[Webkit-unassigned] [Bug 30304] struct RGBA32Buffer is twice as big as the old ImageData of Qt

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=30304





--- Comment #18 from Peter Kasting <pkasting at google.com>  2009-10-28 16:41:08 PDT ---
(In reply to comment #17)
> If you’d like to start doing this in WebKit, I think you should talk with other
> people about it in webkit-dev. It’s not debatable what the current style is.

I guess I don't feel like const-on-locals is something that needs to be
consistent between files.  For example, there are certainly many places where
we use "Foo*" where "const Foo*" would be appropriate, but I've never heard a
complaint of inconsistent style for the people who do use "const Foo*".  We're
not consistent about when we use ?: versus if (), preincrement versus
postincrement, etc.  How do we decide whether this is significant enough that
we should be consistent with other files, versus something trivial enough to
change?

(I note that using a non-const ref is something that's similarly not mentioned
anywhere, but I got picky about it on this bug.  I might be being
hypocritical?)

> And I also think it might be good to start using const for local variables.

I would like that in principle, but it seems over-burdensome to effectively
deprecate much of the codebase by saying that everywhere should move
consistently to using const where possible.  (Although I suppose we've done
something similarly broad with the namespace-indenting-in-headers change.)  I
guess this is really another way of saying the same thing I said in the above
paragraph.

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