kotoshu 0.3.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (210) hide show
  1. checksums.yaml +7 -0
  2. data/.rspec +3 -0
  3. data/.rubocop.yml +18 -0
  4. data/CHANGELOG.md +182 -0
  5. data/CLAUDE.md +172 -0
  6. data/CODE_OF_CONDUCT.md +132 -0
  7. data/LICENSE +31 -0
  8. data/README.adoc +955 -0
  9. data/Rakefile +12 -0
  10. data/SECURITY.md +93 -0
  11. data/examples/01_basic_word_checking.rb +38 -0
  12. data/examples/02_text_document_checking.rb +77 -0
  13. data/examples/03_dictionary_backends.rb +137 -0
  14. data/examples/04_trie_data_structure.rb +146 -0
  15. data/examples/05_suggestion_algorithms.rb +239 -0
  16. data/examples/06_configuration_advanced.rb +287 -0
  17. data/examples/07_multi_language_dictionaries.rb +278 -0
  18. data/exe/kotoshu +6 -0
  19. data/lib/kotoshu/algorithms/capitalization.rb +276 -0
  20. data/lib/kotoshu/algorithms/lookup.rb +876 -0
  21. data/lib/kotoshu/algorithms/ngram_suggest.rb +270 -0
  22. data/lib/kotoshu/algorithms/permutations.rb +283 -0
  23. data/lib/kotoshu/algorithms/phonet_suggest.rb +167 -0
  24. data/lib/kotoshu/algorithms/suggest.rb +575 -0
  25. data/lib/kotoshu/algorithms.rb +14 -0
  26. data/lib/kotoshu/analyzers/semantic_analyzer.rb +295 -0
  27. data/lib/kotoshu/cache/base_cache.rb +596 -0
  28. data/lib/kotoshu/cache/cache.rb +91 -0
  29. data/lib/kotoshu/cache/frequency_cache.rb +224 -0
  30. data/lib/kotoshu/cache/language_cache.rb +454 -0
  31. data/lib/kotoshu/cache/lookup_cache.rb +166 -0
  32. data/lib/kotoshu/cache/model_cache.rb +513 -0
  33. data/lib/kotoshu/cache/suggestion_cache.rb +113 -0
  34. data/lib/kotoshu/cache.rb +40 -0
  35. data/lib/kotoshu/cli/auto_setup.rb +71 -0
  36. data/lib/kotoshu/cli/batch_reporter.rb +315 -0
  37. data/lib/kotoshu/cli/cache_command.rb +356 -0
  38. data/lib/kotoshu/cli/display_formatter.rb +431 -0
  39. data/lib/kotoshu/cli/errors.rb +36 -0
  40. data/lib/kotoshu/cli/interactive_reviewer.rb +319 -0
  41. data/lib/kotoshu/cli/language_resolver.rb +91 -0
  42. data/lib/kotoshu/cli/navigation_manager.rb +272 -0
  43. data/lib/kotoshu/cli/progress_reporter.rb +114 -0
  44. data/lib/kotoshu/cli/status_report.rb +130 -0
  45. data/lib/kotoshu/cli.rb +627 -0
  46. data/lib/kotoshu/commands/cache_command.rb +424 -0
  47. data/lib/kotoshu/commands/check_command.rb +312 -0
  48. data/lib/kotoshu/commands/model_command.rb +295 -0
  49. data/lib/kotoshu/components/passthrough_spell_checker.rb +72 -0
  50. data/lib/kotoshu/components/pos_tagger.rb +98 -0
  51. data/lib/kotoshu/components/spell_checker.rb +73 -0
  52. data/lib/kotoshu/components/synthesizer.rb +60 -0
  53. data/lib/kotoshu/components/tokenizer.rb +58 -0
  54. data/lib/kotoshu/components/whitespace_tokenizer.rb +96 -0
  55. data/lib/kotoshu/configuration/builder.rb +209 -0
  56. data/lib/kotoshu/configuration/resolver.rb +124 -0
  57. data/lib/kotoshu/configuration.rb +702 -0
  58. data/lib/kotoshu/core/exceptions.rb +165 -0
  59. data/lib/kotoshu/core/indexed_dictionary.rb +291 -0
  60. data/lib/kotoshu/core/models/affix_rule.rb +260 -0
  61. data/lib/kotoshu/core/models/result/document_result.rb +263 -0
  62. data/lib/kotoshu/core/models/result/word_result.rb +203 -0
