When loading a CSV file RowIDs can be automatically generated, but I have no idea, why they take the form of strings “Row1”, “Row2”, …, “Row25”. That makes them pretty useless, as it is impossible to sort (correctly) them, and processing them during join operation is very ineffective. Why can’t they be just ordinal LONG numbers?
Generic row IDs can be used for join and starting from v 4.7.0 there is enhancement:
- AP-3276: Support for Natural Sort Order (e.g. “Row9” < “Row10”) in Sorter, TopK nodes and table views
2 Likes
Yes, this has been raised a few times already @torczyk and as @izaychik63 pointed out, there is a ticket that should help with this.
In the mean time, as a workaround, you can use the Counter Generation node to generate sequential numerical “IDs” that you can use as your own Row IDs that you can sort on:
4 Likes
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.