nsThreadManager::get() can return a reference. This lets us remove some
redundant assertions.
nsThreadArray elements can be NotNull<>s.
--HG--
extra : rebase_source : fd49010167101bc15f7f6d01bf95fd63b81d60fb
The main renaming was generated with the following python script:
```
import sys
import re
CAMEL_CASE_REGEX = re.compile(r"(^|_|-)([A-Z])([A-Z]+)")
DISPLAY_REGEX = re.compile(r"\bNS_STYLE_DISPLAY_([^M][A-Z_]+)\b")
def to_camel_case(ident):
return re.sub(CAMEL_CASE_REGEX,
lambda m: m.group(2) + m.group(3).lower(), ident)
def constant_to_enum(constant):
return "StyleDisplay::" + to_camel_case(constant) + ("_" if constant == "NONE" else "")
def process_line(line):
return re.sub(DISPLAY_REGEX,
lambda m: constant_to_enum(m.group(1)), line)
lines = []
with open(sys.argv[1], "r") as f:
for line in f:
lines.append(process_line(line))
with open(sys.argv[1], "w") as f:
for line in lines:
f.write(line)
```
And the following shell commands:
```
find . -name '*.cpp' -exec python display.py {} \;
find . -name '*.h' -exec python display.py {} \;
```
MozReview-Commit-ID: 91xYCbLC2Vf
If we synchronously shut down the prompt, that will exit modal state while we're
still processing our mouse or key event, which happens _before_ Gecko runnable
processing. Exiting modal state will generally post a runnable to run timeouts,
so we can end up processing that runnable before we return from processNextEvent
and unwind to whatever code did the alert() call on the web page. The upshot is
that the webpage will see timeouts fire before the alert() call returns.
If we exit modal state off a Gecko runnable instead, that will ensure that we
return to the nsPrompter code immediately after exiting modal state, see that
the prompt is no longer active, and return to the calling web page code _before_
processing any more runnables.
The new behavior matches the specification, web-platform-tests, Chrome,
and Edge. I couldn't figure out any reason for the old behavior.
MozReview-Commit-ID: 6cktZuN1vCV
This allows other UAs to use it, removes duplicated checks, and
increases the chance of us noticing if the spec changes. Some of the
expected values in our mochitest were contrary to the spec.
I checked the new expected failures against the spec and the other UAs.
I filed a spec bug for one group because it was contrary to all UAs
(although IMO the spec makes more sense and the UAs are buggy), and the
others are fixed in the next patch.
MozReview-Commit-ID: 1j11XgfuErB
When I wrote this initially, I was told it was not possible to inherit
from desktop-build. That may have changed, but I still don't know how
to inherit, so we duplicate the logic here, inviting bugs of this
class. The underlying logic moves slowly so this is an acceptable
approach.
MozReview-Commit-ID: AsWdbp0QK5K
--HG--
rename : testing/docker/desktop-build/dot-config/pip/pip.conf => testing/docker/android-gradle-build/dot-config/pip/pip.conf
extra : rebase_source : b13d97f5faabfaa42d1068b216d2c5ad5052154c
hasFeature() always returning true matches the current DOM spec. SVG 2
has removed requiredFeatures. Chrome has had both of these always
return true since 2014, and they seem to have had no problems.
Even requiredFeatures="" (empty string) now returns true, matching
Chrome.
MozReview-Commit-ID: 1LEu3iK4R94
Mike Conley
Nicholas Nethercote
Rutuja Surve
Emilio Cobos Álvarez
Phil Ringnalda
Mike Shal
Boris Zbarsky
Lee Salzman
Eric Rahm
Wes Kocher
Milan Sreckovic
Ryan VanderMeulen
Aryeh Gregor
Andrew Halberstadt
Nick Alexander
Sebastian Hengst
James Willcox
Jason Laster
Shu-yu Guo
Bob Owen
Zibi Braniecki
Kannan Vijayan