Bug 46805

Summary:

[chromium] notify the WebFrameClient when a new WebFrame was created

Product:

WebKit

Reporter:

jochen

Component:

New Bugs

Assignee:

jochen

Status:

RESOLVED WONTFIX

Severity:

Normal

CC:

dglazkov, fishd, jamesr, jhawkins, webkit.review.bot

Priority:

Version:

528+ (Nightly build)

Hardware:

Other

OS:

OS X 10.5

Attachments:

Description	Flags
Patch	none
Patch	none

jochen

Reported 2010-09-29 07:02:20 PDT

[chromium] notify the WebFrameClient when a new WebFrame was created

Attachments
Patch (2.74 KB, patch) 2010-09-29 07:02 PDT, jochen	no flags	Details Formatted Diff Diff
Patch (2.96 KB, patch) 2010-09-29 08:15 PDT, jochen	no flags	Details Formatted Diff Diff
Show Obsolete (2) View All Add attachment proposed patch, testcase, etc.

jochen

Comment 1 2010-09-29 07:02:51 PDT

Created attachment 69189 [details] Patch

jochen

Comment 2 2010-09-29 07:04:56 PDT

by adding this method to webframe client, it is possible to keep a map from webframes to ids in the renderer. I'll use this id to augment the navigation ipc calls, so the browser has a way of knowing which (sub-)frame of a page is navigating.

WebKit Review Bot

Comment 3 2010-09-29 07:56:16 PDT

Attachment 69189 [details] did not build on chromium: Build output: http://queues.webkit.org/results/4202010

jochen

Comment 4 2010-09-29 08:15:43 PDT

Created attachment 69199 [details] Patch

Darin Fisher (:fishd, Google)

Comment 5 2010-09-29 09:26:06 PDT

Keeping a map containing WebFrames is quite problematic. We've tried to do so for autofill, and it has led to some very difficult to track down crashes. The final resolution to that problem was to stop keeping WebFrame pointers. Instead, we keep WebNode objects. I'd like to better understand how you intend to use this new WebFrameClient method in case there is a way that we can avoid storing WebFrame pointers in an external map.

jochen

Comment 6 2010-09-29 09:34:37 PDT

In the end, I want to tell apart whether two subframe navigations happening on the same page happened in the same iframe.

jochen

Comment 7 2010-09-29 09:56:55 PDT

So I'd map the pointers to integers. I thought that might be cleaner than using the pointer directly as id.

Eric Seidel (no email)

Comment 8 2010-11-07 23:35:37 PST

Comment on attachment 69199 [details] Patch Cleared review? from attachment 69199 [details] so that this bug does not appear in http://webkit.org/pending-review. If you would like this patch reviewed, please attach it to a new bug (or re-open this bug before marking it for review again).

Note You need to log in before you can comment on or make changes to this bug.