| """Extract legal hierarchy from country-law text (Oregon-style, multilingual). |
| |
| Oregon Revised Statutes are stored as title / chapter / section / subsection |
| trees. National gazettes are not that uniform: some use Article/Artikel, |
| some §, some only a single instrument body. |
| |
| This module: |
| |
| * strips leftover HTML chrome so GraphRAG sees legal text, not tags; |
| * detects multilingual heading markers (title/chapter/part/article/section); |
| * splits an instrument into those units when at least two headings exist; |
| * otherwise keeps the whole instrument (never invents a hierarchy). |
| |
| Subsection markers such as ``(a)`` / ``(1)`` are recorded on the parent |
| unit; they are not forced into their own retrieval rows. |
| """ |
|
|
| from __future__ import annotations |
|
|
| from dataclasses import dataclass, field |
| from html.parser import HTMLParser |
| import html as html_lib |
| import json |
| import os |
| from pathlib import Path |
| import re |
| import unicodedata |
| from typing import Any |
|
|
| _WS_RE = re.compile(r"\s+", re.UNICODE) |
| _HAS_TAG_RE = re.compile(r"</?[a-zA-Z][^>]*>") |
|
|
| |
| |
| _LATIN_FOLD = str.maketrans({ |
| **dict.fromkeys("áàâäãåăąāấầẩẫậắằẳẵặảạ", "a"), |
| **dict.fromkeys("ÁÀÂÄÃÅĂĄĀẢẠ", "A"), |
| **dict.fromkeys("éèêëěęēếềểễệẻẹə", "e"), |
| **dict.fromkeys("ÉÈÊËĚĘĒẺẸƏ", "E"), |
| **dict.fromkeys("íìîïĩįīıịỉ", "i"), |
| **dict.fromkeys("ÍÌÎÏĨĮĪİỈỊ", "I"), |
| **dict.fromkeys("óòôöõőøōốồổỗộớờởỡợỏọ", "o"), |
| **dict.fromkeys("ÓÒÔÖÕŐØŌỎỌ", "O"), |
| **dict.fromkeys("úùûüűũūůứừửữựủụ", "u"), |
| **dict.fromkeys("ÚÙÛÜŰŨŪŮỦỤ", "U"), |
| **dict.fromkeys("ýÿỳỹỷỵ", "y"), |
| **dict.fromkeys("ÝŸỶỴ", "Y"), |
| **dict.fromkeys("çćčĉċ", "c"), |
| **dict.fromkeys("ÇĆČĈĊ", "C"), |
| **dict.fromkeys("ďđ", "d"), |
| **dict.fromkeys("ĎĐ", "D"), |
| **dict.fromkeys("ğĝģġ", "g"), |
| **dict.fromkeys("ĞĜĢĠ", "G"), |
| **dict.fromkeys("ñńņň", "n"), |
| **dict.fromkeys("ÑŃŅŇ", "N"), |
| **dict.fromkeys("ŕŗř", "r"), |
| **dict.fromkeys("ŔŖŘ", "R"), |
| **dict.fromkeys("śŝşšș", "s"), |
| **dict.fromkeys("ŚŜŞŠȘ", "S"), |
| **dict.fromkeys("ţťț", "t"), |
| **dict.fromkeys("ŢŤȚ", "T"), |
| **dict.fromkeys("źżž", "z"), |
| **dict.fromkeys("ŹŻŽ", "Z"), |
| **dict.fromkeys("ĺļľł", "l"), |
| **dict.fromkeys("ĹĻĽŁ", "L"), |
| "æ": "a", "Æ": "A", "œ": "o", "Œ": "O", "ß": "s", |
| "ɛ": "e", "Ɛ": "E", "ɔ": "o", "Ɔ": "O", "ŋ": "n", "Ŋ": "N", |
| **dict.fromkeys("ơớờởỡợ", "o"), |
| **dict.fromkeys("ƠỚỜỞỠỢ", "O"), |
| **dict.fromkeys("ưứừửữự", "u"), |
| **dict.fromkeys("ƯỨỪỬỮỰ", "U"), |
| **dict(zip("άέήίόύώϊϋΐΰ", "αεηιουωιυιυ")), |
| **dict(zip("ΆΈΉΊΌΎΏ", "ΑΕΗΙΟΥΩ")), |
| "ς": "σ", |
| "ё": "е", "Ё": "Е", "ї": "и", "Ї": "И", "є": "е", "Є": "Е", "ґ": "г", "Ґ": "Г", |
| }) |
|
|
|
|
| def _latin_fold(text: str) -> str: |
| return text.translate(_LATIN_FOLD) |
|
|
| |
| |
| |
| _HEADING_NUM = ( |
| r"(?:[0-9٠-٩۰-۹]+(?:er|ère|bis|ter|quater|o)?" |
| r"|(?-i:[IVXLCDM]{1,8})(?:er|ère)?(?![A-Za-z])" |
| r"|premier(?:e)?|première|primero|primeira" |
| r"|unique|unico|unica|único|única)" |
| ) |
| _ARTICLE_N = ( |
| r"(?:" |
| r"(?:[LRD]\.?\s*)?" |
| r"(?:" |
| r"[0-9٠-٩۰-۹]+(?:\.[0-9]+)?" |
| r"(?:er|ère|bis|ter|quater|aad|[A-Za-z]|o)?" |
| r"(?:/[0-9A-Za-z]+)?" |
| r"(?:\s+(?:bis|ter|quater))?" |
| r"|(?-i:[IVXLCDM]{1,8})(?:er|ère)?(?![A-Za-z])" |
| r"|[Α-Ω]{1,6}(?![Α-Ωα-ω])" |
| r")" |
| r"(?:[-–](?:[0-9]+|aad|bis|ter|quater|[A-Za-z](?![A-Za-z])))?" |
| r"(?:\([0-9A-Za-z]+\))?" |
| r"|premier(?:e)?" |
| r"|première" |
| r"|unique|unico|unica|único|única" |
| r"|primero|primera|primeiro|primeira" |
| r"|primo(?![A-Za-z])|secondo(?![A-Za-z])|terzo(?![A-Za-z])|quarto(?![A-Za-z])|quinto(?![A-Za-z])|prima(?![A-Za-z])" |
| r"|eerste|första|første" |
| r"|annexe" |
| r"|primera|segunda|tercera|cuarta|quinta" |
| r"|sexta|séptima|octava|novena|décima" |
| r")" |
| ) |
| _HEADING_START = r"(?:^|(?<=[.;:])|(?<=\d[ \t]))[ \t\[\(\u00ab\u00bb\"“”‘’«»‹›\u2039\u203a]*" |
| _HEADING_RE = re.compile( |
| _latin_fold( |
| r"(?im)" + _HEADING_START + r"(?:" |
| r"(?P<title>TITLE|TITRE|TITEL|T[IÍ]TULO|TITOLO|TITLUL|T[IÍ]TUL|LUKU|PEAT[UÜ]KK|" |
| r"РАЗДЕЛ|РОЗДІЛ|DZIA[ŁL]|ΤΙΤΛΟΣ|ДЯЛ)\s+" |
| r"(?P<title_n>" + _HEADING_NUM + r"(?:\.[0-9]+|-(?:[0-9]+|[A-Za-z](?![A-Za-z])))?)" |
| r"|(?P<chapter>CHAPTER|CHAPITRE|KAPITEL|CAP[IÍ]TULO|CAPITOLO|HOOFDSTUK|" |
| r"ROZDZIA[ŁL]|CAPITOLUL|NODA[ĻL]A|GLAVA|ГЛАВА|POGLAVLJE|KAPITOLA|" |
| r"ΚΕΦΑΛΑΙ[ΟO]|CH[UƯ][OƠ]NG|BAB|KABANATA|CAPO|HLAVA|KREU|FEJEZET|" |
| r"B[ÖO]L[ÜU]M|FESIL|CAIBIDIL|PENNOD|KAPITOLU|CHAPIT|ODDIEL|" |
| r"SURA(?:\s+YA)?|KAPITTEL|KAPITTUL|ANDIANY|SASHE(?:\s+NA)?|WASE|" |
| r"ISIGABA|ICANDELO|KGAOLO|BOBI|KUTAA|ORI|CAP\.)\s+" |
| r"(?P<chapter_n>" + _HEADING_NUM + r"(?:\.[0-9]+|-(?:[0-9]+|[A-Za-z](?![A-Za-z])))?)" |
| r"|(?P<part>PART|PARTIE|TEIL|PARTE|DEEL|LIVRE|BOOK|BUCH|TOMO|KNIHA|" |
| r"KODEKS|CODE|ZAKONIK|BAHAGIAN|BAGIAN|SEHEMU|MỤC|MUC|LIBRO|PH[AẦ]N|" |
| r"QAYBTA|WAHANGA|VAEGA|HAP|KAROLO|CHIKAMU|BES|EKITUNDU|ETENI|" |
| r"AKPA|FAANDA|PATTE|ΜΕΡΟΣ)\s+" |
| r"(?P<part_n>" + _HEADING_NUM + r"(?:\.[0-9]+|-(?:[0-9]+|[A-Za-z](?![A-Za-z])))?)" |
| r"|(?P<article>(?:l['’])?(?:(?:EK|GEÇICI|GECICI|GEÇİCİ)\s+)?" |
| r"(?:ART(?:ICLE|IKEL|ÍCULO|ICULO|IGO|ICOLO|IKKEL|YKU[ŁL]|ICOLUL|ICOL|IKOLU|IKULA)?\.?" |
| r"|ART\.|NENI|PASAL|MADDE|MADDƏ|ĐIỀU|ĐIÊU|DIEU|CLANAK|ČLANAK|ČLÁNEK|ČLÁNOK|ČLEN|ČLAN|CLAN" |
| r"|ЧЛАНАК|ЧЛАН|ЧЛЕН|ЧЛ\.|СТАТЬЯ|СТАТТЯ|SZAKASZ|CIKK|ΆΡΘΡΟ|ΑΡΘΡΟ" |
| r"|PYKÄLÄ|GREIN|PANTS|SEKSYEN|CIKK|KIFUNGU|MODDA|ATIK|KUPU|ATIKOL|" |
| r"INGINGO(?:\s+YA)?|NDIME|ABALA|NKEJI|SARIYA|TERE|AHYEDE|PENNAD|ERTHYGEL|HYEDEE|" |
| r"ARTIGU|PAUKU|IRAVA" |
| r"|მუხლი|ՀՈԴՎԱԾ|Հոդված))(?![A-Za-z])" |
| r"(?:\s*(?:n\.?[o°º”'“*´`]|nr|no)\.?\s*)?\s*[.\s:–—-]*\(?\s*" |
| r"(?P<article_n>" + _ARTICLE_N + r")\)?" |
| r"|(?P<section>SUB(?:-)?SECTION|SECTION|SECCI[OÓ]N|SEZIONE|ABSCHNITT|PARAGRAHV|PARAGRAPHE|PARAGRAAF|PARAGRAPH|" |
| r"RECITAL|CONSIDERANDO|" |
| r"ALIN[EÉ]A|ALINEATUL|INCISO|P[AÁ]RRAFO|PUNKT(?![A-Za-z])|SATZ(?![A-Za-z])|LID|" |
| r"ΠΑΡΑΓΡΑΦΟΣ|ABSATZ|APARTADO|COMMA(?![A-Za-z])|FIKRA|STAVAK|STK\.?|АЛ\.|SEKSHON|" |
| r"NUMERAL|CAPOVERSO|NUMMER|" |
| r"AYAT|KHOẢN|KHOAN|" |
| r"RULE(?![A-Za-z])|REGULATION(?![A-Za-z])|SCHEDULE(?![A-Za-z])|FORM(?![A-Za-z])|" |
| r"ORDINANCE|ORDONNANCE|ORDONAN[TŢȚ][AĂÁ]?|ORDIN(?![A-Za-z])|BY-?LAW|DIRECTIVE(?![A-Za-z])|" |
| r"D[EÉ]CISION(?![A-Za-z])|DECISION(?![A-Za-z])|DECIZIE|" |
| r"BESLUIT(?![A-Za-z])|REGELING(?![A-Za-z])|BESCHIKKING(?![A-Za-z])|" |
| r"ODLUKA|PRAVILNIK|HOT[ĂA]R[ÂAÎI]RE|" |
| r"RESOLUCI[OÓ]N(?![A-Za-z])|RESOLU[CÇ][AÃ]O(?![A-Za-z])|PORTARIA(?![A-Za-z])|ACUERDO(?![A-Za-z])|" |
| r"MEDIDA\s+PROVIS[OÓ]RIA|" |
| r"PROCLAMATION(?![A-Za-z])|PROKLAMATION|CIRCULAR(?![A-Za-z])|INSTRUCTION(?![A-Za-z])|" |
| r"GUIDELINE(?![A-Za-z])|MEMORANDUM(?![A-Za-z])|" |
| r"AVISO(?![A-Za-z])|AVIS(?![A-Za-z])|BEKANNTMACHUNG|MEDDELANDE|" |
| r"NORMA(?![A-Za-z])|" |
| r"PRESIDENTIAL\s+DECREE|DECRETO-LEI|DECRETO(?![A-Za-z])|DECREE(?![A-Za-z])|" |
| r"D[EÉ]CRET-LOI|D[EÉ]CRET(?![A-Za-z])|ARR[EÊ]T[EÉ]|" |
| r"DELIBERA[CÇ][AÃ]O|D[EÉ]LIB[EÉ]RATION|COMUNICADO|" |
| r"R[EÈ]GLEMENT(?![A-Za-z])|CIRCULAIRE(?![A-Za-z])|VERORDNUNG(?![A-Za-z])|ERLASS(?![A-Za-z])|" |
| r"PUNTO(?![A-Za-z])|SUB(?:-)?PARAGRAPH|" |
| r"ANNEX(?:E)?(?![A-Za-z])|ANEXO|ANLAGE|BIJLAGE|APPENDIX|ALLEGATO|" |
| r"ZA[ŁL][AĄ]CZNIK|BILAGA|LIITE|P[ŘR][IÍ]LOHA|" |
| r"UST\.|ODST\.|ODS\.|BEK\.|" |
| r"SEC(?:\.|(?=\s+[0-9])))(?![A-Za-z])(?:\s*(?:n\.?[o°º”'“*´`]|nr|no)\.?\s*)?[.\s:]*\(?\s*" |
| r"(?P<section_n>(?:[0-9][0-9A-Za-z.\-]*(?:\([0-9A-Za-z]+\))?|" |
| r"(?-i:[IVXLCDM]{1,8})(?![A-Za-z])|[A-Za-z](?![A-Za-z])))\)?" |
| r"|(?P<section_sym>§+)\s*(?P<section_sym_n>[0-9]+(?:[A-Za-z](?![A-Za-z]))?(?:\s+[A-Za-z](?![A-Za-z]))?(?:\s*\([0-9A-Za-z]+\))?)" |
| r"|(?P<section_pre_n>[0-9]+(?:\s*[a-z])?)\s*\.?\s*(?P<section_pre>§)" |
| r"|(?P<grein_n>[0-9]+[a-z]?)\s*\.\s*(?P<grein>gr|pants)\.?" |
| r"|(?P<straipsnis_n>[0-9]+[a-z]?)\s*[.]?\s*(?P<straipsnis>straipsnis)" |
| r"|(?P<artikulua_n>[0-9]+[a-z]?)\s*[.]?\s*(?P<artikulua>artikulua)" |
| r"|(?P<ledd_n>[0-9]+)\s*[.]?\s*(?P<ledd>ledd)" |
| r"|(?P<kap_n>[0-9]+)\s*(?P<kap>kap)\.?" |
| r"|(?P<luku_n>[0-9]+)\s*(?P<luku>luku)" |
| r"|(?P<fejezet_n>[IVXLCDM0-9]+)\s*\.\s*(?P<fejezet>fejezet)" |
| r"|(?P<peatukk_n>[0-9]+)\s*[.]?\s*(?P<peatukk>peat[uü]kk)" |
| r"|(?P<skyrius_n>[IVXLCDM0-9]+)\s*(?P<skyrius>skyrius)" |
| r"|(?P<nodala_n>[IVXLCDM0-9]+)\s*(?P<nodala>noda[ļl]a)" |
| r"|(?P<kafli_n>[IVXLCDM0-9]+)\s*[.]?\s*(?P<kafli>kafli)" |
| r"|(?P<bob_n>[0-9]+)\s*-\s*(?P<bob>bob)" |
| r"|(?P<tarau_n>[0-9]+)\s*-\s*(?P<tarau>тарау)" |
| r"|(?P<bolum_n>[0-9]+)\s*-\s*(?P<bolum>бөлүм)" |
| r"|(?P<modda_n>[0-9]+)\s*-\s*(?P<modda>модда|modda)" |
| r"|(?P<mn_n>[0-9]+)\s*(?P<mn>(?:дүгээр|дугаар)\s*зүйл)" |
| r"|(?P<kk_n>[0-9]+)\s*-?\s*(?P<kk>бап)" |
| r"|(?P<mom_n>[0-9]+)\s*(?P<mom>momentti)" |
| r"|(?P<order>(?:EXECUTIVE\s+)?ORDER)\s+(?P<order_n>[0-9]+)" |
| r"|(?P<kidogo>KIFUNGU\s+KIDOG[OU])\s*\(?\s*(?P<kidogo_n>[0-9]+)" |
| r"|(?P<disp>DISPOSICI[OÓ]N)(?:\s+(?:adicional(?:es)?|transitoria(?:s)?|derogatoria(?:s)?|final(?:es)?))?\s+" |
| r"(?P<disp_n>" + _ARTICLE_N + r")" |
