@stll/anonymize 1.5.0 → 2.0.0

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 (155) hide show
  1. package/ATTRIBUTION.md +70 -0
  2. package/README.md +86 -48
  3. package/dist/index.d.mts +3 -1201
  4. package/dist/index.mjs +3 -16267
  5. package/dist/index.mjs.map +1 -1
  6. package/dist/native-node.d.mts +138 -0
  7. package/dist/native-node.mjs +3 -0
  8. package/dist/native-node2.d.mts +3 -0
  9. package/dist/native-node2.mjs +728 -0
  10. package/dist/native-node2.mjs.map +1 -0
  11. package/dist/native.d.mts +970 -0
  12. package/dist/native.mjs +230 -0
  13. package/dist/native.mjs.map +1 -0
  14. package/dist/native2.d.mts +2 -0
  15. package/index.cjs +3 -0
  16. package/native-pipeline.cs.stlanonpkg +0 -0
  17. package/native-pipeline.de.stlanonpkg +0 -0
  18. package/native-pipeline.en.stlanonpkg +0 -0
  19. package/native-pipeline.stlanonpkg +0 -0
  20. package/package.json +57 -9
  21. package/scripts/build-native-pipeline-package.mjs +225 -0
  22. package/dist/address-boundaries.mjs +0 -195
  23. package/dist/address-boundaries.mjs.map +0 -1
  24. package/dist/address-prepositions.mjs +0 -182
  25. package/dist/address-prepositions.mjs.map +0 -1
  26. package/dist/address-stop-keywords.mjs +0 -137
  27. package/dist/address-stop-keywords.mjs.map +0 -1
  28. package/dist/address-stopwords.mjs +0 -84
  29. package/dist/address-stopwords.mjs.map +0 -1
  30. package/dist/allow-list.mjs +0 -196
  31. package/dist/allow-list.mjs.map +0 -1
  32. package/dist/clause-noun-heads.mjs +0 -75
  33. package/dist/clause-noun-heads.mjs.map +0 -1
  34. package/dist/common-words-en.mjs +0 -9887
  35. package/dist/common-words-en.mjs.map +0 -1
  36. package/dist/coreference.cs.mjs +0 -14
  37. package/dist/coreference.cs.mjs.map +0 -1
  38. package/dist/coreference.de.mjs +0 -14
  39. package/dist/coreference.de.mjs.map +0 -1
  40. package/dist/coreference.en.mjs +0 -14
  41. package/dist/coreference.en.mjs.map +0 -1
  42. package/dist/coreference.es.mjs +0 -22
  43. package/dist/coreference.es.mjs.map +0 -1
  44. package/dist/coreference.fr.mjs +0 -32
  45. package/dist/coreference.fr.mjs.map +0 -1
  46. package/dist/coreference.it.mjs +0 -27
  47. package/dist/coreference.it.mjs.map +0 -1
  48. package/dist/coreference.pl.mjs +0 -27
  49. package/dist/coreference.pl.mjs.map +0 -1
  50. package/dist/coreference.pt-br.mjs +0 -14
  51. package/dist/coreference.pt-br.mjs.map +0 -1
  52. package/dist/coreference.sk.mjs +0 -27
  53. package/dist/coreference.sk.mjs.map +0 -1
  54. package/dist/currencies.mjs +0 -231
  55. package/dist/currencies.mjs.map +0 -1
  56. package/dist/date-months.mjs +0 -618
  57. package/dist/date-months.mjs.map +0 -1
  58. package/dist/document-structure-headings.mjs +0 -90
  59. package/dist/document-structure-headings.mjs.map +0 -1
  60. package/dist/generic-roles.mjs +0 -244
  61. package/dist/generic-roles.mjs.map +0 -1
  62. package/dist/hotword-rules.mjs +0 -149
  63. package/dist/hotword-rules.mjs.map +0 -1
  64. package/dist/legal-form-leading-clauses.mjs +0 -23
  65. package/dist/legal-form-leading-clauses.mjs.map +0 -1
  66. package/dist/legal-forms.mjs +0 -2115
