[PATCH v2 2/2] user32: Properly handle monochrome icon creation
Zebediah Figura (she/her)
zfigura at codeweavers.com
Mon Aug 30 16:58:03 CDT 2021
On 8/30/21 4:54 PM, Fabian Maurer wrote:
> On Montag, 30. August 2021 23:38:25 CEST you wrote:
>> On 8/30/21 4:31 PM, Fabian Maurer wrote:
>>> On Montag, 30. August 2021 18:26:12 CEST Zebediah Figura (she/her) wrote:
>>>> On 8/21/21 8:18 AM, Fabian Maurer wrote:
>>>>> This fixes a regression from db2b266c57b73e1a16785213ce923b749c84400e.
>>>>> Multiple things here:
>>>>> 1) We must use a color table to avoid undefined behavior
>>>>> 2) We must set the bitmap line by line, in reverse
>>>> This sounds like a sign that the patch should be split...
>>> How exactly? Realistically, you need both for it to work, just having half
>>> of it isn't useful.
>> I must be missing something, but I didn't suggest only submitting one part.
> Sorry, I mean, patches should be able to stand on their own, right? I don't
> see a reason for splitting if just adding the one wouldn't add an improvement,
> although I'm open for it.
Sure, patches should be able to stand on their own, but "improvement"
doesn't have to be user-visible. Though even if it were, I'm not sure
why these wouldn't qualify.
More information about the wine-devel