  63. data/lib/kotoshu/core/models/word.rb +142 -0
  64. data/lib/kotoshu/core/trie/builder.rb +119 -0
  65. data/lib/kotoshu/core/trie/node.rb +94 -0
  66. data/lib/kotoshu/core/trie/trie.rb +249 -0
  67. data/lib/kotoshu/core.rb +28 -0
  68. data/lib/kotoshu/data/common_words/de.yml +1800 -0
  69. data/lib/kotoshu/data/common_words/en.yml +1215 -0
  70. data/lib/kotoshu/data/common_words/es.yml +750 -0
  71. data/lib/kotoshu/data/common_words/fr.yml +1015 -0
  72. data/lib/kotoshu/data/common_words/pt.yml +870 -0
  73. data/lib/kotoshu/data/common_words/ru.yml +484 -0
  74. data/lib/kotoshu/data/common_words_loader.rb +152 -0
  75. data/lib/kotoshu/data_structures/bloom_filter.rb +176 -0
  76. data/lib/kotoshu/debug_logger.rb +146 -0
  77. data/lib/kotoshu/debug_mode.rb +134 -0
  78. data/lib/kotoshu/defaults.rb +86 -0
  79. data/lib/kotoshu/dictionaries/catalog.rb +817 -0
  80. data/lib/kotoshu/dictionary/base.rb +237 -0
  81. data/lib/kotoshu/dictionary/cspell.rb +254 -0
  82. data/lib/kotoshu/dictionary/custom.rb +224 -0
  83. data/lib/kotoshu/dictionary/hunspell.rb +526 -0
  84. data/lib/kotoshu/dictionary/plain_text.rb +282 -0
  85. data/lib/kotoshu/dictionary/repository.rb +248 -0
  86. data/lib/kotoshu/dictionary/unified.rb +260 -0
  87. data/lib/kotoshu/dictionary/unix_words.rb +218 -0
  88. data/lib/kotoshu/documents/asciidoc_document.rb +441 -0
  89. data/lib/kotoshu/documents/document.rb +229 -0
  90. data/lib/kotoshu/documents/location.rb +139 -0
  91. data/lib/kotoshu/documents/markdown_document.rb +389 -0
  92. data/lib/kotoshu/documents/plain_text_document.rb +147 -0
  93. data/lib/kotoshu/embeddings/embedding_pipeline.rb +244 -0
  94. data/lib/kotoshu/embeddings/lru_cache.rb +233 -0
  95. data/lib/kotoshu/embeddings/onnx_runtime_model.rb +388 -0
  96. data/lib/kotoshu/embeddings/protocol.rb +83 -0
  97. data/lib/kotoshu/embeddings/protocols.rb +17 -0
  98. data/lib/kotoshu/embeddings/registry.rb +182 -0
  99. data/lib/kotoshu/embeddings/search.rb +192 -0
  100. data/lib/kotoshu/embeddings/similarity_engine.rb +248 -0
  101. data/lib/kotoshu/embeddings/similarity_search.rb +331 -0
  102. data/lib/kotoshu/embeddings/vocabulary.rb +257 -0
  103. data/lib/kotoshu/embeddings.rb +97 -0
  104. data/lib/kotoshu/fluent_checker.rb +91 -0
  105. data/lib/kotoshu/grammar/pattern_matchers/base_matcher.rb +48 -0
  106. data/lib/kotoshu/grammar/pattern_matchers/double_negative_matcher.rb +105 -0
  107. data/lib/kotoshu/grammar/pattern_matchers/possessive_context_matcher.rb +77 -0
  108. data/lib/kotoshu/grammar/pattern_matchers/vowel_sound_matcher.rb +83 -0
  109. data/lib/kotoshu/grammar/rule.rb +95 -0
  110. data/lib/kotoshu/grammar/rule_engine.rb +111 -0
  111. data/lib/kotoshu/grammar/rule_loader.rb +31 -0
  112. data/lib/kotoshu/grammar.rb +18 -0
  113. data/lib/kotoshu/integrity/audit_log.rb +88 -0
  114. data/lib/kotoshu/integrity/manifest.rb +117 -0
  115. data/lib/kotoshu/integrity/net_http.rb +46 -0
  116. data/lib/kotoshu/integrity.rb +25 -0
  117. data/lib/kotoshu/keyboard/layout.rb +115 -0
  118. data/lib/kotoshu/keyboard/layouts/azerty.rb +57 -0
  119. data/lib/kotoshu/keyboard/layouts/dvorak.rb +56 -0