  67. package/dist/legal-forms.mjs.map +0 -1
  68. package/dist/legal-role-heads.cs.mjs +0 -42
  69. package/dist/legal-role-heads.cs.mjs.map +0 -1
  70. package/dist/legal-role-heads.de.mjs +0 -33
  71. package/dist/legal-role-heads.de.mjs.map +0 -1
  72. package/dist/legal-role-heads.en.mjs +0 -37
  73. package/dist/legal-role-heads.en.mjs.map +0 -1
  74. package/dist/legal-role-heads.es.mjs +0 -54
  75. package/dist/legal-role-heads.es.mjs.map +0 -1
  76. package/dist/legal-role-heads.fr.mjs +0 -72
  77. package/dist/legal-role-heads.fr.mjs.map +0 -1
  78. package/dist/legal-role-heads.it.mjs +0 -68
  79. package/dist/legal-role-heads.it.mjs.map +0 -1
  80. package/dist/legal-role-heads.pl.mjs +0 -84
  81. package/dist/legal-role-heads.pl.mjs.map +0 -1
  82. package/dist/legal-role-heads.pt-br.mjs +0 -63
  83. package/dist/legal-role-heads.pt-br.mjs.map +0 -1
  84. package/dist/legal-role-heads.sk.mjs +0 -80
  85. package/dist/legal-role-heads.sk.mjs.map +0 -1
  86. package/dist/manifest.mjs +0 -69
  87. package/dist/manifest.mjs.map +0 -1
  88. package/dist/names-exclusions.mjs +0 -223
  89. package/dist/names-exclusions.mjs.map +0 -1
  90. package/dist/names-first.mjs +0 -418
  91. package/dist/names-first.mjs.map +0 -1
  92. package/dist/names-nw-ar.mjs +0 -202
  93. package/dist/names-nw-ar.mjs.map +0 -1
  94. package/dist/names-nw-excluded-allcaps.mjs +0 -112
  95. package/dist/names-nw-excluded-allcaps.mjs.map +0 -1
  96. package/dist/names-nw-fil.mjs +0 -202
  97. package/dist/names-nw-fil.mjs.map +0 -1
  98. package/dist/names-nw-id.mjs +0 -210
  99. package/dist/names-nw-id.mjs.map +0 -1
  100. package/dist/names-nw-in.mjs +0 -526
  101. package/dist/names-nw-in.mjs.map +0 -1
  102. package/dist/names-nw-ja-latn.mjs +0 -260
  103. package/dist/names-nw-ja-latn.mjs.map +0 -1
  104. package/dist/names-nw-ko.mjs +0 -162
  105. package/dist/names-nw-ko.mjs.map +0 -1
  106. package/dist/names-nw-th.mjs +0 -188
  107. package/dist/names-nw-th.mjs.map +0 -1
  108. package/dist/names-nw-vi.mjs +0 -151
  109. package/dist/names-nw-vi.mjs.map +0 -1
  110. package/dist/names-nw-zh-latn.mjs +0 -197
  111. package/dist/names-nw-zh-latn.mjs.map +0 -1
  112. package/dist/names-surnames.mjs +0 -113
  113. package/dist/names-surnames.mjs.map +0 -1
  114. package/dist/names-title-tokens.mjs +0 -40
  115. package/dist/names-title-tokens.mjs.map +0 -1
  116. package/dist/person-stopwords.mjs +0 -205
  117. package/dist/person-stopwords.mjs.map +0 -1
  118. package/dist/section-headings.mjs +0 -64
  119. package/dist/section-headings.mjs.map +0 -1
  120. package/dist/sentence-verb-indicators.mjs +0 -232
  121. package/dist/sentence-verb-indicators.mjs.map +0 -1
  122. package/dist/signing-clauses.mjs +0 -78
  123. package/dist/signing-clauses.mjs.map +0 -1
  124. package/dist/stopwords.mjs +0 -9915
  125. package/dist/stopwords.mjs.map +0 -1
  126. package/dist/structural-single-cap-prefixes.mjs +0 -99
  127. package/dist/structural-single-cap-prefixes.mjs.map +0 -1
  128. package/dist/triggers.cs.mjs +0 -569
  129. package/dist/triggers.cs.mjs.map +0 -1
  130. package/dist/triggers.de.mjs +0 -139
