1273346 - [e10s] Additional cursor is always displayed in codemirror instances

Status: VERIFIED DUPLICATE of bug 1276467

(Core :: Layout, defect)

Description

[arni2033]

>>>   My Info:   Win7_64, Nightly 49, 32bit, ID 20160511030221
STR_1:
1. Open https://www.mozilla.org/en-US/firefox/nightly/firstrun/
2. Open style editor (Shift+F7). Resize toolbox to make stylesheet scrollable
3. Scroll stylesheet to the bottom with mouse wheel

STR_2:
1. Open scratchpad, type "window"

AR:  Additional cursor is displayed in both cases
ER:  No additional cursor

This is regression from bug 1265237. Regression range:
> https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=cc9051869d1a9078a821da5d1639299f9e836e0e&tochange=1ee604ae873ffae665e745b3553fb1c406dbdd6a

Comment 1

[arni2033]

Attachment: screenshot 2 - bug 1273346 STR_2.png

Comment 2

[Timothy Nikkel (:tnikkel)]

Does the screenshot show the bug? I can't see a problem in the screenshot. I tried the STR I couldn't see what the problem was. Maybe a little more detail, a screenshot of the problem?

Comment 3

[arni2033]

> I can't see a problem in the screenshot.
Well, now I can see why there're hundreds of bugs in Firefox...
Screenshot is completely relevant. There's 1 white cursor and 1 black cursor. Is it reproducible now?

STR_3:  type "window" in scratchpad -> right-click between "n" and "d" -> press Escape

Comment 4

[arni2033]

Attachment: screenshot 1 - bug 1273346 STR_1.png

Comment 5

[arni2033]

Attachment: screenshot 3 - bug 1273346 STR_3.png

Comment 6

[Timothy Nikkel (:tnikkel)]

(In reply to arni2033 from comment #3)
> > I can't see a problem in the screenshot.
> Well, now I can see why there're hundreds of bugs in Firefox...
> Screenshot is completely relevant. There's 1 white cursor and 1 black
> cursor. Is it reproducible now?

There's no need to get hostile. Please be polite. You'll note that I didn't say you were wrong, I didn't say the bug didn't exist. I simply asked for help in seeing and reproducing the bug.

Since you said "an additional cursor" without any mention that it was a different color I was looking for another white cursor. Further, because it was very late at night and I use Flux (https://justgetflux.com/) the black cursor was almost identical in color to the dark background. Further to that someone (perhaps you?) had marked the comment where that screenshot was uploaded as spam, making me suspect that a mistake had been made uploading the wrong image which didn't show the problem. And I tried both of your STR and I couldn't see the problem.

Comment 7

[Timothy Nikkel (:tnikkel)]

I tried all three STR on mac, and on Windows (32bit and 64bit), I couldn't reproduce in any case.

Comment 8

[arni2033]

It's still unclear, what OS you've tested. My impression is that Firefox developers always use the best hardware and the latest versions of drivers and OS (i.e. Windows 10).
So it's better to test Firefox in safe mode (to be sure) on Win7 (SP 1).
But your results are strange anyway, because this bug is 100% reproducible, unlike some of the others.

Comment 9

[Patrick Brosset <:pbro>]

I'm not sure if bug 1276467 is a duplicate, but it does sound very similar. On Mac, the additional cursor receives a blue outline that looks like a focus style. See https://bug1276467.bmoattachments.org/attachment.cgi?id=8757888

Comment 10

[arni2033]

Why are you unsure? Bug 1276467 just duplicates all scenarios mentioned here. Cursor has blue outline
on Mac, because CodeMirror uses textarea for second cursor (native textareas on Mac have blue outline)

Comment 11

[Patrick Brosset <:pbro>]

(In reply to arni2033 from comment #10)
> Why are you unsure? Bug 1276467 just duplicates all scenarios mentioned
> here. Cursor has blue outline
> on Mac, because CodeMirror uses textarea for second cursor (native textareas
> on Mac have blue outline)
I wasn't sure especially because of the blue outline which wasn't mentioned here. And I didn't know codemirror used a textarea for the additional cursor.

Comment 12

[Milan Sreckovic [:milan] (needinfo for best results)]

Timothy, would you expect your patch from bug 1276467 to deal with this?

Comment 13

[Timothy Nikkel (:tnikkel)]

(In reply to Milan Sreckovic [:milan] from comment #12)
> Timothy, would you expect your patch from bug 1276467 to deal with this?

Yes!

Comment 14

[Mike Conley (:mconley) (:⚙️)]

Bug 1276467 has landed on central and should be in Nightly. arni2033, do you have a moment to confirm that this bug is fixed with a recent Nightly?

Comment 15

[Astley Chen (inactive)]

The regression has been fixed on Nightly '50.0a1 (2016-07-14)'.
But it's still seen on Aurora '49.0a2 (2016-07-14)' and Beta 7 '48.0b7'.

As I know, Timothy'd requested uplift to beta and aurora. It should be fixed after uplift.

Hi Timothy, could you help to confirm whether this bug is duplicate of bug 1276467 ? Thanks.

Comment 16

[Astley Chen (inactive)]

Tested and update result on comment 15.

Comment 17

[Timothy Nikkel (:tnikkel)]

(In reply to Astley Chen [:astley] (UTC+8) from comment #15)
> Hi Timothy, could you help to confirm whether this bug is duplicate of bug
> 1276467 ? Thanks.

Yes, this bug should be a dupe of bug 1276467.

Comment 19

[arni2033]

This bug doesn't occur anymore
Tested on:  Win7_64, Nightly 50, 32bit, ID 20160714030208 (2016-07-14)

Comment 20

[[:philipp]]

in bug 1276467 the fix has landed for 48/49 as well.