I'm struggling a little bit with using trying to build in a variable reader schema. I've built the workbench where it prompts the user for the source shape file. I received a sample of the schema but there is a good chance of a different shapefile schema being input.
I have tried working with the knowledge base but I'm coming up short. The output is a known schema I've created after the analysis.
Does anyone have any ideas?
Have you tried working in dynamic mode? That usually takes care of situations where the ws needs to be adaptable for schema flexibility.
Hope this helps.
I'm 99% sure that's what I need. I've added a generic reader and set it as a published parameter (for the source file.) I've buit the workbench based on the input schema and even though it should all map up properly, I get an error message:
During Translation, some features were read that did not match a reader feature type in the workspace. This can happen if the reader dataset is changed, or a reader feature type removed or renamed.
Weired thing is, the only thing that changed was that I went from a known hard-coded input reader (shapefile) to a generic reader. I used the exact same file so the schema did not change.
My transformers are "errored out" (ie red) when I connect the generic reader.
Any idea how to modify the workbench to accept the generic reader but still accept the unknown schema when the tool starts?
That particular error would suggest that the source feature type you have for the Generic reader is fixed to a certain shape file name. You can deal with that issue by editing that feature type properites and checking Merge Feature Type and then entering * for the wild card. This will let a shape file with any name pass through.
Hope this helps