Hi @julpet​ ,

Sounds like you'll want to use a FeatureWriter and a FeatureReader if you want to keep this all in the same workspace. Using a FeatureWriter is essentially the transformer version of a writer with the benefit of being able to continue processing features instead of ending the translation. The FeatureWriter enables you to writer out features mid-translation.

Following the FeatureWriter, you can use a FeatureReader to read in the data that was just written to "System2". You'll simply need to connect the FeatureReader to the Summary port of the FeatureWriter, set the format, and point it to the dataset (or set the dataset parameter using the _dataset attribute that is created by the FeatureWriter).

This is a great method if you are inserting records into an existing table/dataset as the FeatureReader will read in the entire table. However, if you only need to continue processed the features that are already in the workspace but want to see how they map to the "System3's" schema, you can import the schema in transformers like the AttributeManager. This may be faster since the workspace doesn't have to re-read the data if you are wanting to continue processing the same features.

You have a few different options here but I've attached an example in case that helps.

Hi @chrisatsafe​ ,

thank you for your answer. I will try it today.

So far I just had a short look at your attachement.

Sadly, there are two things not working for me:

1. If there are three systems, I also just want to have three elements in my overview, so that for example I could also take a screenshot of how they are connected.
2. I'd also like to connect my attributes with each other, so that I instantly know how they changed. I mean now I have self-speaking names for the attributes but for example if in my "system1" the attribute names are just numbers and I want to know how they changed into the new names (like first_name, last_name etc.) in the overview you sent me I couldn't really follow the transformation I think.

I hope my approach became a bit more understandle. Maybe it is also not possible to do that with FME then it's also ok.

Hi @julpet​ ,

Ah I see! Okay for visual purposes yes you can do this by expanding the drop down on the writer feature type (as pictured below). By clicking on the drop down, you can drag and drop the attribute mappings from the reader feature type to the corresponding attribute on the writer feature type. Unfortunately this is not an option on the FeatureWriter at the moment.

If you want to visually see the mappings from gsystem 1 to system 2] and system 2 to system 3], you can still do this but it would be best to handle this in two separate workspaces in this case.

Hi @chrisatsafe​ ,

sorry for my late response.

Ok, I see but then it is not possible to have a system in between those two right?

I mean I would like to have the visual purpose but also the transformation into a new file type for example.

Thanks a lot for your help!

I just wanted to be sure that I don't oversee any functionality which might help me.

Correct, if you want to see the attribute connection lines, you wouldn't see the connections between the reader and the FeatureWriter mentioned in the previous post. Using the FeatureWriter method that I mentioned in the previous response, you would still be reading system 1 and writing to system 2 but you would only visually see the connections between system 2 to system 3 on the canvas.

System 1 to 2 would still be mapped but the only visual queue you'd have would be in the transformer you use to map the attributes (i.e. SchemaMapper, AttributeManager, other) or in the FeatureWriter > User Attributes tab.