[webkit-efl] Do we really need a global history implementation?

[Raphael Kubo da Costa]

Hey there.

Mikhail has been working on a few patches [1][2][3][4] adding support
for a global history to the port.

Most of the code in those patches looks fine, but before commenting on
the bugs I would like to discuss whether we actually need a global
history implementation at all.

Only the Apple ports (mac and win) really use the global history
callbacks in WebCoreSupport (Qt implements some bits just to pass the
layout tests [5]).

For us, it means we need to add a few more data structures and emit a
few very specialized signals I doubt users are going to need (such as
"perform,client,redirect" or "navigate,with,data"). The only benefit I
see so far is that a few more tests will be unskipped, but then if we
choose we do not want to support this feature we can just move the
skipped tests to the "Tests which are expected to fail" section.

Does anyone have a different opinion or has anyone needed a global
history implementation when using the port?

[1] https://bugs.webkit.org/show_bug.cgi?id=86343
[2] https://bugs.webkit.org/show_bug.cgi?id=82579
[3] https://bugs.webkit.org/show_bug.cgi?id=86370
[4] https://bugs.webkit.org/show_bug.cgi?id=82724
[5] http://trac.webkit.org/browser/trunk/Source/WebKit/qt/WebCoreSupport/FrameLoaderClientQt.cpp#L802