| r"|(?P<esresol>primero|segundo|tercero|cuarto|quinto|sexto|septimo|octavo|noveno|decimo|" |
| r"erstens|zweitens|drittens|viertens|funftens|" |
| r"premierement|deuxiemement|troisiemement|quatriemement|cinquiemement)\s*:" |
| r")" |
| ) |
| ) |
| |
| _CW_NUM_RE = re.compile( |
| r"(?m)^[ \t]*(?P<cw_n>[0-9]+[A-Za-z]?)\.?\s+(?:\([0-9A-Za-z]+\)\s+)?(?=[A-Z][a-z]+)" |
| ) |
| _MONTH_START_RE = re.compile( |
| r"(?i)^(january|february|march|april|may|june|july|august|september|october|november|december|" |
| r"janvier|f[eé]vrier|marzo|abril|enero|junio|julio)\b" |
| ) |
| _CW_NUM_LANGS = frozenset({"", "en", "hi", "bn", "ne", "ur", "ms", "fil", "sw", "si"}) |
| _LEXICON_HEADING_RE: re.Pattern[str] | None = None |
| _LEXICON_KIND: dict[str, str] = {} |
|
|
| _SUBSECTION_RE = re.compile(r"\(([0-9A-Za-z]{1,6})\)") |
|
|
| _KIND_RANK = { |
| "title": 1, |
| "chapter": 2, |
| "part": 3, |
| "article": 4, |
| "section": 5, |
| "subsection": 6, |
| } |
|
|
| MIN_SPLIT_HEADINGS = 2 |
| MIN_UNIT_CHARS = 40 |
|
|
| |
| _ZH_ART_RE = re.compile( |
| r"第\s*[一二三四五六七八九十百千万零〇两0-9]+\s*[条條](?:之[一二三四五六七八九十百0-9]+)?" |
| ) |
| _AR_ORD = ( |
| r"(?:الحادية|الثانية|الثالثة|الرابعة|الخامسة|السادسة|السابعة|الثامنة|التاسعة)[\s\u0640]*والعشرون|" |
| r"(?:الحادية|الثانية|الثالثة|الرابعة|الخامسة|السادسة|السابعة|الثامنة|التاسعة)[\s\u0640]*عشرة|" |
| r"العشرون|" |
| r"الأولى|الأولي|األولى|الاولى|الثانية|الثالثة|الرابعة|الخامسة|" |
| r"السادسة|السابعة|الثامنة|التاسعة|العاشرة" |
| ) |
| _AR_ORD_M = ( |
| r"الأول|الاول|الثاني|الثالث|الرابع|الخامس|" |
| r"السادس|السابع|الثامن|التاسع|العاشر" |
| ) |
| _AR_ART_RE = re.compile( |
| r"\(?(?:الماد[ةه]|املاد[ةه]|مادة)\s*[-–]?\s*[()]*\s*(?:[0-9٠-٩۰-۹]+|" + _AR_ORD + r")" |
| r"(?:\s*[-–])?(?:\s*[()]*)?(?:\s*مكرر(?:اً|ا)?)?" |
| ) |
| _CITE_TAIL_RE = re.compile( |
| _latin_fold( |
| r"(?i)\s+(?:" |
| r"of\s+(?:this\s+|the\s+)?(?:act|law|agreement|code|constitution|article|section|" |
| r"principal\s+act|ordinance|regulation|decree|chapter|part|title|ruling|order)" |
| r"|thereof\b" |
| r"|hereof\b" |
| r"|hereunder\b" |
| r"|de\s+la\s+(?:loi|ley|presente|présente)" |
| r"|du\s+(?:code|présent|present)" |
| r"|des\s+gesetzes" |
| r"|van\s+(?:deze\s+)?(?:wet|artikel)" |
| r"|i\s+lov\b" |
| r"|cua\s+luat(?:\s+nay)?" |
| r")\b" |
| ) |
| ) |
| _NUMERO_FOLD_RE = re.compile( |
| r"(?i)(?:№|n\.\s*o\b|n\.?[°º]|nÂ[oº°]|\bno\.?\s+(?=\d))\.?\s*" |
| ) |
| _CJK_RANGE_RE = re.compile( |
| r"第\s*[一二三四五六七八九十百千万零〇两0-9]+\s*[条條]\s*(?:至|到|から)\s*" |
| r"第\s*[一二三四五六七八九十百千万零〇两0-9]+\s*[条條](?:まで)?" |
| ) |
| _FA_ORD = ( |
| r"یازدهم|دوازدهم|سیزدهم|چهاردهم|پانزدهم|شانزدهم|هفدهم|هجدهم|نوزدهم|بیستم|" |
| r"اولی?|نخست|دوم|سوم|چهارم|چارم|جارم|پنجم|ششم|هفتم|هشتم|نهم|دهم" |
| ) |
| _FA_ART_RE = re.compile(r"(?:ماده|مادة)\s*(?:[0-9۰-۹]+|" + _FA_ORD + r")") |
| _DA_ART_RE = re.compile(r"(?:Artikel|§)\s*[0-9]+", re.IGNORECASE) |
| _CHROME_LINE_RE = re.compile( |
| r"(?i)^(?:" |
| r"početna stranica\b.*|" |
| r"upute za korištenje\b.*|" |
| r"elektronička pošta\b.*|" |
| r"vsebina uradnega lista\b.*|" |
| r"the incident id is\b.*|" |
| r"ωράριο λειτουργίας\b.*|" |
| r"τηλέφωνο επικοινωνίας\b.*|" |
| r"skip to (?:main )?content\b.*|" |
| r"javascript is (?:required|disabled)\b.*|" |
| r"privacy policy\s*$|" |
| r"traduction française pour information\s*$|" |
| r"pdf\s*$|" |
| r"https?://\S+\s*$|" |
| r"légimonaco\b.*|" |
| r"statutory instruments\s*$|" |
| r"official gazette\s*$|" |
| r"government gazette\s*$|" |
| r"(?:the )?london gazette\s*$|" |
| r"journal officiel(?: de la république(?: [^\n]{0,40})?)?\.?\s*$|" |
| r"bulletin officiel\s*$|" |
| r"gazette du canada\s*$|" |
| r"(?:the )?[\w .'-]{0,50}gazette\s*$|" |
| r"الوقائع المصرية\s*$|" |
| r"الوقائع العراقية\b.*|" |
| r"العدد\s*$|" |
| r"العدد\s+\S.{0,80}الوقائع\b.*|" |
| r".*شعبة الجريدة الرسمية\s*$|" |
| r"(?:the )?(?:republic of sudan gazette|gazette published by authority|official gazette of the republic of sudan)\b.*|" |
| r"الصفحة(?:\s*[0-9٠-٩]+)?\s*$|" |
| r"(?:الجريدة|اجلريدة|جريدة)(?:\s+الر\s*سمية)?(?:\s+العدد(?:\s*\(\s*\)?\s*[0-9٠-٩]+)?)?\s*$|" |
| r"république du s[ée]n[ée]gal\s*$|" |
| r"jamhuuriyadda\b.*|" |
| r"warta kerajaan\s*$|" |
| r"lembaran negara\s*$|" |
| r"berita negara\s*$|" |
| r"công báo\s*$|" |
| r"la gaceta\s*$|" |
| r"no\.?\s*\d+\s+government gazette\b.*|" |
| r"supplement to (?:the |official ).{0,60}gazette\b.*|" |
| r"[a-z0-9 ,.'()-]*supplement to official gazette[a-z0-9 ,.'()-]*$|" |
| r"printed by\b.*|" |
| r"https?://\S+(?:\s+\d+(?:\.\d+)*)?\s*$|" |
| r"©+\s*.*congress\b.*|" |
| r"(?:s|sm)\s+congress\b.*|" |
| r"palikir,\s*pohnpei\b.*|" |
| r"www\.\S+(?:\s+\d+)?\s*$|" |
| r"at\s+www\.\S+\s*$|" |
| r".*research bulletin.*|" |
| r"acts supplement\s*$|" |
| r"supplement\s+(?:nr\.\s*|no\.?\s*)\d+(?:\s+\d{1,2}(?:st|nd|rd|th)\s+[a-z]+,?\s+\d{4}\.?)?\s*$|" |
| r"no\.\s+of\s+\d{4}\.?\s*$|" |
| r"act no\.\s+of\s+\d{4}\.?\s*$|" |
| r"bill for the\s*$|" |
| r"(?:the )?republic of [a-z]+(?:\s+[a-z]+)?\s*$|" |
| r"federal republic of [a-z]+(?:\s+[a-z]+)?\s*$|" |
| r"democratic people's republic of korea\s*$|" |
| r".{0,10}republic of palau\s*$|" |
| r"saint vincent and the grenadines\s*$|" |
| r"\[?\s*član\s+\.\.\.\s*\]?\s*$|" |
| r".*аудиони тинглаш.*|" |
| r"reprint\s*$|" |
| r"reprint authorised by\b.*|" |
| r"parliament building\s*$|" |
| r"oau drive, tower hill\s*$|" |
| r"office of the clerk of parliament\s*$|" |
| r"(?:(?:first|second|third|fourth|fifth)\s+parliament(?:\s+of\s+the\s+(?:first|second|third|fourth|fifth)\s+republic)?\s*){1,2}$|" |
| r"(?:(?:first|second|third|fourth|fifth)\s+meeting of parliament\s*){1,2}$|" |
| r"تفاصيل\s+(?:النظام|اللائحة)\s*$|" |
| r"مجموعة الأنظمة السعودية\s*$|" |
| r"المجلد\s+\S+\s*$|" |
| r"أنظمة\s+\S.{0,60}$|" |
| r"(?:first|second|third|fourth|fifth|sixth|seventh|eighth|ninth|tenth)\s+folder\s*$|" |
| r"this law includes:\s*$|" |
| r"-\s*اختر\s*-\s*$|" |
| r"تاريخ الإصدار\s*$|" |
| r"تاريخ النشر\s*$|" |
| r"الحالة\s*$|" |
| r"ساري\s*$|" |
| r"العربية\s*$|" |
| r"الإسم\s*$|" |
| r"البريد الإلكتروني\s*$|" |
| r"رقم الجوال\s*$|" |
| r"نوع البلاغ\s*$|" |
| r"\d{4}/\d{2}/\d{2}\s+ه\s+الموافق\s*:?\s*\d{2}/\d{2}/\d{4}\s+م\s*$|" |
| r"\d{3,4}\s+ه\s*$|" |
| r"خطأ لغوي\s*$|" |
| r"خطأ في مادة\s*$|" |
| r"يتضمن (?:النظام|التنظيم) ما يلي\s*:?\s*$|" |
| r"عدد مرات التصفح\s*\d*\s*$|" |
| r"نبذة عن النظام\s*$|" |
| r"اسم النظام/اللائحة\b.*|" |
| r"تاريخ (?:إصدار|نشر) النظام/اللائحة\s*$|" |
| r"حالة النظام/اللائحة\b.*|" |
| r"أدوات إصدار النظام\s*$|" |
| r"نص النظام\s*$|" |
| r"مادة معدلة\s*$|" |
| r"مادة ملغية\s*$|" |
| r"اصل الوثيقة\s*$|" |
| r"طباعة\s*$|" |
| r"إبلاغ\s*$|" |
| r"الإصدارات\s*$|" |
| r"اللغات\s*$|" |
| r"الإنجليزية\s*\(English\)\s*$|" |
| r"(?:\d{4}\s+)?revised edition(?:\s+as at\b.*|\s+page\s+\d+)?\s*$|" |
| r"page\s+\d+\s+\d{4}\s+revised edition\s*$|" |
| r"\[note: this version\b.*|" |
| r"\[this is the version of this document\b.*|" |
| r"el peruano\s*$|" |
| r"(?:act\s+\d+\s+of\s+\d{4}\s+)?page\s*\d+(?:\s+(?:of\s+\d+|act\s+\d+\s+of\s+\d{4}|[a-z]))?\s*$|" |
| r"[a-z]\s+page\s*\d+\s*$|" |
| r"(?:section\s+)?pages?\s*$|" |
| r"seite\s+\d+(?:\s+(?:von|of|bis)\s+\d+)?\s*$|" |
| r"bundesrecht konsolidiert\b.*|" |
| r".*ris\.bka\.gv\.at.*$|" |
| r"\{\\rtf\d+\b.*|" |
| r"p[aá]gina\s+\d+\s*$|" |
| r"pagina\s+\d+\s*$|" |
| r"strona\s+\d+\s*$|" |
| r"p\.\s*\d+(?:\s+(?:of|di|von|sur|de)\s+\d+)?\s*$|" |
| r"pag\.\s*\d+(?:\s+(?:di|of|de)\s+\d+)?\s*$|" |
| r"blz\.\s*\d+\s*$|" |
| r"folio\s+\d+\s*$|" |
| r"foglio\s+\d+\s*$|" |
| r"лист\s+\d+\s*$|" |
| r"стр\.\s*\d+\s*$|" |
| r"ص\s*\.?\s*\d+\s*$|" |
| r"עמוד\s+\d+\s*$|" |
| r"trang\s+\d+\s*$|" |
| r"halaman\s+\d+\s*$|" |
| r"sayfa\s+\d+\s*$|" |
| r"oldal\s+\d+\s*$|" |
| r"stran(?:ica)?\s+\d+\s*$|" |
| r"www\.\S+\s*$|" |
| r"downloaded from\b.*|" |
| r"printed on\s+\d.*|" |
| r"generated by\b.*|" |
| r"e-?mail:\s*\S+@\S+\s*$|" |
| r"tel(?:ephone)?:\s*[+\d].*$|" |
| r"for more information visit\b.*|" |
| r"[-–—]\s*\d{1,4}\s*[-–—]\s*$|" |
| r"table of contents\s*$|" |
| r"table des mati[eè]res\s*$|" |
| r"inhaltsverzeichnis\s*$|" |
| r"(?:índice|indice)\s*$|" |
| r"sommaire\s*$|" |
| r".*\barrangement of (?:sections|provisions|regulations|rules)\.?(?:\s+cap\.\s*[\d.]+.*)?\s*$|" |
| r"contents\s*$|" |
| r"section\s*$|" |
| r"[\s.\-;,*/_×|+~•·{}()\\=»«®∞]+$|" |
| r"(?=.*[+|])[+\-|]{3,}\s*$|" |
| r"\[original service \d{4}\].*|" |
| r"statute law of the [a-z ]+$|" |
| r"[A-Z][A-Z .'\-]{0,40}\[ch\.\s*\d+.*|" |
| r"التاريخ:\s*$|" |
| r"republic of the marshall islands\s*$|" |
| r"jepilpilin ke ejukaan\s*$|" |
| r"zastupni[cč]ki dom hrvatskoga sabora\s*$|" |
| r"dr[zž]avni zavod za normizaciju i mjeriteljstvo\s*$|" |
| r"олдинги таҳрирга қаранг\.?\s*$|" |
| r"explanatory note\s*$|" |
| r"\.\s*\(\s*\d+\s*\)\s*$|" |
| r"(?:يناير|فبراير|مارس|أبريل|ابريل|مايو|يونيو|يوليو|أغسطس|اغسطس|سبتمبر|أكتوبر|اكتوبر|نوفمبر|ديسمبر)\s+\d{4}\s*$|" |
| r"laws of (?:the )?[a-z]+(?:\s+(?:&|[a-z]+)){0,4}\s*$|" |
| r"(?-i:LAWS OF (?:THE )?[A-Z]+(?:\s+[A-Z]+){0,6}\s+s\s*\d+\s*$)|" |
| r"(?:í|i)ndice legislativo\s*$|" |
| r"قائمة (?:المحتويات|احملتويات)\s*$|" |
| r"spis tre[sś]ci\s*$|" |
| r"содержание\s*$|" |
| r"πίνακας περιεχομένων\s*$|" |
| r"목차\s*$|" |
| r"目次\s*$|" |
| r"สารบัญ\s*$|" |
