[TYPO3-dev] sysext's

Ernesto Baschny [cron IT] ernst at cron-it.de
Mon Apr 7 18:50:57 CEST 2008


Michael Stucki wrote: on 01.04.2008 21:12:
> Martin Kutschker wrote:
> 
>>> Our SQL queries have always been pretty simple and I think it should be
>>> possible to port them to other systems without using adodb at all...
>> With due respect, but this is a nive approach. No sir, you cannot solve
>> this without ADOdb or another portability layer without changing the DB
>> layout of TYPO3.
> 
> You say it: Try using another portability layer. To me ADOdb looks like
> overkill, so we should probably look out for something smaller.
> 
> (On the other hand, there is not much feedback from people using it, so I
> cannot argue how much slower the overhead is, and if an improvement is
> needed after all...)
> 
>>> For now I would like to focus on indexed_search and see how far we get if
>>> fulltext indexing is implemented...
>> Did you only add a full text index on top of the existing data structure
>> of indexed serach or did you change more? I'm thinking of getting rid of
>> all the single word tables and replacing them with a "document" table
>> (contains rendered page or copy of external document).
> 
> I dropped index_words. To be precise, it remains there but is not used after
> you install the indexed_search_mysql extension.

Stucki, as you seem to have done research on that matter:

Is there a way to get a "words list" from an MySQL FULLTEXT search 
index? For example: give me all words starting with "abc*"?

This would be essencial for ajax-based auto-completions. If this is 
indeed possible, I would also think about using the MySQL FULLTEXT index 
in some other projects.

Cheers,
Ernesto




More information about the TYPO3-dev mailing list