Bug 139089 - Reloading a web view loaded with HTML data loads an empty page
Summary: Reloading a web view loaded with HTML data loads an empty page
Status: NEW
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKit2 (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords: Gtk
Depends on:
Blocks:
 
Reported: 2014-11-28 05:38 PST by Carlos Garcia Campos
Modified: 2014-12-11 10:40 PST (History)
3 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Carlos Garcia Campos 2014-11-28 05:38:42 PST
Noticed while writing a unit test for GTK+ port, after reload only load started is emitted. I'm still not sure if it's a GTK+ specific issue or not, I'll write a unit tests to confirm it.
Comment 1 Carlos Garcia Campos 2014-12-06 07:49:16 PST
Ok, so the problem is no that the finished callback is not emitted, but that the resource load callbacks are not emitted because reloading HTML data always loads a blank page in DocumentLoader::maybeLoadEmpty(). Is this known or expected?
Comment 2 Alexey Proskuryakov 2014-12-11 09:31:32 PST
As far as I know, we do not remember the source of substitute data. Reloading a page loaded from a string doesn't make sense, but reloading a WebArchive or something loaded from appcache certainly does, and that presumably works.

loadHTMLString clients that I know of don't have reload functionality in the first place. What is the scenario where this is needed?
Comment 3 Carlos Garcia Campos 2014-12-11 10:40:39 PST
(In reply to comment #2)
> As far as I know, we do not remember the source of substitute data.
> Reloading a page loaded from a string doesn't make sense, but reloading a
> WebArchive or something loaded from appcache certainly does, and that
> presumably works.
> 
> loadHTMLString clients that I know of don't have reload functionality in the
> first place. What is the scenario where this is needed?

I was writing a unit test where loadHTML is very convenient, but I could use loadURL when reload is required. I was not sure if we really wanted it or not, that's why I opened the bug but didn't try to fix it :-)