Ed Morley [:emorley] Reporter
	Description • 12 years ago

Mozilla/5.0 (Windows NT 6.1; WOW64; rv:16.0) Gecko/16.0 Firefox/16.0a1
http://hg.mozilla.org/mozilla-central/rev/7e4c2abb9fc9

1) Go to https://etherpad.mozilla.org/test-clear-authorship
2) Type some text
3) Press the "Clear Authorship Colors" button on the toolbar

Expected:
Colour highlight on the text you typed disappears.

Actual:
No colour change and this in the error console...

Timestamp: 08/06/2012 10:15:00
Error: TypeError: D is null
Source File: https://etherpad.mozilla.org/test-clear-authorship
Line: 471

Works fine on Chrome, so not sure if this is platform or Etherpad.

	Ed Morley [:emorley] Reporter
	Comment 1 • 12 years ago

mozilla-central...
Last good nightly: 2012-01-16
First bad nightly: 2012-01-17
Pushlog: 
http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=047c8ba7d2e4&tochange=34572943a3e4

mozilla-inbound...
Last good nightly: 2012-01-12
First bad nightly: 2012-01-13
Pushlog: 
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=0dbdc8243364&toch
ange=4a9747a88eab

Moving to Core.

	Boris Zbarsky [:bzbarsky]
	Comment 2 • 12 years ago

Hmm.  This is working for me on Mac with a nightly from a few days ago....

	Alice0775 White
	Comment 3 • 12 years ago

I can reproduce in 
http://hg.mozilla.org/mozilla-central/rev/dc410944aabc
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:16.0) Gecko/16.0 Firefox/16.0a1 ID:20120610030528

The error appears and the command does not work if you do not select any text before click the "Clear Authorship Colors" button.
However, No error appears if some text selected.


Regression window(m-i)
Good:
http://hg.mozilla.org/integration/mozilla-inbound/rev/8c24766efc04
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:12.0a1) Gecko/20120112 Firefox/12.0a1 ID:20120112134124
Bad:
http://hg.mozilla.org/integration/mozilla-inbound/rev/d41fbe450000
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:12.0a1) Gecko/20120112 Firefox/12.0a1 ID:20120112140137
Pushlog:
http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=8c24766efc04&tochange=d41fbe450000

Triggered by:
899a12aeff6c	Boris Zbarsky — Bug 716793. Dispatch synthetic mousemove off the refresh driver, not as fast as we can. r=roc We use Flush_Display here because mousemoves flush out layout, so we want to do the synthetic one after we've done our normal layout flushing

	Ed Morley [:emorley] Reporter
	Comment 4 • 12 years ago

Thank you! :-)

	Boris Zbarsky [:bzbarsky]
	Comment 5 • 12 years ago

Weird.  Is this Windows-specific, then?  Why would changing when synthetic mousemoves fire affect JS execution in any way????

	Alice0775 White
	Comment 6 • 12 years ago

I can reproduce on Ubuntu10.40
Mozilla/5.0 (X11; Linux i686; rv:16.0) Gecko/16 Firefox/16.0a1 ID:20120612030527

	Alice0775 White
	Comment 7 • 12 years ago

(In reply to Boris Zbarsky (:bz) from comment #5)
> Weird.  Is this Windows-specific, then?  Why would changing when synthetic
> mousemoves fire affect JS execution in any way????
One more strange thing
setting prompts.tab_modal.enabled = false helps

	Boris Zbarsky [:bzbarsky]
	Comment 8 • 12 years ago

That's really odd.  I have no idea why this is being an issue.

	Alice0775 White
	Comment 9 • 12 years ago

Local built(backed Bug 716793 out from tip) fixed the issue...

	Timothy Nikkel (:tnikkel)
	Comment 10 • 12 years ago

I don't see the bug on Mac but I do see it on Linux.

A simpler testcase might be helpful here.

	Alice0775 White
	Comment 11 • 12 years ago

See several demo site of Etherpad http://etherpad.org/public-sites/
mopad(etherpad.mozilla.org/ – Etherpad V1 hosted by Mozilla)does not work only .... I guess mopad using something older library?

	Ed Morley [:emorley] Reporter
	Comment 12 • 11 years ago

Has worked for some time, forgot to update the bug.