  120. data/lib/kotoshu/keyboard/layouts/jcuken.rb +59 -0
  121. data/lib/kotoshu/keyboard/layouts/qwerty.rb +54 -0
  122. data/lib/kotoshu/keyboard/layouts/qwertz.rb +57 -0
  123. data/lib/kotoshu/keyboard/registry.rb +146 -0
  124. data/lib/kotoshu/keyboard.rb +60 -0
  125. data/lib/kotoshu/language/detector.rb +242 -0
  126. data/lib/kotoshu/language/identifier.rb +378 -0
  127. data/lib/kotoshu/language/languages/base.rb +256 -0
  128. data/lib/kotoshu/language/normalizer/base.rb +137 -0
  129. data/lib/kotoshu/language/registry.rb +147 -0
  130. data/lib/kotoshu/language/resources/ar/common_words.txt +6753 -0
  131. data/lib/kotoshu/language/resources/ar/confusion_sets.txt +11 -0
  132. data/lib/kotoshu/language/resources/de/common_words.txt +10003 -0
  133. data/lib/kotoshu/language/resources/de/confusion_sets.txt +246 -0
  134. data/lib/kotoshu/language/resources/en/common_words.txt +9979 -0
  135. data/lib/kotoshu/language/resources/en/confusion_sets.txt +871 -0
  136. data/lib/kotoshu/language/resources/es/common_words.txt +9992 -0
  137. data/lib/kotoshu/language/resources/es/confusion_sets.txt +17 -0
  138. data/lib/kotoshu/language/resources/fr/common_words.txt +9993 -0
  139. data/lib/kotoshu/language/resources/fr/confusion_sets.txt +76 -0
  140. data/lib/kotoshu/language/resources/pt/common_words.txt +9977 -0
  141. data/lib/kotoshu/language/resources/pt/confusion_sets.txt +18 -0
  142. data/lib/kotoshu/language/resources/ru/common_words.txt +9951 -0
  143. data/lib/kotoshu/language/resources/ru/confusion_sets.txt +5 -0
  144. data/lib/kotoshu/language/tokenizer/base.rb +170 -0
  145. data/lib/kotoshu/language/tokenizer/french_tokenizer.rb +170 -0
  146. data/lib/kotoshu/language/tokenizer/german_tokenizer.rb +41 -0
  147. data/lib/kotoshu/language/tokenizer/japanese_tokenizer.rb +60 -0
  148. data/lib/kotoshu/language/tokenizer/latin_tokenizer.rb +141 -0
  149. data/lib/kotoshu/language/tokenizer/portuguese_tokenizer.rb +160 -0
  150. data/lib/kotoshu/language/tokenizer/russian_tokenizer.rb +95 -0
  151. data/lib/kotoshu/language/tokenizer/spanish_tokenizer.rb +122 -0
  152. data/lib/kotoshu/language.rb +99 -0
  153. data/lib/kotoshu/languages/de/language.rb +546 -0
  154. data/lib/kotoshu/languages/en/language.rb +448 -0
  155. data/lib/kotoshu/languages/es/language.rb +459 -0
  156. data/lib/kotoshu/languages/fr/language.rb +493 -0
  157. data/lib/kotoshu/languages/ja/language.rb +477 -0
  158. data/lib/kotoshu/languages/pt/language.rb +423 -0
  159. data/lib/kotoshu/languages/ru/language.rb +404 -0
  160. data/lib/kotoshu/languages.rb +43 -0
  161. data/lib/kotoshu/metrics_collector.rb +222 -0
  162. data/lib/kotoshu/metrics_module.rb +110 -0
  163. data/lib/kotoshu/models/context.rb +119 -0
  164. data/lib/kotoshu/models/embedding_model.rb +182 -0
  165. data/lib/kotoshu/models/fasttext_model.rb +220 -0
  166. data/lib/kotoshu/models/nearest_neighbor.rb +87 -0
  167. data/lib/kotoshu/models/onnx_model.rb +333 -0
  168. data/lib/kotoshu/models/semantic_error.rb +165 -0
  169. data/lib/kotoshu/models/suggestion.rb +106 -0
  170. data/lib/kotoshu/models/word_embedding.rb +107 -0
  171. data/lib/kotoshu/paths.rb +53 -0
