>>>>>      bpy.context.scene.objects.active = arm
>>>>>>>
>>>>>>>
>>>>>>> Since the script looks quite well written (aside from noted issues) -
>>>>>>> suggest this go into contrib/ so users can start testing more easily
>>>>>>> and the few remaining issues can be resolved.
>>>>>>>
>>>>>>> On Tue, Dec 27, 2011 at 7:42 AM, jerome<jerome.le.chat at free.fr>      wrote:
>>>>>>>> hi,
>>>>>>>>
>>>>>>>> the .x file importer has now support for armatures and parent-childs
>>>>>>>> relationships importation.
>>>>>>>> download and info :
>>>>>>>> https://github.com/littleneo/directX_blender/wiki
>>>>>>>>
>>>>>>>> you need to enable the 'experimental blah' checkbox in the importer
>>>>>>>> options for armatures/empties, and relation-ship importation.
>>>>>>>> once the script has finished its job, mesh objects need to be parented
>>>>>>>> 'by hand' using alt-p>      Armature Deform or an Armature modifier.
>>>>>>>> (vertices weights are ready)
>>>>>>>>
>>>>>>>> Campbell please let me know for naming convention, it's ok according to me.
>>>>>>>> Thanks for feedback and .x files you'd like to share to improve this.
>>>>>>>>
>>>>>>>> wish you the best,
>>>>>>>>
>>>>>>>> Jerome
>>>>>>>>
>>>>>>>>
>>>>>>>> Le 30/11/2011 01:06, jerome a écrit :
>>>>>>>>> Le 25/11/2011 05:16, Campbell Barton a écrit :
>>>>>>>>>> On Thu, Nov 24, 2011 at 6:52 PM, jerome<jerome.le.chat at free.fr>        wrote:
>>>>>>>>>>> oh ok Campbell,
>>>>>>>>>>>
>>>>>>>>>>> this should be alright then, at least for objs, meshes and mats. (tex
>>>>>>>>>>> slots / img not supported yet but will be)
>>>>>>>>>>> I though you talked about python module namespaces, 'import / from bpy
>>>>>>>>>>> import' etc stuffs :)
>>>>>>>>>>>
>>>>>>>>>>> bpy.data.materials.new(name="MyMat").name = name
>>>>>>>>>>> :)
>>>>>>>>>>>
>>>>>>>>>>> J.
>>>>>>>>>> see found a way around but its not foolproof.
>>>>>>>>>>
>>>>>>>>>> you might still exceed the name length or have a blank script - see:
>>>>>>>>>> http://www.blender.org/documentation/blender_python_api_2_60_5/info_gotcha.html#naming-limitations
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> Bf-python mailing list
>>>>>>>>>> Bf-python at blender.org
>>>>>>>>>> http://lists.blender.org/mailman/listinfo/bf-python
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>> oh the bpy.data.materia... line was just a kind of joke, you can find
>>>>>>>>> the naming function quickly. assuming the name max lenght is 21 this
>>>>>>>>> should be ok (actually 17 + 4 digits maybe a bit paranoid..).
>>>>>>>>>
>>>>>>>>> sorry I was not very available these days and won't until the next week.
>>>>>>>>> nice pointer btw thanks.
>>>>>>>>>
>>>>>>>>> Jerome
>>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Bf-python mailing list
>>>>>>>> Bf-python at blender.org
>>>>>>>> http://lists.blender.org/mailman/listinfo/bf-python
>>>>>> _______________________________________________
>>>>>> Bf-python mailing list
>>>>>> Bf-python at blender.org
>>>>>> http://lists.blender.org/mailman/listinfo/bf-python
>>>> _______________________________________________
>>>> Bf-python mailing list
>>>> Bf-python at blender.org
>>>> http://lists.blender.org/mailman/listinfo/bf-python
>>>
>> _______________________________________________
>> Bf-python mailing list
>> Bf-python at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-python
>
>

More information about the Bf-python mailing list