Actually, GeckoView registers both desktop and mobile actor for `<select>`
element. It is unnecessary to use desktop's `<select>` actor.
But, if GeckoView doesn't register desktop's `<select>` actor,
`layout/forms/test/test_select_reframe.html` will be failure since GV doesn't
handle `mozhidedropdown` that is a event when the control is unbinded from
layout tree.
So we need to handle this for this situation that is one of dismissed cases.
Differential Revision: https://phabricator.services.mozilla.com/D149526
Implements showing a "Paste" popup at the last mouse position (which
overlapped the browser window) when a certain custom event is received.
Creating and dispatchting that event is implemented in one of the
following parts.
Differential Revision: https://phabricator.services.mozilla.com/D135333
Use the correct function name (delMacXAttr vs removeMacXAttr) to remove the com.apple.quarantine extended attribute from downloaded plugin files.
Add test code to ensure extracted plugin files do not have the quarantine attribute.
Differential Revision: https://phabricator.services.mozilla.com/D148739
Before this patch, the content signature verifier
(nsIContentSignatureVerifier/ContentSignatureVerifier) would identify the root
it trusted based on the value of a preference. This patch changes the
implementation to require a specified hard-coded root to trust as with add-on
signature verification.
Depends on D146644
Differential Revision: https://phabricator.services.mozilla.com/D146645
This is a breaking change. It is not possible to control the loading of dumps from a preference anymore.
Dumps are only loaded if server is pointed at PROD. The recommended way to prevent dumps loading in tests is to store a high timestamp in the local DB.
Differential Revision: https://phabricator.services.mozilla.com/D146049
This introduces a breaking change: the buckets cannot be changed via preferences anymore.
Before landing this patch, we should have a released a new version of the Remote Settings DevTools that is compatible with this new API.
Differential Revision: https://phabricator.services.mozilla.com/D145455
These test were intended to use the planned, but dropped, one thread
per TabGroup feature. They used to dispatch timer callbacks to the
system group to test the labelling needed, but since TabGroups were
removed and the event target was changed to the main thread event
target, these tests actually test the same thing as the regular
non-target tests. This because nsITimer is created with a target which
just happens to be the mainThreadEventTarget.
Differential Revision: https://phabricator.services.mozilla.com/D146532
Currently if the pref for using content sig checks with GMP updates is missing
we will use cert pinning. I.e. the default behaviour in the absence of a pref
value is cert pinning. This changes that so that we use content sig by default.
This is done as we're moving to content signatures as default behaviour, so it
makes sense to have it be the default if the pref is missing.
Differential Revision: https://phabricator.services.mozilla.com/D146260
We have a number of users that check for updates but don't seem to
actually update. This should help narrow down what's going wrong.
Differential Revision: https://phabricator.services.mozilla.com/D146216
We have a number of users that check for updates but don't seem to
actually update. This should help narrow down what's going wrong.
Differential Revision: https://phabricator.services.mozilla.com/D146216
This modifies `ActivityStreamProvider.getTopFrecentSites()` so it excludes URLs
with a given search param. There are two questions I considered when writing
this patch:
* The top-sites code is a little complex and there are multiple places where
this logic could be implemented. I chose the bottommost layer, where the top
sites are fetched from Places and some other exclusion logic already exists,
because we don't want these URLs to appear in any list of top sites in Firefox
AFAIK -- not on the new-tab page and not in the urlbar.
* How should we encode this new exclusion logic? We don't want to hardcode a
specific search param. We may not even want to base it on search params at all
but instead make it more general somehow. For now, I did the simple thing and
went ahead and based it on a search param, and I added a new option to
`getTopFrecentSites()` that specifies the param and that falls back to a new
pref. Callers can override the pref fallback by passing in an option.
The pref value and option to `getTopFrecentSites()` supports the following
forms:
* `""` (empty) - Disable this feature
* `"key"` - Search param named "key" with any or no value
* `"key="` - Search param named "key" with no value
* `"key=value"` - Search param named "key" with value "value"
Differential Revision: https://phabricator.services.mozilla.com/D146139
Change tests and snippets in documentation to use `Date.now()` instead of an arbitrary number like `42`.
This way, we make sure the packaged dump isn't loaded on top of the tests data. Indeed, since Bug 1718083 we load the packaged dump if it's newer than local data.
Differential Revision: https://phabricator.services.mozilla.com/D145579
See the comment in the file explaining it. For a case of logging 100k numbers,
this dropped the time per number from 15 microseconds to 9 with the console
closed, and 55 microseconds to 38 with the console open. I think we could shave
off more with a native approach, but I don't know that it's worth it and it's
much more likely for that to introduce bugs.
Differential Revision: https://phabricator.services.mozilla.com/D143782
