+8Hi. I’ve built a data cleaning workspace that performs a number of different tests, creating or retaining different attributes for each. I’ve created an ‘error type’ attribute to fan out by, have added schema scanners and have followed the gist of this. However, I cannot seem to get the writer to respect the varying schemas for each tab. Set up like this:
Pointers appreciated.
Many thanks
+34Do you have a different SchemaScanner per path that will each create a different tab in the xlsx? If you are only using one SchemaScanner for multiple error types that could be the issue, in which case you would probably need to use GroupProcessing to create a schema per error_type.
+26Note that when using a Schema Feature for dynamic writing, the Schema Feature needs to arrive before the data features.
Edit: Noted already done that above but leaving the note in the thread 😉
If routing from a SchemaScanner, then the Parameter “Output Schema Before Data Features” needs to set to “Yes” (and is a little weird this is not the default setting)
Similarly as
+8Do you have a different SchemaScanner per path that will each create a different tab in the xlsx? If you are only using one SchemaScanner for multiple error types that could be the issue, in which case you would probably need to use GroupProcessing to create a schema per error_type.
I do have multiple SchemaScanners, yes.
+34If you are currently just using one xlsx feature type (that is fanned out), you could try copy/paste that writer feature type so that you have more than one on the canvas. So that each SchemaScanner is connected to a writer feature type. I am wondering if there is still an issue with the order of features arriving, if you have multiple schema scanners flowing into one writer feature type.
+8If you are currently just using one xlsx feature type (that is fanned out), you could try copy/paste that writer feature type so that you have more than one on the canvas. So that each SchemaScanner is connected to a writer feature type. I am wondering if there is still an issue with the order of features arriving, if you have multiple schema scanners flowing into one writer feature type.
..or maybe precede the Excel writer with a Sorter?
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.
