NAME

Grapheme::Ngram - n-grams of Unicode Extended Grapheme Clusters

SYNOPSIS

 use Grapheme::Ngram;

 my $class = 'Grapheme::Ngram';
 my @ngrams = $class->ngram($string,$width);

DESCRIPTION

For many applications it's better to work along graphemes.

Building n-grams is one of them.

METHODS

new

  $object = Grapheme::Ngram->new();

ngram

  my $array_ref = $object->ngram($string, $width);

$string ...... string of characters

$width ....... length of the resulting tokens. Default is 1.

$array_ref ... reference to array of ngram tokens

Returns one token with the unmodified $string if the number of graphemes in $string is lower than $width. Returns an empty $array_ref if $string is empty or undef. NOTE: maybe this will be changed in future. Defaults to length = 1 if $width is not an integer larger than 0.

from_tokens

  my @ngram = $object->from_tokens(\@tokens, $width);

Same as ngram but takes tokens. This method is used by ngram.

This allows to use a custom tokenizer for e.g. treating 'sh' also as grapheme:

  my @tokens = $string =~ m/(Sh|sh|\X)/g;

_tokenize

  my @graphemes = $object->_tokenize($string);

This internal method splits $string into a list of graphemes.

SOURCE REPOSITORY

http://github.com/wollmers/Grapheme-Ngram

AUTHOR

Helmut Wollmersdorfer, <helmut.wollmersdorfer@gmail.com>

COPYRIGHT AND LICENSE

Copyright (C) 2014 by Helmut Wollmersdorfer

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

cpanm Grapheme::Ngram

perl -MCPAN -e shell
install Grapheme::Ngram