[Webkit-unassigned] [Bug 206217] Substitute data lost after reload

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=206217

--- Comment #16 from Michael Catanzaro <mcatanzaro at gnome.org> ---
(In reply to Carlos Garcia Campos from comment #15)
> I don't mind, TBH, I was just trying to fix this bug. Current behavior
> doesn't look right for sure,

It doesn't look right when looking at this example, but it's easy to see why we can't change it.

> but maybe we can just ignore reloads instead of
> loading the base URL if we don't want to reload the substituted data.

Say the user hits bug #203620, or a random network failure, or other spurious error, and winds up viewing a network error page. Do you want refresh to be ignored, or to reload the network error page? Of course not: you want it to try reloading the original page. Displaying error pages is even the documented use-case of webkit_web_view_load_alternate_html(). What we need here is better documentation to clarify that the alternate data will not be used when reloaded and that the application is responsible for handling that if required. Or, if this is really important, we could add new API to let the API user choose, but I doubt it's that important.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20200127/244d74ef/attachment-0001.htm>