This fixes Single Locale Repacks for Fennec 57.0 betas.
This broke because our merge scripts rewrote l10n-central to mozilla-beta expecting the l10n repo to be called 'mozilla-beta', which is no longer true with cross-channel l10n. When we patched the configs we didn't see a 'mozilla-beta' entry here, butmissed that the merge-day scripts change it on us.
MozReview-Commit-ID: F7BJzpZg0Xj
--HG--
extra : source : cc652dcb13dba40ae3f263ae89ce3e610a34165f
extra : amend_source : 9fa72e4b2a662a6a63ddfc59dbacf7f03e57dd67
This adds new test verification steps for mochitest/reftest/xpcshell tests
with MOZ_CHAOSMODE=3 (thread scheduling and network thread scheduling).
Enabling all chaos mode features seems to destabilize test verification
so I am only enabling these features for now.
The check for isAbsolutelyPositioned was an old incorrect attempt to fix bug
81290. We have since added the proper fix (not adding offsets of the offset
parent frame) in bug 375003.
MozReview-Commit-ID: 7NgnfrYcs8h
Complications:
- had to copy ReftestManifest into a test zip
- reftest harness was emitting multiple suite_start log entries with --repeat
- some extra path manipulation required to find reftests
The test type is already in the MANIFEST.json file and requiring it in
the expectation data doesn't make much sense. It isn't used execpt
during updates and so people often forget to add it. Therefore it
makes a lot of sense to just use the data from the main manifest.
MozReview-Commit-ID: HyOoN6T28qc
Add a --stability option to wpt-update. If test results are inconsistent
across many runs, the test is disabled for the condition(s) indicated by
the corresponding log files. The --stability option is used to specify
a reason for disabling the tests (typically a bug number), which is
recorded in metadata. Disabled test paths are printed to stdout.
MozReview-Commit-ID: DEIqRkN7NzR
--HG--
extra : rebase_source : ea01b4f3431c8873c140b04e53ce11567ccd5df3
This test was timing out, so split it into firstMatch and alwaysMatch files.
MozReview-Commit-ID: 8E9wZQqdajG
--HG--
extra : rebase_source : e851af039a10e34e0eaa0558388e1442f3bb69ee
This switches to using a timeout of 120s * timeout_multplier, which is
twice the current value. The Marionette startup_timout value is
removed because it turns out that's only used when marionette starts
the Firefox instance and therefore is ignored in this case.
MozReview-Commit-ID: 4VA8yB6M1d5
--HG--
extra : rebase_source : ba9d506d4061d49f2a95fbb24e690841fc902c97
The sizemodechange event is not strongly connected to the
visibilitychange event that the WPT minimize_window.py test is now
using to ascertain whether the window has been successfully iconified.
Because Marionette uses the sizemodechange event it is causing
intermittents such as https://bugzil.la/1397007. You can also read a
lengthy summary I did on the problem in https://bugzil.la/1397007#c11.
The fix for the problem is to wait for the visibilitychange DOM
event content.
MozReview-Commit-ID: B6i33Ee5iMC
According to the spec, if one of the matrices for transform interpolation is
non-invertible, the used animation function must fall-back to a discrete
animation. Add wpt tests so we can ensure the web compability for this behavior.
Note that we don't add 'discrete' type for transform property in property-list.js.
Because the 'discrete' type also tests discrete addition and accumulation,
however, the fallback behavior is just for interpolation and accumulation, not
for addition. So, we add tests in each part of transformListType in property-types.js.
One test is failing in Gecko, so annotate it for Gecko.
MozReview-Commit-ID: 3JGvgqbBqZp
--HG--
extra : rebase_source : 216c471580ad419eb65a6373ea21864fad4b5767
The test type is already in the MANIFEST.json file and requiring it in
the expectation data doesn't make much sense. It isn't used execpt
during updates and so people often forget to add it. Therefore it
makes a lot of sense to just use the data from the main manifest.
MozReview-Commit-ID: HyOoN6T28qc
This makes it possible to subscribe to DOM events in web content, in
order to facilitate fixes for cases such as https://bugzil.la/1397007.
Although the current interface, WebElementEventTarget, is associated with
browser.Context (curBrowser) it is the intention that we in the future
will associate this with representations of the marshaled web element.
This will make it quack and act like an Element's EventTarget, which
has nice parity to it.