| r"[aα]ριθμός\s+\d+[^\n]{0,24}\s*$|" |
| r"[aα]ρ\.\s*\d+(?:/\d+)?[^\n]{0,12}\s*$|" |
| r"številka\s+\d+[^\n]{0,24}\s*$|" |
| r"(?:nr|no|n[uú]m(?:ero|éro|ber)?|broj|č[ií]slo|szám|sayı|номер)\.?\s*\d{2,}(?:/\d+)?[.\s]*$|" |
| r"№\s*\d+\s*$|" |
| r"copyright\s*©?\s*\d{4}.*$|" |
| r"©\s*\d{4}.*$|" |
| r"\d{1,2}[./-]\d{1,2}[./-]\d{2,4}\s*$|" |
| r"\d{4}-\d{2}-\d{2}\s*$|" |
| r"\d{1,2}\s+(?:january|february|march|april|may|june|july|august|september|october|november|december|" |
| r"janvier|f[eé]vrier|mars|avril|mai|juin|juillet|ao[uû]t|septembre|octobre|novembre|d[eé]cembre|" |
| r"enero|febrero|marzo|abril|mayo|junio|julio|agosto|septiembre|octubre|noviembre|diciembre)\s+\d{4}\s*$|" |
| r"end of (?:the )?document\s*$|" |
| r"fin du document\s*$|" |
| r"ende des dokuments?\s*$|" |
| r"fin del documento\s*$|" |
| r"fim do documento\s*$|" |
| r"[-*=._]{3,}\s*$|" |
| r"published\s*$|" |
| r"publi[eé]\s*$|" |
| r"publicado\s*$|" |
| r"veröffentlicht\s*$|" |
| r"\[?seal\]?\s*$|" |
| r"\(seal\)\s*$|" |
| r"l\.s\.?\s*$|" |
| r"god save the (?:king|queen)\s*$|" |
| r"vive la r[eé]publique\s*$|" |
| r"by (?:order|command)\s*$|" |
| r"signed\s*$|" |
| r"diario oficial(?: de la federación)?\s*$|" |
| r"gazzetta ufficiale(?: della repubblica(?: italiana)?)?\s*$|" |
| r"boletín oficial(?: del estado)?\s*$|" |
| r"diário da república\s*$|" |
| r"bundesgesetzblatt\s*$|" |
| r"moniteur belge\s*$|" |
| r"belgisch staatsblad\s*$|" |
| r"staatsblad\s*$|" |
| r"lovtidende\s*$|" |
| r"svensk författningssamling\s*$|" |
| r"uradni list\s*$|" |
| r"dziennik ustaw\s*$|" |
| r"monitorul oficial\s*$|" |
| r"εφημερίδα της κυβερνήσεως\s*$|" |
| r"resmi gazete\s*$|" |
| r"iris oifigiúil\s*$|" |
| r"latvijas vēstnesis\s*$|" |
| r"valstybės žinios\s*$|" |
| r"官報\s*$|" |
| r"国务院公报\s*$|" |
| r"관보\s*$|" |
| r"ราชกิจจานุเบกษา\s*$|" |
| r"(?:เล่ม|เลม)\s*[๐-๙0-9].*ราชกิจจานุเบกษา.*|" |
| r"اطلاق نسخة جديدة من الميزان\s*$|" |
| r"\[\d+/\d{4}(?:\s+wef\s+[\d/]+)?\]\s*$|" |
| r"\[\d{1,2}(?:st|nd|rd|th)\s+[a-z]+\s+\d{4}\]\s+.+\s+\d{1,4}\s*$|" |
| r"\[\d{1,2}(?:st|nd|rd|th)\s+[a-z]+,?\s+\d{4}\]\s*$|" |
| r"(?:18|19|20)\d{2}\s*$|" |
| r"gazette\.?\s*$|" |
| r"cap\.\s*\d+(?:\.\d+)*\s*$|" |
| r".+\scap\.?\s*[\d.]+\s+section\s+\d+\s*$|" |
| r"diario oficial\s*$|" |
| r"diário oficial(?: da união)?\s*$|" |
| r"gaceta oficial(?: de la república)?\s*$|" |
| r"registro oficial\s*$|" |
| r"sbírka zákonů\s*$|" |
| r"zbierka z[^\n]{0,40}\s*$|" |
| r"magyar közlöny\s*$|" |
| r"narodne novine\s*$|" |
| r"službeni glasnik\s*$|" |
| r"службени гласник\s*$|" |
| r"държавен вестник\s*$|" |
| r"monitorul oficial(?: al româniei)?\s*$|" |
| r"norsk lovtidend\s*$|" |
| r"suomen säädöskokoelma\s*$|" |
| r"finlands författningssamling\s*$|" |
| r"riigi teataja\s*$|" |
| r"российская газета\s*$|" |
| r"(?:the )?kenya gazette\s*$|" |
| r"federal negar[ie]t gazett?e\s*$|" |
| r"gazette officielle\s*$|" |
| r"amtsblatt(?: der europäischen union)?\s*$|" |
| r"official journal of the european union\s*$|" |
| r"journal officiel de l'union européenne\s*$|" |
| r"gazzetta ufficiale dell'unione europea\s*$|" |
| r"الجريدة الرسمية\b.*|" |
| r"ваш броузер застарів\b.*|" |
| r"your browser is (?:out of date|outdated|obsolete)\b.*|" |
| r"all rights reserved\b.*|" |
| r"subscribe to\s+(?:our\s+)?(?:newsletter|mailing(?:\s+list)?|updates|feed)\b.*|" |
| r"(?:we use|this (?:site|website) uses)\s+cookies\b.*|" |
| r".{0,80}використовуємо cookies\b.*|" |
| r".{0,80}використання cookies\b.*|" |
| r"продовжуючи відвідування сайту\b.*|" |
| r"al meezan\b.*|" |
| r"الميزان\b.*|" |
| r".{0,40}وزارة العدل(?: و الشؤون القانونية)?\s*$|" |
| r"التشريعات\b.*|" |
| r"page not found\b.*|" |
| r"(?:error\s+)?404(?:\s+not found)?\b.*|" |
| r"access denied\s*$|" |
| r"forbidden\s*$|" |
| r"public notification\s*$|" |
| r"[^\n]{0,40}@mjla\b.*|" |
| r".{0,40}gaceta oficial(?: digital)?\b.*|" |
| r"paragraph\s*$|" |
| r"[^\n]{0,60}slov-lex\s*$|" |
| r"(?:crown\s+)?copyright\s*(?:©\s*)?\d{4}\.?\s*$|" |
| r"copying\s*/\s*unauthori[sz]ed distribution strictly prohibited\.?\s*$|" |
| r"(?:revision|consolidation) date\s*:.*|" |
| r"showing the law as at\s+\d.*|" |
| r"this is a revised edition of the law\b.*|" |
| r"printed under authority by\s*$|" |
| r"(?:the )?regional law revision centre\b.*|" |
| r"continued on next page\b.*|" |
| r"back to (?:page one|the top|top)\s*$|" |
| r"page\s*:\s*\d+\s+date\s*:.*|" |
| r"bwpageid\s*::.*|" |
| r"bwservice\s*::.*|" |
| r".*ron_book\b.*|" |
| r"\d+\s+service\s+\d+\s*$|" |
| r"(?:pdf|word|audio)\s+download\s*$|" |
| r"login with (?:facebook|google|apple|twitter|email)\b.*|" |
| r"(?:view|print|e-?mail)\s*$|" |
| r"last updated(?:\s+on)?\s+\d.*|" |
| r"speakers office\s*$|" |
| r"fsm congress\s*$|" |
| r".*almeezan\b.*|" |
| r"add new section\s*$|" |
| r"section name\s*$|" |
| r".*hijripage\b.*|" |
| r"search in articles\s*$|" |
| r"privacy statement\s*$|" |
| r"terms of use\s*$|" |
| r"rate this website\s*$|" |
| r"feedback via reach\s*$|" |
| r"singapore statutes online\b.*|" |
| r".*reprint is authorised by\b.*|" |
| r"công báo\s*/\s*số\b.*|" |
| r"thời gian ký\s*:.*|" |
| r"lex\.?uz\b.*|" |
| r"\[?(?:окоз|тсз|спит)\s*:.*|" |
| r"\d+(?:\.\d{2}){3,}.*[\u0400-\u04FF].*|" |
| r"issue\s*:\s*\d+\s*,\s*amendment\s*:\s*\d+\b.*|" |
| r"presidential comm\.?\s*no\b.*|" |
| r"(?:january|february|march|april|may|june|july|august|september|october|november|december)\s+\d{1,2},?\s+\d{4}(?:\s+[a-z])?\s*$|" |
| r"(?-i:\d{1,4}\s+[Cc][Aa][Pp]\.\s*\d+\.\d+\b(?!.*\b(?!etc\b)[a-z]{3,}\b).*$)|" |
| r"(?-i:(?!.*\b(?!etc\b)[a-z]{3,}\b).{0,90}\b[Cc][Aa][Pp]\.\s*\d+\.\d+(?:\.\d+)*\s+\d{1,4}\s*$)|" |
| r"(?-i:\d{1,4}\s+[Cc][Aa][Pp]\.\s*\d+\s+(?!.*\b[a-z]{3,}\b).+$)|" |
| r"(?-i:(?!.*\b[a-z]{3,}\b).{0,90}\b[Cc][Aa][Pp]\.\s*\d+\s+\d{1,4}\s*$)|" |
| r"(?-i:\d{1,4}\s+[Cc]hap\.\s*\d+(?::\d+)?\b(?!.*\b[a-z]{3,}\b).*$)|" |
| r"(?-i:\d{1,4}\s+(?:[A-Z][A-Za-z]+(?:\s+|$)){3,12}(?:19|20)\d{2}\s*$)|" |
| r"(?-i:(?:[A-Z][A-Za-z]+(?:\s+|$)){3,12}(?:19|20)\d{2}\s+\d{1,4}\s*$)|" |
| r"constituci[oó]n pol[ií]tica del per[uú]\s+\d+\s*$|" |
| r"\d{1,4}\s+edici[oó]n del congreso\b.*|" |
| r"nr\.\s*\d+\s*\].*\[.*|" |
| r"st\.?\s*christopher and nevis\s*$|" |
| r"official gazette\s+nr\.\s*\d[\d\s/.-]*(?:\s+of\s+[\d/.-]+)?\s*$|" |
| r"www\.\S+(?:\s+(?!(?:for|under|and|the|is|of)\b)[A-Za-z][\w.-]{1,24}){1,2}\s*$|" |
| r".*laws\.africa\b.*|" |
| r".*share widely and freely.*|" |
| r".*no copyright on the legislative content.*|" |
| r"(?-i:^(?:Malawi|Zambia|Kenya|Uganda|Ghana|Namibia|Botswana|Lesotho|Eswatini|Rwanda|Tanzania|Nigeria|Zimbabwe|Mauritius|Seychelles|Liberia|Gambia)\s+.{0,80}\(Chapter\s+\d+\)\s*$)|" |
| r"(?-i:^Act\s+\d+\s+(?:[A-Z][a-z]+\s+){1,6}(?:19|20)\d{2}\s*$)|" |
| r"issue\s*:?\s*\d+(?:\.\d+)?\s+page\s+\d+\s+of\s+\d+\b.*|" |
| r"issue\s*:\s*\d+(?:\.\d+)?\s+\d+\s+\d{1,2}\s+[a-z]+\s+\d{4}\s*$|" |
| r".*edition officielle mise [aà] jour.*|" |
| r"code general des impots\s+\d{4}\s*$|" |
| r"\d{1,2}\.\d{1,2}\.\d{4}\s*-\s*\d{1,2}\.\d{1,2}\.\d{4}\s*$|" |
| r"(?!.*δημοσιευ)(?!.*\bστην\b)(?=.*εφημερ.{0,12}κυβερν)(?=.*(?:τευξ|τεύχ|\(\s*τευ)).{0,140}$|" |
| r"(?:\d{1,4}\s+)?επ[ιί]σημη\s+εφημερ[ιί]δ[αά]\b.*|" |
| r"sistema integrado de gest[aã]o de finan[cç]as p[uú]blicas\s*\(sigfip\)\s*\d+\s*/\s*\d+\s*$|" |
| r"editado por\s*:.*\beditado em\s*:.*|" |
| r"issue\s+\d{1,4}\s*$|" |
| r"\d{1,4}\s*\|\s*page\s*$|" |
| r"page\s*\|\s*\d{1,4}\s*$|" |
| r"\{\{[^}\n]*\}\}\s*$|" |
| r"lap tetej[eé]re\s*$|" |
| r"megnyitottak\s*$|" |
| r"jogszab[aá]lykeres[oő]\s*$|" |
| r".*slu[zž]beni glasnik\b.{0,50}stranica\s+\d+\s*$|" |
| r".*\(unknown\)/ron\b.*|" |
| r"\[the next page is\s+[\d,]+\s*\]\s*$|" |
| r"service\s+\d+\s+[\d,]+\s*$|" |
| r"\[act\s+\d+\s+of\s+\d{4}\s+wef\s+[\d/]+\]\s*$|" |
| r"(?:i{1,3}|iv|vi{0,3}|ix|xi{0,3})\s+s[eé]rie\s*-\s*n[uú]mero\s+\d+(?:\s+(?:[a-zç]+-feira|s[aá]bado|domingo),?\s*\d{1,2}\s+de\s*[a-zç]+\s+de\s+\d{4})?\s*$|" |
| r"(?:i{1,3}|iv|vi{0,3}|ix|xi{0,3})\s*-?\s*s[eé]rie\b.*\d.*|" |
| r"(?:i{1,3}|iv|vi{0,3}|ix|xi{0,3})s[eé]rie\b.*\d.*|" |
| r"\d+\s+(?:[a-zç]+-feira|s[aá]bado|domingo),?\s+\d{1,2}\s+de\s+[a-zç]+\s+de\s+\d{4}(?:\s+p[aá]gina\s+\d+)?\s*$|" |
| r"s[eé]rie\s+[ivx]+\s*,\s*nr\.\s*\d+\b.{0,80}p[aá]gina\s+\d+\s*$|" |
| r"(?:[a-zç]+-feira|s[aá]bado|domingo),?\s+\d{1,2}\s+de\s+[a-zç]+\s+de\s+\d{4}\s+s[eé]rie\s+[ivx]+\s*,\s*nr\.\s*\d+\s*$|" |
| r"\d+\s*[\(\)]+\s*مذكرة تفسيرية\b.*|" |
| r"-\s*\d+\s*-\s*\).*$|" |
| r".*constituteproject\.org.*|" |
| r"pdf generated\s*:.*|" |
| r"[a-z][a-z]+(?:\s+[a-z][a-z]+){0,3}\s+(?:19|20)\d{2}\s+page\s+\d+\s*$|" |
| r"الصفحة\s*\.?\s*[0-9٠-٩]+\s*$|" |
| r"\[state/territory\]\s*$|" |
| r"direction g[eé]n[eé]rale adjointe des imp[oô]ts\s*$|" |
| r"et des domaines\s*$|" |
| r"rio:\s*\d+\s*$|" |
| r"#+\s*journal officiel\b.*|" |
| r"journal officiel\s*-\s*banque des?\s*donn[eé]es juridiques\b.*|" |
| r"(?!.*\b(?:publie|publi[eé]|shall|entre)\b)journal officiel (?:de la r[eé]publique|du faso)\b.{0,50}$|" |
| r"\[(?:\.{3}|…)\]\s*$|" |
| r"bsd:\s*\d+\s*$|" |
| r"\d{1,2}\.?\s+(?:gennaio|febbraio|marzo|aprile|maggio|giugno|luglio|agosto|settembre|ottobre|novembre|dicembre)\s*$|" |
