You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Terje Eggestad (JIRA)" <ji...@apache.org> on 2010/08/04 13:46:18 UTC
[jira] Created: (LUCENE-2587) Highlighter picks wrong offset for
fragment boundaries
Highlighter picks wrong offset for fragment boundaries
------------------------------------------------------
Key: LUCENE-2587
URL: https://issues.apache.org/jira/browse/LUCENE-2587
Project: Lucene - Java
Issue Type: Bug
Components: contrib/highlighter
Affects Versions: 3.0.2
Environment: Java 6 + Lucene 3.0.2
Reporter: Terje Eggestad
Priority: Trivial
I have written a new Fragmenter since we need fragments for hitlines to be on sentence boundaries and not cross paragraphs.
When using it with org.apache.lucene.search.highlight.Highlighter, I get hitlines that starts with ". ", "? ", "! "...
Consider the text "A b c d e. F g h i j! K l m n o. "
which become the tokenstream : (A) (b) (c) (d) (e) (F) (g) (h) (i) (j) (K) (l) (m) (n) (o)
If the fragmenter return isNewFragment() = true on F and K and Highlighter pick the middle fragment, lets say we search on "g" the hitline becomes:
". F <B>g</B> h i j"
The reason, it seems, is that the offset to the fragment boundaries found by taking the endOffset of the last token in a fragment ,
not the startOffset of the first.
TJ
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org