olatool

This is a command-line tool that provides an interface to the Open Linguistic Architecture library.

olatool [options] [INPUTFILE]

If the INPUTFILE is not specified, it will read from stdin.

Command-line options to the olatool:

• --lang=STRING - one or more to specify the language(s) of the input document. Possible values for STRING are:
  • an ISO 639 code (both two or three letter codes accepted), possibly extended with country and variant codes to specify an exact language locale
  • the string mono - it indicates that the document is monolingual; if used alone, the built-in language identifier will be utilised to guess the language; if used together with one specific language code (in a separate --lang option), it is superfluous; if used together with multiple language specifications it is meaningless and will be ignored
  • the string multi - it indicates that the document is multilingual; if used alone, the built-in language identifier will be utilised to guess the languages; if used together with one specific language code (in a separate --lang option), it triggers the language identifier to try to identify other languages than the given; if used together with multiple language specifications the specified languages take precedence, and it will be ignored.
• --statistics

The output depends on the linguistic processing requested, but is always given to stout. The following lists the output for each processing type.

Speller

The speller gives as output the following:

• optionally a header with some statistical information, see below
• input token
• speller response (correct, incorrect, ...)
• possibly a(n) (empty) list of suggestions

The header has the following format (the given headers are just indicative of what could be put in the header):

//Date:
//Time:
//Time elapsed:
//Speller lexicon version:
//...
//Unrecognised words:
//Candidates for lexicalisation:
//...

The format of the rest of the speller output data is as follows:

Hyphenator

dfghjk