When digging into T277366 we noticed that the table which populates the unsubscribe csv needs cleaning up. We could probably skip the more complicated SQL fix by adding a normalise_email fn in the python layer that cleans em up prior to writing them to file.
Description
Description
Related Objects
Related Objects
Event Timeline
Comment Actions
This is one way to see the number of duplicates that only have whitespace differences, ie "user@example.com" and "us er@example.com"
cat Unsubscribes-*.csv | sed -e 's/\s//g' | awk -F, '{print $1}' | sort | uniq -d