603031 - [SeaMonkey] mochitest-chrome: test_idcheck.xul reports warnings/errors/assertions for ChatZilla

Status: NEW

(Other Applications :: ChatZilla, defect)

Description

[Serge Gautherie (:sgautherie)]

I noticed this on my local Windows 2000 with an Opt tinderbox-build:
{{
cz: Initializing ChatZilla {
cz:   Global History failed to initialize: TypeError: Components.classes[GHIST_CONTRACTID] is undefined
cz: } 0.137 sec
cz: Shutting down ChatZilla.
}}

(This test has not been running recently, but previously it reported:)
http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1285355957.1285360820.6315.gz&fulltext=1
Linux comm-central-trunk debug test mochitest-other on 2010/09/24 12:19:17
{{
cz: Initializing ChatZilla {
JavaScript strict warning: chrome://chatzilla/content/menus.js, line 576: reference to undefined property window[labelString + "_ACCESSKEY"]
JavaScript strict warning: chrome://chatzilla/content/static.js, line 256: reference to undefined property Components.classes[GHIST_CONTRACTID]

cz:   Global History failed to initialize: TypeError: Components.classes[GHIST_CONTRACTID] is undefined

JavaScript error: , line 0: Permission denied for <moz-safe-about:blank> to call method XULElement.QueryInterface on <>.
###!!! ASSERTION: JS failed without setting an exception!: 'JS_IsExceptionPending(cx)', file /builds/slave/comm-central-trunk-linux-debug/build/mozilla/js/src/xpconnect/src/xpcwrappedjsclass.cpp, line 329
...

and more
}}

Hopefully, these can be fixed (or documented).
Though I have no idea what these mean wrt running CZ by itself.

Comment 1

[Serge Gautherie (:sgautherie)]

[Mozilla/5.0 (Windows NT 5.0; rv:2.0b8pre) Gecko/20101127 Firefox/4.0b8pre SeaMonkey/2.1b2pre] (comm-central-trunk-win32/1290878667)
{
cz: Initializing ChatZilla {
cz:   Global History failed to initialize: TypeError: Components.classes[GHIST_CONTRACTID] is undefined
cz:   can't find frame for window (stop)
cz: } 0.137 sec
cz: Shutting down ChatZilla.
}

http://mxr.mozilla.org/mozilla-central/search?string=global-history%3B1&case=1&find=%2Fbase%2FnsIGlobalHistory.idl
/docshell/base/nsIGlobalHistory.idl
    * line 73 -- "@mozilla.org/browser/global-history;1"

http://mxr.mozilla.org/comm-central/search?string=GHIST_CONTRACTID&case=1&find=%2Firc%2Fxul%2Fcontent%2Fstatic.js
/mozilla/extensions/irc/xul/content/static.js
    * line 254 -- const GHIST_CONTRACTID = "@mozilla.org/browser/global-history;1";
    * line 256 -- Components.classes[GHIST_CONTRACTID].getService(nsIGlobalHistory);

Bug still there, but I don't know why the class is unavailable.

Comment 2

[Serge Gautherie (:sgautherie)]

(In reply to comment #1)

I evaluated
Components.classes["@mozilla.org/browser/global-history;1"].getService(Components.interfaces.nsIGlobalHistory)
in the error console.

SM 2.0.10,
FF 3.6.14pre:
{
[xpconnect wrapped nsIGlobalHistory]
}

FF 4.0b8pre, SM 2.1b2pre:
{
Error: Components.classes['@mozilla.org/browser/global-history;1'] is undefined
Source File: javascript:%20Components.classes["@mozilla.org/browser/global-history;1"].getService(Components.interfaces.nsIGlobalHistory)
Line: 1
}

I filed bug 615213 !

Comment 3

[Serge Gautherie (:sgautherie)]

(In reply to comment #2)

And I filed (unrelated) bug 615219
"[ChatZilla] Start using nsIGlobalHistory2, maybe stop support for nsIGlobalHistory".

Comment 4

[Serge Gautherie (:sgautherie)]

(In reply to comment #3)
> And I filed (unrelated) bug 615219

Per bug 615213 comment 2, s/unrelated/blocking/ :->

Comment 5

[:Gijs (he/him)]

I don't understand this bug. What does the test check? What is the actual issue?

Comment 6

[Justin Wood (:Callek)]

The test itself is independant of these issues really, just hitting them.

This is also the (currently) only automated test we have that even tries to use ANY bits of cZ.

What it does is tests each .xul window for duplicate id's being present. The main reason the test was created was due to our (SeaMonkey) current prefwindow having tons of overlays and _all_ panes being part of the single window, so we test to be sure that id's in the window don't overlap and cause unforeseen issues.

It was relatively easy to add support for other xul windows (such as cZ) and that is what it is trying to test here. However we get these (other) issue's that serge is filing bugs for.

So more or less *this* bug has very little direct work for the cZ team, but its blockers do require attention.

Comment 7

[Serge Gautherie (:sgautherie)]

Bug 615219 is now fixed, but m-c test suite currently fails before running this test :-|