1181287 - Getting inconsistent mouse click events in Silverlight with FF39

Status: RESOLVED WONTFIX

(Core Graveyard :: Plug-ins, defect, P5)

Description

[Eric]

Attachment: SilverlightMouseTest.zip

User Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/43.0.2357.130 Safari/537.36

Steps to reproduce:

I received a bug report from my support team for a Silverlight application when running under Firefox 39.  In all areas of the application where we detect mouse click events, the system will sometimes receive several LeftButtonDown events.  I created a small demonstration application (attached).  The demo has a simple grid of text items on the top and a list box of mouse events on the bottom.  Left-click in the top grid and notice that the log shows 2 button events and a single button up event.

Continued clicking in the grid usually works fine.  However, if you click in either of Firefox's text boxes (web address or search) and then click back onto the grid, you will see two (or more) button down events with a single button up once again.


Actual results:

This is affecting my application in a variety of ways.  For data grids, my application will sometimes behave as if the user double-clicked a row when they only single clicked it.  This causes unintended actions to occur.  There are other effects of this throughout my application, but this is the easiest to describe.


Expected results:

I've tested this in Chrome and IE, and those browsers appear to be working fine.  Also, this problem did not appear to manifest in earlier versions of Firefox.  We only started seeing this problem with version 39.

Comment 1

[Eric]

Addendum - this issue appears on the Windows version of Firefox.  The issue does not seem to occur on the Mac version.

Comment 2

[Gingerbread Man]

(In reply to Eric from comment #0)
> Also, this problem did not appear to manifest in earlier versions of
> Firefox.  We only started seeing this problem with version 39.

Could you find the regression range?

mozregression --good-release 38 --bad-release 39
http://mozilla.github.io/mozregression/

Comment 3

[Eric]

(In reply to Gingerbread Man from comment #2)
> (In reply to Eric from comment #0)
> > Also, this problem did not appear to manifest in earlier versions of
> > Firefox.  We only started seeing this problem with version 39.
> 
> Could you find the regression range?
> 
> mozregression --good-release 38 --bad-release 39
> http://mozilla.github.io/mozregression/

Thanks for the quick response, Benjamin!

I tried this in 38 and 38.1.  Both of those versions worked normally.  It appears that 39 is where the issue was introduced.

Comment 4

[Benjamin Smedberg]

Eric, yes. What we're looking for now is a regression range in nightly builds between Firefox 38 and 39. The mozregression tool can help you find the regression range.

Comment 5

[Eric]

(In reply to Benjamin Smedberg  [:bsmedberg] from comment #4)
> Eric, yes. What we're looking for now is a regression range in nightly
> builds between Firefox 38 and 39. The mozregression tool can help you find
> the regression range.

Sorry about the misunderstanding, Benjamin.  I used the regression tool and it looks like the last good nightly build was 3-31-2015.  The first bad build was 4-1-2015.

Comment 6

[Benjamin Smedberg]

Hrm. That would be https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=8af276ab8636&tochange=0b88606f8fe7

But I don't see anything suspicious in that range :-(

Perhaps obviously, focus handling is the primary suspect here. Since NPAPI plugins are a legacy technology, we're probably not going to spend the time to diagnose or fix this before we start phasing out NPAPI completely.

Comment 7

[Eric]

(In reply to Benjamin Smedberg  [:bsmedberg] from comment #6)
> Hrm. That would be
> https://hg.mozilla.org/mozilla-central/
> pushloghtml?fromchange=8af276ab8636&tochange=0b88606f8fe7
> 
> But I don't see anything suspicious in that range :-(
> 
> Perhaps obviously, focus handling is the primary suspect here. Since NPAPI
> plugins are a legacy technology, we're probably not going to spend the time
> to diagnose or fix this before we start phasing out NPAPI completely.

Sorry to hear that you weren't able to see a likely culprit.  I certainly understand not wanting to spend too much time with an isolated plug-in bug.  We are working on moving off of Silverlight, but it's going to take us some time to complete the transition.  I'll let our support staff know that we need to have our customers move off of Firefox until we get our systems converted.

Thank you for your looking into this, Benjamin.  I do appreciate your time and effort.

Comment 8

[Benjamin Smedberg]

Heh ok. Which browser are you going to recommend instead? Microsoft Edge and Chrome both won't run Silverlight either.

Comment 9

[Eric]

(In reply to Benjamin Smedberg  [:bsmedberg] from comment #8)
> Heh ok. Which browser are you going to recommend instead? Microsoft Edge and
> Chrome both won't run Silverlight either.

Ya, it's a big problem for us :).  We're moving as quick as we can.

Comment 10

[Virtual_ManPL [:Virtual] 🇵🇱 - (please needinfo? me - so I will see your comment/reply/question/etc.)]

I'm marking this bug as WONTFIX per bug #1269807.

For more information see - https://blog.mozilla.org/futurereleases/2015/10/08/npapi-plugins-in-firefox/