The button was stretching vertically as the console input grew,
which looked weird.
This is fixed by aligning the item of the container differently
and adjusting padding and margins in order to properly aligh
the context selector and the editor switch button.
We take this as an opportunity to set a max width on the button,
since it could be quite large if the context name is long.
Differential Revision: https://phabricator.services.mozilla.com/D67860
--HG--
extra : moz-landing-system : lando
Xray waivers for remote window proxies don't make much sense, so we nuke any
waiver when a window proxy navigates from local to remote. However, this xray
waiver can be the target of a CCW, and dead wrappers are not supposed to the
target of a CCW, so we need to find and nuke any CCWs targetting the waiver.
This requires fixing js::RemapWrapper so that it can be used to fix CCWs to
dead wrappers.
Hopefully this will fix a bunch of intermittent failures in
browser_windowProxy_transplant.js.
Differential Revision: https://phabricator.services.mozilla.com/D66459
--HG--
extra : moz-landing-system : lando
I suspect the number of texture array layers impacts some arithmetic realted to how the GL implementation deals with texture coordinates. These refetests started failing more consistently (though not 100% of the time) with the patch that sets the number of rgba8 linear texture array laters to 16.
Differential Revision: https://phabricator.services.mozilla.com/D67863
--HG--
extra : moz-landing-system : lando
After bug 1582832, DocShell destruction and BrowsingContext detaching happen
in separate operations, leaving a gap where a DocShell has been destroyed, but
its BrowsingContext is still considered attached. During this gap, the usual
invariant that an in-process, attached BrowsingContext always has an
associated DOM window doesn't hold, nor do the usual invariants for outer
window forwarding security checks.
This patch fixes the detach timing so that a child BrowsingContext for a frame
which has been removed is always marked detached at the same time its DocShell
is destroyed.
Co-authored-by: Kris Maglione <maglione.k@gmail.com>
Differential Revision: https://phabricator.services.mozilla.com/D62791
--HG--
extra : moz-landing-system : lando
This test was wrong. The event timings were fire before, when performing the request.
Not when opening the context menu. So, with the previous changeset, this test started to fail.
Differential Revision: https://phabricator.services.mozilla.com/D66659
--HG--
extra : moz-landing-system : lando
We were explicitely waiting for event timings in two tests only,
but these requests are also done in most other tests using waitForNetworkEvents.
Unfortunately, EVENT_TIMINGS aren't always fetched, so it complexify
waitForNetworkEvents a bit. But I think it would prevent pending requests
and helps using a unique wait helper.
Differential Revision: https://phabricator.services.mozilla.com/D66658
--HG--
extra : moz-landing-system : lando
Other tests are typically using waitForNetworkEvents when doing a reload,
with a clear assertion on how many requests should be displayed and awaited for.
Differential Revision: https://phabricator.services.mozilla.com/D66656
--HG--
extra : moz-landing-system : lando
There appears to be a bug in babel-eslint where it still uses the old
`ExperimentalSpreadProperty` and `ExperimentalRestProperty` node types when
babel itself (and hence the AST visitor keys) have switched to `SpreadElement`
and `RestElement`. I suspect there isn't much point in filing a babel-eslint
issue for this, they are currently in the process of working on v11 which does
fix the issue but is currently only available as a beta release.
There are three alternatives:
1. Use the beta babel-eslint.
2. Hack the visitor keys to include the old node types.
3. Traverse the AST changing the nodes from the old types to the new types.
Using a beta version is probably not a great idea so this patch does 2 which is
the simplest of the other options. It does have the drawback however that if
any eslint rules we use rely on the new node type names they will fail to work.
Differential Revision: https://phabricator.services.mozilla.com/D66998
--HG--
extra : moz-landing-system : lando
In the case where there is no overflow, the lhsreg and the destreg can be the same, so the rhs constant needs its own register.
Differential Revision: https://phabricator.services.mozilla.com/D67742
--HG--
extra : moz-landing-system : lando
This patch adds the basic plumbing for vulnerable logins. Correct icons and strings will be fixed in future work. As such, this feature is still disabled.
Differential Revision: https://phabricator.services.mozilla.com/D67557
--HG--
extra : moz-landing-system : lando
CLOSED TREE
Backed out changeset 622591504941 (bug 1616901)
Backed out changeset 034d885d32cb (bug 1623777)
Backed out changeset 610fe9c88f2e (bug 1623647)
Razvan Caliman
Nicolas Chevobbe
Florin Strugariu
Raphael Pierzina
Andrew McCreight
Ryan VanderMeulen
Nicolas Silva
ajitesh
Nika Layzell
aarushivij
undef1nd
Alexandre Poirot
Dave Townsend
Julian Descottes
ffxbld
Iain Ireland
Jared Wein
Sylvestre Ledru
Christian Holler
Mirko Brodesser
Marco Bonardo
Samarjeet
Mihai Alexandru Michis
Adam Hammad
Christoph Kerschbaumer
anjali1903
Jan de Mooij
Martin Stransky
Edgar Chen
Logan Smyth
Simon Giesecke
jbone
Andrea Marchesini
Dorel Luca