-
Hi eulerto,
thanks first of all for the cool library. I am having trouble compiling it on CentOS 6.4. Please see the following output:
[gpadmin@gpdbsne pg_similarity]$ USE_PGXS=1 make
sed 's,MODULE…
-
The current spellchecker makes an n-gram index of your terms, and queries this for spellchecking.
The terms that come back from the n-gram query are then re-ranked by an algorithm such as Levenshtein.…
-
"Hello World" and "jello" should score higher than 0 with a fuzziness of 0.5, says your test.