lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Erik Hatcher <>
Subject Re: Spans, appended fields, and term positions
Date Mon, 21 Nov 2005 20:41:45 GMT

On 21 Nov 2005, at 12:55, Yonik Seeley wrote:
> On 11/21/05, Erik Hatcher <> wrote:
>> Modifying Analyzer as you have suggested would
>> require DocumentWriter additionally keep track of the field names
>> and note when one is used again.
> For position increments, it doesn't have to be tracked. The patch to
> DocumentWriter could also be:
>       int position = fieldPositions[fieldNumber];
> +      if (position>0) position+=analyzer.getPositionIncrementGap 
> (fieldName)

This could be thwarted with tokens using zero position offsets  
though, no?

> Talk of position increments feels like it belongs more in Analyzer.
> Do you actually need to control that per field instance (rather than
> per field name/type?)

For my purpose it is actually a global need - all successive  
instances of fields will have this offset specified, but setting this  
on a per-Field instance is no problem.

> I think your patch straight-forward and probably the simplest for the
> user, and the most flexible (if it's needed to that degree). I just
> don't know if anyone has objections to expanding the Field class even
> more (and with analysis stuff), so I'm throwing out possible
> alternatives.

It is either expand Field slightly, or expand Analyzer.  The Analyzer  
extensions seem fine, but much more general purpose than my need.  So  
being practical, if there is no current use case then we should stick  
with the concrete.

As for Doug's comments, I'll add javadocs for sure, and increase the  
warnings on getBoost() while I'm at it.  In fact, getPositionOffset()  
isn't technically needed (nor getBoost() for that matter), but I'll  
add it just to be complete, but add thorough warnings.


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message