| r"(?-i:(?:[A-Z][A-Za-z]+\s+){1,12}(?:Act|Law|Code),?\s+(?:19|20)\d{2}(?:\s+(?:19|20)\d{2})?\s+nr\.\s*\d+\s+[A-Z]\s+\d+\s*$)|" |
| r"(?-i:^[A-Z]\s+\d{1,4}\s+(?:19|20)\d{2}\s+nr\.\s*\d+\s+(?:[A-Z][A-Za-z]+\s+){1,12}(?:Act|Law|Code),?\s+(?:19|20)\d{2}\s*$)|" |
| r"printed on the orders? of government\s*\.?\s*$|" |
| r".*(?<![A-Za-z])(?:printed|thinted)\b.{0,80}government printing\b.*|" |
| r"to be purchased at the govt\.?\s*publications bureau\b.*|" |
| r"b\.?\s*l\.?\s*r\.?\s*o\.?\s*\d+\s*/\s*\d{4}\s*$|" |
| r"pages authorised\s*$|" |
| r"(?:current\s+)?authorised pages\s*$|" |
| r"\(inclusive\)\s*by\s+lro\.?\s*$|" |
| r"(?-i:(?:[A-Z][a-z]+(?:\s+|$)){1,8}(?:Malawi|Zambia|Kenya|Uganda|Ghana|Namibia|Botswana|Lesotho|Eswatini|Rwanda|Tanzania|Nigeria|Zimbabwe|Mauritius|Seychelles|Liberia|Gambia)\s*$)|" |
| r"code\s+\S+(?:\s+\S+){0,6}\s+\d{1,3}/\d{2,3}\s*$|" |
| r"index\s*$|" |
| r"ministry/program(?:me)?/sub(?:head|program(?:me)?)\s+page\s*$|" |
| r"page\s+\d{1,3}\s*/\s*\d{1,3}\s*$|" |
| r"\[?\s*printed by authority of the\s*$|" |
| r"printed in [a-z][a-z .'-]{0,40}by the government printer\s*$|" |
| r"[\w.+-]+@[\w.-]+\.[a-z]{2,}\s+www\.\S+\s*$|" |
| r"secr[eé]tariat g[eé]n[eé]ral du gouvernement\s+www\.\S+\s*$|" |
| r"site:\s*(?:https?://|www\.)\S+\s*$|" |
| r"justice sector support program\s*\(jssp\).*|" |
| r"serial number:\s*\d+\b.*|" |
| r"copyright government of [a-z]+(?:\s+[a-z]+){0,3}\s*$|" |
| r"this page was intentionally left blank\.?\s*$|" |
| r"this is page \d+ of \d+\s+pages? of the above table\.?\s*$|" |
| r"new provisions(?: in the schedule)? are printed in italics\.?\s*$|" |
| r"go to top page(?: next chapter)?\s*$|" |
| r"(?:\d{4}\s+)?parliamentary series\s+(?:nr\.|no\.?)\s*\d+\s*$|" |
| r"try reloading the page or downloading the pdf\.?\s*$|" |
| r"reload page\s*$|" |
| r"uploaded by:\s*https?://\S+\s*$|" |
| r"t[eé]l[eé]charg[eé] sur www\.\S+\s*$|" |
| r"ibirimo/summary/sommaire\b.*|" |
| r"website:\s*(?:https?://|www\.)\S+\s*$|" |
| r"www\.\S+\s+[\w.+-]+@[\w.-]+\s*$|" |
| r"gazette order page\.?\s*$|" |
| r"(?:lexis finder|esilec profesional)\s*-\s*www\.\S+\s*$|" |
| r"official gazette\s+(?:nr\.|no\.?)\s*(?:special(?:\s+bis)?|\d+\s*bis)\s+of\b.*|" |
| r"s[eé]rie\s+[ivxlcdm]+\s*,\s*nr\.\s*\S+(?:\s+[A-Za-z0-9]+)?\s*$|" |
| r"unofficial translat(?:ed|ion)\b.*|" |
| r"\(?\d[\d\s().-]{6,}\)?\s*\|\s*(?:https?://|www\.)\S+\s*$|" |
| r"volume:\s*\d+\s+issue no:\s*\d+\s+government gazette\s*$|" |
| r"tonga government gazette supplement\s*$|" |
| r"ministry of legal affairs\s+www\.\S+\s*$|" |
| r"\d+\s+www\.\S+\s*$|" |
| r"www\.\S+\s+\d+\s*/\s*\d+\s*$|" |
| r"www\.\S+\s+(?:(?!for\b|is\b|shall\b|the\b)[a-z][a-z .'-]{0,50})$|" |
| r".*/type/page>{2,}.*|" |
| r"subject\.\s*reference\.\s*page\.?\s*$|" |
| r"name reference page no\.?\s*$|" |
| r"sl\s*#\s*law\s*#\s*legislation commencement gazette\s*$|" |
| r"gn\s*=\s*gazette notice\b.*|" |
| r"de blank page\b.*|" |
| r"printed and published by the ministry of justice\b.*|" |
| r"rarotonga,\s*cook islands:\s*printed under the authority\b.*|" |
| r"e-?mail\s*:\s*\S+@\S+.*(?:site web|www\.)\S+.*|" |
| r"διεύθυνση στο διαδίκτυο\s*\(url\)\s*:\s*https?://\S+.*|" |
| r"spletna stran:\s*www\.\S+\s*$|" |
| r"copyright\s*$|" |
| r"copyright and designs act\s+\d{4}\s*$|" |
| r"copyright act\s+\d{4}\s*$|" |
| r"powered by tcpdf\b.*|" |
| r"pour l'acquisition de votre abonnement\b.*|" |
| r"a\s*bonnement au journal officiel\b.*|" |
| r"[eé]dit[eé] par la direction de l'[eé]dition du journal officiel\b.*|" |
| r"\[the inclusion of this page is authori[sz]ed by\b.*|" |
| r"ins\s*=\s*inserted\b.*|" |
| r"[a-z]{2,8}\s*=\s*gazette\b.*|" |
| r"site web\s*:\s*www\.\S+\s*$|" |
| r"web\s*sites?\s*:?\s*-?\s*www\.\S+\s*$|" |
| r".*www\.documents\.gov\.lk\s*$|" |
| r"τηλ\.?:?\s*[\d\s,]+φαξ:?\s*[\d\s]+\s*-\s*www\.\S+\s*$|" |
| r"das dokument ist einsehbar unter:\s*www\.\S+\s*$|" |
| r"sujet articles page\s*$|" |
| r"language is printed on uneven numbered pages\.?\s*$|" |
| r".*www\.nbs\.sk\.?\s*$|" |
| r"\d{2}-\d{4,6}\s+\d{1,3}/\d{1,3}\s*$|" |
| r"\d{1,3}/\d{1,3}\s+\d{2}-\d{4,6}\s*$|" |
| r"[as]/(?:res|prst)/\d+\s*\(\d{4}\)\s*$|" |
| r"ab\s+\d{4},\s*nr\.\s*\d+\s*$|" |
| r"grondslag\s*:\s*[.\-–\s]*$|" |
| r"رقم الهاتف\s*:.*|" |
| r"\+[\u0600-\u06FF]{1,2}\s*$|" |
| r"(?-i:(?!.*\b(?:est|sont|doit|shall)\b)(?!.*\b[a-z]{4,}\b).*\bImprimerie\s+[A-Z].*$)|" |
| r"(?-i:(?!.*\b(?:shall|is|are|means|under|must)\b).{8,120}\(Chapter\s+\d+(?::\d+)?\)\s+[A-Z][a-z]+\s*$)|" |
| r"(?-i:(?!.*\b(?:ACT|LAW|CODE|DECREE|ORDINANCE|CHAPTER|ARTICLE|SECTION|TITLE|PART|SCHEDULE|GENERAL|PROVISIONS|SHORT|INTERPRETATION|PRELIMINARY|AMENDMENT|REPEAL|COMMENCEMENT|DEFINITIONS|REGULATIONS|RULES|ORDER|APPENDIX)\b)(?:[A-ZÁÉÍÓÚÜÑ]{2,}(?:\s+|$)){4,10}\d{1,3}\s*$)|" |
| r"(?-i:(?!.*\b(?:ACT|LAW|CODE|DECREE|ORDINANCE|CHAPTER|ARTICLE|SECTION|TITLE|PART|SCHEDULE|GENERAL|PROVISIONS|SHORT|INTERPRETATION|PRELIMINARY|AMENDMENT|REPEAL|COMMENCEMENT|DEFINITIONS|REGULATIONS|RULES|ORDER|APPENDIX)\b)\d{3,4}\s+(?:[A-ZÁÉÍÓÚÜÑ]{3,}(?:\s+|$)){1,3}\s*$)|" |
| r"\[\d{1,4}\]?\s*$|" |
| r"หน.?า\s+[0-9๐-๙]+\s*$|" |
| r"(?:home|blog|contact us|cookie|accueil|startseite|" |
| r"página inicial|página principal|επικοινωνία|αρχική)\s*$" |
| r")" |
| ) |
| _CHROME_INLINE_RE = re.compile( |
| r"(?i)skip to (?:main )?content|all rights reserved|click here|" |
| r"légimonaco" |
| ) |
| _JA_ART_RE = re.compile(r"第\s*[0-9一二三四五六七八九十百]+\s*[条條]") |
| _CIRCLED_ART_RE = re.compile(r"[①-⑳⑴-⒇❶-❿⓫-⓴]") |
| _CJK_ENUM_RE = re.compile(r"(?m)^[ \t]*[一二三四五六七八九十百]+、") |
| _CJK_PAREN_RE = re.compile(r"(?m)^[ \t]*[（(][一二三四五六七八九十百]+[）)]") |
| _CJK_DIGIT_ENUM_RE = re.compile(r"(?m)^[ \t]*[0-9０-９]+、") |
| _KO_GA_RE = re.compile(r"(?m)^[ \t]*[가나다라마바사아자차카타파하]\.") |
| _JA_KATA_ENUM_RE = re.compile(r"(?m)^[ \t]*[アイウエオカキクケコ]、") |
| _HANGUL_GA = "가나다라마바사아자차카타파하" |
| _KATA_ENUM = "アイウエオカキクケコ" |
| _AR_ABJAD_INDEX = { |
| "أ": "1", "ا": "1", "ب": "2", "ج": "3", "د": "4", "ه": "5", |
| "و": "6", "ز": "7", "ح": "8", "ط": "9", "ي": "10", |
| } |
| _HE_ALEF = "אבגדהוזחטי" |
| _TH_KO = "กขคงจฉชซฌญ" |
| _AR_ABJAD_RE = re.compile(r"(?m)^[ \t]*[(\[]?[أابجدهوزحطي][)\]]?\s*[-.)]") |
| _HE_ALEF_RE = re.compile(r"(?m)^[ \t]*[אבגדהוזחטי]\.") |
| _HE_PEREK_RE = re.compile(r"(?m)^[ \t]*פרק\s+[א-ת]{1,3}['׳]?") |
| _HE_SIMAN_RE = re.compile(r"(?m)^[ \t]*סימן\s+[א-ת]{1,3}['׳]?") |
| _HE_HELEK_RE = re.compile(r"(?m)^[ \t]*חלק\s+[א-ת]{1,3}['׳]?") |
| _HE_TOS_RE = re.compile( |
| r"(?m)^[ \t]*תוספת\s+(?:ראשונה|שנייה|שניה|שלישית|רביעית|חמישית|שישית|[א-ת]['׳]?|[0-9]+)" |
| ) |
| _HE_TAKANA_RE = re.compile(r"(?m)^[ \t]*תקנה\s+[0-9]+") |
| _AR_ANNEX_RE = re.compile(r"(?m)^[ \t]*ملحق(?:\s*رقم)?\s*[0-9٠-٩()]+") |
| _AR_BAB_RE = re.compile( |
| r"(?m)^[ \t]*الباب\s+(?:[0-9٠-٩]+|" + _AR_ORD_M + r"|" + _AR_ORD + r")" |
| ) |
| _AR_FASL_RE = re.compile( |
| r"(?m)^[ \t]*الفصل\s+(?:[0-9٠-٩]+|" + _AR_ORD_M + r"|" + _AR_ORD + r")" |
| ) |
| _AR_KITAB_RE = re.compile( |
| r"(?m)^[ \t]*الكتاب\s+(?:[0-9٠-٩]+|" + _AR_ORD_M + r"|" + _AR_ORD + r")" |
| ) |
| _AR_FARA_RE = re.compile( |
| r"(?m)^[ \t]*الفرع\s+(?:[0-9٠-٩]+|" + _AR_ORD_M + r"|" + _AR_ORD + r")" |
| ) |
| _FA_BAND_RE = re.compile(r"بند\s*[0-9۰-۹]+") |
| _CJK_ZHANG_RE = re.compile(r"第\s*[一二三四五六七八九十百千万零〇两0-9]+\s*章") |
| _KO_JANG_RE = re.compile(r"제\s*[0-9]+\s*장") |
| _TA_ART_RE = re.compile(r"பிரிவு\s*[0-9௦-௯]+") |
| _TE_ART_RE = re.compile(r"ప్రకరణ(?:ము)?\s*[0-9౦-౯]+") |
| _KN_ART_RE = re.compile(r"ಪ್ರಕರಣ\s*[0-9೦-೯]+") |
| _GU_ART_RE = re.compile(r"કલમ\s*[0-9૦-૯]+") |
| _PA_ART_RE = re.compile(r"ਧਾਰਾ\s*[0-9੦-੯]+") |
| _ML_ART_RE = re.compile(r"വകുപ്പ്\s*[0-9൦-൯]+") |
| _CJK_BIAN_RE = re.compile(r"第\s*[一二三四五六七八九十百千万零〇两0-9]+\s*[编編]") |
| _KO_PYEON_RE = re.compile(r"제\s*[0-9]+\s*편") |
| _TH_LAK_RE = re.compile(r"(?m)^[ \t]*ลักษณะ\s*[0-9๐-๙]+") |
| _KM_CHAP_RE = re.compile(r"ជំពូក\s*[0-9០-៩]+") |
| _LO_PART_RE = re.compile(r"ພາກ\s*[0-9໐-໙]+") |
| _MY_CHAP_RE = re.compile(r"အခန်း\s*[0-9၀-၉]+") |
| _KA_CHAP_RE = re.compile(r"თავი\s+[IVXLCDM0-9]+") |
| _HY_CHAP_RE = re.compile(r"Գլուխ\s+[IVXLCDM0-9]+", re.IGNORECASE) |
| _AM_CHAP_RE = re.compile(r"ምዕራፍ\s*[0-9፩-፻]+") |
| _HI_CHAP_RE = re.compile(r"अध्याय\s*[0-9०-९]+") |
| _BN_CHAP_RE = re.compile(r"অধ্যায়\s*[0-9০-৯]+") |
| _UR_CHAP_RE = re.compile(r"باب\s*[0-9۰-۹]+") |
| _FA_FASL_RE = re.compile(r"فصل\s*[0-9۰-۹]+") |
| _NE_CHAP_RE = re.compile(r"परिच्छेद\s*[0-9०-९]+") |
| _CJK_JIE_RE = re.compile(r"第\s*[一二三四五六七八九十百千万零〇两0-9]+\s*[节節]") |
| _KO_JEOL_RE = re.compile(r"제\s*[0-9]+\s*절") |
| _TH_TON_RE = re.compile(r"(?m)^[ \t]*ตอน\s*[0-9๐-๙]+") |
| _SR_PART_RE = re.compile(r"(?:Одељак|ОДЕЉАК)\s+[IVXLCDM0-9]+", re.IGNORECASE) |
| _RU_TITLE_RE = re.compile( |
| r"(?:Раздел|РАЗДЕЛ|Розділ|РОЗДІЛ|Раздзел|РАЗДЗЕЛ)\s+[IVXLCDM0-9]+", |
| re.IGNORECASE, |
| ) |
| _RU_CHAP_RE = re.compile(r"(?:Глава|ГЛАВА)\s+[IVXLCDM0-9]+", re.IGNORECASE) |
| _TH_KO_RE = re.compile(r"(?m)^[ \t]*[กขคงจฉชซฌญ]\.") |
| _HI_DIGIT_ENUM_RE = re.compile(r"(?m)^[ \t]*[०-९]+\s*[.।]") |
| _BN_DIGIT_ENUM_RE = re.compile(r"(?m)^[ \t]*[০-৯]+\s*[.।]") |
| _CYR_AB = "абвгдежзийк" |
| _EL_AB = "αβγδεζηθικ" |
| _KA_AB = "აბგდევზთიკლ" |
| _HY_AB = "աբգդեզէըթժ" |
| _CYR_AB_RE = re.compile(r"(?im)^[ \t]*[абвгдежзийк]\s*[).]") |
| _EL_AB_RE = re.compile(r"(?im)^[ \t]*[αβγδεζηθικ]\s*[).]") |
