History

Tomoko Uchida f03e6aac59 SOLR-14429: Convert .txt files to properly formatted .md files (#1450 )		2020-04-27 08:43:04 +09:00
..
src	SOLR-14013: javabin performance regressions	2019-12-12 23:26:37 +11:00
README.md	SOLR-14429: Convert .txt files to properly formatted .md files (#1450 )	2020-04-27 08:43:04 +09:00
build.gradle	LUCENE-9182: add apache license headers to all .gradle files and enforce in rat task	2020-01-27 12:05:34 -05:00
build.xml	LUCENE-8807: Change all download URLs in build files to HTTPS	2019-05-21 17:06:00 +02:00
ivy.xml	SOLR-11592: Add OpenNLP language detection to the langid contrib	2018-01-17 11:29:17 -05:00

README.md

Apache Solr Language Identifier

Introduction

This module is intended to be used while indexing documents. It is implemented as an UpdateProcessor to be placed in an UpdateChain. Its purpose is to identify language from documents and tag the document with language code. The module can optionally map field names to their language specific counterpart, e.g. if the input is "title" and language is detected as "en", map to "title_en". Language may be detected globally for the document, and/or individually per field. Language detector implementations are pluggable.

Getting Started

Please refer to the module documentation at http://wiki.apache.org/solr/LanguageDetection

Dependencies

The Tika detector depends on Tika Core (which is part of extraction contrib) The Langdetect detector depends on LangDetect library The OpenNLP detector depends on OpenNLP tools and requires a previously trained user-supplied model