Closed Bug 763169 Opened 12 years ago Closed 12 years ago

Update testharness.js

Categories

(Testing :: Mochitest, defect)

Product:
Testing
Component:
Mochitest
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED FIXED
Milestone:
mozilla16

People

(Reporter: Ms2ger, Assigned: Ms2ger)

Details

Attachments

(5 files, 2 obsolete files)

Part a: Update resources from upstream
(deleted), patch
k0scist
: review+
Details | Diff | Splinter Review
Part b: Miscellaneous fixes
(deleted), patch
k0scist
: review+
Details | Diff | Splinter Review
diff -w (part b)
(deleted), patch
Details | Diff | Splinter Review
Part c: Don't log all the passing results
(deleted), patch
k0scist
: review+
philor
: feedback+
Details | Diff | Splinter Review
Part a (2): Update test names in the expected failures for Selection tests
(deleted), patch
k0scist
: review+
Details | Diff | Splinter Review
Attached patch Patch v1 (obsolete) (deleted) — — Splinter Review
And some miscellaneous cleanup while I was around.
Attachment #631638 - Flags: review?(jhammel)
Attached patch Diff -w for testharnessreport.js (obsolete) (deleted) — — Splinter Review
There's been some more changes lately; looking for a rubber stamp here.
Attachment #631638 - Attachment is obsolete: true
Attachment #631639 - Attachment is obsolete: true
Attachment #631638 - Flags: review?(jhammel)
Attachment #633427 - Flags: review?(jhammel)
Attached patch Part b: Miscellaneous fixes (deleted) — — Splinter Review
This avoids some uncaught exceptions.
Attachment #633428 - Flags: review?(jhammel)
Attached patch diff -w (part b) (deleted) — — Splinter Review
Importing some more tests caused buildbot to truncate the log after about 200 000 logged results. This cuts the log size down quite significantly: 15248 INFO Passed: 203746 15249 INFO Failed: 0 15250 INFO Todo: 21478 (with those new tests).
Attachment #633432 - Flags: review?(jhammel)
Attachment #633432 - Flags: feedback?(philringnalda)
I needed to avoid duplicate test names in the tests using idlharness.js, so I added the arguments there. (This change is imported into our tree in part a.) However, that means the expected failures don't match anymore. This patch fixes that.
Attachment #633439 - Flags: review?(jhammel)
Comment on attachment 633427 [details] [diff] [review] Part a: Update resources from upstream wfm; sorry for the delay
Attachment #633427 - Flags: review?(jhammel) → review+
Comment on attachment 633428 [details] [diff] [review] Part b: Miscellaneous fixes lgtm
Attachment #633428 - Flags: review?(jhammel) → review+
Comment on attachment 633432 [details] [diff] [review] Part c: Don't log all the passing results We have to ensure that enough is logged such that buildbot doesn't count XXXX seconds without output, but every hundred seems plenty conservative here (for a significant gain in log streamlining).
Attachment #633432 - Flags: review?(jhammel) → review+
Attachment #633439 - Flags: review?(jhammel) → review+
Comment on attachment 633432 [details] [diff] [review] Part c: Don't log all the passing results The other time besides keeping the 1200 second timer at bay when printing passes is handy is context for bad messages, when a test has 30 possibilities for "length is wrong, 7 == 10".
Attachment #633432 - Flags: feedback?(philringnalda) → feedback+
Thanks https://hg.mozilla.org/mozilla-central/rev/5ac99785f6ac https://hg.mozilla.org/mozilla-central/rev/baaa5d34e5eb https://hg.mozilla.org/mozilla-central/rev/24f645d265d0
Status: ASSIGNED → RESOLVED
Closed: 12 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla16
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: