Bug 100240 – [EFL] Fix fast/history/window-open.html case.

• Status: RESOLVED
• Resolution: WONTFIX
• Component: WebKit EFL

[Kangil Han]

To pass this case, use WebCore::Page::getHistoryLength() API instead of navigation stuff.

[Kangil Han]

Created attachment 170376 [details]
patch

[Kenneth Rohde Christiansen]

Comment on attachment 170376 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=170376&action=review

> Tools/ChangeLog:10
> +        To pass window-open case, use history length API in Page class instead of calculating navigation history.
> +        This is because test case checks if newly opened window has proper web history as '1'.
> +        Previously, implementation added back/forward navigation histories but this is not matched with what test case asks about.

why does it have that? can you go back?

[Kangil Han]

(In reply to comment #2)
> 
> why does it have that? can you go back?

Nope.
window.open() works in another view so we can't go back in this case.

[Mikhail Pozdnyakov]

Comment on attachment 170376 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=170376&action=review

> Tools/DumpRenderTree/efl/TestRunnerEfl.cpp:122
> +    return DumpRenderTreeSupportEfl::webHistoryItemCount(browser->mainView());

At this point it might be OK for us, but as far as I understand item count of shared history should be returned rather than item count of back-forward list of a concrete page..

[Kangil Han]

As discussed on irc with MPozdnyakov, shared history between two different windows is out of scope at this stage. Therefore, I want to keep latest patch as a solution to fix window-open case. :)

[Kenneth Rohde Christiansen]

Comment on attachment 170376 [details]
patch

Then let's do it correctly and add some tests which tests this, comments?

[Kangil Han]

(In reply to comment #6)
> Then let's do it correctly and add some tests which tests this, comments?

IMHO, another bug would be better If we really want shared history between multiple windows.

This is because I don't think 'REGRESSION (r40508): Window opened with window.open has no initial history item (global and back/forward)' phrase in test case would mean globally shared history.

Seems there hasn't been that kind of implementation in WebCore, so nothing can be regressed. :)

[Kangil Han]

@mitz : Would you kindly tell me if I need to implement shared web history to pass window-open.html case?

[Gyuyoung Kim]

Is this bug still valid ?

[Andreas Kling]

Comment on attachment 170376 [details]
patch

Clearing review flag on patches from before 2014. If this patch is still relevant, please reset the r? flag.

[Michael Catanzaro]

Closing this bug because the EFL port has been removed from trunk.

If you feel this bug applies to a different upstream WebKit port and was closed in error, please either update the title and reopen the bug, or leave a comment to request this.