Closed
Bug 470838
Opened 16 years ago
Closed 14 years ago
Write 302 tests for imagelib
Categories
(Core :: Graphics: ImageLib, defect)
Tracking
()
RESOLVED
INVALID
People
(Reporter: joe, Assigned: joe)
References
Details
Attachments
(1 file, 1 obsolete file)
7.18 KB,
patch
|
vlad
:
review+
|
Details | Diff | Splinter Review |
Bug 89419 fixed imagelib's 302 behaviour, but it didn't add any tests. This is obviously a bad thing.
What will be necessary to fix this is both some form of server-side JS that implements the simple state machine required to switch out which image we're redirecting to (examples of this are source/toolkit/components/passwordmgr/test/authenticate.sjs for a 401 and modules/libpr0n/test/reftest/generic/check-header.sjs for checking other headers), as well as some reftest-alike structure that lets us compare the output - WindowUtils.js apparently has something that will be helpful.
Assignee | ||
Updated•16 years ago
|
Assignee: nobody → joe
Assignee | ||
Comment 1•16 years ago
|
||
Attachment #354398 -
Flags: superreview?(vladimir)
Attachment #354398 -
Flags: review?(vladimir)
Assignee | ||
Updated•16 years ago
|
Attachment #354398 -
Attachment is obsolete: true
Attachment #354398 -
Flags: superreview?(vladimir)
Attachment #354398 -
Flags: review?(vladimir)
Assignee | ||
Comment 2•16 years ago
|
||
Comment on attachment 354398 [details] [diff] [review]
Mochitest for 302 images
This patch doesn't work right now. (I suspect httpd.js changes.) I'll fix it and resubmit a new test.
Comment 3•16 years ago
|
||
Probably bug 471158, but I don't see what would be causing a problem for you here, as you're not using your state outside of the single SJS file. Changing the line in netwerk/test/httpserver/httpd.js that sets DEBUG to set it to true, making in testing/mochitest (I think) if you're on Windows, and running that Mochitest (use --test-path=modules/libpr0n/test/mochitest/test_bug89419.html with runtests.py or TEST_PATH=modules/libpr0n/test/mochitest/test_bug89419.html if you're going the make route to running tests) is likely to give useful console spew (to me at least, quite possibly to you as well, but I don't remember what output I'd have made here) for figuring out what's happening.
Assignee | ||
Comment 4•16 years ago
|
||
I am a moron. The reason it didn't work was because the Makefile.in part of the patch had been bitrotted by other testcases I'd added, and blue.png hadn't been copied in. It was a 404, and so onload never got fired on it.
Attachment #364142 -
Flags: review?(vladimir)
Attachment #364142 -
Flags: review?(vladimir) → review+
Assignee | ||
Comment 5•14 years ago
|
||
The test in this bug is bogus, and I'm going to write a correct test as part of bug 552605.
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → INVALID
You need to log in
before you can comment on or make changes to this bug.
Description
•