The other cases when ClearWindowProxy is called seem to be fine.
It is the Unlink case which was causing null .contentWindow with test_mozfiledataurl.html
Fixes bug 1580391 and backs out bug 1581004
Differential Revision: https://phabricator.services.mozilla.com/D48878
--HG--
extra : moz-landing-system : lando
All but browser_bug744745.js seem to pass even without the fixes I
made, which seems odd.
browser_bug1058164.js is a little odd because it passes in {} instead
of a boolean for the useCapture argument. I think this ends up calling
addEventListener(..., {}, false), which should be the equivalent of
addEventListener(..., {}).
Differential Revision: https://phabricator.services.mozilla.com/D49453
--HG--
extra : moz-landing-system : lando
Most of these tests have been disabled for a long time; they run well
in the current test environment.
With the additional tests running, task times increase; I have added one
more test chunk for android mochitest-plain.
These tests were identified from a random sampling of mochitest manifests;
I intend to enable more mochitests in future patches.
Differential Revision: https://phabricator.services.mozilla.com/D48912
--HG--
extra : moz-landing-system : lando
This flips the direction in which the BrowserBridge actor is generally created
such that it is generally created in the parent and sent down to a child
process.
This is done by making the decision about what kind of switch to perform in the
parent, and sending messages down to child processes async to orchestrate these
process changes.
Process launching is changed to use an async `MozPromise`-returning API in this
patch, though the actual process launching still occurs synchronously. A future
patch will enable performing async process launching through the
NewOrUsedBrowserProcess mechanism.
I know of at least a few timing issues which exist with the new logic,
especially around the state of the BrowsingContext during the process
transition. I decided to not try to fix all of these issues in this patch, as
many are complex and will require changing how we manage the lifecycle of
BrowsingContext substantially. I do, however, think that the new logic is more
reliable and has fewer timing issues than the previous logic.
Differential Revision: https://phabricator.services.mozilla.com/D47310
--HG--
extra : moz-landing-system : lando
Also some minor cleanup in nsWindowWatcher, as well as a small fix,
where GetWindowByName forgot to addref its return value (as changed in
Part 1).
Differential Revision: https://phabricator.services.mozilla.com/D48976
--HG--
extra : moz-landing-system : lando
Add a new FIXUP_FLAG_PRIVATE_CONTEXT to nsIURIFixup, make it use the default
private search engine when it's set.
Update consumers to pass the new flag when necessary.
Differential Revision: https://phabricator.services.mozilla.com/D48741
--HG--
extra : moz-landing-system : lando
When we have a parser-created iframe which starts out in-process, transitions
to remote, and then transitions back to in-process, we create separate
DocShells for the first and last in-process loads. Since both are
network-created, and have the same child index, they both try to add
themselves as children to their parent's SHistory at the same index. And since
the entry for the first DocShell already exists at that index when we try to
add the second, that triggers an assertion.
This isn't really ideal, but it is expected given the current state of session
history under Fission. It should hopefully be solved more gracefully when the
Fission-aware session history rewrite is done, but in the mean time, I think
we should just ignore the conflict, since it's expected.
Differential Revision: https://phabricator.services.mozilla.com/D48437
--HG--
extra : moz-landing-system : lando
We attempt to enforce the same (approximate) access checks to Location-based
navigation that we use for loads that use named targeting (e.g., via
window.open), so that a frame that can't be navigated via, e.g., window.open,
also can't be navigated via, e.g., window.parent[1].location = url. For the
in-process case, this is handled by a somewhat hidden call to
CheckLoadingPermissions() in nsDocShell::InternalLoad, where the former checks
whether the principal of whatever JS context happens to be on the stack
subsumes the principal of the target DocShell or any of its ancestors, and
blocks the load if it doesn't.
Since there is no JS context on the stack when we call into the DocShell
loading code in the cross-process case, the check is simply ignored.
So we need to instead do the check in BrowsingContext::LoadURI, where we
already have an explicit accessor, and can simply use the standard access
checks that we use elsewhere.
