Elm Textfield + a barcode scanner

[Chris Van Vranken]

I'm using a barcode scanner to enter data into an elm textfield.  The barcode scanner basically works like a keyboard with a user that makes very rapid keystrokes.  This rapid entry causes random characters to be missing in the textfield.  If the scan is supposed to enter "50494'', one or more of these characters will often be missing.  Using the debugger I see that my model has values for that field that change from "5" to "50" to "504" to "509" to "5094", but it should be "5" to "50" to "504" to "5049" to "50494"  So it ends up replacing "504" with "509" instead of with "5049".  I believe this is a threading issue where two events are firing in parallel instead of in sequence, and both events are starting with the same model data ("50") with the output of one event replacing the output of the other, when instead the output model for one event should be the input for the next event.  Is this a bug in Elm?  How do I fix this?

Thanks,

Chris

[Simon]

not sure whether this is relevant but can you take advantage of the difference between the `oninput` and `onchange` events in a text field?

[Chris Van Vranken]

I'm using the oninput event, the onchange event occurs when the element loses focus.  The contents of the textbox are wrong before the onchange event even fires, so I don't believe the onchange event will be helpful for solving this.  I think maybe I need to prevent elm from modifying the value of the textbox and then read in the final value when the scanner sends the enter key, which it does every time it finishes scanning the barcode.

[Bernardo]

You should definitely try with onchange. But without using oninput because otherwise the field is updated from the model. If you update the model after the field changes there shouldn't be any difference from using a plain html input.

[Francesco Orsenigo]

This is an interesting problem and I wonder if it would happen also with some other form of automated input (say, selenium).

It might very well be a problem with Elm.

[Witold Szczerba]

This was my case with Selenium WebDriver and Chrome. The end-to-end tests of my Elm application are failing because of this.

It was reported back in 2014: https://github.com/elm-lang/elm-compiler/issues/835

The issue is closed, but the bug is still present.

You want to know my workaround? It "just-works" on Firefox, so this is the browser I use for e2e tests…

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

[Francesco Orsenigo]

Witold, could you put together a minimal Elm app + Selenium script that reproduces the problem?
This might be a significant issue, since it would impact the possibility to end-to-end test Elm apps.

--
You received this message because you are subscribed to a topic in the Google Groups "Elm Discuss" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/elm-discuss/Oy35n_BHUGM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to elm-discuss+unsubscribe@googlegroups.com.

[Noah Hall]

A known issue: https://github.com/elm-lang/html/issues/55

The issue is elm-compiler was likely closed due to the fact that it is
not a compiler problem, it is a web-browser problem.

`defaultValue` is the recommended approach

>>> email to elm-discuss...@googlegroups.com.

>>> For more options, visit https://groups.google.com/d/optout.
>>
>>
>> --
>> You received this message because you are subscribed to a topic in the
>> Google Groups "Elm Discuss" group.
>> To unsubscribe from this topic, visit
>> https://groups.google.com/d/topic/elm-discuss/Oy35n_BHUGM/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to

>> elm-discuss...@googlegroups.com.

>> For more options, visit https://groups.google.com/d/optout.
>
>
> --

> You received this message because you are subscribed to the Google Groups
> "Elm Discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to elm-discuss...@googlegroups.com.

[Witold Szczerba]

My issue was with regular input type=text fields.

I will try to find some time to create a project reproducing the issue.

>>> email to elm-discuss+unsubscribe@googlegroups.com.

>>> For more options, visit https://groups.google.com/d/optout.
>>
>>
>> --
>> You received this message because you are subscribed to a topic in the
>> Google Groups "Elm Discuss" group.
>> To unsubscribe from this topic, visit
>> https://groups.google.com/d/topic/elm-discuss/Oy35n_BHUGM/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to

>> elm-discuss+unsubscribe@googlegroups.com.

>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "Elm Discuss" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to elm-discuss+unsubscribe@googlegroups.com.

> For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Elm Discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to elm-discuss+unsubscribe@googlegroups.com.

[Witold Szczerba]

Here it is:

https://github.com/witoldsz/elm-webdriver-problem

And the Reddit topic:

https://www.reddit.com/r/elm/comments/693v43/elmwebdriverproblem/

I hope it will draw a little bit more attention to the issue.

[Matthias Winkelmann]

I have to say, the way issues are managed in the elm ecosystem has frequently left me stumped because i couldn’t find the relevant issue. This problem was reported at least twice, in https://github.com/elm-lang/elm-compiler/issues/727 and https://github.com/elm-lang/elm-compiler/issues/835. Those two threads were closed with a reference to https://github.com/elm-lang/core/issues/332, which is an aggregation of 20+ tickets. That, in turn, was closed without any further activity a year later–with yet another reference to https://github.com/evancz/elm-graphics/issues/4, at which point the ticket no longer has anything to do with the original issues.

Most of these meta-tickets also discourage comments, asking people to open new issues for those. If people do so, chances are low they’ll manage to mention all the relevant tickets, so nobody would ever have a chance to find that information. 

Every other project simply has a ticket per problem. Duplicates get closed with a reference. Issues that can’t be reproduced get closed. Comments often contain useful workarounds, or partial fixes slowly building to a resolution. 

I can imagine that a three-digit number of open issues feels daunting, but I have never looked at a project on github with 1000+ open issues and thought worse of the maintainers. Almost universally, the number of issues is proportional to interest in the project (see tensorflow or Visual Studio Code). 

To unsubscribe from this group and stop receiving emails from it, send an email to elm-discuss...@googlegroups.com.

[Witold Szczerba]

You are so right. I have no idea how is this system supposed to work. 

To unsubscribe from this group and stop receiving emails from it, send an email to elm-discuss+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Rehno Lindeque]

In https://runelm.io/c/nx0 it looks like the trouble is that it is using value attribute no? Just to check, have you tried the defaultValue approach suggested? At least, in the example it doesn't look like value needs to be set at all.

My understanding is that two-way binding is problematic (there's some kind of race with the internal state of the control).

PS I think it would be interesting for someone to create a pure div control for text input some time (but that definitely way more challenging than it sounds).