> Did you read the page at the URL I gave ? It answers the question. yes the page shows an extremely unhelpful comparison: > SELECT count(*) FROM enrondata1 WHERE content MATCH 'linux'; /* 0.03 seconds > */ > SELECT count(*) FROM enrondata2 WHERE content LIKE '%linux%'; /* 22.5 seconds > */ they are apparently putting this forward as some kind of proof of FTS's amazing speed. but it is not an apples-to-apples test. i want to see the test: > SELECT count(*) FROM enrondata1 WHERE content MATCH '*linux*'; /* ???? > seconds */ > SELECT count(*) FROM enrondata2 WHERE content LIKE '%linux%'; /* 22.5 seconds > */
yes they do explain this sortof a little bit but it made me feel like they were saying something like "well, comparing an apple to an orange is not entirely equivalent, for example an apple is not an orange. but they're both fruit!" so i am still left wondering if searching substrings is really any faster using FTS. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