  131. package/dist/triggers.de.mjs.map +0 -1
  132. package/dist/triggers.en.mjs +0 -119
  133. package/dist/triggers.en.mjs.map +0 -1
  134. package/dist/triggers.es.mjs +0 -96
  135. package/dist/triggers.es.mjs.map +0 -1
  136. package/dist/triggers.fr.mjs +0 -275
  137. package/dist/triggers.fr.mjs.map +0 -1
  138. package/dist/triggers.global.mjs +0 -79
  139. package/dist/triggers.global.mjs.map +0 -1
  140. package/dist/triggers.hu.mjs +0 -41
  141. package/dist/triggers.hu.mjs.map +0 -1
  142. package/dist/triggers.it.mjs +0 -74
  143. package/dist/triggers.it.mjs.map +0 -1
  144. package/dist/triggers.pl.mjs +0 -271
  145. package/dist/triggers.pl.mjs.map +0 -1
  146. package/dist/triggers.pt-br.mjs +0 -193
  147. package/dist/triggers.pt-br.mjs.map +0 -1
  148. package/dist/triggers.ro.mjs +0 -59
  149. package/dist/triggers.ro.mjs.map +0 -1
  150. package/dist/triggers.sk.mjs +0 -555
  151. package/dist/triggers.sk.mjs.map +0 -1
  152. package/dist/triggers.sv.mjs +0 -58
  153. package/dist/triggers.sv.mjs.map +0 -1
  154. package/dist/year-words.mjs +0 -62
  155. package/dist/year-words.mjs.map +0 -1
@@ -1,64 +0,0 @@
1
- //#region src/data/section-headings.json
2
- var _comment = "Section heading patterns that mark the end of the header zone. Multilingual: Czech, English, German, French, Italian, Spanish, Slovak, Polish. All patterns require a trailing numeral or roman numeral to avoid false-positives on body text. Each entry specifies its own regex flags; keyword-based patterns use 'iu' for case-insensitive Unicode matching, while the roman-numeral pattern stays case-sensitive to avoid matching lowercase list markers (i., ii.).";
3
- var patterns = [
4
- {
5
- "re": "^\\s*(?:I\\.|1\\.)\\s",
6
- "flags": ""
7
- },
8
- {
9
- "re": "^\\s*(?:Článek|Čl\\.)\\s*\\d",
10
- "flags": "iu"
11
- },
12
- {
13
- "re": "^\\s*(?:Article|Art\\.)\\s*\\d",
14
- "flags": "iu"
15
- },
16
- {
17
- "re": "^\\s*§\\s*\\d",
18
- "flags": ""
19
- },
20
- {
21
- "re": "^\\s*(?:Abschnitt|Kapitel)\\s*\\d",
22
- "flags": "iu"
23
- },
24
- {
25
- "re": "^\\s*(?:Chapitre|Section)\\s*\\d",
26
- "flags": "iu"
27
- },
28
- {
29
- "re": "^\\s*(?:Annexe|Préambule|Titre)\\s*\\d",
30
- "flags": "iu"
31
- },
32
- {
33
- "re": "^\\s*(?:Capitolo|Sezione|Articolo)\\s*\\d",
34
- "flags": "iu"
35
- },
36
- {
37
- "re": "^\\s*Sezione\\s+[IVX]+\\b",
38
- "flags": "iu"
39
- },
40
- {
41
- "re": "^\\s*(?:AMMINISTRAZIONE\\s+AGGIUDICATRICE|ALTRE\\s+INFORMAZIONI|OGGETTO|PROCEDURA)\\b",
42
- "flags": "iu"
43
- },
44
- {
45
- "re": "^\\s*(?:Artículo|Sección)\\s*\\d",
46
- "flags": "iu"
47
- },
48
- {
49
- "re": "^\\s*(?:Článok|Čl\\.|Oddiel|Odsek|Príloha)\\s*\\d",
50
- "flags": "iu"
51
- },
52
- {
53
- "re": "^\\s*(?:Rozdział|Artykuł|Paragraf|Załącznik)\\s*\\d",
54
- "flags": "iu"
55
- }
56
- ];
57
- var section_headings_default = {
58
- _comment,
59
- patterns
60
- };
61
- //#endregion
62
- export { _comment, section_headings_default as default, patterns };
63
-
64
- //# sourceMappingURL=section-headings.mjs.map
@@ -1 +0,0 @@
1
- {"version":3,"file":"section-headings.mjs","names":[],"sources":["../src/data/section-headings.json"],"sourcesContent":[""],"mappings":""}
