PGNet Dev
2020-Nov-02 14:48 UTC
removing 'use_libfts' from dovecot fts_solr config; docs/examples of solr-backend-only tokenizer config for dovecot?
given existing problems with fts_solr = ... use_libfts ... i've rm'd it from dovecot config, and am moving all tokenization to solr config. iiuc, in dovecot + solr context, that's correctly config'd in solr/data/dovecot/conf/schema.xml dovecot release ships with 'starter configs', cp -af /usr/share/doc/dovecot/solr-config-7.7.0.xml /path/to/solr/data/dovecot/conf/solrconfig.xml cp -af /usr/share/doc/dovecot/solr-schema-7.7.0.xml /path/to/solr/data/dovecot/conf/schema.xml that^ schema.xml contains a limited solr.StandardTokenizerFactory config. it can of course be extended; lots of generic solr config info available. is there _documentation_, preferably a good example, of a dovecot-'friendly' solr config for non-"use_libfts" use case? specifically, moving any/all of the libfts front-end tokenizers @ https://doc.dovecot.org/configuration_manual/fts/tokenization/#fts-tokenization https://doc.dovecot.org/settings/plugin/fts-plugin/#plugin-fts-setting-fts-tokenizers fully to a backend solr instance? i suspect that solr config is fairly well insulated from dovecot's; what i'm not sure of is if dovecot's fts has specific requirements to get the backend-only tokenization working.
Apparently Analagous Threads
- v2.3.11.3 solr plugin search via MUA fails to match accented ascii characters; cmd line exec of `doveadm fts lookup` PANICs (assertion failed) [proposed patch]
- v2.3.11.3 solr plugin search via MUA fails to match accented ascii characters; cmd line exec of `doveadm fts lookup` PANICs (assertion failed) [proposed patch]
- v2.3.11.3 solr plugin search via MUA fails to match accented ascii characters; cmd line exec of `doveadm fts lookup` PANICs (assertion failed) [proposed patch]
- v2.3.11.3 solr plugin search via MUA fails to match accented ascii characters; cmd line exec of `doveadm fts lookup` PANICs (assertion failed)
- Dovecot v2.3.18 released