User Defined Java Expression - how to process fields with spaces in their names?

[dave-csc]

I should probably file this as a bug, but maybe there's a way to do it: how can UDJE process fields if their names have spaces (or other non-alphanumeric characters) in them?

Those characters are allowed in Hop field names, but not in Java: when using the GUI the field names are copied as they are, and if they have spaces this results in "Unexpected tokens" when executing the pipeline...

Thanks for your support :)

[sramazzina]

Ciao @dave-csc

don't think of me as boring person :-) but I think that in transforms that integrates scripting languages we must follow the code style rules set in specific formatting languages they refers to, must be kept as they are. Therefore I personally don't suggest any change for the reason of this issue you are experiencing. In any case, don't hate me, but I personally have a more aggressive approach in the terms that even if Hop permit field names with spaces I personally disagree and I suggest our customers not to use spaces in field names and if some fields exists with that kind of names I suggest to rename them as soon as it is possible, if possible. I know that you and all the community could hate me for these opinions :-) but I love you all.

Cheers

[dave-csc]

More than respectable opinion, don't worry... :)

I usually don't like spaces and similar in field names, it happened that I imported an Excel file that has those and I was just wondering if it was possible to process the fields without forcing their names in the transform (btw, so did I in the Excel input).

Thanks 😄