Class: Kotoshu::Components::Tokenizer Abstract

Inherits:
Object
  • Object
show all
Defined in:
lib/kotoshu/components/tokenizer.rb

Overview

This class is abstract.

Subclasses must implement #tokenize

Base class for tokenizers.

Tokenizers split text into individual tokens (words, punctuation). Different languages use different tokenization strategies:

  • Latin scripts: Whitespace + punctuation
  • CJK: Morphological analysis
  • German: Compound word splitting
  • RTL: Right-to-left text handling

Examples:

Tokenizing English text

tokenizer = WhitespaceTokenizer.new
tokens = tokenizer.tokenize("Hello, world!")
# => [
#      { token: "Hello", position: 0, length: 5 },
#      { token: ",", position: 5, length: 1 },
#      { token: "world", position: 7, length: 5 },
#      { token: "!", position: 12, length: 1 }
#    ]

Direct Known Subclasses

WhitespaceTokenizer

Instance Method Summary collapse

Instance Method Details

#tokenize(text) ⇒ Array<Hash>

This method is abstract.

Subclasses must implement

Split text into tokens.

Each token is a hash with:

  • :token (String) - The token text
  • :position (Integer) - Character position in original text
  • :length (Integer) - Token length in characters

Additional keys may be added by subclasses:

  • :pos_tag (String) - Part of speech tag
  • :lemma (String) - Base form / lemma
  • :compound_part (Boolean) - Whether this is a compound word part
  • :script (Symbol) - Script type for multilingual text

Parameters:

  • text (String) —

    The input text

Returns:

  • (Array<Hash>) —

    Array of token hashes

Raises:

  • (NotImplementedError) —

    if not implemented by subclass



43
44
45
 
# File 'lib/kotoshu/components/tokenizer.rb', line 43

def tokenize(text)
  raise NotImplementedError, "#{self.class} must implement #tokenize"
end

#tokenize_to_strings(text) ⇒ Array<String>

Tokenize and return just the token strings.

Convenience method for when you only need the text content.

Parameters:

  • text (String) —

    The input text

Returns:

  • (Array<String>) —

    Array of token strings



53
54
55
 
# File 'lib/kotoshu/components/tokenizer.rb', line 53

def tokenize_to_strings(text)
  tokenize(text).map { |t| t[:token] }
end