I have a table with two columns. From the second column I need to remove some entries for which I use a simple string replacer without actually replacing the string with something. I end up with a table with some empty cells like this:
If I then group by column1 and unique concatenate column2 I get:
The group by on b is okay but the other two leave me with annoying commas at the beginning or the end although the cells seem to be empty. Unfortunately I need result b too (so I can not filter the empty rows out upfront). Is there a way to avoid the strange commas?
Instead of making the cell empty, why not use string replacer to put in the word BLANK. Then after groupby node you can use another string replacer nodes to replace ,BLANK for nothing and BLANK, for nothing.does this work.
there is also the missing value tick box you could disable in the groupby node. Does this help.
the missing value tick box doesn't work but your other suggestion is practical although I have to use three string replacers for ", BLANK," replaced by comma and ", BLANK" and " Blank," replaced by nothing.
The missing values didnt work because the field is blank not a missing or null field.
If the empty fields are left as nulls and not blanks the missing values option in the groupby node should work fine I think, there would still be a comma in between values EG name_1,name_2 but there should no longer be comma's at the begining or the end.