Robert O'Callahan wrote: > On 8/30/07, *fantasai* <[EMAIL PROTECTED] > <mailto:[EMAIL PROTECTED]>> wrote: > > Why does a size change require invalidating the overflow area? If the frame's > descendants have changed size or position, they should have already been > invalidated by the time the frame exists Reflow, no? If that's true the only > part that needs invalidation is the difference in the frame's mRect (and that > really only if there are backgrounds, outlines, or borders). Are you saying > that some frames don't invalidate their children properly, or am I missing > something else here? > > The CSS 'outline' property.
That seems to be covered by CheckInvalidateSizeChange(), which is called by the frame in its Reflow function--at least for blockframes. I think that results in us invalidating the same region twice for such frames. (Is that a bad thing?) ~fantasai _______________________________________________ dev-tech-layout mailing list [email protected] https://lists.mozilla.org/listinfo/dev-tech-layout
