- 07 Jul, 2015 1 commit
-
-
Iván Matellanes authored
-
- 06 Jul, 2015 34 commits
-
-
Huw Davies authored
-
Alexandre Julliard authored
-
Aaryaman Vasishta authored
-
Aaryaman Vasishta authored
-
Thomas Faber authored
-
Thomas Faber authored
-
Olivier F. R. Dierick authored
-
Olivier F. R. Dierick authored
-
Huw Davies authored
-
Dmitry Timoshkov authored
Some custom implementations return it instead of STG_E_INVALIDFUNCTION.
-
Nikolay Sivov authored
-
Nikolay Sivov authored
-
Nikolay Sivov authored
-
Nikolay Sivov authored
-
Nikolay Sivov authored
-
Nikolay Sivov authored
-
Nikolay Sivov authored
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
Also lock/unlock the module only on creation/destruction of the object.
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
-
Alex Henrie authored
75 characters is the maximum length of a line before gettext breaks it, so this convention will cause each translatable string to fit snugly into the po files. It's also about what we were doing anyway, just not consistently. The biggest advantage to consistent line wrapping is that translators will no longer have to guess where they should insert line breaks. I have rewrapped the Catalan, Czech, Danish, German, Spanish, Finnish, French, Croatian, Hungarian, Italian, Lithuanian, Norwegian, Dutch, Polish, Portuguese, Romanian, Russian, Slovene, Swedish, Turkish, and Ukranian translations so as to not burden their maintainers. The Bulgarian, Greek, Esperanto, Farsi, Hindi, Malayalam, Oriya, Punjabi, Romansh, Slovak, Serbian, Telugu, Thai, and Walloon translations did not need to be rewrapped. I removed Hebrew "translations" of cmd.rc that were just copies of the English strings, and marked cmd.rc:142 as fuzzy if the translation mentioned wine.conf. Single-line translations that were 80 characters long or less were not wrapped. The Arabic, Japanese, Korean, and Chinese translations still need to be rewrapped.
-
Sebastian Lackner authored
-
Sebastian Lackner authored
When the number of elements per bucket is too small, then try to reduce the number of threads by merging buckets. This is to ensure that the number of running wait queue threads doesn't get too big.
-
Sebastian Lackner authored
-
Sebastian Lackner authored
-
Sebastian Lackner authored
-
Sebastian Lackner authored
To implement waiting for an arbitrary number of handles, we group them in buckets up to (MAXIMUM_WAIT_OBJECTS - 1) objects, and then assign a dedicated wait queue thread. The last handle is used to notify about changes.
-
Sebastian Lackner authored
-
- 03 Jul, 2015 5 commits
-
-
Nikolay Sivov authored
-
Nikolay Sivov authored
-
Michael Stefaniuc authored
-
Michael Stefaniuc authored
Also lock/unlock the module only on creation/destruction of the object.
-
Michael Stefaniuc authored
-