Just to be clear, it isn't really "mine", just an adaption of the many
excellent contributions, from which I too have learned.
A huge pleasure and fun exercise no less!
On 2014/02/08 14:35, Stephan Beal wrote:
On Sat, Feb 8, 2014 at 12:39 PM, RSmith <[email protected]> wrote:
SELECT name, max(vid=2)-max(vid=1) FROM v GROUP BY name;
So much more succint than my original, like I predicted :)
Indeed!!! This one wins if i am able to refactor it for use with the much
more complex structure i'm actually working with (the fossil SCM's vfile
table - my example is a simplified form to help me get my head around the
SQL).
FWIW, sqlite3's ".stats" say yours is overall more efficient:
Virtual Machine Steps: 242
vs the WITH variant i posted:
Virtual Machine Steps: 308
Thanks again!
_______________________________________________
sqlite-users mailing list
[email protected]
http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
