Closed
Bug 1026254
Opened 10 years ago
Closed 10 years ago
No error page when trying to load a chrome: URL using an unregistered package
Categories
(Toolkit :: Startup and Profile System, enhancement)
Tracking
()
RESOLVED
FIXED
mozilla33
Tracking | Status | |
---|---|---|
firefox33 | --- | fixed |
People
(Reporter: neil, Assigned: neil)
References
()
Details
Attachments
(1 file, 1 obsolete file)
3.03 KB,
patch
|
benjamin
:
review+
|
Details | Diff | Splinter Review |
If you try to load a chrome: URL for a missing file in a registered package, you get the "Problem loading page" display. But if you try to load a chrome: URL for a missing package, nothing happens. This is because the chrome protocol returns NS_ERROR_UNEXPECTED in this case. We can make it return NS_ERROR_FILE_NOT_FOUND instead which will allow the doc shell to display an error page. There are a few bugs reporting hidden blank chrome documents showing up in the Window menu; if these are actually caused by invalid chrome: URLs then the error page would give us more information. (It should then be possible to scrape the error page using DOM Inspector for example.)
Assignee | ||
Comment 1•10 years ago
|
||
> NS_ERROR_UNEXPECTED
Actually NS_ERROR_FAILURE or NS_ERROR_NOT_AVAILABLE, depending.
Assignee | ||
Comment 2•10 years ago
|
||
Assignee | ||
Comment 3•10 years ago
|
||
Try tells me that we have a test for the error code, so I'll need to patch that too.
Assignee | ||
Updated•10 years ago
|
Attachment #8441035 -
Attachment description: 1026254.diff → Proposed patch
Assignee | ||
Comment 4•10 years ago
|
||
Attachment #8441035 -
Attachment is obsolete: true
Attachment #8441035 -
Flags: review?(benjamin)
Attachment #8441200 -
Flags: review?(benjamin)
Updated•10 years ago
|
Attachment #8441200 -
Flags: review?(benjamin) → review+
Assignee | ||
Comment 5•10 years ago
|
||
https://hg.mozilla.org/integration/mozilla-inbound/rev/0b04aefdd3e3
Comment 6•10 years ago
|
||
This landed with the wrong bug number - and the bug number it uses is a private bug, so I can't comment there to point this way. I don't suppose you could backout and reland (in one push, as two commits, and with DONTBUILD in the topmost commit), so hg blame points here? :-) https://hg.mozilla.org/mozilla-central/rev/0b04aefdd3e3
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
status-firefox33:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → mozilla33
Assignee | ||
Comment 7•10 years ago
|
||
This was backed out as changeset b85caf0e485a because the patch for bug 1026008 caused leaks. Re-landed as https://hg.mozilla.org/integration/mozilla-inbound/rev/c843ae1b1eb2
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Comment 8•10 years ago
|
||
https://hg.mozilla.org/mozilla-central/rev/c843ae1b1eb2
Status: REOPENED → RESOLVED
Closed: 10 years ago → 10 years ago
Flags: in-testsuite+
Resolution: --- → FIXED
You need to log in
before you can comment on or make changes to this bug.
Description
•