summaryrefslogtreecommitdiff
path: root/redo/imagelist_windows.c
diff options
context:
space:
mode:
authorPietro Gagliardi <[email protected]>2014-08-29 20:46:39 -0400
committerPietro Gagliardi <[email protected]>2014-08-29 20:46:39 -0400
commitb0564aa91b01976df34d13cf2ad84272cbc1ca46 (patch)
tree8c87bddbd5c19d8081c9d1d5cc1450c41b645af0 /redo/imagelist_windows.c
parent1ff16049ab76df6813437eedd36887c40ecdd818 (diff)
Settled alpha support in Windows imagelists; now need to settle premultiplied....
Diffstat (limited to 'redo/imagelist_windows.c')
-rw-r--r--redo/imagelist_windows.c4
1 files changed, 2 insertions, 2 deletions
diff --git a/redo/imagelist_windows.c b/redo/imagelist_windows.c
index 08046e7..4402268 100644
--- a/redo/imagelist_windows.c
+++ b/redo/imagelist_windows.c
@@ -30,8 +30,8 @@ HIMAGELIST newImageList(int width, int height)
{
HIMAGELIST il;
- // TODO does this strip alpha?
- // sinni800 in irc.freenode.net/#go-nuts suggests our use of *image.RGBA makes this not so much of an issue
+ // this handles alpha properly; see https://web.archive.org/web/20100512144953/http://msdn.microsoft.com/en-us/library/ms997646.aspx#xptheming_topic13 and http://stackoverflow.com/a/2640897/3408572
+ // TODO alpha-premultiplied? http://stackoverflow.com/a/641836/3408572 implies it is, http://trac.wxwidgets.org/ticket/9050#comment:1 implies it isn't
il = (*fv_ImageList_Create)(width, height, ILC_COLOR32, 20, 20); // should be reasonable
if (il == NULL)
xpanic("error creating image list", GetLastError());