Union if one of inputs has error

[Andrzej]

I have build workflow that unions data from two different inputs. Before union each part has its own workflow. It can happen that one workflow will produce an error. I would like to union data anyway then, just without data from input with error. I have added 3rd text input to make sure we always have at least two inputs. The error is produced because column names in one input can change and then workflow will produce an error. I am attaching example what happens when columns of one input changed. Correct column names are [Country] and [City]. 

[Jean-Balteryx]

Hi @Andrzej ,

For the Union tool you don't need a dummy input. it works even with only one input.

As for the error, if you set the following option as "Ignore", it won't raise an error when columns names differ.

[Pingu · Accepted solution]

That is only when the fields differ... but here the issue is that there are no fields at all due to the parse error of one input stream. I tried to adjust the workflow, but was not able to do it. I was thinking about a formula or filter that checks if there are any fields or not. If not create the dummy fields. But this will only work if the field names are always the same, I think.

[atcodedog05]

Hi @Andrzej & @Pingu 

Since the error is in the input stream to union tool. There no way make the workflow work after this error. Upon error alteryx would stop all later operations which is in the same input stream.

[Andrzej]

In my case column names should always be the same in 2nd workflow. Creating text input only with column names and afterwards union it before an error can emerge worked. I only had to use data cleansing to original dataset to replace all nulls with 0/empty. And after union filter out all null fields