Rob van Maris <[EMAIL PROTECTED]> wrote:
> > Actually, I would have implemented it already for postgresql 
> > (openFTS) and mysql (support is native I think), but I still 
> > don't really know how this functionality then should be 
> > called from e.g. taglib. I would expect a pseudo operator 
> > like 'FTS' or so, but it seams that it is needed to use 
> > LegacyConstraint.
> 
> Not correct. 
> 
> See the javadoc of org.mmbase.storage.search.legacy.ConstraintParser,

Ah well, something with legacy and constraints, any how.

> for the syntax of constraints that can be parsed into Constraint objects
> (non-legacy). That means this syntax can safely be used in the taglib
> and bridge methods.

Yes, but in taglib/bridge there is no need to construct strings first.
(Though it is possible)

> Here you'll find the syntax of the string-search-condition, that is
> translated into a StringSearchConstraint. It may look a bit cumbersome
> at first, but it's just a generalization of the various pseudo-operators
> like FTS, used by different vendors.


I do more or less understand it now. I think I could also construct a
BasicStringSearchConstraint, isn't it? So I don't have to use the parsing of
ConstraintParser?

So, I then only need a way to implement it e.g. for mysql (which has a
rather straightforward way of doing it). I think that would perhaps involve
extending ChainedSqlHandler, in wich getSuccessor().appendConstraintToSql
must be called?

Remains only the question where I should then plug it in, since I do not
find how that would be instantiated or configured in StorageManagerFactory.
Also storage_1_1.dtd only sais that the SqlHandler is a class, not 2 or more
classes. Do I overlook something?
[A
 MIchiel



-- 
Michiel Meeuwissen 
Mediapark C101 Hilversum  
+31 (0)35 6772979
nl_NL eo_XX en_US
mihxil'
 [] ()

Reply via email to