lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mark Miller" <markrmil...@gmail.com>
Subject Re: QueryParser syntax French Operator : DONE!
Date Fri, 13 Oct 2006 18:16:02 GMT
Great work Patrick. I was unfamiliar with the use of TOKEN_MGR_DECLS. Looks
like a powerfull feature for dynamic token selection. Thanks a lot,

- Mark

On 10/13/06, Patrick Turcotte <Patrick.Turcotte@usherbrooke.ca> wrote:
>
> Mark Miller wrote:
> > Could you say in a few words what you did to accomplish this? I know
> that
> > you mentioned you used a resource bundle, but what part of the code
> reads
> > this resource bundle? What method did you use to get by the JavaCC
> > issues?
> >
>     Basically:
>
>     * I used TOKEN_MGR_DECLS to declare fields in the TokenManager class.
>
>     * When setting a Locale or setting the useLocalizedOperators options
>       (new method) in QueryParser, I fetch, with ResourceBundle, the
>       strings to match against and set (with token_source.fieldName)
>       values to be used by the TokenManager class
>     * I expanded the token <TERM> to do further processing (as per
>       JavaCC syntax), which create a TokenLixicalActions(Token
>       matchedToken) in the TokenManager class with syntax:
>
> <TERM: <_TERM_START_CHAR> (<_TERM_CHAR>)* >
> {
>     // THIS Will end up in TokenLixicalActions(Token matchedToken) method
>     // DEFINE java processing with matchedToken.image to get to the
> matched string, set matchedToken.kind accordingly.
>     // USES fields set by the QueryParser to decide on behavior
> }
>
>     Hope this answer your question.
>
> Patrick
> > thanks,
> >
> > -Mark
> >
> > On 10/13/06, Patrick Turcotte <Patrick.Turcotte@usherbrooke.ca> wrote:
> >>
> >> Hello!
> >>
> >> This may not be the best place for this message, sorry if this is the
> >> case, but since this is the result of a question I asked here, I
> decided
> >> to post it here. If I'm in error, please refer me to the best
> procedure.
> >> Thanks!
> >>
> >> I've completed the desired "patch". I now have a version of the
> >> QueryParser that can "understand" the AND, OR and NOT keyword in other
> >> languages.
> >>
> >> 1) Patched from revision 454769 of lucene 2.1dev (trunk)
> >> 2) The "ant test" target is still successful when the modified
> >> QueryParser is used
> >> 3) It doesn't break actual code
> >> 4) The default behavior is the same as before
> >> 5) It has to be deliberately activated
> >> 6) It use ResourceBundle to find the keywords translation
> >> 7) Comes with FRENCH translation
> >> 8) Comes with JUnit testCases
> >> 9) Adds 1 public method to QueryParser
> >>
> >> So, now, how do I go about sending it all to the those who can decide
> to
> >> integrate it? Where? In what format? Etc.
> >>
> >> Thanks,
> >>
> >> Patrick Turcotte
> >>
> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> >> For additional commands, e-mail: java-user-help@lucene.apache.org
> >>
> >>
> >
>
>
> --
> Patrick Turcotte
> ------------------------------------
> Patrick.Turcotte@USherbrooke.ca
> (819) 821-8000, poste 63234
> 1-800-267-8337 poste 63234
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message