[Webkit-unassigned] [Bug 82287] MemoryCache should adopt our standard RefCounted model for object lifetime

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=82287





--- Comment #16 from Brady Eidson <beidson at apple.com>  2012-05-22 16:37:31 PST ---
(In reply to comment #15)
> (From update of attachment 143386 [details])
> LGTM.  Please add some explanation to the top of CachedResource.h as well as CachedResourceHandle.h to explain when one is supposed to use a CachedResourceHandle, vs. a RefPtr.  You can do that in a separate patch, but I believe it's important to document this distinction in the code now that it's possible to confuse these two.

If this is a concern, then why are we doing this bizarre halfway step?

Understanding the MemoryCache is already difficult enough.  Making it even more confusing (but relying on comments for clarity) doesn't seem worthwhile.

Why can't we replace all uses of CachedResourceHandle with RefPtrs, or go 100% CachedResourceHandle with no raw pointers?

Wouldn't either of those be better than "confusing"?

(full disclosure: have not looked at the patch, have just been paying attention to the worrisome comments)

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