This commit also fixes a race in `test_error_reporting.html`, where the
push service would initialize and attach its listeners after
`sessionstore-windows-restored`. Even though the test replaces the real
service with a mock, the former keeps listening for pref changes.
When the test calls `setupPrefs` to enable the push connection, the
real service tries to connect to the push server, which asserts in
automation because non-local connections aren't allowed.
We work around this by ensuring that `replacePushService` and
`restorePushService` always wait for the service to shut down before
replacing it with a mock, or restoring the real implementation.
Finally, this commit removes a test that's no longer relevant, since
we don't need to fetch all subscriptions at startup.
Differential Revision: https://phabricator.services.mozilla.com/D20085
--HG--
extra : moz-landing-system : lando
If configured to show the profile manager on startup we want to defer doing that
until after attempting to remote to an existing Firefox. So in the default
startup case (not first run, not when a profile is selected on the command line
and not when the profile manager is requested on the command line) let default
profile selection complete, check for an existing Firefox and only then check
that we're configure to show the profile manager on startup and show it then.
Differential Revision: https://phabricator.services.mozilla.com/D23410
--HG--
extra : moz-landing-system : lando
When the dns resolution for an url's hostname fails, the webRequest API seems to be
receiving (from time to time but more often on macos and windows) more then one call
to the HttpObserverManager.observeActivity method for the activityType
nsISocketTransport.STATUS_RESOLVING and nsISocketTransport.STATUS_RESOLVED
while channel wrapper errorString property is still null.
The observeActvity method goal seems to fire fallback WebRequest errors for error conditions
that are not fired as ChannelWrapper error events, but the "NS_ERROR_UNKNOWN_HOST" error
is also being fired right after the fallback webRequest error event fired as
"NS_ERROR_NET_ON_RESOLVED" (or "NS_ERROR_NET_ON_RESOLVING").
This behavior consistently triggers the two intermittent failures on the test
test_ext_webrequest_basic.html while executing the test_webRequest_frames
task
"expected error message received in onErrorOccurred - Expected: NS_ERROR_UNKNOWN_HOST, Actual: NS_ERROR_NET_ON_RESOLVED"
(when the fallback NS_ERROR_NET_ON_RESOLVED is fired first)
and then
"received onErrorOccurred"
(because the "NS_ERROR_UNKNOWN_HOST" error is also being fired
right after the fallback error message).
Differential Revision: https://phabricator.services.mozilla.com/D22587
--HG--
extra : moz-landing-system : lando
When I added AddonTestUtils.initMochiTest to an existing test at
browser/components/preferences/in-content/tests/browser_extension_controlled.js
, the test started to fail often on debug builds, with errors like
"leaked 2 window(s) until shutdown".
Fix this by clearing the global that was saved by initMochiTest.
Differential Revision: https://phabricator.services.mozilla.com/D23502
--HG--
extra : moz-landing-system : lando
We need to collect how many Confluence instances are loaded and how percentage
of that is old versions which require split keypress event model.
Differential Revision: https://phabricator.services.mozilla.com/D17908
--HG--
extra : moz-landing-system : lando
If `signon.management.overrideURI` is set, open that URI in a new tab instead of opening passwordManager.xul in a window.
Differential Revision: https://phabricator.services.mozilla.com/D23204
--HG--
extra : moz-landing-system : lando
Also, remove a test that's no longer relevant, since we don't need to
fetch all subscriptions at startup.
Differential Revision: https://phabricator.services.mozilla.com/D20085
--HG--
extra : moz-landing-system : lando
Since we now have a method on nsFrameLoaderOwner/MozFrameLoaderOwner
that can update remoteness, we should no longer need to unbind and
rebind browser elements to the tree to change their remoteness
attributes. We can just call the method and have the Frameloaders
rebuilt in the backend.
We're still getting some test breakage in Marionette and browser
chrome with this patch. Putting this behind a pref so the fission
team can still work with it while the tests are being fixed.
Depends on D22790
Differential Revision: https://phabricator.services.mozilla.com/D22791
--HG--
extra : moz-landing-system : lando
Just set the RemoteType in the options object argument, instead of
doubling up the information.
Differential Revision: https://phabricator.services.mozilla.com/D23250
--HG--
extra : moz-landing-system : lando
Originally we stored the new information about installation defaults in
installs.ini since older versions of Firefox would throw away any new data in
profiles.ini any time they made changes to the profiles. That does however mean
we have to load two files on startup.
This changes things so that we save all the data in profiles.ini as well as a
version tag and still save the install data into installs.ini. An older version
will throw away the install data and version tag from profiles.ini but leave
installs.ini alone. On startup if the version tag is gone from profiles.ini then
we reload the install data from installs.ini and put it back into profiles.ini.
At some point in the future where we don't care about supporting older versions
of Firefox we can just drop installs.ini entirely.
A lot of the changes here involve moving to loading profiles.ini into an
in-memory ini, keeping it up to date and flushing it to disk. This means that we
no longer throw away any information in the ini file that this version does not
understand allowing the possibility of adding new data to this file in the
future.
Differential Revision: https://phabricator.services.mozilla.com/D22576
--HG--
extra : rebase_source : 9264bfebc7b0781ee903ef2772686d4cf920d1c9
extra : source : d4feb17faf013134f5eac8b5e19b714c56410973
Originally we stored the new information about installation defaults in
installs.ini since older versions of Firefox would throw away any new data in
profiles.ini any time they made changes to the profiles. That does however mean
we have to load two files on startup.
This changes things so that we save all the data in profiles.ini as well as a
version tag and still save the install data into installs.ini. An older version
will throw away the install data and version tag from profiles.ini but leave
installs.ini alone. On startup if the version tag is gone from profiles.ini then
we reload the install data from installs.ini and put it back into profiles.ini.
At some point in the future where we don't care about supporting older versions
of Firefox we can just drop installs.ini entirely.
A lot of the changes here involve moving to loading profiles.ini into an
in-memory ini, keeping it up to date and flushing it to disk. This means that we
no longer throw away any information in the ini file that this version does not
understand allowing the possibility of adding new data to this file in the
future.
Differential Revision: https://phabricator.services.mozilla.com/D22576
--HG--
extra : rebase_source : d097342b0c26fb92e5236e83035b87bb7da84321
Bug 1530771 - Replaced instances where "&" (ampersand) was used instead of "and" to join two words in the preferences menu, and added Oxford comma where appropriate in certain labels to ensure consistency. r?jaws, r?flod
Differential Revision: https://phabricator.services.mozilla.com/D21472
--HG--
extra : moz-landing-system : lando
Don't prompt to save the same username and password combination in the same document.
Differential Revision: https://phabricator.services.mozilla.com/D22039
--HG--
extra : moz-landing-system : lando
