Serge Gautherie (:sgautherie) Reporter
	Description • 16 years ago

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20090223 SeaMonkey/2.0b1pre] (experimental/_m-c_, home, optim default) (W2Ksp4)
(http://hg.mozilla.org/mozilla-central/rev/f6cdd2d6a9ea
 +http://hg.mozilla.org/comm-central/rev/720d3a1ea63d)

(Not sure why these tests are not run in this build.)

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20090228 SeaMonkey/2.0b1pre] (experimental/_m-c_, home, optim default) (W2Ksp4)
(http://hg.mozilla.org/mozilla-central/rev/f7f62131998d
 +http://hg.mozilla.org/comm-central/rev/7ea34ef19dc4)

{
chrome://mochikit/content/browser/xpinstall/tests/browser_enabled.js
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_enabled.js | installTrigger should have been enabled
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_enabled.js | installTrigger should have been enabled - Got , expected true
chrome://mochikit/content/browser/xpinstall/tests/browser_enabled2.js
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_enabled2.js | Timed out
...
chrome://mochikit/content/browser/xpinstall/tests/browser_localfile2.js
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_localfile2.js | installTrigger should have failed
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_localfile2.js | installTrigger should have failed - Got , expected exception
chrome://mochikit/content/browser/xpinstall/tests/browser_navigateaway.js
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_navigateaway.js | Timed out
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_navigateaway.js | Install should succeed
}

And while the tests runs, I get multiple error dialogs
saying "SeaMonkey could not install the file at .../xpinstall/..."
which I need to manually cancel.

	Serge Gautherie (:sgautherie) Reporter
	Comment 1 • 16 years ago

(In reply to comment #0)
> [Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20090228
> SeaMonkey/2.0b1pre] (experimental/_m-c_, home, optim default) (W2Ksp4)
> (http://hg.mozilla.org/mozilla-central/rev/f7f62131998d
>  +http://hg.mozilla.org/comm-central/rev/7ea34ef19dc4)

And there is a 5th:
{
chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js | Exception thrown - ReferenceError: BrowserOpenAddonsMgr is not defined
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js | Install should succeed
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js | Install should succeed
}


PS:
This test suite reports a new leak (maybe related), but let's fix the failures first...

	Serge Gautherie (:sgautherie) Reporter
	Comment 2 • 16 years ago

(In reply to comment #0)
> [Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20090223
> SeaMonkey/2.0b1pre] (experimental/_m-c_, home, optim default) (W2Ksp4)
> (http://hg.mozilla.org/mozilla-central/rev/f6cdd2d6a9ea
>  +http://hg.mozilla.org/comm-central/rev/720d3a1ea63d)
> 
> (Not sure why these tests are not run in this build.)

Ah:
the tests did land on the 12th, but were disabled after that,
then reenabled on the 24th !

	Frank Wein [:mcsmurf]
	Comment 3 • 16 years ago

What do I have to do with this bug here?

	Dave Townsend [:mossop] Assignee
	Comment 4 • 16 years ago

Most of these tests depend on UI that is present in Firefox, so it isn't much of a surprise that they fail in Seamonkey. I wasn't even aware that Seamonkey ran the browser chrome tests. Is there some standard way to workaround these issues?

	Serge Gautherie (:sgautherie) Reporter
	Comment 5 • 16 years ago

(In reply to comment #3)
> What do I have to do with this bug here?

I would have answered: have a look;
but comment 4 seems to explain it already ;->

(In reply to comment #4)
> Is there some standard way to workaround these issues?

Let's ask KaiRo ;-)

	Robert Kaiser
	Comment 6 • 16 years ago

I don't know about any standard workaround other than not placing Firefox-specific tests in toolkit/core code.

	Serge Gautherie (:sgautherie) Reporter
	Comment 7 • 16 years ago

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20090326 SeaMonkey/2.0b1pre] (experimental/_m-c_, home, optim default) (W2Ksp4)
(http://hg.mozilla.org/mozilla-central/rev/58c11ba7a31d
 +http://hg.mozilla.org/comm-central/rev/de7070b99471)

"Only" 3 failures remain:
{
chrome://mochikit/content/browser/xpinstall/tests/browser_enabled.js
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_enabled.js | installTrigger should have been enabled
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_enabled.js | installTrigger should have been enabled - Got , expected true
chrome://mochikit/content/browser/xpinstall/tests/browser_enabled2.js
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_enabled2.js | Timed out

chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js
TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js | Exception thrown - ReferenceError: BrowserOpenAddonsMgr is not defined
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js | Install should succeed
TEST-PASS | chrome://mochikit/content/browser/xpinstall/tests/browser_opendialog.js | Install should succeed
}

Mossop, could you "ifdef" the (parts of) the tests that are Firefox specific, like
http://mxr.mozilla.org/mozilla-central/search?string=BrowserOpenAddonsMgr&case=on

	Dave Townsend [:mossop] Assignee
	Comment 8 • 16 years ago

(In reply to comment #7)
> Mossop, could you "ifdef" the (parts of) the tests that are Firefox specific,
> like
> http://mxr.mozilla.org/mozilla-central/search?string=BrowserOpenAddonsMgr&case=on

I don't know of anything to ifdef against, but that won't help either, I need to fix these so they don't depend on the UI.

