Since I have added Windows backend for lock orientation by Part 1. fix, I
would like to use it in `screen.orientation.lock()`.
Since `dom.screenorientation.allow-lock` is still false on all platfroms even
if this fix is applied, orientation lock isn't turned on as default.
If lock orientation isn't supported in the device, the backend will throw
`NS_ERROR_DOM_NOT_SUPPORTED_ERR`.
Depends on D162451
Differential Revision: https://phabricator.services.mozilla.com/D162452
Since Windows tablet mode has a orientation lock API, this patch
implements orientation lock backend for Windows tablet mode.
`GetAutoRotationState` API recognizes whether orientation API is supported on
the device. So this fix uses this API to check orientation API capability.
Differential Revision: https://phabricator.services.mozilla.com/D162451
WorkerDispatcher.invoke will instantiate a new "Task" method on each call.
Ideally, we should only use WorkerDispatch.task(), unless the related method is called once.
Differential Revision: https://phabricator.services.mozilla.com/D162564
From what I can tell switching to `withFirefoxView` and opening a new window to run these tests in + closing that window before disabling the theme messed with the timing of these tests in such a way that we started getting that vsync error shown in the logs. I've switched to running all these tests in the same window, which looks good from a couple of try pushes:
https://treeherder.mozilla.org/jobs?repo=try&revision=9a5f379ecc48bda08eab67210278b547f558f856https://treeherder.mozilla.org/jobs?repo=try&revision=76b909924455fda54165ffeaa00b3654b4decd09
I'll file a follow up bug to change the default behaviour of `withFirefoxView` is to run tests in the same window, as this has come up before.
Also worth noting that even if this patch lands and fixes the new frequent intermittent we will likely still see the less frequent intermittents that happened prior to the change for Bug 1780441.
Differential Revision: https://phabricator.services.mozilla.com/D162617
This is an RAII class that does nothing in opt builds. Some compilers warn
about its use but that can be fixed by adding an empty destructor.
Differential Revision: https://phabricator.services.mozilla.com/D162984
Just add the minimal stuff in order to be able to view a pdf in using pdf.js
(whatever the content disposition is).
There are still some things to plug (e.g. find stuff), we must change the UI
and add support for pinch-to-zoom, hence we set the pref pdfjs.disabled to true.
Differential Revision: https://phabricator.services.mozilla.com/D162122
Ideally we would use the window mediator to just find new browser windows that
are in the process of opening but while we can find the windows they just appear
as about:blank with no way to verify that they are browser windows.
This just takes the straightforward approach of forcing code that opens browser
windows to register them with the BrowserWindowTracker and provides a simple
shared API for opening browser windows that does this.
Differential Revision: https://phabricator.services.mozilla.com/D161076
When one Firefox instance finishes starting up it releases the startup lock and
deletes the lock directory. Other instances that are waiting to startup are in
the meantime polling to acquire the lock but with the directory gone their next
attempt just throws a failure and so they assume they don't have the lock yet
and keep polling until they hit the five second timeout.
With this patch before each attempt to acquire the lock we make sure that the
lock directory exists first.
Differential Revision: https://phabricator.services.mozilla.com/D161075
For some reason, in most cases, there is no redirection, but once in a
while, there is, and that breaks the download.
At the same time, make the download messages tell the url that is being
downloaded rather than where it is downloaded to.
Differential Revision: https://phabricator.services.mozilla.com/D162952
