Commit Graph

420 Commits

Author SHA1 Message Date
Henrik Skupin
620e1de054 Bug 1320073 - Rename marionette-client to marionette-harness and release version 4.0.0. r=gps,maja_zf,mtseng,SingingTree
MozReview-Commit-ID: L6V3b5h8RWm
2016-12-12 13:05:34 +01:00
Ehsan Akhgari
9916b175a4 Bug 1318879 - Remove the app related code from mozprofile; r=ahal 2016-11-21 18:56:08 -05:00
Wes Kocher
f34c85c8a8 Backed out changeset 0cafbf944d89 (bug 1318879) for robocop failures a=backout 2016-11-21 13:01:21 -08:00
Ehsan Akhgari
b138925a19 Bug 1318879 - Remove the app related code from mozprofile; r=ahal 2016-11-21 14:59:05 -05:00
Ehsan Akhgari
03cac476b4 Bug 1312209 - Make some small fixes to runtests.py in case a test manifest isn't found; r=ted
Before these changes we get a bunch of Python errors.
Now we get a human readable warning saying the test
manifest does not exist.
2016-11-15 08:31:35 -05:00
Andrew Halberstadt
5c9549f857 Bug 1020516 - Add timestamp messages when dumping buffered messages in mochitest, r=ted
When dumping buffered messages in mochitest, this adds a new log line for every second's
worth of messages. Alternatively we could print the timestamp in brackets at the end of
each buffered message, but I found that in practice there tends to be many log messages
generated each second. This looks better for that case, but will look much worse if logs
are generated more than a second apart. I would be ok with either method.

MozReview-Commit-ID: Jkd9hOlmiGZ
2016-11-02 13:42:58 -04:00
Andrew Halberstadt
adf31cb049 Bug 1020516 - Only buffer mochitest log messages that happen between a TEST-START/TEST-END, r=ted
I believe the current behaviour of buffering messages outside of a test is wrong. For example:

1. INFO logged -> gets buffered
2. TEST-START logged -> doesn't get buffered
3. Test fails and buffer dumped -> INFO looks like it happened after the TEST-START

I think this is very confusing as developers trying to debug a test will think that INFO message
is happening somewhere inside their test when in reality it isn't.

MozReview-Commit-ID: Jkd9hOlmiGZ
2016-11-02 13:42:58 -04:00
Andrew Halberstadt
4a195c2f56 Bug 1020516 - Pass options.quiet into mochitest MessageLogger via constructor, r=ted
For some reason we have a 'buffering' parameter in the MessageLogger constructor, but
then rather than using this, the mochitest harness modifies state after instantiation.
Using the constructor is easier to understand, and simplifies some of the logic in the
next couple of patches.

MozReview-Commit-ID: Jkd9hOlmiGZ
2016-11-02 13:42:57 -04:00
Ryan VanderMeulen
112d5746d5 Merge autoland to m-c. a=merge 2016-10-29 09:12:45 -04:00
Randell Jesup
be3119063b Bug 1311397: bustage fix with space rs=tomcat
on a CLOSED TREE
2016-10-28 03:49:33 -04:00
Randell Jesup
f76f9db4ad Bug 1311397: fix DataChannel logging for mochitests/etc r=erahm 2016-10-28 03:25:10 -04:00
Geoff Brown
34c650c140 Bug 1313187 - Increase default timeout on ASAN browser mochitests to 90 seconds; r=jmaher 2016-10-26 16:33:10 -06:00
Andrew Halberstadt
7a19b7b241 Bug 1313265 - Log messages containing 'TEST-UNEXPECTED-FAIL' at the error level in mochitest, r=jgraham
We were previously logging these at the info level as a really hacky way to make sure that the mozharness
output parser was able to detect them. Since bug 1261194 switched mochitests to use the StructuredOutputParser,
we now need to log them at the error level or they won't get picked up anymore.

MozReview-Commit-ID: 99n9YizlEDH
2016-10-27 15:03:38 -04:00
Andrew Halberstadt
ef275026dc Bug 1308202 - Merge MochitestBase and MochitestDesktop into a single class, r=jmaher
MozReview-Commit-ID: 737L08LhLDp
2016-10-06 10:16:06 -04:00
Andrew Halberstadt
0b5479bf7c Bug 1308202 - Remove all b2g and mulet related code from testing/mochitest, r=jmaher
MozReview-Commit-ID: BfIJLp6oGHO
2016-10-06 10:05:27 -04:00
Prabhjyot Singh Sodhi
55ca3ed6d4 Bug 1280571 - Add testing/mochitest to the flake8 linter, r=ahal
MozReview-Commit-ID: INBZaZzcPmF
2016-09-28 08:28:04 -04:00
Greg Mierzwinski
b701385044 Bug 1278402 - Add linux64-ccov and linux64-jsdcov as code coverage build platforms to taskcluster. r=dustin,jmaher
These builds can be run on taskcluster to obtain per-test (JSDebugger) code coverage with the linux64-jsdcov build and overall (GCOV) code coverage with the linux64-ccov build. The linux64-jsdcov build also needed to have leak checking disabled for debug mode.

