AUTO_LEXER Language Support

At index time, AUTO_LEXER automatically detects the language of the document, and tokenizes and stems the document appropriately.

AUTO_LEXER Dictionary

To specify an AUTO_LEXER dictionary, use the name of the dictionary you created instead of the file name for the dictionary.

At query time, the language of the query is inherited from the query template. If the query template is not used, or if no language is specified in the query template, then the language of the query is inherited from the session language.

Note: The dictionary data is not processed until the index or policy creation time or until the ALTER INDEX time. Errors in dictionary data format are caught at the index or policy creation time or at the ALTER INDEX time, and are reported as the “DRG-13710: Syntax Error in Dictionary” error.

AUTO_LEXER Component Versions

Starting with Oracle Database 26ai, the AUTO_LEXER component supports version ANL6, which is shipped with the Oracle Database installation by default.

The earlier version (ANL1) of the AUTO_LEXER component is available as an optional download patch. If you want to use version ANL1 to retain the prior language behavior for backward compatibility, then you can download ANL1 from My Oracle Support. After downloading the component, you must set Event 30579 Level 1048576 at the SYSTEM level.

Languages Distribution Model

• By default, Oracle Text ships language data files for only some of the languages supported for the AUTO_LEXER component. You can download data files for all other languages on demand from My Oracle Support using optional download patches. This language patch mechanism helps you control the installed languages and thus reduce the size of the database distribution for on-premises deployments.

• These language data files are included with the Oracle Database installation by default:

  Arabic	Korean
  Bokmal (Norwegian)	Nynorsk (Norwegian)
  Catalan	Persian
  Croatian	Polish
  Czech	Portuguese
  Danish	Romanian
  Dutch	Russian
  English	Serbian
  Finnish	Simplified Chinese (see Note)
  French	Slovak
  German	Slovenian
  Greek	Spanish
  Hebrew	Swedish
  Hungarian	Thai
  Italian	Traditional Chinese (see Note)
  Japanese	Turkish

  Note: Due to the limitation of 30 characters for the string, Traditional Chinese must be specified as trad_chinese. Simplified Chinese must be specified as simp_chinese.

• You can download these language data files from My Oracle Support using optional download patches:

  Afrikaans	Indonesian
  Basque	Latvian
  Belarusian	Lithuanian
  Bulgarian	Macedonian
  Estonian	Malay
  Galician	Ukrainian
  Hindi	Urdu
  Icelandic	-