Bug 1603127 - Replaced mozilla::Tuple with std::tuple and applied structured bindings in mozilla/Encoding.h. r=hsivonen

Differential Revision: https://phabricator.services.mozilla.com/D129920

Bug 1737865 - [devtools] Do not treat xpcshell targets as parent process targets r=ochameau,nchevobbe

Differential Revision: https://phabricator.services.mozilla.com/D130212

We represent rects in WR with the Box2D format which uses a (p0, p1) representation.
When evaluating the size of these, the subtraction p1-p0 can potentially overflow the
i32 range, so we need to guard against this. Bug 1730695 consequently allows SWGL
to see much larger transforms than previously, which makes this a possibility.

Differential Revision: https://phabricator.services.mozilla.com/D130562

While working on this screenshot marker, I noticed that the lambda initialization could directly copy or move its needed captures, instead of going through separate copies on the stack, which was probably done before C++14 lambda were available to us.

Differential Revision: https://phabricator.services.mozilla.com/D130436

There were two structs claiming to be "CompositorScreenshot", which is not allowed. It was working because the JSON output was compatible, and handled in a special way in the front-end.
But it's better to combine the two structs into one, for consistency with other markers, and lower maintainability risks.

Differential Revision: https://phabricator.services.mozilla.com/D130435

Essentially, we just want to make sure that one of our windows has focus before
we run this handler, as it will likely want to focus a specific window. We can
focus any window we want as long as *one* of our windows has been focused
already with ::SetForegroundWindow. This patch just does that.

Additionally, per cmartin's request, I added MOZ_DIAGNOSTIC_ASSERTs before some
of the early exits here, as I don't think they should be reachable? But please
do let me know if you'd prefer to keep them out or if I'm wrong about them
being reachable.

Differential Revision: https://phabricator.services.mozilla.com/D130222

Differential Revision: https://phabricator.services.mozilla.com/D130467

Bug 1739847 - disable browser/components/urlbar/tests/unit/test_unitConversion.js until DST issues get fixed to suppress permanent failures. a=Aryx

Differential Revision: https://phabricator.services.mozilla.com/D129854

It could make code more explicit.

Differential Revision: https://phabricator.services.mozilla.com/D130434

Differential Revision: https://phabricator.services.mozilla.com/D130521

Bug 1738736 - Update dav1d to new version 3fd2ad938ace8d2d0ab86e4108d46a7722691073 from 2021-11-01 15:14:21 r=jbauman

Differential Revision: https://phabricator.services.mozilla.com/D130107

Differential Revision: https://phabricator.services.mozilla.com/D130477

Differential Revision: https://phabricator.services.mozilla.com/D130335

Bug 1737217 - Enable -Wunused-but-set-parameter warnings on clang. r=firefox-build-system-reviewers,glandium

But don't enable on gcc because it reports false positives, such as:

js/src/jsapi-tests/testJitABIcalls.cpp:445:61: warning: parameter 'lineno' set but not used [-Wunused-but-set-parameter]
js/src/jsapi-tests/testJitABIcalls.cpp:461:56: warning: parameter 'base' set but not used [-Wunused-but-set-parameter]

Even though those parameters are used here:

https://searchfox.org/mozilla-central/source/js/src/jsapi-tests/testJitABIcalls.cpp#445,450,461,464

Differential Revision: https://phabricator.services.mozilla.com/D129351

This change fixes the following -Wunused-but-set-parameter warning:

netwerk/test/fuzz/FuzzingStreamListener.cpp:18:49: warning: parameter 'aOffset' set but not used [-Wunused-but-set-parameter]

Differential Revision: https://phabricator.services.mozilla.com/D129350

Backed out changeset 4053cb4ba9ad (bug 1737263) for causing Windows graphics crashes (bug 1739661). CLOSED TREE

Differential Revision: https://phabricator.services.mozilla.com/D130146

Depends on D130547

Differential Revision: https://phabricator.services.mozilla.com/D130548

Differential Revision: https://phabricator.services.mozilla.com/D130547

This should mitigate the code size impact. Also make get_resolved_value
non-generic to avoid monomorphizing it multiple times.

Differential Revision: https://phabricator.services.mozilla.com/D130354

Differential Revision: https://phabricator.services.mozilla.com/D130107

Depends on D130532

Differential Revision: https://phabricator.services.mozilla.com/D130533

Differential Revision: https://phabricator.services.mozilla.com/D130532

I ran into problems while working on tests for bug 1737923 and bug 1737928,
where pref changes were causing UrlbarQuickSuggest to recreate its results map
and keyword tree and they both ended up becoming empty at random points during
the tests. With the way UrlbarQuickSuggest is currently written, there's no way
for tests to prevent that from happening, so I came back to this bug and finally
finished it.