  172. data/lib/kotoshu/personal_dictionary.rb +94 -0
  173. data/lib/kotoshu/plugins/plugin.rb +61 -0
  174. data/lib/kotoshu/plugins/registry.rb +120 -0
  175. data/lib/kotoshu/project_config.rb +76 -0
  176. data/lib/kotoshu/readers/aff_data.rb +356 -0
  177. data/lib/kotoshu/readers/aff_reader.rb +375 -0
  178. data/lib/kotoshu/readers/condition_checker.rb +142 -0
  179. data/lib/kotoshu/readers/dic_reader.rb +118 -0
  180. data/lib/kotoshu/readers/file_reader.rb +347 -0
  181. data/lib/kotoshu/readers/lookup_builder.rb +299 -0
  182. data/lib/kotoshu/readers/readers.rb +6 -0
  183. data/lib/kotoshu/readers.rb +9 -0
  184. data/lib/kotoshu/resource_bundle.rb +30 -0
  185. data/lib/kotoshu/resource_manager.rb +295 -0
  186. data/lib/kotoshu/results/result.rb +165 -0
  187. data/lib/kotoshu/scripts/fasttext_to_onnx.py +275 -0
  188. data/lib/kotoshu/source_registry.rb +74 -0
  189. data/lib/kotoshu/spellchecker/parallel_checker.rb +90 -0
  190. data/lib/kotoshu/spellchecker.rb +298 -0
  191. data/lib/kotoshu/string_metrics.rb +153 -0
  192. data/lib/kotoshu/suggestions/context.rb +55 -0
  193. data/lib/kotoshu/suggestions/generator.rb +175 -0
  194. data/lib/kotoshu/suggestions/pipeline.rb +135 -0
  195. data/lib/kotoshu/suggestions/strategies/base_strategy.rb +296 -0
  196. data/lib/kotoshu/suggestions/strategies/composite_strategy.rb +140 -0
  197. data/lib/kotoshu/suggestions/strategies/edit_distance_strategy.rb +671 -0
  198. data/lib/kotoshu/suggestions/strategies/keyboard_proximity_strategy.rb +228 -0
  199. data/lib/kotoshu/suggestions/strategies/ngram_strategy.rb +130 -0
  200. data/lib/kotoshu/suggestions/strategies/phonetic_strategy.rb +329 -0
  201. data/lib/kotoshu/suggestions/strategies/semantic_strategy.rb +316 -0
  202. data/lib/kotoshu/suggestions/strategies/symspell_strategy.rb +275 -0
  203. data/lib/kotoshu/suggestions/suggestion.rb +174 -0
  204. data/lib/kotoshu/suggestions/suggestion_set.rb +238 -0
  205. data/lib/kotoshu/version.rb +5 -0
  206. data/lib/kotoshu.rb +493 -0
  207. data/script/validate_all_dictionaries.rb +444 -0
  208. data/sig/kotoshu.rbs +4 -0
  209. data/test_oop.rb +79 -0
  210. metadata +298 -0
@@ -0,0 +1,167 @@
1
+ # frozen_string_literal: true
2
+
3
+ module Kotoshu
4
+ module Algorithms
5
+ # Phonetic suggestion algorithm provides suggestions based on phonetical
6
+ # (pronunciation) similarity.
7
+ #
8
+ # Ported from Spylls (Python) phonet_suggest.py
9
+ #
10
+ # Requires .aff file to define PHONE table (extremely rare in known dictionaries).
11
+ #
12
+ # Internally:
13
+ # 1. Selects words from dictionary similarly to ngram_suggest
14
+ # (and reuses its root_score)
15
+ # 2. Scores their phonetic representations (calculated with metaphone)
16
+ # with phonetic representation of misspelling
17
+ # 3. Chooses the most similar ones with final_score (ngram-based comparison)
18
+ module PhonetSuggest
19
+ MAX_ROOTS = 100
20
+
21
+ class << self
22
+ # Main entry point for phonetic suggestions.
23
+ #
24
+ # Note that both this method and NgramSuggest.suggest iterate through
25
+ # the whole dictionary. Hunspell optimizes by doing it all in one
26
+ # loop. Spylls (and Kotoshu) splits them for clarity.
