You are viewing a plain text version of this content. The canonical link for it is here.
Posted to java-user@lucene.apache.org by "Johannes.Lichtenberger" <Jo...@uni-konstanz.de> on 2012/07/28 17:44:00 UTC
Revisioned Lucene Index
Hello,
I'm currently working on revisioned index structures for a treebased
storage system[1] and I want to provide an index-structure for fulltext
search (more or less on XML text-nodes). Either I'm going to implement a
Radix/PATRICIA-tree or I'm opting for Lucene. I thought about adding a
node <=> Field mapping , that would probably be sufficient as we do not
need a tree-structure in order to support revisioning strategies (full
dump, incremental, differential). However, I guess Lucene uses some kind
of tree-structure itself though perhaps it would be more appropriate to
map these nodes to our "nodes".
Any suggestions?
kind regards,
Johannes
[1] https://github.com/JohannesLichtenberger/sirix
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
AW: Revisioned Lucene Index
Posted by Lutz Fechner <LF...@hubwoo.com>.
Hi,
Lucene is storing it's data pretty much flat. You have Documents representing a seach result. This Documents are created during the indexing process you have to implement.
For XML data I would recommend to store the Xpath of the indexed data in a field in the lucene documents in order to get the nodes addessed again in the XML tree.
Best Regards
Lutz
--------------------------
...via my BlackBerry Wireless Handheld
----- Originalnachricht -----
Von: Johannes.Lichtenberger [mailto:Johannes.Lichtenberger@uni-konstanz.de]
Gesendet: Saturday, July 28, 2012 05:44 PM
An: java-user@lucene.apache.org <ja...@lucene.apache.org>
Betreff: Revisioned Lucene Index
Hello,
I'm currently working on revisioned index structures for a treebased
storage system[1] and I want to provide an index-structure for fulltext
search (more or less on XML text-nodes). Either I'm going to implement a
Radix/PATRICIA-tree or I'm opting for Lucene. I thought about adding a
node <=> Field mapping , that would probably be sufficient as we do not
need a tree-structure in order to support revisioning strategies (full
dump, incremental, differential). However, I guess Lucene uses some kind
of tree-structure itself though perhaps it would be more appropriate to
map these nodes to our "nodes".
Any suggestions?
kind regards,
Johannes
[1] https://github.com/JohannesLichtenberger/sirix
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org