This revision makes these changes:

* Serialize all access to RS, so for example we're not trying to handle a sync
  listener callback from RS at the same time we're initializing the
  UrlbarQuickSuggest instance, or we're not trying to do two concurrent syncs
  from RS (`onEnabledUpdate`) when quick suggest prefs change back to back. This
  also allows tests to safely access/test the UrlbarQuickSuggest instance
  (combined with `readyPromise`) without the possibility of the data changing at
  the same time
* Add a `readyPromise` that consumers (tests) can use so they can be certain
  that no updates are currently ongoing
* Simplify the methods related to syncing from RS. Now there's only one method,
  `_queueSettingsSync`
* Don't unnecessarily recreate the RS client (`_rs`) and sync from it every time
  a quick suggest pref changes or something else causes `onEnabledUpdate` to be
  called
* Tear down the RS client (`_rs`) when quick suggest is disabled so that it's
  not taking up memory (hopefully?) and it's not listening for syncs that are
  unnecessary when the user turns off suggestions
* Move `SUGGESTION_SCORE` from KeywordTree to UrlbarQuickSuggest. It was
  supposed to be there all along but I added it to the wrong class in D124132!
  :-/
* Make a few other style/cosmetic tweaks to UrlbarQuickSuggest
* Make some unrelated trivial improvements to test_quicksuggest_merino.js

Depends on D128970

Differential Revision: https://phabricator.services.mozilla.com/D130429

BarProp.visible on content context now returns the negation of "is popup" of
the top-level browsing context, regardless of the actual visibility of each UI
parts.

browser_test_toolbars_visibility.js is modified to verify:
  * content context follows the updated spec
  * chrome context still sees the actual visibility

There was pre-existing issue in the test that getToolbarsFromBrowserContent was
reading chrome-context value when it's supposed to read content-context value.
Fixed it to inject script to make sure it reads content-context value.

Differential Revision: https://phabricator.services.mozilla.com/D129412

Add BrowsingContext::FieldValues.mIsPopupRequested, and pass "is popup"
value calculated before opening window/tab to BrowsingContext::CreateDetached.

Other code path that is unrelated to content-priv window.open uses the
default value false.

Differential Revision: https://phabricator.services.mozilla.com/D129411

Removed "width" feature from the popup condition, and removed related parameters
(aWidthSpecified, and aSizeSpec) from functions.

Also added "popup" feature that explicitly specify whether to request popup or
not.
This is only for content context, and it behaves differently than existing
"popup" feature for chrome context that makes the window no-style.

Differential Revision: https://phabricator.services.mozilla.com/D129410

Bug 1737832 - Simplify the chromeFlag calculation in window.open for nsGlobalWindowOuter::Print and nsDocShell::PerformRetargeting. r=smaug

This simplifies the chromeFlag calculation for:
  * chrome-priv case, by removing not-fully-chrome-priv case
  * content case, by removing presenceFlag calculation
both by removing the special case for
nsGlobalWindowOuter::Print and nsDocShell::PerformRetargeting,
that has chrome caller but no chrome parent.

Code path for nsGlobalWindowOuter::Print and nsDocShell::PerformRetargeting
now uses the content case instead of the chrome-priv case, but the resulting
flag doesn't change (CHROME_ALL) for those 2 consumers.

Differential Revision: https://phabricator.services.mozilla.com/D129530

Backed out 4 changesets (bug 1737832, bug 1701001) for causing geckoview failures on noopener-noreferrer-BarProp.window.html. CLOSED TREE

Backed out changeset ff439d9c0391 (bug 1701001)
Backed out changeset b0e47b1f4d39 (bug 1701001)
Backed out changeset 0e2ca3cc650e (bug 1701001)
Backed out changeset 712b897cbc76 (bug 1737832)

Differential Revision: https://phabricator.services.mozilla.com/D130205

Differential Revision: https://phabricator.services.mozilla.com/D129886

Differential Revision: https://phabricator.services.mozilla.com/D130190

This fixes a bracketing error that was causing the sdpSemantics not to be recorded when no ICE configuration was supplied.

Depends on D130316

Differential Revision: https://phabricator.services.mozilla.com/D130522

Bug 1739266 - P1 - do not delete the sdpSemantics field before it can be passed to about:webrtc;r=jib

Differential Revision: https://phabricator.services.mozilla.com/D130316

BarProp.visible on content context now returns the negation of "is popup" of
the top-level browsing context, regardless of the actual visibility of each UI
parts.

browser_test_toolbars_visibility.js is modified to verify:
  * content context follows the updated spec
  * chrome context still sees the actual visibility

There was pre-existing issue in the test that getToolbarsFromBrowserContent was
reading chrome-context value when it's supposed to read content-context value.
Fixed it to inject script to make sure it reads content-context value.

Differential Revision: https://phabricator.services.mozilla.com/D129412

Add BrowsingContext::FieldValues.mIsPopupRequested, and pass "is popup"
value calculated before opening window/tab to BrowsingContext::CreateDetached.

Other code path that is unrelated to content-priv window.open uses the
default value false.

Differential Revision: https://phabricator.services.mozilla.com/D129411

Removed "width" feature from the popup condition, and removed related parameters
(aWidthSpecified, and aSizeSpec) from functions.

Also added "popup" feature that explicitly specify whether to request popup or
not.
This is only for content context, and it behaves differently than existing
"popup" feature for chrome context that makes the window no-style.

Differential Revision: https://phabricator.services.mozilla.com/D129410