[Bf-committers] Compilation error

Jürgen Herrmann shadowrom at me.com
Tue Jun 25 14:25:47 CEST 2013 

This compile error drives me insane!
I just tried to construct a simple repro case for MS...
And what shall I say it's gone!
After effortlessly trying to reproduce this with simple own code I tried with recent blender trunk (with reverted if 0 in makesrna of course) and it's gone ...

To quote Edgar Allan Poe: "I became insane, with long intervals of horrible sanity"

/Jürgen

Am 23.06.2013 um 16:13 schrieb Jürgen Herrmann <shadowrom at me.com>:

> Hi Brecht, 
> 
> now everything compiles fine, no more errors...
> We should really consider splitting this Monster into smaller headers.
> 
> Thanks for the help ;)
> 
> /Jürgen
> 
> -----Ursprüngliche Nachricht-----
> Von: bf-committers-bounces at blender.org
> [mailto:bf-committers-bounces at blender.org] Im Auftrag von Brecht Van Lommel
> Gesendet: Sonntag, 23. Juni 2013 15:59
> An: bf-blender developers
> Betreff: Re: [Bf-committers] Compilation error
> 
> I've removed the comments for now. This file is ridiculously large anyway,
> will need some changes if we ever make this a public API.
> 
> On Sat, Jun 22, 2013 at 8:24 PM, Jürgen Herrmann <shadowrom at me.com> wrote:
>> I can't stop laughing ...
>> I think found a bug in the compiler ...
>> RNA_blender_cpp.h is too long. That's it!
>> I fooled around with the file and deleted a random comment line from 
>> the file. After that it compiled fine ;) I asked myself it is possible to
> tell makesrna not to generate comments for this header as a workaround?
>> I'll report the bug to MS tomorrow, they should definitely take a look at
> this.
>> 
>> /Jürgen
>> 
>> Am 22.06.2013 um 16:40 schrieb Brecht Van Lommel
> <brechtvanlommel at pandora.be>:
>> 
>>> Ok, I also couldn't yet find the culprit from the error log and code 
>>> you posted, and still can't redo it here locally. I'll try some more 
>>> to figure this out but might not have enough time to do it this 
>>> weekend, so maybe monday.
>>> 
>>> On Sat, Jun 22, 2013 at 4:21 PM, Jürgen Herrmann <shadowrom at me.com>
> wrote:
>>>> I give up, I just can't find the error. I tested 3 PCs with VS 2012 Pro
> and Express with and without Update 1, 2 and 3 RC.
>>>> It is strange but I think it's some sort of strange Bug in VS 2012.
>>>> Some change made in 57608 seems to trigger this bug, every revision
> before is ok.
>>>> I'll restart the Buildbots after my System is back in a useable 
>>>> state, I think on Monday, just in case someone wants to test ;)
>>>> 
>>>> /Jürgen
>>>> 
>>>> -----Ursprüngliche Nachricht-----
>>>> Von: bf-committers-bounces at blender.org 
>>>> [mailto:bf-committers-bounces at blender.org] Im Auftrag von Jürgen 
>>>> Herrmann
>>>> Gesendet: Samstag, 22. Juni 2013 11:37
>>>> An: bf-blender developers
>>>> Betreff: Re: [Bf-committers] Compilation error
>>>> 
>>>> Status update:
>>>> 
>>>> Seems to be a problem with VS2012 Pro.
>>>> I did a complete clean reinstall of my computer (Windows, VS 2012 and
> all the prerequisites) over night and it still fails to compile.
>>>> 
>>>> /Jürgen
>>>> 
>>>> Am 21.06.2013 um 15:24 schrieb Jürgen Herrmann <shadowrom at me.com>:
>>>> 
>>>>> Alright,
>>>>> 
>>>>> I repaired my VS2012 installation, rebooted my system reinstallest he
> Updates for VS2012, rebooted again.
>>>>> I deleted an redownloaded the sources from svn, cleaned my build
> directories and started everything from scratch...
>>>>> Something is really strange here :(
>>>>> 
>>>>> Win64 and Win32 fail building with scons and CMake, all get the same
> error:
>>>>> http://www.pasteall.org/43377
>>>>> 
>>>>> That’s the RNA_blender_cpp.h generated:
>>>>> http://www.pasteall.org/43376
>>>>> 
>>>>> /Jürgen
>>>>> 
>>>>> -----Ursprüngliche Nachricht-----
>>>>> Von: bf-committers-bounces at blender.org 
>>>>> [mailto:bf-committers-bounces at blender.org] Im Auftrag von Jürgen 
>>>>> Herrmann
>>>>> Gesendet: Freitag, 21. Juni 2013 14:48
>>>>> An: bf-blender developers
>>>>> Betreff: Re: [Bf-committers] Compilation error
>>>>> 
>>>>> In in this case I'll first try to find the error on my side. I use vs
> 2012 professional and compile with the x64 native tool chain. Maybe using
> the x64 cross toolchain solves the problem.
>>>>> And a windows typical reboot might help too ;)
>>>>> 
>>>>> 
>>>>> Am 21.06.2013 um 14:07 schrieb Brecht Van Lommel
> <brechtvanlommel at pandora.be>:
>>>>> 
>>>>>> Hi,
>>>>>> 
>>>>>> I'm not sure what is causing this, could you post the 
>>>>>> RNA_blender_cpp.h file? It might give a clue. I don't see anything 
>>>>>> specifically wrong at the line numbers from that build log, in the 
>>>>>> RNA_blender_cpp.h that is generated here.
>>>>>> 
>>>>>> I also tried this with a visual studio 2012 express build (64 
>>>>>> bit), and didn't get any error.
>>>>>> 
>>>>>> Thanks,
>>>>>> Brecht.
>>>>>> 
>>>>>> On Fri, Jun 21, 2013 at 9:02 AM, Jürgen Herrmann <shadowrom at me.com>
> wrote:
>>>>>>> Ok, this gets strange...
>>>>>>> 
>>>>>>> when I do this in rna_scene.c Line 120-128:
