Looking over the results on test.winehq.org, the user32 menu tests are the worst in terms of failure count (though I suppose crashing tests are worse): they fail over 1000 times on NT4, over 600 times on Win9x, and over 300 times on Vista. The only operating systems on which they seem to succeed are Windows 2000 and XP, and they never succeeded for me on Windows XP, so I'm not sure the succeeding results are even representative.
Basically, they're the worst offender in terms of reliable results. Is there anyone who knows anything about them that can take a look at the test failures? --Juan
P.S. The crypt32 encode tests are the next-worst offender. I'm working on 'em.
2008/10/10 Juan Lang juan.lang@gmail.com:
Looking over the results on test.winehq.org, the user32 menu tests are the worst in terms of failure count (though I suppose crashing tests are worse): they fail over 1000 times on NT4, over 600 times on Win9x, and over 300 times on Vista. The only operating systems on which they seem to succeed are Windows 2000 and XP, and they never succeeded for me on Windows XP, so I'm not sure the succeeding results are even representative.
Basically, they're the worst offender in terms of reliable results. Is there anyone who knows anything about them that can take a look at the test failures?
I've looked in on them from time to time.
The Vista failures are due to the logic that is used to calculate the expected values for the size of menu items is tuned specifically to 2003 and earlier. Vista is doing that calculation differently in the presence of Aero, so you get the failures. Windows 2008 does not have these failures, as it is rendering similar to the "Classic" Windows 2000 look.
These size differences cannot be handled with error margins since the values are completely different. Also, the tests have some involved calculations that are used to determine the widths (this is likely so that it can cope with things like font size variance).
I don't know about the failures on other versions though.
- Reece
On Fri, Oct 10, 2008 at 12:51 PM, Juan Lang juan.lang@gmail.com wrote:
Looking over the results on test.winehq.org, the user32 menu tests are the worst in terms of failure count (though I suppose crashing tests are worse): they fail over 1000 times on NT4, over 600 times on Win9x, and over 300 times on Vista. The only operating systems on which they seem to succeed are Windows 2000 and XP, and they never succeeded for me on Windows XP, so I'm not sure the succeeding results are even representative.
Basically, they're the worst offender in terms of reliable results. Is there anyone who knows anything about them that can take a look at the test failures?
The last time I talked to Alexandre about the failing menu tests (test 25, 26, 27), he said it was because of a race condition. I don't have any more information than that though.
-
James Hawkins -
Juan Lang -
Reece Dunn