[Gdal-dev] GDAL changes affecting GRASS?

[William Kyngesburye]

Other things that changed around the same time (I didn't notice it  
until later, so I can't be absolutely sure when it went wrong):

- switched from JasPer to UUID JasPer

- Largefile support (that's when I started using GDAL CVS - end of  
July.  Unfortunately I didin't keep a snapshot of that source to test  
it now)

- Xerces added

- HDF5 added

- GEOS updated


I can do some more testing - leave those and others out, barebones  
GDAL (I considered that today).  I'll see what happens.


On Nov 29, 2005, at 6:55 PM, Frank Warmerdam wrote:

> On 11/29/05, William Kyngesburye <woklist at kyngchaos.com> wrote:
>> I've been having a problem with GRASS lately, and it seems to be
>> related to GDAL.  I have a GRASS bug open for it: #3585.  Noone has
>> provided any help so far.
>>
>> The key trigger combination is: GDAL 1.3.1 (and 1.3.0, possibly the
>> beta), GRASS 6.0/6.1, and built with GCC 3.3 on Mac OS 10.3.  There
>> is no problem building with GCC 4 on Mac OS 10.4, or with GCC 3.3 and
>> GDAL 1.2.6.  Or if I build GRASS without GDAL.  Yes, the GCC 3.3 and
>> GCC 4 builds each have their dependencies built with the
>> corresponding GCC version.
>>
>> The problem is that I get bus errors in a few module - specifically,
>> simple general commands such as g.remove, g.rename and g.copy (but
>> not others, like g.list).
>>
>>
>> So, the question for GDAL - what changed from 1.2.6 to 1.3.x that
>> might affect GRASS this way and only for GCC 3.3-built.
>
> William,
>
> I can't think of any obvious reason for this.  I think you will need
> to dig deeper before we can do anything about this.
>

-----
William Kyngesburye <kyngchaos at kyngchaos.com>
http://www.kyngchaos.com/

"I ache, therefore I am.  Or in my case - I am, therefore I ache."

- Marvin