Subject: RE: [htdig3-dev] [3.2.0b2] AND operator not working as it should?
From: Quim Sanmarti (email@example.com)
Date: Fri Aug 25 2000 - 05:09:15 PDT
> I reviewed your parser.cc patches to 3.2.0b2 and they look good. The
> only catch is that your patches are to the 3.2.0b2 code and not the
> CVS branch:
OK -- there's also a newer version, 18.104.22.168, I guess I should use that.
Otherwise just tell me, I'll redo.
Remark that I must modify parser.h too. I used version 22.214.171.124.
> ... the conflict is with Arthur's patch--you
> try to do the same things in certain places in different areas...
Concerning Arthur's patch... it will not work if the first word in the
phrase has isIgnore set. I'd rather keep my version, that includes also the
optimization that stops lookups if some two words in a phrase are nowhere
found to be contiguous.
On the other hand, I must withdraw the other modification in perform_phrase
that concerned ignored words in a phrase. It's plain wrong, since words
marked to be ignored are not indexed, so the location counter is not
incremented by them (*). Sorry, my tests were bad, I shouldn't have posted
I annex the diffs. If what I did is no good, just tell me. I'll correct
them as needed.
(*): Why? In fact this may have strange effects on an eventual 'near' query
operator, or proximity scoring... to be reviewed.
// Joaquim Sanmarti
// GTD Ingenieria de sistemas y software industrial, S.A.
// c/Rosa Sensat 9-11
// 08005 Barcelona SPAIN
// Tel. +34 93 225 77 00
// Fax. +34 93 225 77 08
To unsubscribe from the htdig3-dev mailing list, send a message to
You will receive a message to confirm this.
This archive was generated by hypermail 2b28 : Fri Aug 25 2000 - 05:14:01 PDT