>>>>>>> 
>>>>>>> EnumPropertyItem proportional_editing_items[] = {  
>>>>>>> {PROP_EDIT_OFF, "DISABLED", ICON_PROP_OFF, "Disable", 
>>>>>>> "Proportional Editing disabled"},  {PROP_EDIT_ON, "ENABLED", 
>>>>>>> ICON_PROP_ON, "Enable", "Proportional Editing enabled"},
> {PROP_EDIT_PROJECTED, "PROJECTED", ICON_PROP_ON, "Projected (2D)",
>>>>>>>                       "Proportional Editing using screen space 
>>>>>>> locations"},  {PROP_EDIT_CONNECTED, "CONNECTED", ICON_PROP_CON,
> "Connected",
>>>>>>>                       "Proportional Editing using connected 
>>>>>>> geometry only"}/*,  {0, NULL, 0, NULL, NULL}*/ };
>>>>>>> 
>>>>>>> (You can comment out any of these items to make it work...)
>>>>>>> 
>>>>>>> It compiles fine. But I just don't see how this is related to the 
>>>>>>> RNA_blender_cpp.h file When I recompile RNA_blender_cpp.h is
> regenerated but the contents doesn't seem to change at all, but the build
> error is gone.
>>>>>>> 
>>>>>>> 
>>>>>>> Am 21. Juni 2013 um 08:01 schrieb Jürgen Herrmann <shadowrom at me.com>:
>>>>>>> 
>>>>>>>> I tried to isolate the Problem and it seems to be r57608 that breaks
> the compilation on VC 2012.
>>>>>>>> 
>>>>>>>> 
>>>>>>>> Am 20. Juni 2013 um 23:24 schrieb Jürgen Herrmann
> <shadowrom at me.com>:
>>>>>>>> 
>>>>>>>>> Hi,
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> somewhere between r57600 and r57620 a strange error creeped 
>>>>>>>>> into my CMake
>>>>>>>>> Builds:
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> http://www.pasteall.org/43357
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> This seems to be a problem with the RNA generated files, when I 
>>>>>>>>> change RNA_blender_cpp.h lines 35299 – 35306 like this:
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> /* Interpolation */
>>>>>>>>> 
>>>>>>>>> enum ts_interpolation_enum {
>>>>>>>>> 
>>>>>>>>> interpolation_NONE = 0,
>>>>>>>>> 
>>>>>>>>> interpolation_BILINEAR = 1,
>>>>>>>>> 
>>>>>>>>> interpolation_BICUBIC = 2,
>>>>>>>>> 
>>>>>>>>> };
>>>>>>>>> 
>>>>>>>>> inline ts_interpolation_enum interpolation(void);
>>>>>>>>> 
>>>>>>>>> inline void interpolation(ts_interpolation_enum value);
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> and RNA_blender_cpp.h line 53860 to this:
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> ENUM_PROPERTY(ts_interpolation_enum, TransformSequence,
>>>>>>>>> interpolation)
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> Everything compiles fine.
>>>>>>>>> 
>>>>>>>>> I don’t get the problem though, interpolation_enum is used in 
>>>>>>>>> many classes within RNA_blender_cpp.h. but this one misbehaves 
>>>>>>>>> O_o
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> Building with scons/MinGW64 seems to be fine.
>>>>>>>>> 
>>>>>>>>> VC 2008 Win64 buildbot compiles fine too (I started it
>>>>>>>>> accidentally)
>>>>>>>>> 
>>>>>>>>> VC 2012 Win64 buildbot fails too…
>>>>>>>>> 
>>>>>>>>> So it seems to be a problem with the VC 11 compiler in some way 
>>>>>>>>> :/
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> /Jürgen
>>>>>>>>> 
>>>>>>>>> _______________________________________________
>>>>>>>>> Bf-committers mailing list
>>>>>>>>> Bf-committers at blender.org
>>>>>>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>>>>>> _______________________________________________
>>>>>>>> Bf-committers mailing list
>>>>>>>> Bf-committers at blender.org
>>>>>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>>>>> _______________________________________________
>>>>>>> Bf-committers mailing list
>>>>>>> Bf-committers at blender.org
>>>>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>>>> _______________________________________________
>>>>>> Bf-committers mailing list
>>>>>> Bf-committers at blender.org
>>>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>>> _______________________________________________
>>>>> Bf-committers mailing list
>>>>> Bf-committers at blender.org
>>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>>> 
>>>>> _______________________________________________
>>>>> Bf-committers mailing list
>>>>> Bf-committers at blender.org
>>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>> _______________________________________________
>>>> Bf-committers mailing list
>>>> Bf-committers at blender.org
>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>> 
>>>> _______________________________________________
>>>> Bf-committers mailing list
>>>> Bf-committers at blender.org
>>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>> _______________________________________________
>>> Bf-committers mailing list
>>> Bf-committers at blender.org
>>> http://lists.blender.org/mailman/listinfo/bf-committers
>> _______________________________________________
>> Bf-committers mailing list
>> Bf-committers at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-committers
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers
> 
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers

More information about the Bf-committers mailing list