| _KA_AB_RE = re.compile(r"(?m)^[ \t]*[აბგდევზთიკლ]\s*[).]") |
| _HY_AB_RE = re.compile(r"(?m)^[ \t]*[աբգդեզէըթժ]\s*[).]") |
| _MARK_AFTER = r"(?:\s+|(?=[^\x00-\x7f]))" |
| _LATIN_LETTER_RE = re.compile( |
| r"(?m)^[ \t]*(?:\(\s*(?P<paren>[a-z])\s*\)|(?P<bare>[a-z])\)|" |
| r"(?P<roman>viii|iii|vii|ii|iv|ix|vi|i|v|x)\.)" |
| + _MARK_AFTER |
| ) |
| _BULLET_RE = re.compile(r"(?m)^[ \t]*[•·●‣⁃*#]\s+") |
| _DASH_NUM_RE = re.compile(r"(?m)^[ \t]*-\s*(?P<bn>[0-9]{1,3})\s+(?=[^\W\d_])") |
| _LATIN_NUM_LIST_RE = re.compile( |
| r"(?m)^[ \t]*(?:" |
| r"\(\s*(?P<nparen>[0-9]{1,3})\s*\)|" |
| r"(?P<nclose>[0-9]{1,3})\)|" |
| r"(?P<nord>[0-9]{1,3})[º°o]|" |
| r"(?P<ndot>[0-9]{1,3})\.-|" |
| r"(?P<ncolon>[0-9]{1,3}):|" |
| r"(?P<ndash>[0-9]{1,3})-|" |
| r"(?P<dbl>([a-z])\2)\)" |
| r")" |
| + _MARK_AFTER |
| ) |
| _ROMAN_LOWER = { |
| "i": "1", "ii": "2", "iii": "3", "iv": "4", "x": "10", |
| "v": "5", "vi": "6", "vii": "7", "viii": "8", "ix": "9", |
| } |
| _JA_KOU_RE = re.compile(r"第[0-9一二三四五六七八九十百]+項") |
| _ZH_KUAN_RE = re.compile(r"第[一二三四五六七八九十百千万零〇两0-9]+款") |
| _KO_ART_RE = re.compile(r"제\s*[0-9]+\s*조") |
| _KO_HANG_RE = re.compile(r"제\s*[0-9]+\s*항") |
| _KO_HO_RE = re.compile(r"제\s*[0-9]+\s*호") |
| _TH_PARA_RE = re.compile(r"วรรค\s*[0-9๐-๙]+") |
| _TH_KHO_RE = re.compile(r"(?m)^[ \t]*ข้อ\s*[0-9๐-๙]+") |
| _TH_HUAT_RE = re.compile(r"(?m)^[ \t]*หมวด\s*[0-9๐-๙]+") |
| _FA_TAB_RE = re.compile(r"تبصره\s*[0-9۰-۹]+") |
| _HE_PARA_RE = re.compile(r"פסקה\s*[0-9]+") |
| _AR_PARA_RE = re.compile(r"الفقرة\s*[0-9٠-٩]+") |
| _AR_QARAR_RE = re.compile(r"قرار(?:\s*رقم)?\s*[0-9٠-٩]+") |
| _AR_DECREE_RE = re.compile(r"(?:مرسوم|أمر)(?:\s*رقم)?\s*[0-9٠-٩]+") |
| _RU_INST_RE = re.compile(r"(?:Указ|Постановление|Приказ)\s+[0-9]+", re.IGNORECASE) |
| _KO_RANGE_RE = re.compile( |
| r"제\s*[0-9]+\s*조\s*부터\s*제\s*[0-9]+\s*조(?:까지)?" |
| ) |
| _HE_ART_RE = re.compile( |
| r"(?:סעיף\s*(?:[0-9]+|(?!קטן)[א-ת]{1,3}['׳]?)|\.\s*[0-9]+(?=[\u0590-\u05FF]))" |
| ) |
| _HE_KATAN_RE = re.compile(r"סעיף\s+קטן\s*\(?\s*[א-ת]'?\)?") |
| _TH_ART_RE = re.compile(r"มาตรา\s*[0-9๐-๙]+") |
| _HI_ART_RE = re.compile(r"(?:धारा|ধারা)\s*[0-9০-৯]+") |
| _RU_ART_RE = re.compile(r"(?:Статья|СТАТЬЯ|Стаття|СТАТТЯ)\s+[0-9]+(?:-[0-9]+)?", re.IGNORECASE) |
| _KA_ART_RE = re.compile(r"მუხლი\s*[0-9]+") |
| _HY_ART_RE = re.compile(r"Հոդված\s*[0-9]+", re.IGNORECASE) |
| _AM_ART_RE = re.compile(r"አንቀጽ\s*[0-9፩-፻]+") |
| _BN_ART_RE = re.compile(r"ধারা\s*[0-9০-৯]+") |
| _LO_ART_RE = re.compile(r"ມາດຕາ\s*[0-9໐-໙]+") |
| _KM_ART_RE = re.compile(r"មាត្រា\s*[0-9០-៩]+") |
| _MY_ART_RE = re.compile(r"ပုဒ်မ\s*[0-9၀-၉]+") |
| _UZ_ART_RE = re.compile(r"[0-9]+\s*-\s*(?:модда|modda)", re.IGNORECASE) |
| _MN_ART_RE = re.compile(r"[0-9]+\s*(?:дүгээр|дугаар)\s*зүйл", re.IGNORECASE) |
| _KK_ART_RE = re.compile(r"[0-9]+\s*-?\s*бап", re.IGNORECASE) |
| _NE_ART_RE = re.compile(r"दफा\s*[0-9०-९]+") |
| _TI_ART_RE = re.compile(r"ዓንቀጽ\s*[0-9፩-፻]+") |
| _BO_ART_RE = re.compile(r"དོན་ཚན་\s*[0-9༠-༩]+") |
| _UG_ART_RE = re.compile(r"ماددا\s*[0-9۰-۹]+") |
| _AR_POINT_RE = re.compile(r"(?m)^[ \t]*(?:أولا|ثانيا|ثالثا|رابعا|خامسا)\s*[:/]") |
| _AR_POINT_NUM = (("أولا", "1"), ("ثانيا", "2"), ("ثالثا", "3"), ("رابعا", "4"), ("خامسا", "5")) |
| _SI_ART_RE = re.compile(r"වගන්තිය\s*[0-9෦-෯]+") |
| _UR_ART_RE = re.compile(r"دفعہ\s*[0-9۰-۹]+") |
| _KY_ART_RE = re.compile(r"[0-9]+\s*-?\s*берене", re.IGNORECASE) |
| _BE_ART_RE = re.compile(r"Артыкул\s*[0-9]+", re.IGNORECASE) |
|
|
| |
| _SPACED_KEYWORD_RE = re.compile( |
| r"(?iu)\b((?:[A-ZÁÉÍÓÚÜÑÀÈÌÒÙÂÊÎÔÛÄËÏÖÜÃÕÇ]\s+){3,}" |
| r"[A-ZÁÉÍÓÚÜÑÀÈÌÒÙÂÊÎÔÛÄËÏÖÜÃÕÇ])\b" |
| r"(?=\s*[.:\-–—]?\s*[0-9IVXLCDM])" |
| ) |
|
|
|
|
| def _squeeze_spaced_keyword(match: re.Match[str]) -> str: |
| return re.sub(r"\s+", "", match.group(1)) |
|
|
|
|
| _ZW_RE = re.compile( |
| r"[\ufeff\u200b\u200c\u200d\u00ad\u2060\u200e\u200f\u202a-\u202e\u2066-\u2069]" |
| ) |
| |
| _STRIP_MARKS_RE = re.compile( |
| "[" |
| "\u0300-\u036f" |
| "\u0483-\u0489" |
| "\u0591-\u05bd\u05bf\u05c1\u05c2\u05c4\u05c5\u05c7" |
| "\u0610-\u061a\u064b-\u065f\u0670\u06d6-\u06ed" |
| "\u08d3-\u08e1\u08e3-\u08ff" |
| "]" |
| ) |
| _SCRIPT_CITE_TAIL_RE = re.compile( |
| r"(?i)^\s*(?:" |
| r"من\s*(?:الدستور|القانون|الأمر|المرسوم|القرار)|" |
| r"của\s+luật|" |
| r"של\s+(?:חוק|פקודה)" |
| r")" |
| ) |
|
|
| |
| _LATIN1_UTF8_RE = re.compile(r"[\u00c2\u00c3][\u0080-\u00bf]") |
|
|
|
|
| def _repair_latin1_utf8(text: str) -> str: |
| if "\u00c2" not in text and "\u00c3" not in text: |
| return text |
|
|
| def repl(match: re.Match[str]) -> str: |
| try: |
| return match.group(0).encode("latin-1").decode("utf-8") |
| except UnicodeDecodeError: |
| return match.group(0) |
|
|
| return _LATIN1_UTF8_RE.sub(repl, text) |
|
|
|
|
| |
| _LATIN_FALLBACK_LANGS = frozenset( |
| {"ka", "km", "hy", "am", "lo", "my", "si", "th", "he", "ne", "fa"} |
| ) |
|
|
|
|
| class _HTMLText(HTMLParser): |
| SKIP = {"script", "style", "noscript", "svg", "nav", "footer", "header"} |
|
|
| def __init__(self) -> None: |
| super().__init__(convert_charrefs=True) |
| self.parts: list[str] = [] |
| self.skip = 0 |
|
|
| def handle_starttag(self, tag, attrs): |
| if tag in self.SKIP: |
| self.skip += 1 |
| if tag in {"br", "p", "tr", "div", "li", "h1", "h2", "h3"} and self.skip == 0: |
| self.parts.append("\n") |
|
|
| def handle_endtag(self, tag): |
| if tag in self.SKIP and self.skip: |
| self.skip -= 1 |
| if tag in {"p", "div", "li", "tr"} and self.skip == 0: |
| self.parts.append("\n") |
|
|
| def handle_data(self, data): |
| if self.skip == 0: |
| self.parts.append(data) |
|
|
|
|
| def has_html_tags(text: str) -> bool: |
| return bool(text and _HAS_TAG_RE.search(text)) |
|
|
|
|
| def strip_html(raw: str) -> str: |
| """Turn leftover HTML into visible legal text. No-op if there are no tags.""" |
| if not raw or not _HAS_TAG_RE.search(raw): |
| return raw |
| parser = _HTMLText() |
| try: |
| parser.feed(raw) |
| parser.close() |
| text = "".join(parser.parts) |
| except Exception: |
| text = re.sub(r"(?is)<script.*?>.*?</script>", " ", raw) |
| text = re.sub(r"(?is)<style.*?>.*?</style>", " ", text) |
| text = re.sub(r"(?is)<[^>]+>", " ", text) |
| text = html_lib.unescape(text).replace("\xa0", " ") |
| text = re.sub(r"[ \t]+", " ", text) |
| text = re.sub(r"\n[ \t]+", "\n", text) |
| text = re.sub(r"\n{3,}", "\n\n", text) |
| return text.strip() |
|
|
|
|
| _ROMAN_PAGE_RE = re.compile( |
| r"(?i)(?:i{1,3}|iv|vi{0,3}|ix|xi{0,3}|xiv|xv)$" |
| ) |
|
|
|
|
| def _is_roman_page_line(line: str, nxt: str) -> bool: |
| """A lone i/ii/iii between blocks is a page stamp, not a list item.""" |
| if _ROMAN_PAGE_RE.fullmatch(line.strip()) is None: |
| return False |
| nxt = nxt.strip() |
| if not nxt: |
| return True |
| return not (nxt[:1].islower() or nxt.startswith("(")) |
|
|
|
|
| def _is_page_num_line(line: str) -> bool: |
| token = line.strip() |
| if 1 <= len(token) <= 4 and all(unicodedata.category(ch) == "Nd" for ch in token): |
| return True |
| |
| parts = token.split("/") |
| if len(parts) == 2 and all(part.isascii() and part.isdigit() for part in parts): |
| left, right = int(parts[0]), int(parts[1]) |
| return 1 <= left <= right <= 30 |
| return False |
|
|
|
|
| def _is_facing_page_line(line: str) -> bool: |
| """'736 737' between blocks is a two-page spread, not a citation.""" |
| parts = line.split() |
| if len(parts) != 2 or not all(part.isascii() and part.isdigit() for part in parts): |
| return False |
| left, right = int(parts[0]), int(parts[1]) |
| return 100 <= left and right == left + 1 and right <= 9999 |
|
|
|
|
| _PAGE_OF_RE = re.compile(r"(?i)(\d{1,2})\s+of\s+(\d{1,2})") |
|
|
|
|
| def _is_page_of_line(line: str) -> bool: |
| """'3 of 5' between blocks is a page count, not a citation.""" |
| match = _PAGE_OF_RE.fullmatch(line.strip()) |
| if match is None: |
| return False |
| left, right = int(match.group(1)), int(match.group(2)) |
| return 1 <= left <= right <= 30 |
|
|
|
|
| def learned_patterns_path() -> Path: |
| override = os.environ.get("COUNTRY_LAWS_LEARNED_PATTERNS") |
| if override: |
| return Path(override) |
| return Path(__file__).resolve().parent / "data" / "learned_line_patterns.json" |
|
|
|
|
| _LEARNED: tuple[float, re.Pattern[str] | None] | None = None |
|
|
|
|
| def learned_line_pattern() -> re.Pattern[str] | None: |
| """Whole-line rules added by the normalize loop. Compiled once per file change.""" |
| global _LEARNED |
| path = learned_patterns_path() |
| if not path.is_file(): |
| return None |
| mtime = path.stat().st_mtime |
| if _LEARNED is not None and _LEARNED[0] == mtime: |
| return _LEARNED[1] |
| try: |
| payload = json.loads(path.read_text(encoding="utf-8")) |
| except (OSError, json.JSONDecodeError): |
| _LEARNED = (mtime, None) |
| return None |
| parts: list[str] = [] |
| for item in payload.get("patterns") or []: |
| if not isinstance(item, str): |
| continue |
| try: |
| re.compile(item, re.IGNORECASE) |
| except re.error: |
| continue |
| parts.append(item) |
| compiled = re.compile("|".join(f"(?:{part})" for part in parts), re.IGNORECASE) if parts else None |
| _LEARNED = (mtime, compiled) |
| return compiled |
|
|
|
|
