Closed dreadedhamish closed 1 year ago
pwang347
commented
1 year ago Hi @dreadedhamish, thanks for the feedback!
Just wondering if you've tried the Filter operation to remove unwanted rows:
Would this help your use case? Or perhaps were you hoping to have the conflicting outliers be detected automatically at startup?
dreadedhamish
commented
1 year ago Filter! Yeah that helps alot - I don't know why I didn't notice it. Could a "Drop filtered rows" checkbox be added?
When there is a conflict it would be good to get some more details rather than just the conflicting cell value.
pwang347
commented
1 year ago Filter! Yeah that helps alot - I don't know why I didn't notice it. Could a "Drop filtered rows" checkbox be added?
When there is a conflict it would be good to get some more details rather than just the conflicting cell value.
Awesome! Yes, we should be able to add a "Drop filtered rows" option to use the inverse condition - we'll take a look at adding that in the upcoming releases.
Regarding conflicts, could you elaborate a bit on what sort of details you'd be interested to see?
pwang347
commented
1 year ago Small update - we've merged a change to optionally drop the filtered rows and it will be available in next week's pre-release!
pwang347
commented
1 year ago Closing the issue for now since the main ask was addressed, feel free to open a new issue regarding the conflicting cell value if needed. Thanks!
Loving Data Wrangler so far.
I'm currently cleaning up a source that uses int64 for one column, but they include an "Other" row that has "ZZZZZZZ" as the value. I'm envisioning cleaning up data but opening a source, changing the type of columns and seeing what errors are produced.
At this stage it would be handy to have a preset operation for Drop Row based on value, regex, length etc... Bonus points would be exposing in the Data Wrangler tab the conflicting row (rather than just the value) with some shortcuts (add a step - drop this row, drop rows like this etc...)