Hello devs,
I see some XP [1] and W2k3 [2] boxes fail on every GetStandardColorSpaceProfileA() using valid parameters until SetStandardColorSpaceProfileA() is called for the first time. Then all those GetStandardColorSpaceProfileA() tests succeed.
I guess that mean SPACE_RGB color profile isn't registered on those boxes. Failures are just because of missing GLE codes, but IMHO we need to test this both cases. (when SPACE_RGB is registered and when it's not)
The problem is that if we test the unregistered color profile and we register it later using *A functions, then repeating same tests for the *W functions won't do OK, as I found no way to unregister Standard Color Space Profile. Any ideas on how to do that?
[1] http://test.winehq.org/data/200512221000/xp_punt/mscms:profile.txt [2] http://test.winehq.org/data/200512221000/2003_w2k3/mscms:profile.txt
On Friday 23 December 2005 14:11, Saulius Krasuckas wrote:
I see some XP [1] and W2k3 [2] boxes fail on every GetStandardColorSpaceProfileA() using valid parameters until SetStandardColorSpaceProfileA() is called for the first time. Then all those GetStandardColorSpaceProfileA() tests succeed.
In that case these tests have served their purpose. We should remove them because it is now clear that applications cannot depend on any color profile being registered for a standard color space in a Windows installation.
-Hans
* On Fri, 23 Dec 2005, Hans Leidekker wrote:
- On Friday 23 December 2005 14:11, Saulius Krasuckas wrote:
I see some XP [1] and W2k3 [2] boxes fail on every GetStandardColorSpaceProfileA() using valid parameters until SetStandardColorSpaceProfileA() is called for the first time. Then all those GetStandardColorSpaceProfileA() tests succeed.
In that case these tests have served their purpose.
Are you sure they have given all they can already? I am not.
We should remove them because it is now clear that applications cannot depend on any color profile being registered for a standard color space in a Windows installation.
Tests confirms such unability only for SPACE_RGB. To see if that's right in all other cases we should test all known SPACE_* constants. No?
OTOH, I am not sure how should behave Wine itself - I guess it is registering some profiles during Wine installation process now.
And my question remains unanswered - is there a way to unregister standard color space profile under win32?
-
Hans Leidekker -
Saulius Krasuckas