The WebElementEventTarget works by passing IPC messages to
a backend service in the content frame script implemented by
ContentEventObserverService. This registers event handlers for specific
DOM events, and relays these events back to the browser-specific message
handler in chrome.
This will enable chrome code to wait for events to occur in content,
like this:
await new Promise(resolve => {
webElement.addEventListener("visibilitychange", resolve, {once: true});
contentBrowser.minimize();
});
MozReview-Commit-ID: B8MMHyG8n04
--HG--
extra : rebase_source : 25ac77fbcfb09435325719ac61614c4a8ce37996
This uploads a JSON summary of the results, without extra logging or
expectation data or anything. It is mostly useful for comparing the
results from two runs e.g. in a dashboard.
MozReview-Commit-ID: Ac45NVBxhy8
--HG--
extra : rebase_source : 2c7a82d6e46b8c9f7af8ad34559e51d84ba4e4ba
The variables (basicShapeSVGBoxValues, etc.) in property_database.js are
moved to the beginning of the file so that they're defined before usage.
MozReview-Commit-ID: 7L3obIY1alP
--HG--
extra : rebase_source : 6c3dff5ecbdad8ef6cf1a49953e4ad1001620b6c
With shared JSM globals, |this| is an NSVO, not a global.
MozReview-Commit-ID: V1j4tipw3m
--HG--
extra : rebase_source : b696dfb881e58c6262b7715ded12422edbdec6d3
This patch re-organizes control flow in the browser-chrome harness so that
extra memory cleanup only happens just before the browser is closed, even
when running with --repeat.
We're getting relatively frequent intermittent shutdown crashes when stream
filters are added during ts_paint tests. Skipping favicon.ico stream filters
should prevent that until we have a more reliable fix.
We try to find fix_macosx_stack.py on the wrong path, which leads us can't run
wpt tests without setting stack fix directory manually (--stackfix-dir).
In this patch, we fix this issue by pointing the path to @topobjdir/dist/bin.
MozReview-Commit-ID: 8JzWWgVM6fM
--HG--
extra : rebase_source : a7822998d7668b976284852bdca86e70d615b611
Currently only the Mochitests on Linux use a startup_timeout of 180s
for Marionette. This would also be necessary for the other platforms
(MacOS and Windows).
MozReview-Commit-ID: TGO7XDLXdI
--HG--
extra : rebase_source : 29219cc51404cebf22e894966294eae687bb5d68
On Windows TaskCluster workers Marionette could be ready within 1s.
As such the currently used timeout is too long, and needs to be
further reduced.
MozReview-Commit-ID: KbLVdpB71cu
--HG--
extra : rebase_source : 2b9139e4b7e2c9d5815e7cc481086ca779385538
This patch makes it possible to disable mochitest-devtools tests on linux64-jsdcov through the 'coverage' flag.
MozReview-Commit-ID: 1gkgbFo19MZ
--HG--
extra : rebase_source : 9c6d311cb467e4424d42848e1caf024f9702b8cd
This is a liberal port of the old geckodriver release instructions
that are on the GitHub wiki:
https://github.com/mozilla/geckodriver/wiki/Releasing
Because the project's canonical home has moved to mozilla-central,
it is better to keep all documentation in one place. Documentation
close to the source code is the best.
I have improved the release instructions in a number of ways, in
particular by describing how to make a release to GitHub/Travis from
mozilla-central.
DONTBUILD
MozReview-Commit-ID: 6f87DLDM59E
--HG--
extra : rebase_source : c144ac8ee2486a38d7eb26c91276e841be50fc81
Shield experiments interfer with core features of Firefox which can
cause tests run via Marionette or geckodriver to fail. Resetting the
API URL will make sure that no experiments can be downloaded.
MozReview-Commit-ID: HnE49jfnen7
--HG--
extra : rebase_source : 2fa6cf86ddce798eba88a41d227768f149cccc70
Shield experiments interfer with core features of Firefox which can
cause tests run via Marionette or geckodriver to fail. Resetting the
API URL will make sure that no experiments can be downloaded.
MozReview-Commit-ID: HnE49jfnen7
--HG--
extra : rebase_source : 9cec29085032318e9e8467d6d754afe0890ae645
Currently the mochitest and reftest runners are forcing a timeout of 60s
(or 900s for valgrind or debug builds) when calling `start_session` of
Marionette. While this method still offers a timeout parameter, the
timeout should be set via the `startup_timeout` argument for Marionette.
