Skip to main content

Hello,

I am working on input datasets having different schemas. In the end of my translation, I have to inspect obtained results. I had therefore to use a testfilter and 3 (AttributeManagers + Inspectors) as shown in the screenshot. The attributes managers perform some schema mapping depending on input features (this include removing and adding attributes)

 

Is there a way to reduce the amount of used Attribute managers and Inspectors ? My workspace is getting really big and I need to optimize it.

Thank you very much.

If you need to visually simplify the workspace, there are two simple solutions:

  1. Collapsing bookmarks, introduced in FME 2018
  2. Custom transformer

They both have their advantages and disadvantages, so be sure to skim through the docs before committing to either.


Depending on exactly what you do in the AttributeManagers you might get away with replacing all 3 and the TestFilter with a single AttributeManager and make use of conditional statements (basically a Testfilter inside the AttributeManager). That might be a beast to set up and maintain though.

Perhaps a SchemaMapper could be useful too, again it depends on what you're doing exactly.


Depending on exactly what you do in the AttributeManagers you might get away with replacing all 3 and the TestFilter with a single AttributeManager and make use of conditional statements (basically a Testfilter inside the AttributeManager). That might be a beast to set up and maintain though.

Perhaps a SchemaMapper could be useful too, again it depends on what you're doing exactly.

SchemaMapper would definitely move the complexity of the renaming out into an external Excel or CSV file. And you'd have only a couple transformers as your reward. A set of tutorials (this thing is powerful) starts at https://knowledge.safe.com/articles/1136/schemamapper-transformer-tutorial.html

 

 


Reply