Differential Revision: https://phabricator.services.mozilla.com/D48443
--HG--
extra : moz-landing-system : lando
We were keeping nsDocShell::mHistoryId and nsDocShell::mOSHE as keys. They
weren't quite good because:
1. While loading an iframe, they were being registered twice with the same
ids(for about:blank and the real URL) sometimes.
2. It wasn't possible to access to the parent mHistoryId and mOSHE from a child
processes if the parent is in a different process. That may not be the case for
now, but it will be after fission.
So we had to find other IDs to:
1. Determine the Tab of the frames.
2. Determine the URLs of the frames.
For the first use case, we were using nsDocShell::mHistoryId for that purpose
but that was wrong. The closest thing that we can get to a tab ID is
BrowsingContext ID because they don't change after a navigation. But iframes
have different BrowsingContext's, so we still need to create a tree to
construct a tab content. That can be either in the front-end or capture time.
For the second use case, we were using a key pair of mHistoryId and mOSHE. We
now chose to keep inner window IDs for that purpose. Inner window IDs are
unique for each navigation loads because inner window correspond to each JS
window global objects. That's why we can use that without any problem. But one
problem is that we cannot handle `history.pushState` and `history.replaceState`
changes with that change since window global objects won't change during those.
But that was the best thing we can do after fission. So this will be a small
sacrifice for us to keep that functionality working after fission.
In that patch we also remove the registration/unregistration calls. We are
going to add those calls in the next patch.
Differential Revision: https://phabricator.services.mozilla.com/D47065
--HG--
extra : moz-landing-system : lando
These still fail or timeout because of missing platform features, but at least
the tests will pass once those platform features are fixed after this.
Differential Revision: https://phabricator.services.mozilla.com/D48221
--HG--
extra : moz-landing-system : lando
The key change here is to use SpecialPowers.spawn to access `location.href`
and `location.reload()` for remote windows in the correct proceses. The
remaining changes are refactorings to make it easier to encorporate async
operations like `SpecialPowers.spawn` in the test logic.
Differential Revision: https://phabricator.services.mozilla.com/D48220
--HG--
extra : moz-landing-system : lando
This flips the direction in which the BrowserBridge actor is generally created
such that it is generally created in the parent and sent down to a child
process.
This is done by making the decision about what kind of switch to perform in the
parent, and sending messages down to child processes async to orchestrate these
process changes.
Process launching is changed to use an async `MozPromise`-returning API in this
patch, though the actual process launching still occurs synchronously. A future
patch will enable performing async process launching through the
NewOrUsedBrowserProcess mechanism.
I know of at least a few timing issues which exist with the new logic,
especially around the state of the BrowsingContext during the process
transition. I decided to not try to fix all of these issues in this patch, as
many are complex and will require changing how we manage the lifecycle of
BrowsingContext substantially. I do, however, think that the new logic is more
reliable and has fewer timing issues than the previous logic.
Differential Revision: https://phabricator.services.mozilla.com/D47310
--HG--
extra : moz-landing-system : lando
This test was skipped on Android for a long time, then recently enabled by
bug 1582884. Current failures are infrequent, but let's avoid them anyway.
Differential Revision: https://phabricator.services.mozilla.com/D47795
--HG--
extra : moz-landing-system : lando
This flips the direction in which the BrowserBridge actor is generally created
such that it is generally created in the parent and sent down to a child
process.
This is done by making the decision about what kind of switch to perform in the
parent, and sending messages down to child processes async to orchestrate these
process changes.
Process launching is changed to use an async `MozPromise`-returning API in this
patch, though the actual process launching still occurs synchronously. A future
patch will enable performing async process launching through the
NewOrUsedBrowserProcess mechanism.
I know of at least a few timing issues which exist with the new logic,
especially around the state of the BrowsingContext during the process
transition. I decided to not try to fix all of these issues in this patch, as
many are complex and will require changing how we manage the lifecycle of
BrowsingContext substantially. I do, however, think that the new logic is more
reliable and has fewer timing issues than the previous logic.
Differential Revision: https://phabricator.services.mozilla.com/D47310
--HG--
extra : moz-landing-system : lando