Reason is that Marionette doesn't control the browser instance and
is getting called right after the application gets launched. As such
slow running builds can cause timeouts once it takes longer than 60s
for the Marionette server to get started.
By using `startup_timeout` the timeout will even be configurable via
the command line and mozharness config files.
MozReview-Commit-ID: EV7GklBcJjU
--HG--
extra : rebase_source : 9d3c623c49deb92d68c40ba4410c812c864e06bd
When the binary gets launched we do not immediately kill the process if the
connection to Marionette server cannot be established within the given amount
of seconds. Instead "_handle_socket_failure" is getting called because
the utility method `raise_for_port` inappropriately uses the `@do_process_check`
decorator.
By removing the decorator the initial connection attempt can be handled
differently. As such the process if handled by Marionette will be immediately
killed. Currently we are waiting for the process to quit itself within 120s,
which will actually never happen due to no active session.
Further `start_session` defaults to a timeout of 60s which itself is problematic
for test harnesses using Marionette but controlling the binary themselves. In
those cases timeouts can happen often for slow starting browser processes like
debug builds. Instead this timeout should default to the `startup_timeout` value.
MozReview-Commit-ID: BZvX5KT45mK
--HG--
extra : rebase_source : 875807b75f0575a10dc9de3588237002fa6a5821
By logging the used browser arguments from geckodriver only the
-marionette argument ends up in the log. Instead mozrunner should
be used which knows about any of them.
MozReview-Commit-ID: J9px0pWSwQm
--HG--
extra : rebase_source : 36730f7c2ecffa7a28dda3d16404f523bd075d35
This just adds two basic tests, one for a passing test and another for a
failing one. In mochitest, we use privileged APIs to also tests crashes,
assertions, asan and leaks. But these APIs aren't available to reftests
so I'm not sure how we can test these things.
I figure it's not worth holding the framework up on this though, I'll file
a follow-up to figure out something to do for that.
MozReview-Commit-ID: 59TSbsugT5T
--HG--
extra : rebase_source : 72ecd817017c8b7d55eab879db4f6ad5fecc54c0
This includes code for downloading a Firefox binary, downloading + setting up a tests.zip and
running output through mozharness' output parsers. This is all stuff that will also be required
for the reftest selftests.
I couldn't think of a better location to put this stuff, suggestions welcome.
MozReview-Commit-ID: 59TSbsugT5T
--HG--
extra : rebase_source : a328f6bc90e73fe23f9054933cd01a30065419f6
This change rides the train along with the change to build beta from
l10n-central.
Not changing 52 esr config, as that continues to build from the
repository it's currently building from.
ESR will be picked up when the next ESR is riding the trains from central.
MozReview-Commit-ID: F9Wo2mcZ3PA
--HG--
extra : rebase_source : dbf6c39cc5ab51bf4970caa4d62a80060b6b8c7c
Take into account that a weak referenced element might have been destroyed
in the element staleness check.
An error is thrown when the reference object has been destroyed when
getting a weakrefs' pointer. We catch this, but element.isStale does
not take into account that the el argument in this case can be null,
or in this revision of the patch, undefined.
MozReview-Commit-ID: 7sr4YGhAotS
--HG--
extra : rebase_source : 995eeef4ec1b19bf100cb95c4dd343e947b1cf52
This test causes both a failure and an error and we need to investigate
how to reuse the wptrunner session configuration when implicitly starting
a session.
MozReview-Commit-ID: 5k7nfevamZY
--HG--
extra : rebase_source : 6c00bf4505bd7da0b97dccd3eda61adc9efa874e
The WindowProxy screenX/screenY/outerWidth/outerHeight fields are
platform-independent longs, meaning the bitness of the system is not
taken into account. This is defined in WebIDL as "a signed integer type
that has values in the range [−2147483648, 2147483647]". This can be
represented with an i32 internally.
Additionally, the WebDriver specification says that we should accept
JSON Numbers as input. This means the input value can be either a float
or an integer. rustc_serialize handles this very badly and we first
need to extract the input as f64 because this is the most liberal type
(i.e. an integer can be coerced to a float without problem).
We then we cast the number to an i64 which trims off any decimals without
worrying that the number wraps. We then run bounds checks of an i32
and error if the number will not fit within the set range.
This patch expects the output from Marionette to be conforming.