| def strip_chrome_lines(text: str) -> str: |
| """Drop gazette-site chrome lines. Never invents replacement legal text.""" |
| if not text: |
| return "" |
| raw_lines = text.splitlines() |
| multi = sum(1 for ln in raw_lines if ln.strip()) > 1 |
| learned = learned_line_pattern() |
| kept = [] |
| for idx, ln in enumerate(raw_lines): |
| stripped = ln.strip() |
| if not stripped: |
| continue |
| if _CHROME_LINE_RE.match(stripped) or _CHROME_INLINE_RE.search(ln): |
| continue |
| if learned is not None and learned.match(stripped): |
| continue |
| |
| |
| if multi and ( |
| _is_page_num_line(stripped) |
| or _is_facing_page_line(stripped) |
| or _is_page_of_line(stripped) |
| ): |
| continue |
| nxt = "" |
| for later in raw_lines[idx + 1 :]: |
| if later.strip(): |
| nxt = later |
| break |
| if multi and _is_roman_page_line(stripped, nxt): |
| continue |
| kept.append(ln) |
| return "\n".join(kept).strip() |
|
|
|
|
| _HYPHEN_BREAK_RE = re.compile(r"(?<=[^\W\d_])-[ \t]*\n+[ \t]*([^\W\d_])") |
|
|
|
|
| def _join_hyphen_breaks(text: str) -> str: |
| """Join a hyphenated word split across lines. Leave a following capital header.""" |
|
|
| def repl(match: re.Match[str]) -> str: |
| nxt = match.group(1) |
| if nxt.isupper(): |
| return match.group(0) |
| return nxt |
|
|
| return _HYPHEN_BREAK_RE.sub(repl, text) |
|
|
|
|
| def normalize_legal_text(value: Any) -> str: |
| """NFKC + HTML strip. Keeps newlines so heading detection still works.""" |
| if value is None: |
| return "" |
| text = unicodedata.normalize("NFKC", str(value)).replace("\x00", "") |
| if text.startswith("ÿþ") or text.startswith("þÿ"): |
| text = text[2:] |
| text = re.sub(r"[\x00-\x08\x0b\x0e-\x1f\x7f]", " ", text) |
| text = text.replace("\ufffd", " ") |
| text = re.sub(r"[\ue000-\uf8ff]", "", text) |
| text = _repair_latin1_utf8(text) |
| text = re.sub(r"\(cid:\d+\)", " ", text) |
| text = _ZW_RE.sub("", text).replace("ـ", "") |
| text = _STRIP_MARKS_RE.sub("", text) |
| text = re.sub(r"[\u2010-\u2015\u2212\ufe58\ufe63\uff0d]", "-", text) |
| text = re.sub(r"[\u00a0\u1680\u2000-\u200a\u202f\u205f\u3000]", " ", text) |
| text = text.replace("\u2028", "\n").replace("\u2029", "\n").replace("\f", "\n") |
| text = _join_hyphen_breaks(text) |
| text = _NUMERO_FOLD_RE.sub("nr. ", text) |
| text = re.sub(r"المادة\s*-\s*اختر\s*-\s*", "", text) |
| text = re.sub(r"(?m)^[ \t]*©+\s*", "", text) |
| text = re.sub(r"\b((?:19|20)\d{2}) \1\b", r"\1", text) |
| text = re.sub(r"(?<=[A-Za-z])!+(?=\d)", " ", text) |
| text = re.sub(r":!+", ":", text) |
| text = re.sub(r"(?m)!+\s*$", "", text) |
| text = _SPACED_KEYWORD_RE.sub(_squeeze_spaced_keyword, text) |
| text = re.sub(r"(?s)<!--.*?-->", " ", text) |
| text = strip_html(text) |
| text = html_lib.unescape(text).replace("\xa0", " ") |
| text = re.sub(r"_{3,}", " ", text) |
| text = re.sub(r"\.{4,}", " ", text) |
| text = re.sub(r"(?m)^[ \t]*#{1,6}[ \t]+", "", text) |
| text = strip_chrome_lines(text) |
| text = text.replace("\xa0", " ") |
| text = re.sub(r"~\$\S{0,16}", " ", text) |
| text = re.sub(r"\*\*([^*]{1,80})\*\*", r"\1", text) |
| text = re.sub(r"__([^_]{1,80})__", r"\1", text) |
| text = re.sub(r"'{3}([^']{1,80})'{3}", r"\1", text) |
| text = re.sub(r"^={2,6}\s*(.+?)\s*={2,6}\s*$", r"\1", text, flags=re.MULTILINE) |
| text = re.sub(r"[ \t]+", " ", text) |
| text = re.sub(r"\n[ \t]+", "\n", text) |
| text = re.sub(r"[ \t]+\n", "\n", text) |
| text = _join_hyphen_breaks(text) |
| text = re.sub(r"\n{3,}", "\n\n", text) |
| return text.strip() |
|
|
|
|
| @dataclass |
| class StructureUnit: |
| kind: str |
| number: str |
| heading: str |
| body: str |
| title_number: str = "" |
| chapter_number: str = "" |
| part_number: str = "" |
| article_number: str = "" |
| section_number: str = "" |
| subsections: tuple[str, ...] = () |
| hierarchy_path: str = "" |
| char_start: int = 0 |
| char_end: int = 0 |
|
|
| def to_dict(self) -> dict[str, Any]: |
| return { |
| "kind": self.kind, |
| "number": self.number, |
| "heading": self.heading, |
| "body": self.body, |
| "title_number": self.title_number, |
| "chapter_number": self.chapter_number, |
| "part_number": self.part_number, |
| "article_number": self.article_number, |
| "section_number": self.section_number, |
| "subsections": list(self.subsections), |
| "hierarchy_path": self.hierarchy_path, |
| "char_start": self.char_start, |
| "char_end": self.char_end, |
| } |
|
|
|
|
| def _cursor_path(cursor: dict[str, str]) -> str: |
| parts = [] |
| for key, label in ( |
| ("title", "Title"), |
| ("chapter", "Chapter"), |
| ("part", "Part"), |
| ("article", "Article"), |
| ("section", "Section"), |
| ): |
| value = cursor.get(key) or "" |
| if value: |
| parts.append(f"{label} {value}") |
| return " > ".join(parts) |
|
|
|
|
| def _subsection_tokens(text: str) -> tuple[str, ...]: |
| seen: list[str] = [] |
| for match in _SUBSECTION_RE.finditer(text): |
| token = match.group(1) |
| if token not in seen: |
| seen.append(token) |
| if len(seen) >= 40: |
| break |
| return tuple(seen) |
|
|
|
|
| def check_exclusive_cover(text: str, units: list[StructureUnit]) -> bool: |
| """True iff units are contiguous, non-overlapping, and cover *text* exactly.""" |
| if not units: |
| return not text |
| ordered = sorted(units, key=lambda u: u.char_start) |
| if ordered[0].char_start != 0 or ordered[-1].char_end != len(text): |
| return False |
| prev = 0 |
| for unit in ordered: |
| if unit.char_start != prev or unit.char_end < unit.char_start: |
| return False |
| if text[unit.char_start : unit.char_end] != unit.body: |
| return False |
| prev = unit.char_end |
| return True |
|
|
|
|
| def _script_regex(language: str): |
| from .profiles import iso_lang |
|
|
| lang = iso_lang(language) |
| if lang in {"zh", "zh-cn", "zh-tw"}: |
| return _ZH_ART_RE |
| if lang == "ar": |
| return _AR_ART_RE |
| if lang == "fa": |
| return _FA_ART_RE |
| if lang == "ja": |
| return _JA_ART_RE |
| if lang == "ko": |
| return _KO_ART_RE |
| if lang == "he": |
| return _HE_ART_RE |
| if lang == "th": |
| return _TH_ART_RE |
| if lang == "hi": |
| return _HI_ART_RE |
| if lang == "bn": |
| return _BN_ART_RE |
| if lang in {"ru", "uk", "be"}: |
| return _RU_ART_RE |
| if lang == "ka": |
| return _KA_ART_RE |
| if lang == "hy": |
| return _HY_ART_RE |
| if lang == "am": |
| return _AM_ART_RE |
| if lang == "ti": |
| return _TI_ART_RE |
| if lang == "bo": |
| return _BO_ART_RE |
| if lang == "ug": |
| return _UG_ART_RE |
| if lang == "lo": |
| return _LO_ART_RE |
| if lang == "km": |
| return _KM_ART_RE |
| if lang == "my": |
| return _MY_ART_RE |
| if lang == "uz": |
| return _UZ_ART_RE |
| if lang == "mn": |
| return _MN_ART_RE |
| if lang == "kk": |
| return _KK_ART_RE |
| if lang == "ne": |
| return _NE_ART_RE |
| if lang == "si": |
| return _SI_ART_RE |
| if lang == "ta": |
| return _TA_ART_RE |
| if lang == "te": |
| return _TE_ART_RE |
| if lang == "kn": |
| return _KN_ART_RE |
| if lang == "gu": |
| return _GU_ART_RE |
| if lang == "pa": |
| return _PA_ART_RE |
| if lang == "ml": |
| return _ML_ART_RE |
| if lang == "ur": |
| return _UR_ART_RE |
| if lang == "ky": |
| return _KY_ART_RE |
| if lang == "be": |
| return _BE_ART_RE |
| return None |
|
|
|
|
| def _detected_script_regexes(text: str, language: str) -> list: |
| found: list = [] |
| primary = _script_regex(language) |
| if primary is not None: |
| found.append(primary) |
| if re.search(r"[①-⑳⑴-⒇❶-❿⓫-⓴]", text): |
| found.append(_CIRCLED_ART_RE) |
| if re.search(r"[\u4e00-\u9fff]", text): |
| found.extend((_ZH_ART_RE, _JA_ART_RE, _ZH_KUAN_RE, _JA_KOU_RE, _CJK_ZHANG_RE, _CJK_BIAN_RE, _CJK_JIE_RE, _CJK_ENUM_RE, _CJK_PAREN_RE, _CJK_DIGIT_ENUM_RE)) |
| if re.search(r"[\uac00-\ud7af]", text): |
| found.extend((_KO_ART_RE, _KO_HANG_RE, _KO_HO_RE, _KO_GA_RE, _KO_JANG_RE, _KO_PYEON_RE, _KO_JEOL_RE)) |
| if re.search(r"[\u30a0-\u30ff]", text): |
| found.append(_JA_KATA_ENUM_RE) |
| if re.search(r"[\u0600-\u06ff]", text): |
| found.extend(( |
| _AR_ART_RE, _FA_ART_RE, _FA_TAB_RE, _FA_BAND_RE, _FA_FASL_RE, _AR_PARA_RE, _AR_QARAR_RE, _AR_DECREE_RE, |
| _AR_ABJAD_RE, _AR_ANNEX_RE, _AR_BAB_RE, _AR_FASL_RE, _AR_KITAB_RE, _AR_FARA_RE, _UR_CHAP_RE, _UG_ART_RE, _AR_POINT_RE, |
| )) |
| if re.search(r"[\u0590-\u05ff]", text): |
| found.extend(( |
| _HE_ART_RE, _HE_PARA_RE, _HE_ALEF_RE, _HE_PEREK_RE, _HE_SIMAN_RE, |
| _HE_HELEK_RE, _HE_TOS_RE, _HE_TAKANA_RE, _HE_KATAN_RE, |
| )) |
| if re.search(r"[\u0e00-\u0e7f]", text): |
| found.extend((_TH_ART_RE, _TH_PARA_RE, _TH_KO_RE, _TH_KHO_RE, _TH_HUAT_RE, _TH_LAK_RE, _TH_TON_RE)) |
| if re.search(r"[\u0900-\u097f]", text): |
| found.extend((_HI_ART_RE, _HI_DIGIT_ENUM_RE, _HI_CHAP_RE, _NE_CHAP_RE)) |
| if re.search(r"[\u0980-\u09ff]", text): |
| found.extend((_BN_ART_RE, _BN_DIGIT_ENUM_RE, _BN_CHAP_RE)) |
| if re.search(r"[\u0b80-\u0bff]", text): |
| found.append(_TA_ART_RE) |
| if re.search(r"[\u0c00-\u0c7f]", text): |
| found.append(_TE_ART_RE) |
| if re.search(r"[\u0c80-\u0cff]", text): |
| found.append(_KN_ART_RE) |
| if re.search(r"[\u0a80-\u0aff]", text): |
| found.append(_GU_ART_RE) |
| if re.search(r"[\u0a00-\u0a7f]", text): |
| found.append(_PA_ART_RE) |
