[Webkit-unassigned] [Bug 52699] Allow Fixed Length values to be floating point

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Fri Feb 18 12:55:50 PST 2011


https://bugs.webkit.org/show_bug.cgi?id=52699





--- Comment #13 from Rik Cabanier <cabanier at gmail.com>  2011-02-18 12:55:50 PST ---
(In reply to comment #12)
> I would just make the Length always be a float and not have it be both.  Also the bitfield is completely useless now if you use the full 32 bits for the number (it's not saving you any space), so you can just turn those into bools.
> 
> Lots of style errors too.  You might want to flag for review just so you can see the style errors and fix them.

I was trying to minimize changes to the code. By having it behave differently only when the object is constructed with a float, we keep existing behavior.

My first attempt was to move to just float, but it actually changes the result of a large number of operations. It resulted in a lot of changes all over the code and numerous problems and differences in the layout tests.
Float also has less significant digits than int...

The bitfield is actually still necessary because part of the table logic multiplies by 128. If the value was maxint, it results in overflows which cause incorrect layout.

I will look at the style errors.

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



More information about the webkit-unassigned mailing list