I have used a PythonCaller to flatten a xml dataset and have many handfulls of attributes that need to be exposed. Cutting and pasting into the AttributeExposer does not feel appealing. Surely there must be an easier way to do this? After all the features already have these attributes, only they are set to be Unexposed. It should be just a hide-unhide setting that is needed?
Question
Fastest/easiest way to expose attributes?
+20- Supporter
You can expose them in the PythonCaller directly in the field ‘Attributes to Expose’ at the bottom of the transformer.
Without knowing your data it is tough to tell, but maybe it is handy to create a list and filling it, so you only have one attribute to expose. Afterwards you can split up your list to your liking
+16- Enthusiast
Thing is that all workarounds that I know of are more or less convoluted.
Maybe this is more of a feature request. Maybe the easy solution I was hopping for does not exist.
FME “knows” that a given feature has certain hidden attributes and the values of those. These could easily be exposed with check boxes, ore something similar. But instead one has to manually type the correct name of the attribute to expose it. The former is fast and efficient, the later is cumbersome.
+46- Influencer
In the AttributeExposer you can add attributes from the FeatureCache, so if you’ve run it once, you can easily expose all hidden attributes. Ideally, it’d be great to have this functionality within the python caller (and other transformers, eg. JSON extractor) to avoid the additional AttributeExposer
