Currently the shadowSize parameter of the setShadow and getShadow functions of GraphicsContext is used to pass the offset of the shadow specified in the style of the elements.
Created attachment 65404 [details] Proposed patch Replace the names in the APIs and the locations where the functions are used.
After changing the names I've realized that this condition could be wrong but the old name could make it seem correct: + drawIntoBitmap = !offset.isEmpty() || blur; It is in this file: WebCore/platform/graphics/win/FontCGWin.cpp static void drawGDIGlyphs Could someone check if in Windows negative offsets of the text actually work, or no offsets but blurring. We could open a new bug in this case.
Attachment 65404 [details] did not build on qt: Build output: http://queues.webkit.org/results/3806070
Created attachment 65419 [details] Proposed patch
(In reply to comment #2) > After changing the names I've realized that this condition could be wrong but the old name could make it seem correct: > > + drawIntoBitmap = !offset.isEmpty() || blur; > > It is in this file: WebCore/platform/graphics/win/FontCGWin.cpp static void drawGDIGlyphs > > Could someone check if in Windows negative offsets of the text actually work, or no offsets but blurring. We could open a new bug in this case. I’ve asked someone to check, and indeed in <data:text/html,%3Cdiv%20style=%22-webkit-font-smoothing:%20none%22%3E%3Cdiv%20style=%22text-shadow:%202px%202px%20red%22%3Etest%3C/div%3E%3Cdiv%20style=%22text-shadow:%20-2px%20-2px%20red%22%3Etest%3C/div%3E%3C/div%3E> the bottom line is missing the shadow. Good catch!
Thanks for the test :). Opened a bug for the other issue: bug 44619. I'll land this patch tomorrow.
Landed http://trac.webkit.org/changeset/66082