[
https://issues.apache.org/jira/browse/LUCENE-2968?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13008175#comment-13008175
]
Paul Elschot commented on LUCENE-2968:
--------------------------------------
I had a look at the precedence parser. It would probably have been worthwhile
to use it from the start, but now it would be an almost complete rewrite, which
is too much for relatively small changes.
OTOH, after looking at the Surround code into some depth again after a few
years, I would do a few things differently when now.
The ...Query names in the surround.query package would indeed better be
...Node, just as in the precedence parser. Also the check for allowed distance
subqueries would be better done in the parser itself by restricting the syntax.
Another thing that I don't really like now is the lack of weights in span query
scoring. It works without weighting now, but there is room for improvement.
Which brings me to Spans. It is an abstract class nowadays (and no more an
interface, so a float score() or float weight() could at least be added. But
there has also been talk about removing Spans entirely, how is the situation
there?
Anyway, only adding support for SpanNot in Surround should be straightforward,
I'll give it a try.
> SurroundQuery doesn't support SpanNot
> -------------------------------------
>
> Key: LUCENE-2968
> URL: https://issues.apache.org/jira/browse/LUCENE-2968
> Project: Lucene - Java
> Issue Type: Improvement
> Reporter: Grant Ingersoll
> Priority: Minor
>
> It would be nice if we could do span not in the surround query, as they are
> quite useful for keeping searches within a boundary (say a sentence)
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
