***
Bug 1523957 - Part 0: Whitespace changes in tree.js, r=bgrins
***
Bug 1523957 - Part 1.1: Naively convert tree to a custom element, r=bgrins
***
Bug 1523957 - Part 1.2: Properly mixin BaseControl and extend XULTreeElement, r=bgrins
***
Bug 1523957 - Part 1.3: Use Shadow DOM for slotting children, r=bgrins
***
Bug 1523957 - Part 1.4: Properly handle key events in combination with shift modifiers, r=bgrins
***
Bug 1523957 - Part 1.5: Lint for the lint gods in tree.js, r=bgrins, standard8
***
Bug 1523957 - Part 2.1: Naively convert places-tree to a custom element, preserving history, r=bgrins
***
Bug 1523957 - Part 2.2: Lint for the lint gods in places-tree.js, r=bgrins
***
Bug 1523957 - Part 2.3: Fix browser_passwordmgr_editing.js, r=MattN
***
Bug 1523957 - Part 2.4: Fix browser_bookmarksProperties.js, r=bgrins
***
Bug 1523957 - Part 2.5: Make the edit bookmark tree lazily constructed so there aren't any trees left in the browser window markup, r=bgrins
***
Bug 1523957 - Part 2.6: Add eslint globals to places-tree.js, r=bgrins, standard8
This test was intermittently timing out, mostly on debug builds, because there
was a chance that from time to time the sidebar could have been loaded for the
first time while the background page of the test extension was still being loaded.
The test was sending an extension test message as soon as the sidebar was loaded
for the first time, and so if the background page didn't yet added its test message
listener, the test cases did never get started and the entire test was getting stuck.
The changes included in this patch prevents this intermittent failures by ensuring
that the test always waits for the background page and the sidebar to be both loaded
and ready before sending the test message that starts to run the test cases.
Differential Revision: https://phabricator.services.mozilla.com/D20657
This patch adds a test for ensuring the letterboxing works as we expect.
It will open a tab and resize its window into several different sizes
and to see if the margins are correctly apply. And it will also check
that no margin should apply to a tab with chrome privilege.
This patch implements the window letterboxing. The implementation
is based on adding margins around the browser element to round the
content viewport size. Whenever the browser content is resized, the
RFPHelper will adjust margins around it. But it won't add any margins
for an empty browser or a browser loads a content with the system
principal.
The letterboxing is hidden behind a hidden pref
"privacy.resistFingerprinting.letterboxing." By default, it will use
stepping size 200x100 to round content window. And we can customize
the set of dimensions used for deciding the size of the rounded
content viewport by the pref
"privacy.resistFingerprinting.letterboxing.dimensions". This pref
should be formated as 'width1xheight1, width2xheight2, ...'. We will
find the dimensions which can fit into the real content size and have
the smallest margins to be the rounded content viewport size. For example
, given the set "400x200, 500x300, 800x500" and the real content size
"600x300", we would round the content size into 500x300.
Current window state in the sessionstore system includes `sizeMode` which can be "normal", "minimized", "maximized". However, the OS also remembers whether the window was "normal" or "maximized" before minimization to restore it appropriately. With this fix, sessionstore does likewise.
Differential Revision: https://phabricator.services.mozilla.com/D13234
Include a boolean reporting whether these prefs are enabled on startup. inlcude extra information when user interacts with the control panel reporting if anything is being blocked.
Differential Revision: https://phabricator.services.mozilla.com/D19843
Include a boolean reporting whether these prefs are enabled on startup. inlcude extra information when user interacts with the control panel reporting if anything is being blocked.
Differential Revision: https://phabricator.services.mozilla.com/D19843
Replaced new Function() by CustomEvent carrying original click event as sourceEvent.
Adapted all oncommand listeners to use event.sourceEvent instead of event.
Differential Revision: https://phabricator.services.mozilla.com/D18847
