On Saturday 08 May 2010 06:08:28 am Milan Crha wrote:
> Hi there,
> I'm still in process of testing the PT_DOUBLE bug, with not much
> success, but that might be Evolution's issue, I'll investigate more.
Tell me more about what you're seeing. Is there a bugzilla entry somewhere?
Can you get a network capture?
> But meanwhile, I'm offering you this patch. It slightly simplifies
> cast_mapi_SPropValue function, and adds not as that many handled types,
> but few it does. With this patch applied I can read my items from the
> server with no crash, where after revision 1762 I couldn't. This patch
> is applicable to revision 1775.
I didn't like the macro (just a stylistic preference), and didn't apply that
bit. I copied your support for PT_MV_UNICODE and PT_MV_LONG directly, and
implemented PT_I8 based on your patch (just without the macro).
Committed as r1779.
I'm undecided on this bit:
+ case PT_NULL:
+ case PT_OBJECT:
+ case PT_MV_SHORT:
+ /* types, which cannot be stored in mapi_SPropValue */
+ printf ("cast_mapi_SPropValue: Cannot cast prop value 0x%x to
mapi
type\n", sprop->ulPropTag & 0xFFFF);
+ break;
Can you explain what you're trying to do here?
[In a previous email you basically asked "why not just do them all"? I don't
mind doing it, I just don't like code I don't have any tests for. I just don't
trust untested code, and don't want to commit it. That is why I'm relying on
your testing for this.]
Brad
_______________________________________________
devel mailing list
[email protected]
http://mailman.openchange.org/listinfo/devel
