Bug 32110 - Allow custom memory allocation control for WebCore's FrameLoaderClient
Summary: Allow custom memory allocation control for WebCore's FrameLoaderClient
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebCore Misc. (show other bugs)
Version: 528+ (Nightly build)
Hardware: Other All
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-12-03 02:30 PST by Zoltan Horvath
Modified: 2009-12-11 08:19 PST (History)
1 user (show)

See Also:


Attachments
Patch (1.40 KB, patch)
2009-12-03 02:30 PST, Zoltan Horvath
no flags Details | Formatted Diff | Diff
Patch (1.27 KB, patch)
2009-12-11 04:57 PST, Zoltan Horvath
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Zoltan Horvath 2009-12-03 02:30:31 PST
Created attachment 44221 [details]
Patch

Inherits the following class from Noncopyable because it is
instantiated by 'new' and no need to be copyable:

class name              - instantiated at: WebCore/'location'

class FrameLoaderClient - (its child class) svg/graphics/SVGImage.cpp:229
Comment 1 WebKit Review Bot 2009-12-03 02:31:08 PST
style-queue ran check-webkit-style on attachment 44221 [details] without any errors.
Comment 2 Eric Seidel (no email) 2009-12-03 12:58:19 PST
Comment on attachment 44221 [details]
Patch

Are you using a script to find these?
Comment 3 Adam Barth 2009-12-03 23:28:18 PST
Comment on attachment 44221 [details]
Patch

Are you sure this is right?  This is just an abstract interface.
Comment 4 Zoltan Horvath 2009-12-04 05:03:24 PST
(In reply to comment #2)
> (From update of attachment 44221 [details])
> Are you using a script to find these?

No, I'm using a static source analyzer tool (called Columbus, developed by our department) and check the results by 'hand'. 

(In reply to comment #3)
> (From update of attachment 44221 [details])
> Are you sure this is right?  This is just an abstract interface.

This is right, but not necessary, so okay, in the future at the case of abstract interfaces I won't go up to the superclass in the inheritance tree.
Comment 5 Zoltan Horvath 2009-12-04 05:16:39 PST
Hmm. I checked.
This is the instantiation line:

static FrameLoaderClient* dummyFrameLoaderClient =  new EmptyFrameLoaderClient;

So, I have to inherit either EmptyFrameLoaderClient or FrameLoaderClient which are both abstract interfaces. FrameLoaderClient is the super class of EmptyFrameLoaderClient, so firt FrameLoaderClient should be inherited.

(If some classes inherit from Noncopyable it also inherit from FastAllocBase, so the customization of the new/delete operator will be done.)
Comment 6 Zoltan Horvath 2009-12-11 04:57:20 PST
Created attachment 44674 [details]
Patch
Comment 7 Zoltan Horvath 2009-12-11 04:58:30 PST
Ignore my last comments. You're right! New patch updated and filed!
Comment 8 WebKit Review Bot 2009-12-11 05:00:42 PST
style-queue ran check-webkit-style on attachment 44674 [details] without any errors.
Comment 9 Zoltan Horvath 2009-12-11 08:19:43 PST
Comment on attachment 44674 [details]
Patch

Clearing flags on attachment: 44674

Committed r51990: <http://trac.webkit.org/changeset/51990>
Comment 10 Zoltan Horvath 2009-12-11 08:19:54 PST
All reviewed patches have been landed.  Closing bug.