This patch add comments to IndexedDBHelper.jsm to clarify how txn.result is set.
The comments should clear up any confusion caused by previous commit message
(fix for bug 1409641).
--HG--
extra : rebase_source : 57a6f3eb569224a9e432803ef67d80f57219f7ab
When the namespace changes (e.g. due to a tab move between windows), we
may get stale transaction requests that we need to ignore. In
WebRenderBridgeParent::RecvSetDisplayList, we would automatically send
any unsent transaction data when exiting the method, but this did not
take into account the staleness. This patch ensures we only flush the
data if we actually want it.
The transaction in question that was observed and causing crashes was
UpdateImageBuffer.
We were checking for cross-compartment wrappers in the Debugger constructor, but this patch also fixes addDebuggee and addAllGlobalsAsDebuggees.
Differential Revision: https://phabricator.services.mozilla.com/D3495
--HG--
extra : rebase_source : 3346baa677b4eae1ed8b7b13d93c1c8c89753d97
Without a session being created first the "send_command" method
inappropriately checks for the "session not created" error, which
only gets raised by the "new session" command. For all the other
commands the "invalid session id" error has to be handled.
--HG--
extra : rebase_source : de7ad6ae0ef9af925f6f36c70cb4a85c97da4b0c
If a command is used before creating a new session, an
"invalid session id" error has to be returned by the
driver.
--HG--
extra : rebase_source : c45f33e9e39d876f5fd77561b57ab1fe664452d1
The Marionette client raises a custom MarionetteException
instead of passing through the InvalidSessionIdException as
returned by the Marionette server.
--HG--
extra : rebase_source : 99fd02062e858f31e25d2ca0fb80940586772fb2
This test is basically a copy of its -001 variant, with some "float:left"
sprinkled around on contained descendants.
Before this patch, this test had an additional arbitrary sizing difference as
compared to the -001 version -- there's one element that arbitrarily has class
"outer" in the -002 test whereas it has class "inner-lg" in the -001 version.
These classes have different sizing characteristics, which makes a difference
to whether scrollbars show up, because this element is not contained (though it
is a layout container itself).
This patch undoes this arbitrary difference and also adds a "float" class to
make it easier to see which elements we've sprinkled float styling onto.
Differential Revision: https://phabricator.services.mozilla.com/D3826
--HG--
extra : rebase_source : 2630420b4dfc1a307cbfe20447e66fbfecf7b0aa
HTMLEditor::RefereshEditingUI() works only with enabled UIs. Therefore, if
UI is disabled while it's visible, it keeps shown. This is too bad if web
apps tries to disable the Gecko specific UIs after we show some of them.
This patch adds HTMLEditor::HideAnonymousEditingUIsIfUnnecessary() to hide
unnecessary UIs and makes RefereshEditingUI() call it always.
First, HTMLEditor::HideInlineTableEditingUI() always returns NS_OK. So, we
can change its return type to void.
Additionally, it removes each UI from the DOM tree one by one. However, each
mutation could cause showing same UI again. In such case,
ShowInlineTableEditingUI() overwrites each UI with newly created element.
Then, HTMLEditor cannot remove the old UI anymore. Therefore, this patch
moves all members of the UI into local variables first.
HTMLEditor::CheckSelectionStateForAnonymousButtons() is called a lot internally.
Especially, its virtual call cost may make damage to our performance since
it's called from a selection listener.
So, we should create non-virtual method, RefereshEditingUI() for internal use.
Add a frontend for the Oreo auto-fill API in SessionTextInput, which
processes events from Gecko and provides consumer APIs that match the
Oreo auto-fill APIs. GeckoView then forwards the necessary calls to
SessionTextInput.
Differential Revision: https://phabricator.services.mozilla.com/D3538
Add an auto-fill frontend that listens to events from Gecko. It
populates accessibility nodes for input fields and sends accessibility
events, in order to support auto-fill clients that use accessibility
services to perform auto-fill.
Differential Revision: https://phabricator.services.mozilla.com/D3254
Add an auto-fill backend in GeckoViewContent.js that detects fields on
the page and sends information about the fields to Java through events.
Differential Revision: https://phabricator.services.mozilla.com/D3253
Make the session store event listeners inline, because it makes the code
more readable, and also because auto-fill requires a pageshow listener
that is always registered, so the existing pageshow listener needs to be
moved elsewhere.
Differential Revision: https://phabricator.services.mozilla.com/D3252
Make a distinction between a session being active (i.e. being visible)
and it being focused. More than one session may be active at a time, but
only one session is focused at a time. This means the focused session is
always active, but an active session may not be focused.
Also, manage setting of active/focused states in GeckoView itself, so
consumers don't generally have to worry about these states.
Differential Revision: https://phabricator.services.mozilla.com/D3251
Move the AccessibilityNodeProvider implementation under
SessionAccessibility, to reduce the indent of the code.
Also make all methods in SessionAccessibility.Settings static to make
the code easier to follow.
Differential Revision: https://phabricator.services.mozilla.com/D3250
As bug 1467344 comment 40, it appears we don't currently have any tests that
cover resuming a paused animation.
Differential Revision: https://phabricator.services.mozilla.com/D3752
--HG--
extra : rebase_source : 7fef846a6e1aafa41923df2b29122e04fbfc0c0c
