Hi @ScottF,
thanks for your suggestion. Unfortunately it’s not only about reading files. The CSV writer node – worth to note that a common file writer node as an equivalent to the reader node seems to be missing – has some advantages like writing compressed files. The CSV Reader can also read from compressed files too which the common file reader node can not.
About the CSV file writer node. In case the separator, like a pipe, is already present in the data, it removes prior to writing the data which also causes sever data regressions.
While trying to evaluate results with the regular file reader node I also faced this issue (data preview worked):
WARN File Reader 3:25:0:15 Errors loading flow variables into node : Coding issue: Cannot create URL of data file from '/private/var/folders/jg/zcb9yqn90w54f3szwrsnqrl80000gn/T/knime_CSV write read35895/knime_tc_awjdlvx6aea2/default-setting-escaped-quote.csv' in filereader config
Edit: Above issue likely caused by Problem with the File Reader Node when using workflow variables (again)
Edit #2: I Found a super simple work around to prevent the CSV writer replacing already present quotes. It’s by simply replacing quotes by quotes. To summarize. The CSV writer issue of removing quotes is triggered by its default functionality. It always removes quotes even if no replace value was provided. Same goes for the separator.
The problem is, this input field is not accessible. Cursor won’t enter as it seems to be deactivated. Please not the slight difference in its Appearance.
Kind regards
Mike