MozReview-Commit-ID: ASgrU2X7RQV
2016-07-23 11:27:49 -04:00
Wes Kocher
9ab4cf93f6 Backed out changeset 9c949b216b37 (bug 1278402) for enabling jobs outside of try that shouldn't be enabled outside of try a=backout CLOSED TREE 2016-09-15 11:43:17 -07:00
Greg Mierzwinski
e226db2f37 Bug 1278402 - Add linux64-ccov and linux64-jsdcov as code coverage build platforms to taskcluster. r=dustin,jmaher
These builds can be run on taskcluster to obtain per-test (JSDebugger) code coverage with the linux64-jsdcov build and overall (GCOV) code coverage with the linux64-ccov build. The linux64-jsdcov build also needed to have leak checking disabled for debug mode.

MozReview-Commit-ID: ASgrU2X7RQV
2016-07-23 11:27:49 -04:00
Andrew Halberstadt
733f14d1da Bug 1300163 - Fix hidden tracebacks in mochitest runtests.py, r=armenzg
I had a mistake in my code that resulted in an exception being thrown when starting the
mochitest servers. However, the log just showed a timeout. It was very confusing.

After some digging, I found that there's a large try/finally block (with no except) in
runtests.py. The exception wasn't being displayed because of the finally block. This
try/finally block had an inner try/except/finally that wrapped part of the code. This
patch simply removes that inner block and dedents the 'except' clause to attach to the
original try/finally.

MozReview-Commit-ID: 4lsKGpizfH7
2016-09-02 14:39:34 -04:00
Andrew Halberstadt
9e6b28f4ea Bug 1293295 - Replace all mochitest 'flavor' options with a single --flavor argument, r=jmaher
This accomplishes three things:

1) Easier to use CLI when running without the benefit of testing/mochitest/mach_commands.py
2) Guarantees these arguments are mutually exclusive
3) Simplifies a bunch of logic in the test harness

The primary motivation for this change is to slightly improve the UX when running mochitest
from a taskcluster interactive loaner. However, this is more of a bandaid solution that was
easy to implement before the proper fix in bug 1293259 can be landed.

MozReview-Commit-ID: IeHBGrJ0Sji
2016-08-08 11:48:49 -04:00
Andrew Halberstadt
cfc6df1ca7 Bug 1288827 - Allow running mochitest from test-package without specifying --appname, r=armenzg
Because it is now possible for options.app to get set after 'parse_args' time, we need to make sure
the argument validation happens later. To accomplish this we pass in the parser instance to
'run_test_harness' and do parser.validate there. This unfortunately requires some minor uses of
global to accomplish easily due to how mach handles parsers.

MozReview-Commit-ID: s3Js1aZlSE
2016-07-20 11:13:30 -04:00
Carsten "Tomcat" Book
2d709b2fa1 Backed out changeset 6dcb52d270c7 (bug 1288827) for breaking mochitests 2016-07-28 17:04:42 +02:00
Andrew Halberstadt
e77f6c00c3 Bug 1288827 - Allow running mochitest from test-package without specifying --appname, r=armenzg
Because it is now possible for options.app to get set after 'parse_args' time, we need to make sure
the argument validation happens later. To accomplish this we pass in the parser instance to
'run_test_harness' and do parser.validate there. This unfortunately requires some minor uses of
global to accomplish easily due to how mach handles parsers.

MozReview-Commit-ID: s3Js1aZlSE
2016-07-20 11:13:30 -04:00
Geoff Brown
60b824adee Bug 1286697 - Initialize mochitest urlOpts to avoid duplicate options on subsequent runs; r=jmaher 2016-07-14 10:36:30 -06:00
Ralph Giles
1eec789c7b Bug 1275439 - Switch debugging to MOZ_LOG. r=erahm
This is shorter than MOZ_LOG_MODULES and equally clear.

Add a deprecation warning to encourge folks to migrate,
and update references in the test runner.