27
+ #
28
+ # @param misspelling [String] The misspelled word
29
+ # @param dictionary_words [Array<Hash>] Dictionary entries with stem and flags
30
+ # @param table [Hash] Phone table with :rules hash mapping first char to rule list
31
+ # @yield [String] Each suggestion
32
+ #
33
+ # The table structure should have:
34
+ # - :rules => Hash mapping first character to array of rule hashes
35
+ # Each rule has: :search (Regexp), :replacement (String),
36
+ # :start (Boolean), :end (Boolean)
37
+ def suggest(misspelling, dictionary_words:, table:, &block)
38
+ misspelling_lower = misspelling.downcase
39
+ misspelling_ph = metaphone(table, misspelling_lower)
40
+
41
+ scores = []
42
+
43
+ # First, select words from dictionary whose stems are similar to misspelling
44
+ # This cycle is exactly the same as the first cycle in ngram_suggest
45
+ dictionary_words.each do |word|
46
+ stem = word[:stem] || word
47
+
48
+ # Skip words with length difference > 3
49
+ next if (stem.length - misspelling.length).abs > 3
50
+
51
+ # First, calculate "regular" similarity score, just like in ngram_suggest
52
+ nscore = NgramSuggest.root_score(misspelling_lower, stem)
53
+
54
+ # Check alternative spellings if available
55
+ if word[:alt_spellings]
56
+ word[:alt_spellings].each do |variant|
57
+ nscore = [nscore, NgramSuggest.root_score(misspelling_lower, variant)].max
58
+ end
59
+ end
60
+
61
+ next if nscore <= 2
62
+
63
+ # Calculate metaphone score
64
+ word_ph = metaphone(table, stem.downcase)
65
+ score = 2 * StringMetrics.ngram(3, misspelling_ph, word_ph, longer_worse: true)
66
+
67
+ # Use heap-like behavior: keep only MAX_ROOTS best results
68
+ if scores.size >= MAX_ROOTS
69
+ # Remove the worst score if we're at capacity
70
+ scores.sort!.shift if scores.first && scores.first[0] < score
71
+ end
72
+
73
+ scores << [score, stem] if scores.size < MAX_ROOTS || scores.empty? || score > scores.first[0]
74
+ end
75
+
76
+ # Sort by score descending
77
+ guesses = scores.sort.reverse
78
+
79
+ # Finally, sort suggestions by simplistic string similarity metric
80
+ guesses2 = guesses.map do |score, word|
81
+ final_scr = final_score(misspelling_lower, word.downcase)
82
+ [score + final_scr, word]
83
+ end.sort.reverse
84
+
85
+ guesses2.each do |_, sug|
86
+ yield sug
87
+ end
88
+ end
89
+
90
+ # Calculate score of suggestion against misspelling.
91
+ #
92
+ # @param word1 [String] Misspelling
93
+ # @param word2 [String] Candidate suggestion
94
+ # @return [Float] Final score
95
+ def final_score(word1, word2)
96
+ 2 * StringMetrics.lcslen(word1, word2) -
97
+ (word1.length - word2.length).abs +
98
+ StringMetrics.leftcommonsubstring(word1, word2)
99
+ end
100
+
101
+ # Metaphone calculation.
102
+ #
103
+ # Production in Kotoshu is currently implemented naively as just
104
+ # "search and replace" for rules. To see what potentially should be done,
105
+ # look at aspell's original description:
106
+ # http://aspell.net/man-html/Phonetic-Code.html
107
+ #
108
+ # @param table [Hash] Phone table with :rules hash
109
+ # @param word [String] Word to calculate metaphone for
110
+ # @return [String] Metaphone representation
111
+ def metaphone(table, word)
112
+ return word if table.nil? || table.empty?
113
+
114
+ rules = table[:rules] || {}
115
+ pos = 0
116
+ word_upper = word.upcase
117
+ result = +''
118
+
119
+ while pos < word_upper.length
120
+ char = word_upper[pos]
121
+ matched = false
122
+
123
+ # Get rules for this character
124
+ char_rules = rules[char] || []
125
+ char_rules.each do |rule|
126
+ match_result = match_rule(rule, word_upper, pos)
127
+ next unless match_result
128
+
129
+ result += rule[:replacement]
130
+ pos += match_result
131
+ matched = true
132
+ break
133
+ end
134
+
135
+ pos += 1 unless matched
136
+ end
137
+
138
+ result
139
+ end
140
+
141
+ # Check if a rule matches at the given position.
142
+ #
143
+ # @param rule [Hash] Rule hash with :search (Regexp), :start, :end
144
+ # @param word [String] The word to match against
145
+ # @param pos [Integer] Position in word
146
+ # @return [Integer, nil] Length of match, or nil if no match
147
+ def match_rule(rule, word, pos)
148
+ # Check start constraint
149
+ return nil if rule[:start] && pos > 0
150
+
151
+ # Try to match
152
+ match_data = if rule[:end]
153
+ # Full match from position
154
+ rule[:search].match(word[pos..])
155
+ else
156
+ # Regular match from position
157
+ rule[:search].match(word, pos)
158
+ end
159
+
160
+ return nil unless match_data
161
+
162
+ match_data.to_s.length
163
+ end
164
+ end
165
+ end
166
+ end
167
+ end