198036 - rh 9.0: Midas demo not fully functional and displays alert: "This demo is not supported on your level of mozilla" under Red Hat 8.1 beta (Phoebe)

Status: RESOLVED FIXED

(Core :: DOM: Editor, defect)

Description

[Michiel Toneman]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3) Gecko/20030313
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3) Gecko/20030313

Under Red Hat 8.1 beta (Phoebe 8.0.94), midas doesn't work as expected and
displays an alert: "This demo is not supported on your level of mozilla". The
exact same RPM, when installed on Red Hat 8.0 works fine. 

Creating a brand new profile did not solve the problem for me.

Reproducible: Always

Steps to Reproduce:
1.
2.
3.




[michiel@dhcp-43 michiel]$ rpm -qi mozilla
Name        : mozilla                      Relocations: /usr
Version     : 1.3                               Vendor: Red Hat, Inc.
Release     : 0_rh8_gtk2                    Build Date: Fri 14 Mar 2003 05:44:11
AM CET
Install Date: Mon 17 Mar 2003 12:01:25 PM CET      Build Host:
sylvester.devel.redhat.com
Group       : Applications/Internet         Source RPM:
mozilla-1.3-0_rh8_gtk2.src.rpm
Size        : 26849032                         License: MPL
Signature   : (none)
Packager    : Red Hat, Inc. <http://bugzilla.redhat.com/bugzilla>
Summary     : A Web browser.
Description :
Mozilla is an open-source Web browser, designed for standards
compliance, performance, and portability.

Comment 1

[kinmoz]

--> brade

Comment 2

[Mike Kaply [:mkaply]]

Is there anything on the Javascript console?

Comment 3

[Christopher Blizzard (:blizzard)]

I've seen this too and I wonder why it happens.  I would have figured that it
would have worked in 1.3.

Is there something missing from the package manifests?  How about something
missing from the default prefs?  The rpms are driven the the package manifest
files so if there's a problem it's usually there.

Comment 4

[Mike Kaply [:mkaply]]

Here is the relevant code:

  document.getElementById('edit').contentWindow.document.designMode = "on";
  try {
    document.getElementById('edit').contentWindow.document.execCommand("undo",
false, null);
  }  catch (e) {
    alert("This demo is not supported on your level of Mozilla.");
  }

Is it possible undo is failing on redhat for some other reason?

Could you try bringing up composer and seeing if undo works at all?

Also, we can remove the catch block and then see what the console says.

Comment 5

[Christopher Blizzard (:blizzard)]

Undo works.  And there are no messages on my JS console.

Comment 6

[Kathleen :Brade]

confirming based on comment 3

I'd look for a missing DLL such as: composer, txmgr or similar

There are no "extra" files such as css or html or similar to allow midas to
work; only a missing dll would seem to cause the problem.  If the composer
application works (including) undo, I'm really stumped why this would be a
problem unless there is some capability pref difference?

Comment 7

[Christopher Blizzard (:blizzard)]

Error: uncaught exception: [Exception... "Component returned failure code:
0xbfffc558 [nsIDOMNSHTMLDocument.execCommand]"  nsresult: "0xbfffc558
(<unknown>)"  location: "JS frame :: file:///tmp/Midas%20Demo.html :: Start ::
line 214"  data: no]

Comment 8

[Christopher Blizzard (:blizzard)]

What's the entry point for the undo command so I can start debugging?

Comment 9

[Michiel Toneman]

This problem is also present on a pre-final, non-gtk2, non-xft build dated 7
March on Red Hat 8.1 beta. Again, it works fine on Red Hat 8.0 using this build.

Mozilla 1.3
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3) Gecko/20030307

So I think we can rule out XFT and GTK2 as the culperates.

Random guess:

Could this be a compiler/glibc issue? Phoebe uses glibc 2.3.1.

Comment 10

[Mike Kaply [:mkaply]]

http://lxr.mozilla.org/seamonkey/source/editor/libeditor/base/nsEditorCommands.cpp#86

see if you get there and then step into the Undo.

Maybe it is something dumb like Undo throws an error if there is nothing to
Undo. For some reason this wasn't happening on Windows.

Comment 11

[Mike Kaply [:mkaply]]

Attachment: Fix - uninitialized variable. argh.

Comment 12

[Mike Kaply [:mkaply]]

r=blizzard (since it was your idea :), sr=kin

Fix checked in.

Comment 13

[kinmoz]

*** Bug 198554 has been marked as a duplicate of this bug. ***

Comment 14

[Kathleen :Brade]

*** Bug 198556 has been marked as a duplicate of this bug. ***

Comment 15

[Mike Kaply [:mkaply]]

*** Bug 198013 has been marked as a duplicate of this bug. ***

Comment 16

[Tracy Walker [:tracy]]

mass remove verifyme requests greater than 4 months old