CVS error
Steve Lime
steve.lime at DNR.STATE.MN.US
Tue Feb 7 12:42:10 EST 2006
For the moment I would just roll back to your solution of this morning. At least
it's consistent with the rest of the code. We can consider a global parameter name
change for "template" for 5.0 if necessary. (this hasn't been a huge problem to
date other than the build issues)
Steve
>>> Daniel Morissette <dmorissette at DMSOLUTIONS.CA> 02/07/06 11:20 AM >>>
Petter Reinholdtsen wrote:
> [Daniel Morissette]
>>
>>#ifndef __cplusplus
>> char *template;
>>#else
>> char *_template;
>>#endif
>
>
> Why not rename it to _template in C, and thus avoid having to document
> different name for different languages? Perhaps include a '#define
> template _template' for backward compatibility with existing C
> programs, and announce that the define is going away in the future.
>
Um... if we do that then we should do it accross the board to be
consistent and not just for this new gmlItemObj, but then that may have
implications in SWIG MapScript which is generated directly from the C
header files. i.e. we may end up with _template in the scripted
languages. That's why this #ifdef was used in the past.
What do others think?
Daniel
--
------------------------------------------------------------
Daniel Morissette dmorissette at dmsolutions.ca
DM Solutions Group http://www.dmsolutions.ca/
------------------------------------------------------------
More information about the mapserver-dev
mailing list