May 2024 - Wine-GitLab - list.winehq.org

[PATCH v2 0/2] MR5634: win32u: Ignore WM_MOUSEMOVE when tracking scrollbar and arrow/track is clicked
by Fabian Maurer (＠DarkShadow44) May 28, 2024

May 28, 2024

This fixes a regression from bb496ea8. Also closer mimics windows behavior, moving the mouse should not reset the scrolling timer. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=56582 -- v2: win32u: Only set scroll timer if it's not running win32u: Refactor out scroll timer handling https://gitlab.winehq.org/wine/wine/-/merge_requests/5634

2 3

[PATCH v2 0/1] MR5650: msvcrt,winedump: Demandle std::nullptr_t datatype.
by eric pouech (＠epo) May 28, 2024

May 28, 2024

Signed-off-by: Eric Pouech <epouech(a)codeweavers.com> -- v2: msvcrt,winedump: Demangle std::nullptr_t datatype. https://gitlab.winehq.org/wine/wine/-/merge_requests/5650

4 3

Re: [PATCH v2 0/2] MR5646: ucrtbase: Handle empty list in variadic template.
by Piotr Caban (＠piotr) May 28, 2024

May 28, 2024

On Wed May 15 15:19:50 2024 +0000, eric pouech wrote: > to go one step further native also unmangles without whining something > like `"?AU?$my_iter(a)H$0A@$$VJ@@"` into `struct my_iter<int,0,long>` > (which doesn't make any sense in a C++ way either) > so it looks like the $$V sequence is just a marker for the end of a > variadic template, that is unmangled into nothing, and doesn't seem to > have (at least the ucrtbase version I've used) other visible action > so the attached patch should be enough > [err](/uploads/977e89694bc0fcda26a3ba0f1ccb46b7/err) This marker is specific to templates so it doesn't really belong to `demangle_datatype` function. The patch you have proposed e.g. incorrectly demangles `??__K_l@@YA?AUCC@@I$$VI(a)Z`. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5646#note_70553

1 0

[PATCH v3 0/5] MR5604: Draft: winemac.drv: Some dock icon hiding improvements
by Tim Clem (＠tclem) May 28, 2024

May 28, 2024

The most important thing here is a workaround for an OS bug: rapidly hiding and showing the dock icon can result in multiple icons for the same app, and those icons will ignore NSApp.applicationIconImage. Making sure transitions happen no less than 1 second apart from one another seems to fix it. *Edit:* v2 update: 0.5 seconds is not always long enough. A good test case is the Notepad++ installer: after you select a language, it rapidly hides and shows some windows, and without a longer delay the icon will wind up with the default system terminal-looking icon. One second is probably overkill, but this ought to be a somewhat uncommon case. Also, hiding a dock icon will deactivate an app. That's not desirable if it still has visible windows, so attempt to force a reactivation. That's often not successful on Sonoma due to the cooperative app activation heuristics, but it's the best we can do, and it should be a rare case. Also, there is some logic in -transformProcessToForeground: that we often want to happen regardless of whether the app is getting a dock icon (e.g. App Nap and activation). So I've renamed that function to more accurately communicate what it does, and consolidate it with -tryToActivateIgnoringOtherApps:. -- v3: winemac.drv: Consolidate foregrounding and dock icon behavior. winemac.drv: Remove incorrect documentation for orderBelow: and rename a parameter. winemac.drv: Reactivate the app if needed after hiding its dock icon. winemac.drv: Enforce a delay between dock icon hides and shows. winemac.drv: Factor menu creation out of transformProcessToForeground:. https://gitlab.winehq.org/wine/wine/-/merge_requests/5604

2 7

Re: [PATCH v2 0/2] MR5646: ucrtbase: Handle empty list in variadic template.
by eric pouech (＠epo) May 28, 2024

May 28, 2024

On Wed May 15 13:11:30 2024 +0000, Piotr Caban wrote: > I forgot to mention that native accepts multiple $$V sequences, so e.g. > `?AU?$my_iter(a)H$0A@$$V$$V@@` is also parsed (but probably doesn't make > sense and nothing is mangled this way). to go one step further native also unmangles without whining something like `"?AU?$my_iter(a)H$0A@$$VJ@@"` into `struct my_iter<int,0,long>` (which doesn't make any sense in a C++ way either) so it looks like the $$V sequence is just a marker for the end of a variadic template, that is unmangled into nothing, and doesn't seem to have (at least the ucrtbase version I've used) other visible action so the attached patch should be enough [err](/uploads/977e89694bc0fcda26a3ba0f1ccb46b7/err) -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5646#note_70538

1 0

[PATCH 0/6] MR5656: mshtml: Use DISPEX_IDISPATCH_IMPL (part 3)
by Jacek Caban (＠jacek) May 28, 2024

May 28, 2024

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/5656

4 8

[PATCH v6 0/11] MR5444: jscript: Do not allocate dispex_prop_t for indexed props.
by Gabriel Ivăncescu May 28, 2024

May 28, 2024

-- v6: jscript: Remove PROP_IDX dispex props by handling them in object prop methods. jscript: Move filling the PROTREF into a helper. jscript: Simplify get_flags to only check whether it's enumerable. jscript: Get rid of on_put in the object vtbl. jscript: Inline prop_put. jscript: Inline prop_get. jscript: Inline invoke_prop_func and invoke PROTREFs using their vtbl method. jscript: Inline delete_prop. jscript: Use mandatory methods in the object vtbl to operate on props found jscript: Use mandatory methods in the object vtbl to operate on props mshtml/tests: Test redefining a writable indexed prop. https://gitlab.winehq.org/wine/wine/-/merge_requests/5444

2 13

Re: [PATCH v3 0/4] MR5644: d3dx9: New tests and a compressed image loading bugfix. - approved
by Matteo Bruni (＠Mystral) May 28, 2024

May 28, 2024

This merge request was approved by Matteo Bruni. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5644

1 0

Re: [PATCH v2 0/2] MR5646: ucrtbase: Handle empty list in variadic template.
by Piotr Caban (＠piotr) May 28, 2024

May 28, 2024

On Wed May 15 12:40:06 2024 +0000, Piotr Caban wrote: > Yes I was thinking about something like you have attached first but I > didn't like few things: > - it requires adding new flag > - it's a special marker that is only valid in template, it can't be > used as datatype in other places > - your code is also not eating final '@' correctly but it can be > addressed easily I forgot to mention that native accepts multiple $$V sequences, so e.g. `?AU?$my_iter(a)H$0A@$$V$$V@@` is also parsed (but probably doesn't make sense and nothing is mangled this way). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/5646#note_70526

1 0

[PATCH 0/6] MR5638: wined3d: Remove texture NPOT fixup.
by Elizabeth Figura (＠zfigura) May 28, 2024

May 28, 2024

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/5638

4 8