You can do in SQL by using UNION and subselect, something similar to this (not tested):
select a1, count(a1) from (SELECT c1 FROM t1 UNION SELECT c2 FROM t1 UNION SELECT c2 FROM t1) as a1 GROUP BY a1; 2013/8/28 davedigerati <hypr...@gmail.com> > humbly bumping... > > On Sunday, August 25, 2013 1:32:35 PM UTC-4, davedigerati wrote: >> >> Is there a way to SELECT from my (SQLite) db with example structure/data >> that looks like this: >> >> sue | dog | cat | dog >> joe | cat | rat | dog >> joe | dog | dog | rat >> joe | cat | cat | cat >> tom | rat | dog | dog >> >> a counted, grouped-by set for JOE only that looks like: >> cat - 4 >> dog - 3 >> rat - 2 >> >> I've been playing with selecting Joe's records, then looping over them >> building subtotals in a dict, but this seems a huge waste of effort and my >> data is actually much larger so I am worried about performance as I try to >> scale this. From what I read the groupby is for by fields only, not by >> counted values? Can anyone point me in the right direction? thanks! >> > -- > > --- > You received this message because you are subscribed to the Google Groups > "web2py-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to web2py+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. > -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to web2py+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
