Currently in the user parameters you can setup a choice and import your values from an existing file. This means that all your values are hardcoded in the workflow, if you lookup table changes, the workflow doesn't change with it.
Ideally if you start the workflow, you provide the details for the lookup table (either file or in my case a database) and the values are looked up on the fly and in the second parameter you select your choice.
Maybe this is a bit in line with:
+28I would very much like a way to dynamically generate a choice parameter at runtime. Even better than just a file, or database query would be the ability to ingest a webservice (json).
Im also after a dynamic choice published parameter, so users can select features from a dataset but the dataset can be queried to check what features to offer to the end user in the choice. Once source dataset is selected the choice can be populated on the fly, before that is would ask for source dataset. Or if the choice can just be populated from a database query, it doesnt have to be the source data
+2This question is similar: https://knowledge.safe.com/questions/4412/dynamic-published-parameter.html
+1I would like to populate a choice list from sql or python or both to generate a valid list of operation numbers from a list. The underlying table for that list is being updated constantly so updating the parameter manually before each run is not an option. Currently, I can use pyodbc to get that list but no idea how to populate a choice list in the GUI from these dynamically generated unique values.
+16My specific use case follows @jdh suggestion, where a user can identify a Utility network feature service and will return a JSON output to populate dynamic user parameters.
So the first parameter is to select a utility network name feature service (1 or more) - of string type - pass this service into an http call to query the feature service to return a JSON result of all its Subnetwork features (layer id=500002).
From the JSON output, three additional parameters are visible and their choices are populated as the user picks the required, contingent attribute values for Domain, Tier, and Subnetwork name, in that order.
Once these required parameters are populated, a group parameter is visible for additional parameters used for the query string parameter of the HTTPCaller for the Export Subnetwork.
This workflow will also work for the Trace endpoint and other API endpoints for the Utility network dataset.
+11Really feel this should be possible, including in the style of a linked transformer where you have a parameter reused between workspaces as we do. Manually importing multiple times or working around in other currently-known ways is a pain point, and the values for one of our parameters will change frequently over the next year or two.
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.
