NYCPHP Meetup

[Jon Baer]

Im not sure this does justice w/o any demo but Id never plan on doing  
a search like that ... the idea was that one field w/ key:value pairs  
would be split and put into a searchable index via Zend_Search ...
http://framework.zend.com/manual/en/zend.search.html

The query (or report) from the index search into the db is trivial  
since the pk keys are exchangeable.  This is really a 2 layer  
approach not using MySQL for FT indexing.  Any other option gives you  
way more IR functions than MySQL does.  Although it would be fair to  
point out that it is pluggable: ( <http://dev.mysql.com/doc/refman/ 
5.1/en/plugin-full-text-plugins.html> ) But still pretty new.

- Jon

On May 15, 2007, at 2:01 PM, csnyder wrote:

> On 5/15/07, Jon Baer <jonbaer at jonbaer.com> wrote:
>>  I normally include @ least one column of YAML-based data
>> (a misc field of sorts).
>
> At first I thought this would work with a MySQL FULLTEXT index, but of
> course it won't because the field names would be in more than 50% of
> the records and would therefore be considered stopwords.
>
> Nifty as this is, then, it only works if all querying is handled by
> your application or an application using a platform that includes its
> own sufficiently reliable search index. Clients who want/need to
> create their own SQL queries are out of luck, or forced to use LIKE
> expressions.
>
> -- 
> Chris Snyder
> http://chxo.com/
> _______________________________________________
> New York PHP Community Talk Mailing List
> http://lists.nyphp.org/mailman/listinfo/talk
>
> NYPHPCon 2006 Presentations Online
> http://www.nyphpcon.com
>
> Show Your Participation in New York PHP
> http://www.nyphp.org/show_participation.php