Losing Browser Events on Firefox (slow user machine or firebug open)

[Phillip Patti]

I am noticing that in my Lift application, with a user on a slow machine using Firefox 41.0.2 (latest as of the time-stamp of this post)(though this has been happening for a while), I am losing some browser events.
Examples include:

• user clicks a button, button doesn't click (no mouse/click events fire in the browser)
  
• user is typing in a text input element, letters typed don't get actually get inputted (no keyUp, keyDown, or keyPress events fire in the browser)

An input/event seems to get lost maybe 1 in 20 times/actions.

I have been reproducing this on a slightly faster machine by having firebug running.

I can reproduce this in the Lift 2.6 Scala 2.10 lift_basic project from http://liftweb.net/download.

• Navigate to lift_basic project
• sbt
• container:start
• navigate to app in Firefox on slow machine or on decent machine with firebug running
• go to Login, Sign Up, or Lost Password pages
• focus on a text input element
• start typing by alternating characters (such as typing: ababababab...)
• notice that sometimes a character input is lost (resulting in something such as: abaabab...)

Alternatively, create a simple page that has a <button /> which flips something, maybe a background color or the visibility of some element, (or prints to the console) on clicking the button. Notice you will lose clicks sometimes (your event/flip/console.log won't fire).

I have been trying to narrow down the issue using super simple pages, and it seems to happen for me once jquery is included on the page.
For example:
an html page in the Lift app consisting of only:

<script type="text/javascript" src="http://code.jquery.com/jquery-2.1.4.min.js"></script>
<input></input>

exhibits the issue. A page in the Lift app of only:

<input></input>

does not exhibit the issue.

Couldn't find related issues in Googling, in the Firefox bug tracker, nor in this Google group.
Any help or advice is much appreciated!

Best,
Phil

[Antonio Salazar Cardozo]

Is *solely* including jquery enough to cause this? i.e., does a page with no Lift JS and

only jquery also exhibit this behavior? If so, I'd say this isn't really in the domain of the

Lift list, as it's an issue unrelated to Lift.

On the other hand, if it only happens when Lift-related stuff is on the page, it'd be useful

to know exactly what Lift-related stuff is there.

Thanks,

Antonio

[Phillip Patti]

I only notice this on a Lift page. A static page (not in a server/web-container) of:

<script type="text/javascript" src="http://code.jquery.com/jquery-2.1.4.min.js"></script>
<input></input>

does not exhibit the behavior. I haven't tried this on other frameworks or web containers, though I will try that later (was leaning towards not trying this out because I haven't seen the issue before during regular web browsing and didn't find any reports of others encountering the issue).

... if it only happens when Lift-related stuff is on the page, it'd be useful

to know exactly what Lift-related stuff is there.

You can reproduce this on a simple Lift hosted web page, consisting of only the jQuery include and the input element (no other elements, no templates, no snippets). On the server, I was just using lift_basic (whatever it was hooking up by default in build.sbt and boot). Not sure if I am answering the question correctly. Just let me know.

Best,
Phil

[Joe Barnes]

Phil,

Set up a github repo with the 2.6 template including some buttons and stuff where you are seeing this so we can be certain we're all looking at the same thing.  Also, include a non-lift static page that you mention doesn't exhibit this behavior.  

Thanks,

Joe

[Phillip Patti]

Hi Joe,

I have created:
https://github.com/ppatti/dropped_browser_events

I included a small Lift app, a small Play app, and a static page.

Best,
Phil

[Peter Petersson]

Hi Phil

Thank you for taking the time to set up the sample project I took it for a spin doing some high speed two finger alternate typing in Lift 
qnqnqnqnqnqnqnqqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnq
with firebug window open showing the console.

Unfortunately I was not able to reproduce what you are describing but maybe my comp is to fast.
Lets see if someone else can reproduce.

best regards Peter

--
--
Lift, the simply functional web framework: http://liftweb.net
Code: http://github.com/lift
Discussion: http://groups.google.com/group/liftweb
Stuck? Help us help you: https://www.assembla.com/wiki/show/liftweb/Posting_example_code

---
You received this message because you are subscribed to the Google Groups "Lift" group.
To unsubscribe from this group and stop receiving emails from it, send an email to liftweb+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Peter Petersson]

hmmm just a though, maybe what you are noticing is due to some GC kicking in, it dose how ever not explain why you not see the same thing in Play

best regards Peter

On 11/02/2015 07:05 PM, Phillip Patti wrote:

--

[Phillip Patti]

Hi Peter, thanks for trying it out!

There are a few missed letters in your post. Did you copy and paste that output from the input field?
qnqnqnqnqnqnqnqqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqnqqnqnqnqnqnqnqnnqnqnqnqnqnqnqnqnqnqnqnq

When you say GC, is that the heartbeat check that Lift does for the session?

Best,

Phil

[Peter Petersson]

Hi Philip

Yes it is the copy and paste from the input, I did not notice them. No I was thinking it could be garbage collection kicking in and interrupting.

/Peter

[Joe Barnes]

We can turn gc off in 2.6, right?

Joe

You received this message because you are subscribed to a topic in the Google Groups "Lift" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/liftweb/Gw919LobyjU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to liftweb+u...@googlegroups.com.

[Phillip Patti]

Added

LiftRules.enableLiftGC = false;

to boot (I think that is the right way to turn off the GC). But the issue is still present.
Hmmm. Hopefully I will get some more time to look at this later this week, but please share ideas if you have any!

Best,
Phil