On Mon, 2006-06-05 at 11:45 -0700, Chris Hostetter wrote:
> thought "Wrapper" was an official pattern name
It is.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
: > you might be able to go with a Delegator/Wrapper .. subclass
: > QueryParser to inherit the grammer, override all of the normal methods
: > you override in a subclass to include your field aliasing, but instead
: > of defering to super defer to another instance of QueryParser
: > specified at
On Mon, 2006-06-05 at 13:45 -0400, Yonik Seeley wrote:
> Karl, taking this in context with your other threads, it seems like
> you are tring to insert extra functionality at many different points
> in lucene. Would some type of AOP work for you?
I'm not sure I know what you mean. Try to decouple
On Mon, 2006-06-05 at 10:41 -0700, Chris Hostetter wrote:
>
> you might be able to go with a Delegator/Wrapper .. subclass
> QueryParser to inherit the grammer, override all of the normal methods
> you override in a subclass to include your field aliasing, but instead
> of defering to super defer
Karl, taking this in context with your other threads, it seems like
you are tring to insert extra functionality at many different points
in lucene. Would some type of AOP work for you?
-Yonik
On 6/5/06, karl wettin <[EMAIL PROTECTED]> wrote:
On Mon, 2006-06-05 at 12:58 -0400, Erik Hatcher wrot
: There is a problem with this. Java does not allow multiple inheritance
: and there are other subclasses of QueryParser. Perhaps I can fix it with
: a decorator instead.
you might be able to go with a Delegator/Wrapper .. subclass QueryParser
to inherit the grammer, override all of the normal me
On Mon, 2006-06-05 at 12:58 -0400, Erik Hatcher wrote:
>
> Huh? Could you elaborate on what won't work about subclassing?
I would for instance have to add the same chunk of code to my
QueryParser- and my MultiFieldQueryParser-subclass.
On Jun 5, 2006, at 11:07 AM, karl wettin wrote:
On Mon, 2006-06-05 at 10:55 +1000, Daniel Noll wrote:
Erik Hatcher wrote:
On Jun 4, 2006, at 5:57 PM, karl wettin wrote:
I was thinking it could be nice if the query parser handled
aliases by
Rather than hacking QueryParser, simply do some r
On Mon, 2006-06-05 at 10:55 +1000, Daniel Noll wrote:
> Erik Hatcher wrote:
> >
> > On Jun 4, 2006, at 5:57 PM, karl wettin wrote:
> >> I was thinking it could be nice if the query parser handled aliases by
> >
> > Rather than hacking QueryParser, simply do some regex replaces on the
> > string
: Indeed! I agree - I was temporarily blinded by the desire to hack
: QueryParser itself in perhaps a one-off way that I didn't consider
: the subclassing option.
if all you care about is aliasing term, and phrase queries yiu might even
be able to use the DisjunctionMaxQueryParser i wrote for So
On Jun 4, 2006, at 8:55 PM, Daniel Noll wrote:
Erik Hatcher wrote:
On Jun 4, 2006, at 5:57 PM, karl wettin wrote:
I was thinking it could be nice if the query parser handled
aliases by
passing a Map to the parser. The data could be compiled from the
index.
"name", "nam", "na" and "n" all
Erik Hatcher wrote:
On Jun 4, 2006, at 5:57 PM, karl wettin wrote:
I was thinking it could be nice if the query parser handled aliases by
passing a Map to the parser. The data could be compiled from the index.
"name", "nam", "na" and "n" all trigger on "name", taking there is no
other field st
On Jun 4, 2006, at 5:57 PM, karl wettin wrote:
I was thinking it could be nice if the query parser handled aliases by
passing a Map to the parser. The data could be compiled from the
index.
"name", "nam", "na" and "n" all trigger on "name", taking there is no
other field starting with an "n"
I was thinking it could be nice if the query parser handled aliases by
passing a Map to the parser. The data could be compiled from the index.
"name", "nam", "na" and "n" all trigger on "name", taking there is no
other field starting with an "n".
Did anyone implement this, or should I hack one u
14 matches
Mail list logo