| if re.search(r"[\u0d00-\u0d7f]", text): |
| found.append(_ML_ART_RE) |
| if re.search(r"[\u0e80-\u0eff]", text): |
| found.extend((_LO_ART_RE, _LO_PART_RE)) |
| if re.search(r"[\u1780-\u17ff]", text): |
| found.extend((_KM_ART_RE, _KM_CHAP_RE)) |
| if re.search(r"[\u1000-\u109f]", text): |
| found.extend((_MY_ART_RE, _MY_CHAP_RE)) |
| if re.search(r"[\u1200-\u137f]", text): |
| found.extend((_AM_ART_RE, _AM_CHAP_RE, _TI_ART_RE)) |
| if re.search(r"[\u0f00-\u0fff]", text): |
| found.append(_BO_ART_RE) |
| if re.search(r"[\u10a0-\u10ff]", text): |
| found.extend((_KA_ART_RE, _KA_AB_RE, _KA_CHAP_RE)) |
| if re.search(r"[\u0530-\u058f]", text): |
| found.extend((_HY_ART_RE, _HY_AB_RE, _HY_CHAP_RE)) |
| if re.search(r"[\u0400-\u04ff]", text): |
| found.extend((_CYR_AB_RE, _RU_TITLE_RE, _RU_CHAP_RE, _SR_PART_RE)) |
| if re.search(r"[\u0370-\u03ff]", text): |
| found.append(_EL_AB_RE) |
| if "модда" in text.casefold() or re.search(r"\bmodda\b", text, re.IGNORECASE): |
| found.append(_UZ_ART_RE) |
| if re.search(r"зүйл", text, re.IGNORECASE): |
| found.append(_MN_ART_RE) |
| if re.search(r"бап", text, re.IGNORECASE): |
| found.append(_KK_ART_RE) |
| if "दफा" in text: |
| found.append(_NE_ART_RE) |
| if "परिच्छेद" in text: |
| found.append(_NE_CHAP_RE) |
| if "වගන්තිය" in text: |
| found.append(_SI_ART_RE) |
| if "دفعہ" in text: |
| found.append(_UR_ART_RE) |
| if re.search(r"берене", text, re.IGNORECASE): |
| found.append(_KY_ART_RE) |
| if re.search(r"артыкул", text, re.IGNORECASE): |
| found.append(_BE_ART_RE) |
| if re.search(r"статья|стаття", text, re.IGNORECASE): |
| found.append(_RU_ART_RE) |
| if re.search(r"указ|постановление|приказ", text, re.IGNORECASE): |
| found.append(_RU_INST_RE) |
| |
| out: list = [] |
| seen: set[int] = set() |
| for regex in found: |
| key = id(regex) |
| if key not in seen: |
| seen.add(key) |
| out.append(regex) |
| return out |
|
|
|
|
| def _lexicon_heading_re() -> re.Pattern[str]: |
| """Heading matcher generated from HEADING_LEXICON so new languages are data.""" |
| global _LEXICON_HEADING_RE, _LEXICON_KIND |
| if _LEXICON_HEADING_RE is not None: |
| return _LEXICON_HEADING_RE |
| from .profiles import HEADING_LEXICON |
|
|
| words: list[str] = [] |
| kinds: dict[str, str] = {} |
| for token, hits in HEADING_LEXICON.items(): |
| token = (token or "").strip() |
| if len(token) < 2: |
| continue |
| kinds_found = [kind for kind, _lang in hits if kind in {"article", "section", "title", "chapter", "part"}] |
| if not kinds_found: |
| continue |
| folded = _latin_fold(token) |
| words.append(folded) |
| kinds[folded.casefold()] = kinds_found[0] |
| words.sort(key=len, reverse=True) |
| escaped = [re.escape(w) for w in words] |
| _LEXICON_HEADING_RE = re.compile( |
| _latin_fold( |
| r"(?im)" + _HEADING_START + r"(?:l['’])?(?P<lex>" |
| + "|".join(escaped) |
| + r")(?![A-Za-z])(?:\s*(?:n\.?[o°º”'“*´`]|nr|no)\.?\s*)?\s*[.\s:–—-]*\(?\s*(?P<lex_n>" |
| + _ARTICLE_N |
| + r")\)?" |
| ) |
| ) |
| _LEXICON_KIND = kinds |
| return _LEXICON_HEADING_RE |
|
|
|
|
| def _plausible_cw_number(raw: str) -> bool: |
| """Drop year-like '1902. Amendment' titles from keyword-less numbering.""" |
| digits = re.sub(r"\D", "", raw or "") |
| if not digits: |
| return False |
| value = int(digits) |
| if value <= 0 or value > 2000: |
| return False |
| if 1800 <= value <= 2099: |
| return False |
| return True |
|
|
|
|
| def _fold_digits_in_number(raw: str) -> str: |
| """Map Nd digits (١, १, ๑, …) to ASCII so retrieval numbers stay decimal.""" |
| if not raw: |
| return raw |
| out: list[str] = [] |
| changed = False |
| for ch in raw: |
| if unicodedata.category(ch) == "Nd": |
| digit = unicodedata.digit(ch) |
| if digit is not None: |
| mapped = str(digit) |
| out.append(mapped) |
| if mapped != ch: |
| changed = True |
| continue |
| out.append(ch) |
| return "".join(out) if changed else raw |
|
|
|
|
| def _unicode_digit_runs(text: str) -> list[str]: |
| """ASCII-fold Nd runs (Thai ๑, Myanmar ၁, Arabic-Indic, …).""" |
| runs: list[str] = [] |
| cur: list[str] = [] |
| for char in text: |
| if unicodedata.category(char) == "Nd": |
| cur.append(str(unicodedata.digit(char))) |
| elif cur: |
| runs.append("".join(cur)) |
| cur = [] |
| if cur: |
| runs.append("".join(cur)) |
| return runs |
|
|
|
|
| _AR_ORDINAL_NUM = ( |
| ("الحادية عشرة", "11"), |
| ("الثانية عشرة", "12"), |
| ("الثالثة عشرة", "13"), |
| ("الرابعة عشرة", "14"), |
| ("الخامسة عشرة", "15"), |
| ("السادسة عشرة", "16"), |
| ("السابعة عشرة", "17"), |
| ("الثامنة عشرة", "18"), |
| ("التاسعة عشرة", "19"), |
| ("الحادية والعشرون", "21"), |
| ("الثانية والعشرون", "22"), |
| ("الثالثة والعشرون", "23"), |
| ("الرابعة والعشرون", "24"), |
| ("الخامسة والعشرون", "25"), |
| ("السادسة والعشرون", "26"), |
| ("السابعة والعشرون", "27"), |
| ("الثامنة والعشرون", "28"), |
| ("التاسعة والعشرون", "29"), |
| ("العشرون", "20"), |
| ("الأولى", "1"), |
| ("الأولي", "1"), |
| ("األولى", "1"), |
| ("الاولى", "1"), |
| ("الثانية", "2"), |
| ("الثالثة", "3"), |
| ("الرابعة", "4"), |
| ("الخامسة", "5"), |
| ("السادسة", "6"), |
| ("السابعة", "7"), |
| ("الثامنة", "8"), |
| ("التاسعة", "9"), |
| ("العاشرة", "10"), |
| ("الأول", "1"), |
| ("الاول", "1"), |
| ("الثاني", "2"), |
| ("الثالث", "3"), |
| ("الرابع", "4"), |
| ("الخامس", "5"), |
| ("السادس", "6"), |
| ("السابع", "7"), |
| ("الثامن", "8"), |
| ("التاسع", "9"), |
| ("العاشر", "10"), |
| ) |
| _HE_VALUES = { |
| "א": 1, "ב": 2, "ג": 3, "ד": 4, "ה": 5, "ו": 6, "ז": 7, "ח": 8, "ט": 9, |
| "י": 10, "כ": 20, "ך": 20, "ל": 30, "מ": 40, "ם": 40, "נ": 50, "ן": 50, |
| "ס": 60, "ע": 70, "פ": 80, "ף": 80, "צ": 90, "ץ": 90, "ק": 100, |
| "ר": 200, "ש": 300, "ת": 400, |
| } |
| _HE_ORDINAL_NUM = ( |
| ("ראשונה", "1"), |
| ("שנייה", "2"), |
| ("שניה", "2"), |
| ("שלישית", "3"), |
| ("רביעית", "4"), |
| ("חמישית", "5"), |
| ("שישית", "6"), |
| ) |
| _ETH_VALUES = { |
| "፩": 1, "፪": 2, "፫": 3, "፬": 4, "፭": 5, "፮": 6, "፯": 7, "፰": 8, "፱": 9, |
| "፲": 10, "፳": 20, "፴": 30, "፵": 40, "፶": 50, "፷": 60, "፸": 70, "፹": 80, "፺": 90, |
| "፻": 100, "፼": 10000, |
| } |
| _FA_ORDINAL_NUM = ( |
| ("دوازدهم", "12"), |
| ("سیزدهم", "13"), |
| ("چهاردهم", "14"), |
| ("پانزدهم", "15"), |
| ("شانزدهم", "16"), |
| ("هفدهم", "17"), |
| ("هجدهم", "18"), |
| ("نوزدهم", "19"), |
| ("یازدهم", "11"), |
| ("بیستم", "20"), |
| ("چهارم", "4"), |
| ("چارم", "4"), |
| ("جارم", "4"), |
| ("پنجم", "5"), |
| ("ششم", "6"), |
| ("هفتم", "7"), |
| ("هشتم", "8"), |
| ("نهم", "9"), |
| ("دهم", "10"), |
| ("سوم", "3"), |
| ("دوم", "2"), |
| ("اولی", "1"), |
| ("اول", "1"), |
| ("نخست", "1"), |
| ) |
| _CJK_DIGIT = { |
| "零": 0, "〇": 0, "一": 1, "二": 2, "三": 3, "四": 4, |
| "五": 5, "六": 6, "七": 7, "八": 8, "九": 9, |
| } |
|
|
|
|
| def _cjk_numeral_value(text: str) -> str | None: |
| body = re.sub(r"[第条條項款章编編节節之조항호편절、]", "", text) |
| if not body or not all(ch in _CJK_DIGIT or ch == "十" or ch == "百" for ch in body): |
| return None |
| if body == "十": |
| return "10" |
| if body.startswith("十"): |
| return str(10 + _CJK_DIGIT.get(body[1], 0)) |
| if "百" in body: |
| left, right = body.split("百", 1) |
| if not left: |
| hundreds = 100 |
| elif left in _CJK_DIGIT: |
| hundreds = _CJK_DIGIT[left] * 100 |
| else: |
| return None |
| if not right: |
| return str(hundreds) |
| rest = _cjk_numeral_value(right) |
| if rest is None: |
| return None |
| return str(hundreds + int(rest)) |
| if "十" in body: |
| left, right = body.split("十", 1) |
| tens = _CJK_DIGIT.get(left, 0) * 10 |
| ones = _CJK_DIGIT.get(right, 0) if right else 0 |
| return str(tens + ones) |
| if body in _CJK_DIGIT: |
| return str(_CJK_DIGIT[body]) |
| return None |
|
|
|
|
| def _circled_to_int(char: str) -> str | None: |
| if not char: |
| return None |
| code = ord(char[0]) |
| if 0x2460 <= code <= 0x2473: |
| return str(code - 0x2460 + 1) |
| if 0x2474 <= code <= 0x2487: |
| return str(code - 0x2474 + 1) |
| if 0x2488 <= code <= 0x249B: |
| return str(code - 0x2488 + 1) |
| if 0x2776 <= code <= 0x277F: |
| return str(code - 0x2776 + 1) |
| if 0x24EB <= code <= 0x24F4: |
| return str(code - 0x24EB + 11) |
| return None |
|
|
|
|
| def _ethiopic_numeral_value(text: str) -> str | None: |
| chars = [ch for ch in text if ch in _ETH_VALUES] |
| if not chars: |
| return None |
| body = "".join(chars) |
| if "፼" in body: |
| left, right = body.split("፼", 1) |
| mul = _ethiopic_numeral_value(left) if left else "1" |
| rest = _ethiopic_numeral_value(right) if right else "0" |
| if mul is None or rest is None: |
| return None |
| return str(int(mul) * 10000 + int(rest)) |
| if "፻" in body: |
| left, right = body.split("፻", 1) |
| mul = sum(_ETH_VALUES[ch] for ch in left) if left else 1 |
| rest = sum(_ETH_VALUES[ch] for ch in right) |
| return str(mul * 100 + rest) |
| return str(sum(_ETH_VALUES[ch] for ch in chars)) |
|
|
|
|
| def _hebrew_letter_num(raw: str) -> str | None: |
| """Map פרק א' / תוספת שניה / סעיף יא to decimal. Long Hebrew words stay unmapped.""" |
| cleaned = re.sub(r"['׳״\"`]", "", raw) |
| for word, num in _HE_ORDINAL_NUM: |
| if word in cleaned: |
| return num |
| katan = re.search(r"סעיף\s+קטן\s*\(?\s*([א-ת])", cleaned) |
| if katan: |
