OPTIONS

-d dictionary

use that dictionary. Several dictionaries may be given. At least one dictionary must be specified. Dictionaries are automata built using fsa_build or fsa_ubuild.

-e edit_distance

if the word is not in the dictionaries, look for words that lie within the specified distance. The edit distance  is a number of basic editing operations needed to transform one string into another. Those operations are: inserting one character, deleting one character, changing one character, and transposing two adjacent characters.

-i input_file

specifies a file with words to be corrected. More than one file can be specified in this way (i.e. the option can be used more than once). In absence of this option, standard input is used.

-l language_file

specifies a file that holds language specific information, i.e. (for now) characters that form words, and pairs of (lowercase, uppercase) characters for case conversion. If the option is not specified, latin letters with standard case conversions wil be used.

-r character_class_file

specifies a file that information about relation between single characters and two-character sequences. If a two-letter sounds the same as one (different) letter, that relation can be established in this file, and the replacement of the first with the second will be treated as being one edit distance  unit apart. The first character of this file is a comment character - all lines beginning with it will be ignored. Data lines consist of two columns. Both columns specify either a single character or a two-character sequence, but the sum of the number of characters should be three. The first column specifies something that may (incorrectly) appear in text, the second one - the correct form.

-v

prints version details.

Software at http://www.pg.gda.pl/~jandac/fsa.html