asked on May 24, 2024
Hi,
I found an issue on 'select' requests having some filter and expanded nodes in a custom data source pivot.
Attached a sample to reproduce the issue.
Exact steps to reproduce the issue:
1. expand year 2024
2. expand node Area 1
3. filter by Area 2
4. notice the 'select' requests
5. the last request contains an incoehrent filter (highlighted in attached schreeenshot filter.png)
In the sample this is indeed not a big issue since the pivot seems to show the correct data to the user but:
- the sample is based on your sample implementation of customDataSource api, which, in that case, responds with 'empty' data (notice that the dataset is very small).
- we have a more complex implementation that we use also for other UI reports. This kind of payload is interpreted as 'empty filter' that for us is equal to 'no filter', so in that case we respond with the whole dataset, causing an UI issue: user filters some fields but the whole dataset is shown anyway
But mostly, this issue is causing us highly critical issues for large datasets (whole dataset in the response... paged... so a lot of 'select' requests, high conusiming client side). It is quite impossible for us to distinguish 'empty filter' and 'no filter' because, as I said, we use the same endpoint for other reports.
My point is: regardless of our own implemetation this is a critical performance issue anyway, because of unuseless calls of 'select'. Moreover, the more nodes you have expanded the more calls to the select are triggered.... and with largedataset it is very easy to reach hundreds of useless calls.
Could you please fix this issue?
Thanks in advance
Ilaria
⋅ May 28, 2024
Hello, Ilaria!
Thank you for reaching out to us.
Our team has reproduced the described behavior with filter requests in the custom data source API on our side. Our Dev team will research possible solutions for the behavior and get back to you with the results, with an ETA June 24th.
We will notify you about any updates on the matter.
Feel free to reach out in case of any other questions.
Kind regards,
Solomiia
⋅ May 28, 2024
Hello Solomiia,
thank you first of all. Could you please be more specific? June 24th is the ETA for the fix or just for a feeedback? We need to commit for a date of the fix, to share with our customers, since this is a quite critical issue. Could you give me some feedback for an ETA for the fix, before June 24th, please?
Thanks in advance,
Ilaria
⋅ May 29, 2024
Hello, Ilaria!
Thank you for getting back to us.
Kindly note that our Dev team will thoroughly inspect the described behavior until the 24th of June. If our developers find a possible fix, we'll release it within the given ETA.
We will notify you about any updates on the matter.
Kind regards,
Solomiia
⋅ June 25, 2024
Hello, Ilaria!
Hope you are doing well.
Our team is happy to announce that the /select request for the pivot table was improved. If an expanded member is filtered out using the selection filter, it will not appear in the query.
The fix is included in the 2.9.80 version of Flexmonster: https://www.flexmonster.com/release-notes/version-2-9-80/.
You are welcome to update the component: https://www.flexmonster.com/doc/updating-to-the-latest-version/.
Please let us know if the fix works well for you.
Kind regards,
Solomiia
⋅ July 1, 2024
Hi Solomiia,
the fix seems to work fine.
By the way we noticed that there are two identical 'select' requests (with type 'drilldown'). In the specific sample it is not a big issue but for biggest dataset this is a performance issue.
Could you please fix it?
Thanks in advance.
Ilaria
⋅ July 1, 2024
Hello, Ilaria!
Thank you for your feedback.
We are glad to hear the fix works well.
Regarding your comment about duplicated 'select' requests in custom data source API, our Dev team will research possible improvements of the described behavior.
We'll get back to you with the research updates with the ETA July 22.
Feel free to reach out to us in case of any other questions.
Kind regards,
Solomiia
⋅ July 23, 2024
Hello, Ilaria!
Hope you are having a great week.
Our team is happy to inform you that the issue with duplicate /select requests having the "drillDown" querytype was fixed.
The fix is included in the 2.9.82 version of Flexmonster: https://www.flexmonster.com/release-notes/version-2-9-82/.
You are welcome to update the component: https://www.flexmonster.com/doc/updating-to-the-latest-version/.
Please let us know if the fix works well for you.
Looking forward to hearing your feedback.
Kind regards,
Solomiia
⋅ July 30, 2024
Hello, Ilaria!
Hope everything is well.
Our team is wondering if you had a chance to test the fix with duplicate /select requests having the "drillDown"querytype. Could you please let us know if everything works well after the fix?
Looking forward to hearing from you.
Kind regards,
Solomiia
⋅ August 1, 2024
Hi, Solomiia!
Just letting you know that your released fix was tested and works well as expected.
Thank you for your work.
Best regards,
Edinson
⋅ August 1, 2024
Hello, Edinson!
Thank you for your feedback.
We are glad to hear that the fix works well for you.
Do not hesitate to reach out to us if any other questions arise.
Best regards,
Solomiia
