☝️Small business or a startup? See if you qualify for our special offer.
+
Get Free Trial
Get Free Trial

Behavior of the table data, selects, aggregations and filters when using Custom Datasource API

posted 1 day ago

Answered
Veronika Andreeva asked 1 day ago

Hi Flexmonster Team,

I'm currently working on implementing the pivot table using Custom Datasource API and have a few questions about the general behavior of the table, particularly regarding the select requests/responses and aggregations logic.

At the moment, I'm relying on manual testing and observing UI results based on the only provided example here:
https://jsfiddle.net/flexmonster/j1c6rd24/,

However, I’ve encountered some unclear behaviors and would greatly appreciate your clarification, or ideally, access to any relevant written documentation

1. Aggregation Logic During Expand

I've observed that during an expand action, the response from the select request seems to only return new aggregations, while the table still displays all data, including previous values.

  • Is there some kind of caching mechanism involved that retains previous aggregations?
  • Could you please explain the expected behavior in this scenario?

2. General Table Behavior Within a Session

Could you provide more detail on how the table behaves with regard to:

  • Behavior of the select request - which properties are expected in the json object in which specific cases 
  • Expanding/unexpanding of nodes within the same session - which properties are passed to the request?
  • Adding and removing filters in the same session - which properties are passed to the request? When is a new request triggered?
  • When and how the state of the table is expected to persist or reset?

3. Filter Selection Persistence

I've noticed that filter value selections don't always persist between interactions. Specifically:

  • Changing the values of a filter alone does not seem to persist the selection when exiting the table or page.
  • However, if a subsequent action is performed, such as rearranging fields in rows/columns or report filters, the earlier filter selection is retained.
  • It seems that only certain user actions trigger a persistence of the current filter state. Is this the intended behavior?
  • If so, what actions are expected to trigger state persistence?

Thank you,
Veronika

 

1 answer

Public
Solomiia Andrusiv Solomiia Andrusiv ⋅ Flexmonster ⋅ 2 hours ago

Hello, Veronika!

Thank you for reaching out to us.

You are welcome to check the following docs when implementing custom data source API:

From our experience, the provided documentation is enough to cover all the basic steps of the implementing process.

Regarding your questions, please find more detailed comments from our Tech team below:

1. Aggregation Logic During Expand

  • Is there some kind of caching mechanism involved that retains previous aggregations?
    Yes, we are caching the data that is already displayed on the grid, and only send small /select requests for each expanded node.
  • Could you please explain the expected behavior in this scenario?
    Flexmonster client is sending all the details about the data it needs in the /select request. So, the server just needs to provide the client with the data specified in the request; everything else is handled by Flexmonster for the simpler implementation process.

2. General Table Behavior Within a Session

  • Behavior of the select request - which properties are expected in the json object in which specific cases 
    We recommend checking our select request structure in the docs:
  • Expanding/unexpanding of nodes within the same session - which properties are passed to the request?
    On expand, we recommend checking the docs provided above and sending only the data for the requested expanded node to the client.
    On collapse of the node, we don't request anything as Flexmonster doesn't need additional data to show.
  • Adding and removing filters in the same session - which properties are passed to the request? When is a new request triggered?
    Please check the following guide, which explains in detail how to handle filters when implementing custom data source API: https://www.flexmonster.com/doc/implementing-filters/
  • When and how the state of the table is expected to persist or reset?
    For Flexmonster, the state is updated when the data visualization on the grid is changed.
    For the custom data source API, it means that when we change the whole report, data source, options, or slice(which includes filtering, reordering fields, expanding, drilling, etc.), the new requests are sent to the server. Please note that sorting is a client-side operation that doesn't trigger any additional requests.

3. Filter Selection Persistence

By default, Flexmonster doesn't save the state of the grid after each user interaction. So, if the page was reopened or reloaded, we will show the grid configuration specified in the initial report object. Kindly note that this behavior is consistent for all data sources, including custom data source API.
You can implement caching of the last user actions on your own by using our getReport() API call to save the last configurations and then setReport() to apply them on the next load.

Hope you will find our answer helpful.
Do not hesitate to ask if there are any further questions.

Kind regards,
Solomiia

Please login or Register to Submit Answer