@@ -1,232 +0,0 @@
1
- //#region src/data/sentence-verb-indicators.json
2
- var _comment = "Verb-like tokens that signal sentence context inside a legal-form match span. When one of these appears between a role-head opening ('Vendor', 'Predávajúci') and the legal-form suffix ('Inc.', 's.r.o.'), the match is a swept sentence fragment and gets trimmed back to the real organisation name. Lowercased; matched case-insensitively. Organised per-language because verb morphology is language-specific; at runtime the legal-form detector unions every language's list into a single set because match spans don't carry a language tag.";
3
- var cs = [
4
- "je",
5
- "jsou",
6
- "byl",
7
- "byla",
8
- "byly",
9
- "bude",
10
- "není",
11
- "vlastní",
12
- "vlastníkem",
13
- "prodává",
14
- "prodal",
15
- "prodala",
16
- "koupil",
17
- "koupila",
18
- "kupuje",
19
- "platí",
20
- "hradí",
21
- "podepsal",
22
- "podepsala",
23
- "podepisuje",
24
- "uzavírá",
25
- "uzavřel",
26
- "uzavřela",
27
- "zavazuje",
28
- "dluží"
29
- ];
30
- var en = [
31
- "is",
32
- "are",
33
- "was",
34
- "were",
35
- "owns",
36
- "grants",
37
- "sells",
38
- "sold",
39
- "buys",
40
- "bought",
41
- "has",
42
- "holds",
43
- "agrees",
44
- "pays",
45
- "paid",
46
- "owes",
47
- "signs",
48
- "signed",
49
- "leases",
50
- "rents",
51
- "entered",
52
- "made",
53
- "executed",
54
- "retained",
55
- "hired",
56
- "engaged",
57
- "appointed",
58
- "named",
59
- "sued"
60
- ];
61
- var de = [
62
- "ist",
63
- "sind",
64
- "war",
65
- "waren",
66
- "besitzt",
67
- "gewährt",
68
- "verkauft",
69
- "kauft",
70
- "hat",
71
- "unterzeichnet",
72
- "zahlt"
73
- ];
74
- var fr = [
75
- "est",
76
- "sont",
77
- "était",
78
- "étaient",
79
- "sera",
80
- "seront",
81
- "possède",
82
- "possèdent",
83
- "détient",
84
- "détiennent",
85
- "vend",
86
- "vendent",
87
- "achète",
88
- "achètent",
89
- "signe",
90
- "signent",
91
- "paie",
92
- "paient",
93
- "accorde",
94
- "accordent",
95
- "loue",
96
- "louent",
97
- "s'engage",
98
- "s’engage",
99
- "s'engagent",
100
- "s’engagent",
101
- "conclut",
102
- "concluent",
103
- "doit",
104
- "doivent"
105
- ];
106
- var sk = [
107
- "je",
108
- "sú",
109
- "bol",
110
- "bola",
111
- "boli",
112
- "bude",
113
- "nie je",
114
- "uzatvára",
115
- "uzatvoril",
116
- "uzatvorila",
117
- "uzaviera",
118
- "uzavrel",
119
- "uzavrela",
120
- "podpisuje",
121
- "podpísal",
122
- "podpísala",
123
- "vlastní",
124
- "vlastníkom",
125
- "vlastníkmi",
126
- "predáva",
127
- "predal",
128
- "predala",
129
- "kupuje",
130
- "kúpil",
131
- "kúpila",
132
- "platí",
133
- "hradí",
134
- "zaväzuje",
135
- "dlží",
136
- "prenajíma",
137
- "najíma"
138
- ];
139
- var es = [
140
- "actúa",
141
- "actúan",
142
- "adquiere",
143
- "adquirió",
144
- "celebra",
145
- "celebran",
146
- "celebró",
147
- "compra",
148
- "compró",
149
- "concede",
150
- "contrata",
151
- "contrató",
152
- "debe",
153
- "es",
154
- "fue",
155
- "fueron",
156
- "firma",
157
- "firman",
158
- "firmó",
159
- "ha",
160
- "han",
161
- "otorga",
162
- "otorgan",
163
- "otorgó",
164
- "paga",
165
- "pagó",
166
- "posee",
167
- "compromete",
168
