|
Thanks itimpi, but I am aware of that - I was deleting the redundant PDF copies because it is faster and I do not think it is stuffing up any links to the epub versions I want (I could be wrong of course :-)).
Still, I have written large client server applications with much larger record sets and I have never had performance issues like Calibre....
Still, something like
Delete from xyz where author = 'Blah' should be relatively fast even if there are a few table links - assumnig the join fields are indexed
|