> Date: Mon, 08 Nov 1999 21:06:33 +0100 (MET)
> From: [EMAIL PROTECTED]
> On Mon, 8 Nov 1999, Henri Bergius wrote:
>> We had native searching back in Midgard 0.1, but Jukka
>> decided to drop it when he was cleaning Midgard up for the
>> first public release.
>>
>> The way those features were implemented in early versions
>> of Midgard would've caused even more MySQL dependency
>> issues than we have now.
> LIKE statements? I've used that one before. Not ideal.
Depending on different sql "dialects" is probably really a bad idea.
Especially if midgard will ever interface with non-sql databases.
> Another option would be to feed the contents of the database
> (in suitable form) to a indexing "backend" such as SWISH++.
I've also thought about this. This is probably the best idea.
Realtime searching is especially for large databases a performance drop.
> Let others do our work, better. The challenge is going
> to be interfacing.
> Swish may actually not be featureful enough (no multilang
> support for one) but this was the only GPL indexer I could
> find on short notice.
A separate program (for cron e.g.) which can be called to update
the index would be cool.
Dont know if there is a indexing system which can easily be modified to
work with a database or with the kind of data in midgard databases.
CYa,
Mario <[EMAIL PROTECTED]> http://delysid.org http://piss.at/
--
This is The Midgard Project's mailing list. For more information,
please visit the project's web site at http://www.midgard-project.org
To unsubscribe the list, send an empty email message to address
[EMAIL PROTECTED]