	Dave Townsend [:mossop] Assignee
	Comment 9 • 16 years ago

Basically all of the failures are really because the tests aren't cleaning up after themselves properly. In Firefox this isn't a problem but a quirk in how SeaMonkey's tabbrowser operates means it is there.

I've also removed the dependency on the notification bar, although SeaMonkey emulates it well enough it is probably cleaner to keep these tests as backend as possible, besides it makes for some better tests I think.

The tests still leave a spew of errors in the console under SeaMonkey. This is because the browser-notificationbox there doesn't get destroyed when tabs are closed, however since the binding is detached they no longer have the observer method that is called when the xpinstall-install-blocked event occurs. This is something you probably want to fix in SeaMonkey.

	Dave Townsend [:mossop] Assignee
	Comment 10 • 16 years ago

SeaMonkey is using some non-standard urls for the xpinstall progress dialog which causes the tests to fail.

	Robert Kaiser
	Comment 11 • 16 years ago

(In reply to comment #9)
> Basically all of the failures are really because the tests aren't cleaning up
> after themselves properly. In Firefox this isn't a problem but a quirk in how
> SeaMonkey's tabbrowser operates means it is there.

I basically suspected the first part, the second part might mean that SeaMonkey can/should do better there. As you know what's really up there, could you please file a bug against SeaMonkey Tabbed Browsing so we know to possibly look into that?

	Dave Townsend [:mossop] Assignee
	Comment 12 • 16 years ago

(In reply to comment #11)
> (In reply to comment #9)
> > Basically all of the failures are really because the tests aren't cleaning up
> > after themselves properly. In Firefox this isn't a problem but a quirk in how
> > SeaMonkey's tabbrowser operates means it is there.
> 
> I basically suspected the first part, the second part might mean that SeaMonkey
> can/should do better there. As you know what's really up there, could you
> please file a bug against SeaMonkey Tabbed Browsing so we know to possibly look
> into that?

I've filed bug 485688, though I'm not sure whether there is a lot that can be done to help.

	Ian Neal
	Comment 13 • 16 years ago

Comment on attachment 369672 [details] [diff] [review]
seamonkey prefs patch

Kairo probably knows the code this affects better than me

	Serge Gautherie (:sgautherie) Reporter
	Comment 14 • 16 years ago

(In reply to comment #7)
> "Only" 3 failures remain:

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20090328 SeaMonkey/2.0b1pre] (experimental/_m-c_, home, optim default) (W2Ksp4)
(http://hg.mozilla.org/mozilla-central/rev/262d44d6e425
 +http://hg.mozilla.org/comm-central/rev/054adad6539d)

I get 5 again.
Anyway...

	Robert Kaiser
	Comment 15 • 16 years ago

Comment on attachment 369672 [details] [diff] [review]
seamonkey prefs patch

We must have copied those URLs from somewhere, possibly Firefox used them some time ago but has since moved to the ones without "?view=installs"... In any case, thanks for looking into this bug!

	Dave Townsend [:mossop] Assignee
	Comment 16 • 16 years ago

Landed:

http://hg.mozilla.org/mozilla-central/rev/bdf78f1e12cd
http://hg.mozilla.org/comm-central/rev/6e63890cc878

	Dave Townsend [:mossop] Assignee
	Comment 17 • 16 years ago

Landed on the 1.9.1 branch:

http://hg.mozilla.org/releases/mozilla-1.9.1/rev/786b02938a98

	Serge Gautherie (:sgautherie) Reporter
	Comment 18 • 16 years ago

[Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.9.2a1pre) Gecko/20090417 SeaMonkey/2.0b1pre] (experimental/_m-c_, home, optim default) (W2Ksp4)
(http://hg.mozilla.org/mozilla-central/rev/9e06217fc471
 +http://hg.mozilla.org/comm-central/rev/...)

V.Fixed

	Serge Gautherie (:sgautherie) Reporter
	Comment 19 • 16 years ago

(In reply to comment #1)
> PS:
> This test suite reports a new leak (maybe related), but let's fix the failures
> first...

I filed bug 489077.