[Webkit-unassigned] [Bug 38402] Paginate and column-break at layout time rather than when painting

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=38402


Dave Hyatt <hyatt at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
         AssignedTo|webkit-unassigned at lists.web |hyatt at apple.com
                   |kit.org                     |




--- Comment #4 from Dave Hyatt <hyatt at apple.com>  2010-09-10 13:15:32 PST ---
Created an attachment (id=67226)
 --> (https://bugs.webkit.org/attachment.cgi?id=67226)
Implement internal pagination of floats

There are two big remaining issues with this patch:

(1) Being willing to insert a "strut" between the top of a block and its first line box creates really ugly visual artifacts, especially if the block has a border or background.  Similar to how margins can collapse out through the top of a block, we need to push pagination info for first lines out through block ancestors, so that all of those blocks move to the next page/column instead.  This is a very complicated problem, but I think solving it is essential to treating drop caps floats as unsplittable (because the line and the rest of the block go together).  As a first pass I may only push through to the immediate enclosing block.

(2) There is no pagination of unsplittable floats yet.  Those need to push down to the next page when they don't fit, and they need to bring the line they are on along with them.

I don't expect (2) to be all that difficult, but (1) is going to be a monster.  Once these problems are solved, though, I'll be in a position to land, since all the multicol tests will be passing.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.