Bug 42098 - The WebCore::Document::head() method should be const.
Summary: The WebCore::Document::head() method should be const.
Status: RESOLVED WONTFIX
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebCore Misc. (show other bugs)
Version: 528+ (Nightly build)
Hardware: PC OS X 10.5
: P2 Normal
Assignee: Nobody
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-07-12 11:50 PDT by Jay Civelli
Modified: 2010-07-12 16:39 PDT (History)
0 users

See Also:

Attachments
Initial patch (1.50 KB, patch)
2010-07-12 11:52 PDT, Jay Civelli 		no flags Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Jay Civelli 2010-07-12 11:50:53 PDT 
The WebCore::Document::head() method is not const, it should.
Comment 1 Jay Civelli 2010-07-12 11:52:29 PDT 
Created attachment 61251 [details]
Initial patch
Comment 2 Darin Adler 2010-07-12 12:38:23 PDT 
Comment on attachment 61251 [details]
Initial patch

Why should this be const? You have a const pointer to a document and you get a non-const pointer to one of the elements of that document? I think we should not use const so much on classes like Document* that are part of a tree. It doesn't make sense to have a const pointer to something but then call firstChild()->parent() and get back a non-const pointer to the same thing. I suggest going the opposite direction and phasing out use of const Document*.
Comment 3 Jay Civelli 2010-07-12 13:36:56 PDT 
I came up across this using the Chromium WebKit API (trying to pass a WebDocument as a const& and then call head() on it which ends up calling Document::head()) and thought the const had somehow been forgotten since body() was const.
You are right, it does not make sense if the constness can be circumvented.
Closing the bug.