Skip to Main Content
Spotfire Ideas Portal
Status Unlikely to implement
Product Spotfire
Categories Library
Created by Guest
Created on Sep 30, 2021

API to extract the data table dependencies

Need an API to extract/update the graph of dependencies between the diffrerent data tables in Spotfire 10.10.x. I need to know the order of refresh/reload of the different data tables in a document.


We created a custom data source that loads data from our databases into Spotfire table. We also provided the users a set of tools to tie this data table with other data tables (loaded from the same DB). In some cases, if the user build these tables in the wrong order, we get logical problems ... for example, table 1 needs information from table 2 but table 2 was loaded after table 1 ==> the solution to such "weird" scenario is to update the dependency grah in Spotfire to force table2 to load before table1.

  • Attach files
  • Guest
    Reply
    |
    Nov 30, 2021

    Thanks Magnus ... I understand that data functions solve the problem and we are heading in this direction. But the amount of needed to move all our custom data sources to data functions is enormous.

    This problem of datatable dependencies is getting more critical now with the new parallel loading feature in SF11.4! since SF does not know that we have cusotm data sources that depend on other tables ... the same template working in SF10.10 (because of sequential loading) is failing now in SF11.4 with the following error:


    The data should already be loaded for producer with key be110c5a-ee61-439d-9c0d-7295b58a9890.


    This is a major blocker for the adoption of SF11.4 ... please advise.


    Note: do we have any API to create On-Demand custom data sources in SF? can we inherit from an On-Demand abstract data source to implement our custom data sources instead of inheriting from the more CustomDataSource?


  • Admin
    Magnus Rylander
    Reply
    |
    Oct 26, 2021

    The loading order dependency between data sources is not stored in the model.

    If you have data tables that are dependent on other data tables, the recommendation is to load them through custom data functions (see https://community.tibco.com/wiki/wrap-data-source-inside-data-function).