SSIA, tested with both Epiphany with 1.11.92 WebKit2GTK+ and GtkLauncher master.
Since webkit handles redirection itself now, this is probably a ResourceHandleSoup problem, not a libsoup problem. Assuming it's related to HTTP redirection at all, rather than HTML/JS-based reloading.
Not a Soup or WebKitGtk+ bug per se as I can reproduce it with Chromium as well. There is something wrong in that page anyway because FF for example redirects to an "Error - Page not found"
Fun! I thought it was specific to soup because I could not reproduce with Chrome.
Is there anyone to follow this issue ?
If not, can I pick this one ?
As I checked the CallStack, this problem is happened by touch event in WebCore.
But IOS will be OK, because they've implemented with different way.
(In reply to comment #4)
> Is there anyone to follow this issue ?
> If not, can I pick this one ?
Sure, go ahead.
Created attachment 229200 [details]
Simple page to reproduce it easily.
Created attachment 229201 [details]
Can you explain how page reloading related to touch events?
Sorry, I don't catch your point because it's not related to reloading.
I think you can reproduce it easily on some browser based on WebKit except IOS as I attached a sample page,
It's very simple.
document.addEventListener(name, handler, true);
var event = document.createEvent('TouchEvent'); // <-- create TouchEvent
event.initTouchEvent(name); // <-- there is not any attribute except event name
document.dispatchEvent(event); // <-- but try to dispatch touch event
According to touch-event spec,
URL : www.w3.org/TR/touch-events/
changedTouches of type TouchList, readonly
a list of Touches for every point of contact which contributed to the event.
For the touchstart event this must be a list of the touch points that just became active with the current event. For the touchmove event this must be a list of the touch points that have moved since the last event. For the touchend and touchcancel events this must be a list of the touch points that have just been removed from the surface.
targetTouches of type TouchList, readonly
a list of Touches for every point of contact that is touching the surface and started on the element that is the target of the current event.
touches of type TouchList, readonly
a list of Touches for every point of contact currently touching the surface.
I thought these attributes should be necessarily required in TouchEvent
Perhaps you uploaded the patch to the wrong bug?
(In reply to comment #9)
> Sorry, I don't catch your point because it's not related to reloading.
I'm a bit confused because the bug is "http://www.proprietariosdobrasil.org.br/ constantly reloads"
(In reply to comment #11)
> I'm a bit confused because the bug is "http://www.proprietariosdobrasil.org.br/ constantly reloads"
which of browser do you reproduce it ?
I think crash problem is effect to constantly reload. isn't it ?
Comment on attachment 229201 [details]
Can you make a test for this?
Created attachment 230081 [details]
(In reply to comment #13)
> (From update of attachment 229201 [details])
> Can you make a test for this?
I've attached a test cast. Thank you for your feedback.
Comment on attachment 230081 [details]
View in context: https://bugs.webkit.org/attachment.cgi?id=230081&action=review
> + // TouchEvent should have own attributes of touches, targetTouches and changedTouches before dispatch.
This comment is redundant with the code.
Created attachment 230117 [details]
Patch for landing
Comment on attachment 230117 [details]
Patch for landing
Clearing flags on attachment: 230117
Committed r167805: <http://trac.webkit.org/changeset/167805>
All reviewed patches have been landed. Closing bug.