[Bug 10649] Regression in RegQueryValueExA when called in unorthodox manner

[wine-bugs at winehq.org]

http://bugs.winehq.org/show_bug.cgi?id=10649





--- Comment #12 from Juan Lang <juan_lang at yahoo.com>  2007-12-12 22:08:02 ---
I just tried your test on WinXP, and it succeeds there.  Also, size remains at
4 after the call, so you may want to add a test for that as well.

Regarding your comment #4:
> the problem is that the early type setting change was
> presumably to match some undocumented behaviour too and not knowing exactly
> what that behaviour was, I can't test that.

This is to match win9x when called for a non-existent value, as shown in the
tests at line 570:
http://source.winehq.org/source/dlls/advapi32/tests/registry.c#L570

It's possible to modify RegQueryValueExA only to set the type to 0 when the
value isn't found/can't be read, which might be clearer than your approach.


-- 
Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.