MozReview-Commit-ID: CVdUZWh2t8X
--HG--
extra : rebase_source : 6780985703d5d5291eae508f9eb2fdec16c261cc
When maximizing the window we must restore it from iconified state or
exit fullscreen first. Likewise for minimizing the window, we must
exit fullscreen. For fullscreening the window we need to also restore
the window.
MozReview-Commit-ID: AOQX2cV2C75
--HG--
extra : rebase_source : 53f310cc261a9a411bc46a04dd50a160a930b5af
If the browser process is not running it doesn't make sense to try to
connect to it for another 60s. Instead error out immediately.
MozReview-Commit-ID: 64DTZfEfzQj
--HG--
extra : rebase_source : 9951b1a64cbd8b387c2bc5e1152de0a975dbbe77
This change, though unstubstantial, will make the Markdown documents
under testing/geckodriver more readable in text editors.
DONTBUILD
MozReview-Commit-ID: 21yQUExNJhU
--HG--
extra : rebase_source : cb51d1a0ce93c561c79fabb941b56192ec711480
This patch gently removes support for __exposedProps__ by changing
ExposedPropertiesOnly::check() to always return false, while still
failing silently in deny for some kinds of access.
The tests that I changed all involve testing the behavior with
__exposedProps__. I adjusted them to expect it to fail, or to adjust
the error message they get when they fail. That seemed better than
deleting them entirely.
Note that test_bug1065185.html had a bug, so that it never executed
the first case. I fixed that, and then fixed up the test to work when
__exposedProps__ is not supported.
This also removes various bits of the test framework that use
__exposedProps__, but don't actually need to.
MozReview-Commit-ID: 8fvkAmITmXY
--HG--
extra : rebase_source : ef7e2c55adc12511f17f3865ebb46c343875f0b3
The marionette.defaultPrefs.port preference was changed to
marionette.port, but because we currently do not run tests in CI we
missed updating the test.
MozReview-Commit-ID: LKstRYmJcMO
--HG--
extra : rebase_source : 1633b5b82c3c8725ff66423119d7c476fa942b01
This tests both that the settings have the desired effect and that switching
between sharing enabled and sharing disabled without a startup cache flush
does not cause any issues.
Tests for user pref changes are currently non-fatal, since they're known not
to work reliably.
MozReview-Commit-ID: 1ZFwyiNf3da
--HG--
extra : rebase_source : c38bd92d2137c90f8c4d202b7009612b45ff4be9
Notable changes
* ensure we run dump-symbols and upload actions on all platforms
* On android:
* add configuration and support for aarch64
* set min_sdk levels to match Fennec builds
* use a full copy of the r11c ndk (our truncated one was missing toolchains we needed) and set NDKROOT when calling build
* ensure the tooltool provided sdk is on the PATH
* on linux copy tooltool.py into the mock environment, so we can get dump_syms from tt
* remove macosx32 config as we've deprecated that in Firefox builds
* update dump_syms to recent m-c, notably for aarch64 support on linux
* on linux rev e365137fa61bfd729617ba1ebf9f1ed79facd1f2 (via try 0f72a5c28be1cdc2f3bdfaafdf3826254f6ba077)
* on mac rev e365137fa61bfd729617ba1ebf9f1ed79facd1f2 (via compile on a bld-lin-r5)
* on windows rev a4a448ba7f187069fce916ee234a06cbb0d06f80 (via try dc8b121e3c08e8022d62c0fa1951dd3dc4d6f7cc)
* switch to Visual Studio 2015 Update 3 on win32/win64 to match Firefox
* many updates to environement variables
* painful to get win64 right to run win32 dump_syms.exe, but that's why the x86 redist is on teh PATH
* unwind the changes to get_output_from_command() in v1.6 patch to avoid affecting other builds, and use query_env() which has this support already
* add a scp_upload_directory since we don't have rsync on windows, use that to talk to the ffxbld upload host (not a long term solution but OK for now)
Applies on top of https://reviewboard.mozilla.org/r/64022/diff/4#index_header
MozReview-Commit-ID: B3NiWFvr2oR
Currently defaults for startup_timeout and socket_timeout are defined
at two different places (Marionette driver and harness). As of now it's
even the case that startup_timeout has different values. While Marionette
driver uses 120s, the harness only uses 60s.
As result all jobs which are based on the Marionette harness fail if
Firefox starts-up slowly like for debug builds.
