Hey there,
first question:
Have you considered changing the “Mode” from “File” to “Folder” and to simply read all the files at once?
If column structure is the same and sheet names are always the same you can skip the initial loop for reading the data (you can configure the node so that the file path is added to each row - so if you need to process each file’s content individually you can later iterate over that file path column using a group loop)
I tried to recreate your scenario different ways, however did not manage to replicate the error…
I tried to start with a file that has only integers in the column and transform to double and then have a mix of files with only int, only double or mix of both.
Also started with a double only column (w/o touching transformation tab) and then using the other files - also no issue…
Here’s the workflow both with the loop and the read all in one…
ExcelReaderSchema.knwf (109.5 KB)