I am working on building a component that will ingest a file containing zip codes, do some cleaning, then join with another file with corresponding latitude and longitude points. The matched rows will be the output, while the unmatched rows are written to a file on the local device for manual cleaning. The idea is the process should be reproducible with any input data containing zip codes. The issues I am experiencing are 1) When I use a new file, only the first two configurations in the component dialog correspond to the new document while the column selection only remembers the previous data. The input file is not being passed downstream. 2) I am having trouble using a flow variable to write a new file to the local device.
Thank you in advance.
I would upload the workflow but there is data (Lat/Longs) embedded in it which is too large to share.
Hi @mmays,
The problem is that the component does not know what the Column Expressions will produce, but it does know what will come in. So it can show the correct data for the Column Selection Configurations, but not the Column Filter Configuration. Is it possible to move that to the beginning of the component?
Kind regards,
Alexander
The issue with putting it in front of the column expressions is that for the joiner to work as desired I need the columns manipulated which includes creating a new column which then does not appear in the column filter configuration.
Is there a better configuration node option to select columns for the joiner?
Hi @mmays,
I do not think that is possible. The Column Expressions depend on the Column Selection Configuration nodes, but your Column Filter Configuration depends on the Column Expressions. So when the dialog is shown, how can it know which values to show in the Column Filter Configuration? You could split your component up into two parts: one only with the Column Expressions and the Column Selection Configs and one with the rest.
Kind regards,
Alexander
