Gantt widget - from version R2 2022 - change event triggers on everything

[Vedad]

Hi,

I am trying to upgrade from R3 2021 to latest version of the library. 

However, I noticed that from version 2022.1.301 - whatever happens on gantt list, change event is triggered, what before wasn't case.

For example, if I select task A, and try to collapse another phase, change is triggered, selection of task A is lost.

Behavior is reproducible in the following Dojo, just change library version to any before 2022.1.301  to see old behavior and to that one or after for a new behavior.

If new behavior is not a bug, please could you advise on how to persist selection of the task on these side actions (collapse phase for example).

If you need any info, I will be happy to help.

Regards,

Vedad

[Martin]

Hello Vedad,

Unfortunately I am not sure I see a difference in the behaviour with an older version ( I went back to 2021 R1 ). The change event triggers the same way and the selection is always lost when expanding/collapsing a  Task. Could you please let me know if I am missing a step to observe the problem?

[Vedad]

Hello Martin, 

Please try to do the following and observe log:
1. Select any task in the gantt

2. Click the arrow to collapse phase

3. Change the library to 2022.1.119, run the dojo

4. Repeat steps 1 and 2 - observe sequence of events in the log

If you use latest library version, log looks like this:

log

Gantt data bound

Gantt change triggered ::

Gantt data binding

Gantt selection change :: UI and Interaction

Gantt change triggered ::

Gantt data bound

Gantt change triggered ::

Gantt data binding

 

If you change the version to 2022.1.119 and do the same sequence of actions (starting with running dojo), you get the following

Gantt data bound

Gantt data binding

Gantt selection change :: Architecture

Gantt change triggered ::

Gantt data bound

Gantt data binding

 

As you can see, in new versions, change event is triggered 3x, while in older ones only one time.

Basically, now change event is triggered all the time (my log: Gantt change triggered ::) - even on collapse button click - which is not selection change.

If there is anything I can do to assist, please let me know.

Thank you and all the best

Vedad

[Martin]

Thank you for the additional information. You can see that the selection is lost when you collapse the parent node. That means a change in the state of the node and therefore the event is correctly triggered. The behaviour in the 2022.1.119 version appears to be a bug. Let me know if that clears things up.

[Vedad]

HI Martin,

thank you for your answer.

But this means that selection is lost no matter what you do. This wasn't case in any previous versions before 2022.1.119.

I have the same code for at least two years and with behavior based on change event. And this was never the case.

If we look at the documentation for the change event here, it says event triggers when user selects task. From my perspective, clicking on collapse icon cannot be considered as task selection, so I don't see the point of triggering change event.

I mean, if you decided to change this behavior, then its ok, but then description of the change event has to be changed, because it is not correct anymore. Beside this, this means I have to have the way to make the difference between selection of the item and clicking collapse icon (or whatever else would trigger the event). In this case, documentation should provide info on how to do that.

THank you

Vedad

 

 

 

[Martin]

Hello, Vedad, 

Let me discuss the selection part with a developer to check if the behaviour is expected. I will get back to you with further information.

[Vedad]

Hello Martin,

sure, please do. If you need any additional info from me, please let me know.

Thanks

Vedad

[Martin]

Hello, Vedad,

I have logged a bug about the lost selection in the Gantt. You can track the item in our Feedback Portal where you can also share any further thoughts.

[Vedad]

Hi Martin,

Any update on this issue? Link on feedback portal says its unplanned, which doesn't make much sense since it is a bug that changed behavior of the component?

Thank you very much.

Vedad

[Martin]

Hi, the issue is still not scheduled for fixing. Since the bug was not introduced with a specific Kendo version, it cannot be considered as a high-priority bug (regression). As I mentioned in a previous reply, I was not able to reproduce a different behaviour with an older version - the bug was ever present. Let me know if I am missing something.

[Vedad]

Hi Martin, 

I am bit puzzled, it is reproducible quite easy. If you do the steps I mentioned (and shown in the pictures in previous replies), and read console you will notice that change event in new versions occurs between data bind and data bound which was not case before.

It is still reproducible within dojo I provided.

If you want, we you can schedule online session/call and I will show and explain it to you.

I am holding update of our system for a quite while now, waiting this to be fixed.

 

Thank you and regards

Vedad

[Martin]

Hi Vedad,

Let me provide some further clarification as there seems to be some misunderstanding.

The behaviour is confirmed as a bug, and our workflow is to have an item in our Feedback Portal where people can vote for it. That shows us how the bug affects other users and how severe it is. The more people it affects, the higher the priority.

In order for a bug to be considered a regression ( that is a bug of highest priority ), the bug needs to be introduced with a specific Kendo version. You mentioned that the bug did not occur in previous versions, but when I tested the example with a previous version, the behaviour was still the same ( which is still a bug ). If you show me a Dojo example where the bug does not reproduce and the configuration is working as expected, I will increase the bug's priority so that it would be fixed in a next release. Until then, there is nothing I can do to move the bug up in our bug fixing queue. I hope that clears the matter.

[Vedad]

Hi Martin, 

thank you for the follow up on the matter.

This issue perfectly resembles regression, version 2022.1.119 worked in one way, version 2022.1.301 introduced behavior that affects change event (event is triggered even when it shouldn't be).

As I wrote in original ticket, bug was introduced with the version 2022.1.301.

If you compare behavior of the change event  (hence kendo.log screenshots in first ticket) between this version, and previous one, 2022.1.119 you will notice that in version 1.119 when you load the widget or click expand/collapse on group, sequence is following:

1. dataBinding
2. dataBound

If you observe the same steps on 1.301, we have following:

1. dataBinding
2. change 
3. dataBound

This means, that change event is triggered even when selection is not changed. If i clicked collapse button for category below, I didn't select anything, therefore change event shouldn't trigger. 

just do the following steps and observe kendo log in dojo:

1. Open dojo i provided originally, run it.

2. observe log - notice 3 events (bind, change, bound)

3. click to collapse any group

4. observe log - notice 3 events (bind, change, bound)

5. On the left corner dropdown select kendo ui version: Kendo UI 2022 R1 (version should be 2022.1.119) 

6. Run dojo and observe log - shows two events (biding and bound - NO change event)

7. Click collapse on any group 

8. Observe log - notice 2 events (bind, bound - NO change event)

If you do these steps, you will be able to see it clearly, if this is not enough, please let me know and we really should try to schedule some quick call and I can show it to you.

Thanks a lot and have a nice day.

Vedad

 

 

[Martin]

Hello, Vedad, 

Thank you for the clarification. I will log a separate bug about the regression and will link it here a bit later.

[Martin]

Hello, Vedad,

Here is the link to the new bug report in our Feedback Portal. Please follow it for further development on the matter.

I have updated your Telerik Points as a thank you.

Regards,
Martin
Progress Telerik

[Vedad]

Hello Martin, 

thank you very much.

Now fingers crossed for quick resolution.

Cheers

Vedad