[LeonB]

I have had it happen in real data from real systems - but it’s not good — “true duplicates” are a sign that something is missing from your schema or something has gone wrong earlier in the pipeline.

Trying to delete duplicates (but leave 1 behind) is tricky in itself. I recorded notes on it one time here — using “row_number()” to act as the iniquitie, https://til.secretgeek.net/sql_server/delete_duplicate_rows....

[LeonB]

(I just noticed that I emitted the word “iniquitie” above, where I thought I’d written “uniquifier” — and have just confirmed that autocorrect was responsible. Strangely, in context, the word “iniquitie” sort of “appears” to be an appropriate word, when its meaning is certainly not. I found that interesting.)