You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@jena.apache.org by "Code Ferret (JIRA)" <ji...@apache.org> on 2018/10/29 14:32:10 UTC
[jira] [Commented] (JENA-1621) Upgrade Lucene to 7.5.x
[ https://issues.apache.org/jira/browse/JENA-1621?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16667289#comment-16667289 ]
Code Ferret commented on JENA-1621:
-----------------------------------
The indexes have to be upgraded following the directions at https://lucene.apache.org/solr/guide/7_5/indexupgrader-tool.html, as you’ve noted on dev@jena.apache.org.
The indexes for Lucene versions earlier than 6.x can not be converted directly to 7.x, thus users running on Jena prior to 3.3.0 (released 21 May 2017) will have to reload their data when moving to a Jena with Lucene 7.x.
There are quite a few changes noted at https://lucene.apache.org/core/7_5_0/MIGRATE.html that could impact Jena users. In particular:
Changed SPI lookups for codecs and analysis changed ([LUCENE-7873|https://issues.apache.org/jira/browse/LUCENE-7873])
Due to serious problems with context class loaders in several frameworks (OSGI, Java 9 Jigsaw), the lookup of Codecs, PostingsFormats, DocValuesFormats and all analysis factories was changed to only inspect the current classloader that defined the interface class (lucene-core.jar). Normal applications should not encounter any issues with that change, because the application classloader (unnamed module in Java 9) can load all SPIs from all JARs from classpath.
. . .
and
CustomAnalyzer resources ([LUCENE-7883|https://issues.apache.org/jira/browse/LUCENE-7883])##
Lucene no longer uses the context class loader when resolving resources in CustomAnalyzer or ClassPathResourceLoader. Resources are only resolved against Lucene's class loader by default. Please use another builder method to change to a custom classloader.
and
AnalyzingQueryParser removed ([LUCENE-7355|https://issues.apache.org/jira/browse/LUCENE-7355])
The functionality of AnalyzingQueryParser has been folded into the classic QueryParser, which now passes terms through Analyzer#normalize when generating queries.
This last one is already noted implicitly in [Jean-Marc Vanel’s note|http://mail-archives.us.apache.org/mod_mbox/jena-users/201810.mbox/%3CCANwvFKBHygwGchb1MZWxN1RNW-N6qb9zyQaMkmh-VoZgpoAVEQ@mail.gmail.com%3E].
I’m not sure what all needs to be checked to determine how the class loader changes may impact Fuseki and other Jena deployments.
There have been quite a few improvements and fixes in the geo-spatial functionality.
It seems like a good idea to try and migrate rather than getting stuck in a situation where there’s a larger mountain to climb and the index updater for Lucene 8.x won’t work with Jena versions using Lucene 6.4.1.
I’m willing to look more deeply at this. I have a bug fix for jena-text highlighting that I’ll raise an issue and a PR for next week.
> Upgrade Lucene to 7.5.x
> -----------------------
>
> Key: JENA-1621
> URL: https://issues.apache.org/jira/browse/JENA-1621
> Project: Apache Jena
> Issue Type: Improvement
> Components: Text
> Affects Versions: Jena 3.9.0
> Reporter: Andy Seaborne
> Priority: Major
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)