File Reader -> Cell Spliter by position

[Mihnea]

Hi all,
I need to load a .dat file and then assign it headers based on position.
I have 3 columns to assign. I did that and it worked well until I wanted to test in case a row would be empty for one column (I tested for column 1 empty in original .dat file). If a row is empty, the flow skews to the left. I believe Somewhere the nodes trim the spaces delivered in col 1? Would you know how to prevent that?

how original data in file looks like

how data looks in the File Reader node: the empty ro5 and 6 is trimmed

The flow starts with file reader then applying positions

Thank you,

[takbb]

Hi @Mihnea , and welcome to the KNIME community.

Splitting by position is only going to work if the character positions are always the same, so it might be that a different approach is necessary.

Would you be able to upload a sample workflow (with the pictured nodes) along with a copy of a small example datafile so we can try to see what is occurring in this case, and then we can make suggestions.

edit: Actually I just tried it with a file of my own using File Reader and I see it loses the whitespace at the beginning, even with the “remove quotes and trim white space” option not selected.

Have you tried the LINE READER node instead?

thanks.

[Mihnea]

@takbb

e you tried the LINE READER node instead?

Hi @takbb ,
No, didn’t try that one but will try it now and revert.

Thanks!

[Mihnea]

That did the trick, thanks!