[Snowball-discuss] Snowball goes Zope

[Teodor Sigaev]

[skip]

>>>>Andreas, having a chance to ask Zope core developer, I'd like to know
>>>>where to get more info about fulltext you developed ?
>>>>Did you look on our fulltext search for PostgreSQL ?
>>>>(openfts.sourceforge.net). It's specially designed for dynamic
>>>>
> content.
> 
>>>quote: "Currently, OpenFTS is implemented as a collection of
>>>
> PERL-scripts."
> 
>>>I don't regard that as an scalable solution :-)
>>>
>>Hmm, it should be corrected :-)
>>
>>OpenFTS is based on special data type implemented in PostgreSQL
>>using C. Perl and Tcl are used as a glue languages for external
>>
> interfaces.
> 
>>Actually, OpenFTS utilize GiST (Generalised Search Tree) available in
>>PostgreSQL (we  maintain this pice of code). It's extremely fast for
>>small and medium size collections.
>>
>>
> 
> How large do you define "medium size". Let's say, I have 3 GB of text.
> Can GiST handle that ?
> 
> Andreas
> 

Sorry, we never measure text for indexing in bytes :). We measure number 
of parts (piece) of text, such as mail message etc. Our test base 
contains ~500000 mail messages from PostgrSQL, Perl, Mason and Linux 
mailing lists and works fine.

BTW It seems to me this thread is offtopic here. Let's move our 
discussion to openfts mailing list :) 
(http://lists.sourceforge.net/lists/listinfo/openfts-general)


-- 
Teodor Sigaev
teodor@stack.net


_______________________________________________
Snowball-discuss mailing list
Snowball-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/snowball-discuss