MozReview-Commit-ID: Dl4sBG1H7NA
--HG--
extra : rebase_source : 959facabebc371beee23b4de345ddd2495913bb7
Currently defaults for startup_timeout and socket_timeout are defined
at two different places (Marionette driver and harness). As of now it's
even the case that startup_timeout has different values. While Marionette
driver uses 120s, the harness only uses 60s.
As result all jobs which are based on the Marionette harness fail if
Firefox starts-up slowly like for debug builds.
MozReview-Commit-ID: Dl4sBG1H7NA
--HG--
extra : rebase_source : 688338b1782deaf08eb01c7c5d4ca01ba03328f5
To allow resetting the default no proxy exclusion entries in
Firefox tests have to pass an empty noProxy list. This should
also be correctly applied.
MozReview-Commit-ID: ABmYdPvoSvx
--HG--
extra : rebase_source : 313fe0a918a4a9bfe5204e5777568ee5f89744d3
Currently the listener for addon installs misses a check for the addon id,
to only resolve the promise when it has been called for the expected addon.
This can cause race-conditions if other addons are getting installed at the
same time.
The same applies to uninstall which doesn't wait at all until the operation
has been completed.
MozReview-Commit-ID: 5GsomMoAVZ1
--HG--
extra : rebase_source : a1b43adb2239b0c28cbee1d843f4b6c666a07f0a
This method works by running a long-running script and catching the exception when
it is interrupted. But the exception changed so we must make a corresponding change here.
MozReview-Commit-ID: EdZZAOVZ0Sw
In general these exceptions are the result of something unexpected in
the environment, but shouldn't themseslves cause the test to error.
MozReview-Commit-ID: 5XjJoT4UwnC
To allow the selection of the new webdriver spec conforming
"clickElement" method over the old legacy Selenium one, the
"moz:webdriverClick" capability can be used. Setting it's value
to true will enable the conforming click.
MozReview-Commit-ID: HMvFFEJDvpG
--HG--
extra : rebase_source : 34bb71f7d64d08968cd6b52760cf4d36a61cb29d
To allow geckodriver to temporarily opt-in for using the webdriver
conforming click, a new custom capability has to be created for
Marionette. If not specified the legacy clickElement method will
be used instead.
MozReview-Commit-ID: LuyTjLJXMGL
--HG--
extra : rebase_source : 8e133fb2d9767ea580468013fedff5233fd8eb93
There are DOM attributes to query a window's minimized and fullscreen
states, but there is no reliable way to tell if the window is maximized.
To workaround this, we set the window's size before every test to ensure
it transitioned to the normal window state.
MozReview-Commit-ID: DAT0E4rhmjY
--HG--
extra : rebase_source : 0e9e77ad340c39b656f523085613f43f1d6bd94a
The window state concept was removed from WebDriver level 1.
MozReview-Commit-ID: 6cogstpDc0a
--HG--
extra : rebase_source : 539df9fed0761b8e9609f60c3f9b77b946829147
Some of the HTTP status codes have changed in the WebDriver specification,
as shown in:
d678562e3d
MozReview-Commit-ID: 4DLkYDXckew
--HG--
extra : rebase_source : 37c595265ead9591a6618ca44000bfb0f401da75
Sebastian Hengst
Andrea Marchesini
Wes Kocher
Justin Wood
James Graham
Geoff Brown
Joel Maher
John Dai
Nicholas Nethercote
Thomas Wisniewski
Hiroyuki Ikezoe
Miko Mynttinen
Ian MacLeod
Jessica Jong
Boris Zbarsky
Kim Moir
Rob Wood
Maja Frydrychowicz
Bevis Tseng
Andreas Tolfsen
Jeremy Chen
Alexandre Poirot
Tom Prince
Ryan VanderMeulen
Kartikaya Gupta
Felipe Gomes
Henrik Skupin
Prathiksha
Andrew Halberstadt
Gabor Krizsanits
Rail Aliiev
Axel Hecht
Christoph Kerschbaumer
Ben Kelly
Ting-Yu Lin
Nick Thomas
Andrew McCreight
Jonathan Kew
Kris Maglione
Greg Mierzwinski
Steve Armand
Christian Holler
Ed Lee
Johan Lorenzo
Matt Woodrow
James Cheng
David Burns
Boris Chiou
Ho-Pang Hsu
Anthony Miyaguchi
Michael Layzell
Ionut Goldan
Myk Melez
Chris Manchester
Emilio Cobos Álvarez
Cosm
Nils Ohlmeier [:drno]