- "comprometen",
169
- "comprometió",
170
- "ser",
171
- "son",
172
- "suscribe",
173
- "tiene",
174
- "tienen",
175
- "transfiere",
176
- "vende",
177
- "vendió"
178
- ];
179
- var it = [
180
- "è",
181
- "sono",
182
- "era",
183
- "erano",
184
- "possiede",
185
- "concede",
186
- "vende",
187
- "compra",
188
- "ha",
189
- "firma",
190
- "firmato",
191
- "paga",
192
- "deve",
193
- "ratifica",
194
- "stipula"
195
- ];
196
- var pl = [
197
- "jest",
198
- "są",
199
- "był",
200
- "była",
201
- "było",
202
- "były",
203
- "będzie",
204
- "będą",
205
- "nie jest",
206
- "oświadcza",
207
- "oświadczają",
208
- "sprzedaje",
209
- "sprzedają",
210
- "kupuje",
211
- "kupują",
212
- "zawiera",
213
- "zawierają",
214
- "podpisuje",
215
- "płaci",
216
- "wynajmuje"
217
- ];
218
- var sentence_verb_indicators_default = {
219
- _comment,
220
- cs,
221
- en,
222
- de,
223
- fr,
224
- sk,
225
- es,
226
- it,
227
- pl
228
- };
229
- //#endregion
230
- export { _comment, cs, de, sentence_verb_indicators_default as default, en, es, fr, it, pl, sk };
231
-
232
- //# sourceMappingURL=sentence-verb-indicators.mjs.map
@@ -1 +0,0 @@
1
- {"version":3,"file":"sentence-verb-indicators.mjs","names":[],"sources":["../src/data/sentence-verb-indicators.json"],"sourcesContent":[""],"mappings":""}
@@ -1,78 +0,0 @@
1
- //#region src/data/signing-clauses.json
2
- var _comment = "Signing clause patterns. Captures the place name from contract signing locations. Each entry: prefix (before place), suffix (after place), prepositions (allowed inside multi-word place names).";
3
- var patterns = [
4
- {
5
- "lang": "cs",
6
- "prefix": "(?:V|Ve)\\s+",
7
- "suffix": "\\s*,?\\s*dne",
8
- "prepositions": [
9
- "nad",
10
- "pod",
11
- "u",
12
- "ve",
13
- "na"
14
- ]
15
- },
16
- {
17
- "lang": "sk",
18
- "prefix": "(?:V|Vo)\\s+",
19
- "suffix": "\\s*,?\\s*dňa",
20
- "prepositions": [
21
- "nad",
22
- "pod",
23
- "pri"
24
- ]
25
- },
26
- {
27
- "lang": "de",
28
- "prefix": "",
29
- "suffix": "\\s*,\\s*den",
30
- "prepositions": [
31
- "am",
32
- "an",
33
- "im"
34
- ]
35
- },
36
- {
37
- "lang": "fr",
38
- "prefix": "(?:Fait\\s+)?[Àà]\\s+",
39
- "suffix": "\\s*,?\\s*le",
40
- "prepositions": []
41
- },
42
- {
43
- "lang": "en",
44
- "prefix": "(?:Signed|Executed)\\s+in\\s+",
45
- "suffix": "",
46
- "prepositions": []
47
- },
48
- {
49
- "lang": "pl",
50
- "prefix": "(?:W|We)\\s+",
51
- "suffix": "\\s*,?\\s*dnia",
52
- "prepositions": [
53
- "nad",
54
- "pod",
55
- "przy"
56
- ]
57
- },
58
- {
59
- "lang": "it",
60
- "prefix": "(?:Fatto\\s+)?[Aa]\\s+",
61
- "suffix": "\\s*,?\\s*(?:il|lì)",
62
- "prepositions": []
63
- },
64
- {
65
- "lang": "es",
66
- "prefix": "(?:Firmado\\s+)?[Ee]n\\s+",
67
- "suffix": "\\s*,?\\s*(?:a|el)",
68
- "prepositions": ["de", "del"]
69
- }
70
- ];
71
- var signing_clauses_default = {
72
- _comment,
73
- patterns
74
- };
75
- //#endregion
76
- export { _comment, signing_clauses_default as default, patterns };
77
-
78
- //# sourceMappingURL=signing-clauses.mjs.map
@@ -1 +0,0 @@
1
- {"version":3,"file":"signing-clauses.mjs","names":[],"sources":["../src/data/signing-clauses.json"],"sourcesContent":[""],"mappings":""}