MozReview-Commit-ID: HYY3Q9tSu13
2016-05-16 15:49:20 -07:00
Geoff Brown
d905d04224 Bug 1274395 - Delete pending crash reports before running tests; r=jmaher 2016-05-25 14:48:42 -06:00
Gijs Kruitbosch
310d327ef8 Bug 1270144 - use error() rather than warning() in order to ensure this shows up on the treeherder summary, rs=emorley-over-irc
MozReview-Commit-ID: COtfpoU14vg
2016-05-25 15:14:56 +01:00
Andrew Halberstadt
30b4b782ef Bug 1274002 - Disable e10s when running mochitest-chrome locally with mach, r=jmaher
MozReview-Commit-ID: 1r2aZtnB3lo
2016-05-18 15:17:13 -04:00
Gijs Kruitbosch
48db4ef098 Bug 1270144 - fix message from timed out tests to indicate the following crash is caused by us terminating the browser process, r=ahal,RyanVM
MozReview-Commit-ID: Hyr7xkZdbqc
2016-05-10 22:53:17 +01:00
Jonathan Kingston
f80381dabe Bug 1252068 - Adding in reset prefs button on SSLNetError page. r=gijs
MozReview-Commit-ID: BJQ0cYlrS6O
2016-05-04 17:10:45 +01:00
Byron Campen [:bwc]
94efd94959 Bug 1231981 - Part 3: Set up TURN server for webrtc mochitests, when configured to. r=drno
MozReview-Commit-ID: CVbAYPrwpuB
2016-03-04 15:45:57 -06:00
Byron Campen [:bwc]
146f940956 Bug 1231981 - Part 2.1: Only run the websocket/process bridge for media tests. r=ahal
MozReview-Commit-ID: 3GBAGTSR08c
2016-04-27 13:13:40 -05:00
Byron Campen [:bwc]
80f9d382ca Bug 1231981 - Part 2: A websocket-to-process bridge script that can be used by JS to launch an ICE server for testing. r=ahal
MozReview-Commit-ID: FbfNzyw9SZp
2016-03-15 17:02:00 -05:00
Wes Kocher
7721e2c661 Backed out changesets baa6c68ec413,5b8583070c9e,2942bb31152c,dd2ad6863f53,c68772362633,c8db6695b991 (bug 1231981) a=backout
MozReview-Commit-ID: EoGyz071RoN
2016-04-27 10:24:37 -07:00
Byron Campen [:bwc]
e66ab5d91c Bug 1231981 - Part 5: Install python packages we need in the virtualenv if not present. r=gps
MozReview-Commit-ID: 9PUHLMxHhQP
2016-03-16 15:40:17 -05:00
Byron Campen [:bwc]
bb1364fbce Bug 1231981 - Part 3: Set up TURN server for webrtc mochitests, when configured to. r=drno
MozReview-Commit-ID: CVbAYPrwpuB
2016-03-04 15:45:57 -06:00
Byron Campen [:bwc]
fdd7508c86 Bug 1231981 - Part 2: A websocket-to-process bridge script that can be used by JS to launch an ICE server for testing. r=ahal
MozReview-Commit-ID: FbfNzyw9SZp
2016-03-15 17:02:00 -05:00
Carsten "Tomcat" Book
f602f39509 Backed out changeset de3f062c9a99 (bug 1231981) 2016-04-25 16:28:55 +02:00
Carsten "Tomcat" Book
7c16cbfe9a Backed out changeset a9fcd0e98cc3 (bug 1231981) 2016-04-25 16:28:53 +02:00
Carsten "Tomcat" Book
90b3aaa279 Backed out changeset 5a0d061f7548 (bug 1231981) 2016-04-25 16:28:50 +02:00
Byron Campen [:bwc]
c15bf3c87c Bug 1231981 - Part 5: Install python packages we need in the virtualenv if not present. r=gps
MozReview-Commit-ID: 9PUHLMxHhQP
2016-03-16 15:40:17 -05:00
Byron Campen [:bwc]
fedcf2a88c Bug 1231981 - Part 3: Set up TURN server for webrtc mochitests, when configured to. r=drno
MozReview-Commit-ID: CVbAYPrwpuB
2016-03-04 15:45:57 -06:00
Byron Campen [:bwc]
13f29cc725 Bug 1231981 - Part 2: A websocket-to-process bridge script that can be used by JS to launch an ICE server for testing. r=ahal
MozReview-Commit-ID: FbfNzyw9SZp
2016-03-15 17:02:00 -05:00
Dan Minor
379f2a5aee Bug 1264772 - Use subsuite name for chunk-by-runtime; r=ahal
We currently hard code the runtime name to be mochitest-plain. This
changes things to use the subsuite name if available.

MozReview-Commit-ID: JZKewOD81M8
2016-04-14 13:57:05 -04:00
Joel Maher
14b85fefb0 Bug 1262954 - Explicitly set the 'layers.acceleration.disabled' pref to true on Windows XP e10s test runs. r=ahal
MozReview-Commit-ID: 2KsBZDNvsdU
2016-04-08 14:46:25 -04:00
Andrew Halberstadt
85e76aff37 Bug 1243083 - Enable e10s by default when running mochitests, r=jmaher
MozReview-Commit-ID: NlTGwdQ2nw
2016-04-01 10:59:12 -04:00
Ryan VanderMeulen
8fc96437cd Merge inbound to m-c. a=merge 2016-03-19 11:22:28 -04:00
Ryan VanderMeulen
8a1d569031 Bug 1257708 - Move dom/push/test back into mochitest-plain instead of a separate test suite. 2016-03-19 11:02:04 -04:00
Honza Bambas
24d2808e57 Bug 1248565 - Make test infra use MOZ_LOG_FILE and MOZ_LOG_MODULES. r=jmaher 2016-03-11 08:27:00 -05:00