| ch = katan.group(1) |
| return str(_HE_VALUES[ch]) if ch in _HE_VALUES else ch |
| match = re.search(r"(?:פרק|סימן|חלק|תוספת|סעיף)\s*(?!קטן)([א-ת]{1,3})", cleaned) |
| if not match: |
| return None |
| total = sum(_HE_VALUES.get(ch, 0) for ch in match.group(1)) |
| return str(total) if total else None |
|
|
|
|
| def _script_hit_kind(regex: re.Pattern[str]) -> str: |
| if regex is _RU_TITLE_RE or regex is _TH_LAK_RE: |
| return "title" |
| if regex in ( |
| _HE_PEREK_RE, _AR_BAB_RE, _TH_HUAT_RE, _CJK_ZHANG_RE, _KO_JANG_RE, |
| _RU_CHAP_RE, _KM_CHAP_RE, _MY_CHAP_RE, _KA_CHAP_RE, _HY_CHAP_RE, _AM_CHAP_RE, |
| _HI_CHAP_RE, _BN_CHAP_RE, _UR_CHAP_RE, _FA_FASL_RE, _NE_CHAP_RE, |
| ): |
| return "chapter" |
| if regex in (_HE_HELEK_RE, _AR_KITAB_RE, _CJK_BIAN_RE, _KO_PYEON_RE, _LO_PART_RE, _SR_PART_RE): |
| return "part" |
| if regex in ( |
| _HE_SIMAN_RE, |
| _HE_TOS_RE, |
| _HE_TAKANA_RE, |
| _HE_PARA_RE, |
| _HE_KATAN_RE, |
| _AR_ANNEX_RE, |
| _AR_PARA_RE, |
| _AR_FASL_RE, |
| _AR_FARA_RE, |
| _JA_KOU_RE, |
| _ZH_KUAN_RE, |
| _KO_HANG_RE, |
| _KO_HO_RE, |
| _TH_PARA_RE, |
| _TH_KHO_RE, |
| _TH_TON_RE, |
| _FA_TAB_RE, |
| _FA_BAND_RE, |
| _AR_POINT_RE, |
| _CJK_JIE_RE, |
| _KO_JEOL_RE, |
| ): |
| return "section" |
| return "article" |
|
|
|
|
| def _script_article_number(raw: str) -> str: |
| stripped = raw.strip() |
| for word, num in _AR_POINT_NUM: |
| if stripped.startswith(word): |
| return num |
| compact = raw.strip().strip("()[] \t\u3001\uff08\uff09.、-–—") |
| circled = _circled_to_int(compact) |
| if circled: |
| return circled |
| if compact and len(compact) <= 3: |
| first = compact[0] |
| if first in _HANGUL_GA: |
| return str(_HANGUL_GA.index(first) + 1) |
| if first in _KATA_ENUM: |
| return str(_KATA_ENUM.index(first) + 1) |
| if first in _AR_ABJAD_INDEX: |
| return _AR_ABJAD_INDEX[first] |
| if first in _HE_ALEF: |
| return str(_HE_ALEF.index(first) + 1) |
| if first in _TH_KO: |
| return str(_TH_KO.index(first) + 1) |
| low = first.casefold() |
| if low in _CYR_AB: |
| return str(_CYR_AB.index(low) + 1) |
| if low in _EL_AB: |
| return str(_EL_AB.index(low) + 1) |
| if first in _KA_AB: |
| return str(_KA_AB.index(first) + 1) |
| if first in _HY_AB: |
| return str(_HY_AB.index(first) + 1) |
| cjk_early = _cjk_numeral_value(compact) |
| if cjk_early: |
| return cjk_early |
| he = _hebrew_letter_num(raw) |
| if he: |
| return he |
| digits = _unicode_digit_runs(raw) |
| if digits: |
| if len(digits) >= 2 and re.search(r"[0-9٠-٩۰-۹]+-[0-9٠-٩۰-۹]+", raw): |
| return "-".join(digits[:2]) |
| return digits[0] |
| eth = _ethiopic_numeral_value(raw) |
| if eth: |
| return eth |
| compact = re.sub(r"[\s\u0640]+", "", raw) |
| for word, num in _AR_ORDINAL_NUM + _FA_ORDINAL_NUM: |
| needle = re.sub(r"[\s\u0640]+", "", word) |
| if needle and needle in compact: |
| return num |
| cjk = _cjk_numeral_value(compact) |
| if cjk: |
| return cjk |
| tail = re.search(r"([0-9٠-٩۰-۹]+|[IVXLCDM]{1,8})$", compact) |
| if tail: |
| return tail.group(1) |
| return compact |
|
|
|
|
| def _in_range_boilerplate(text: str, start: int) -> bool: |
| for regex in (_KO_RANGE_RE, _CJK_RANGE_RE): |
| for match in regex.finditer(text): |
| if match.start() <= start < match.end(): |
| return True |
| return False |
|
|
|
|
| def _kind_number_from_heading_match(match: re.Match) -> tuple[str, str]: |
| for name in ("title", "chapter", "part", "article", "section"): |
| if match.group(name): |
| return name, (match.group(f"{name}_n") or "").strip() |
| if match.group("section_sym"): |
| return "section", (match.group("section_sym_n") or "").strip() |
| if match.group("section_pre"): |
| return "section", (match.group("section_pre_n") or "").strip() |
| if match.group("grein"): |
| return "article", (match.group("grein_n") or "").strip() |
| if match.group("straipsnis"): |
| return "article", (match.group("straipsnis_n") or "").strip() |
| if match.group("artikulua"): |
| return "article", (match.group("artikulua_n") or "").strip() |
| if match.group("ledd"): |
| return "section", (match.group("ledd_n") or "").strip() |
| if match.group("kap"): |
| return "chapter", (match.group("kap_n") or "").strip() |
| if match.group("luku"): |
| return "chapter", (match.group("luku_n") or "").strip() |
| if match.group("fejezet"): |
| return "chapter", (match.group("fejezet_n") or "").strip() |
| if match.group("peatukk"): |
| return "chapter", (match.group("peatukk_n") or "").strip() |
| if match.group("skyrius"): |
| return "chapter", (match.group("skyrius_n") or "").strip() |
| if match.group("nodala"): |
| return "chapter", (match.group("nodala_n") or "").strip() |
| if match.group("kafli"): |
| return "chapter", (match.group("kafli_n") or "").strip() |
| if match.group("bob"): |
| return "chapter", (match.group("bob_n") or "").strip() |
| if match.group("tarau"): |
| return "chapter", (match.group("tarau_n") or "").strip() |
| if match.group("bolum"): |
| return "chapter", (match.group("bolum_n") or "").strip() |
| if match.group("modda"): |
| return "article", (match.group("modda_n") or "").strip() |
| if match.group("mn"): |
| return "article", (match.group("mn_n") or "").strip() |
| if match.group("kk"): |
| return "article", (match.group("kk_n") or "").strip() |
| if match.group("disp"): |
| return "article", (match.group("disp_n") or "").strip() |
| if match.group("mom"): |
| return "section", (match.group("mom_n") or "").strip() |
| if match.group("order"): |
| return "section", (match.group("order_n") or "").strip() |
| if match.group("kidogo"): |
| return "section", (match.group("kidogo_n") or "").strip() |
| if match.group("esresol"): |
| word = _latin_fold(match.group("esresol") or "").casefold() |
| return "section", { |
| "primero": "1", "segundo": "2", "tercero": "3", "cuarto": "4", |
| "quinto": "5", "sexto": "6", "septimo": "7", "octavo": "8", |
| "noveno": "9", "decimo": "10", |
| "erstens": "1", "zweitens": "2", "drittens": "3", "viertens": "4", "funftens": "5", |
| "premierement": "1", "deuxiemement": "2", "troisiemement": "3", |
| "quatriemement": "4", "cinquiemement": "5", |
| }.get(word, word) |
| return "", "" |
|
|
|
|
| @dataclass |
| class _Hit: |
| start: int |
| kind: str |
| number: str |
| heading: str |
|
|
|
|
| _PAREN_XREF_RE = re.compile( |
| r"(?i)^\s*(?:\([a-z0-9]{1,4}\))?\s+" |
| r"(?:is|are|or|and|of|have|has|was|were|by|to|for|from|in|that|shall)\b" |
| ) |
|
|
|
|
| def _paren_letter_xref(raw: str, number: str, rest: str) -> bool: |
| """Skip 'paragraph (i) or …' / 'subsection (l)(b) is …' cross-references.""" |
| if len(number) != 1 or not number.isalpha() or "(" not in raw: |
| return False |
| return _PAREN_XREF_RE.match(rest) is not None |
|
|
|
|
| def _collect_hits(text: str, *, language: str, latin: bool) -> list[_Hit]: |
| hits: list[_Hit] = [] |
| folded = _latin_fold(text) if latin else text |
| if latin: |
| for match in _HEADING_RE.finditer(folded): |
| kind, number = _kind_number_from_heading_match(match) |
| if not kind: |
| continue |
| if _CITE_TAIL_RE.match(folded[match.end() : match.end() + 48]): |
| continue |
| if _paren_letter_xref(match.group(0), number, folded[match.end() : match.end() + 48]): |
| continue |
| for gname in ( |
| "title_n", "chapter_n", "part_n", "article_n", "section_n", |
| "section_sym_n", "section_pre_n", "grein_n", "straipsnis_n", |
| "artikulua_n", "ledd_n", "kap_n", "luku_n", |
| "fejezet_n", "peatukk_n", "skyrius_n", "nodala_n", |
| "kafli_n", "bob_n", "tarau_n", "bolum_n", |
| "modda_n", "mn_n", "kk_n", "mom_n", "order_n", "kidogo_n", "disp_n", |
| ): |
| try: |
| start, end = match.span(gname) |
| except IndexError: |
| continue |
| if start >= 0: |
| number = text[start:end] |
| break |
| number = _fold_digits_in_number(number.strip().rstrip(".-:–—")) |
| line_end = text.find("\n", match.start()) |
| if line_end < 0: |
| line_end = len(text) |
| heading = re.sub(r"\s+", " ", text[match.start() : line_end]).strip()[:240] |
| hits.append(_Hit(match.start(), kind, number, heading)) |
| for match in _lexicon_heading_re().finditer(folded): |
| if _CITE_TAIL_RE.match(folded[match.end() : match.end() + 48]): |
| continue |
| token = (match.group("lex") or "").casefold() |
| kind = _LEXICON_KIND.get(token, "article") |
| start, end = match.span("lex_n") |
| number = _fold_digits_in_number( |
| (text[start:end] if start >= 0 else match.group("lex_n") or "").strip().rstrip(".-:–—") |
| ) |
| if not number: |
| continue |
| line_end = text.find("\n", match.start()) |
| if line_end < 0: |
| line_end = len(text) |
| heading = re.sub(r"\s+", " ", text[match.start() : line_end]).strip()[:240] |
| hits.append(_Hit(match.start(), kind, number, heading)) |
| for match in _LATIN_LETTER_RE.finditer(text): |
| letter = match.group("paren") or match.group("bare") |
| if letter: |
| number = str(ord(letter) - 96) |
| else: |
| number = _ROMAN_LOWER.get(match.group("roman") or "", "") |
| if not number: |
| continue |
| line_end = text.find("\n", match.start()) |
| if line_end < 0: |
| line_end = len(text) |
| heading = re.sub(r"\s+", " ", text[match.start() : line_end]).strip()[:240] |
| hits.append(_Hit(match.start(), "section", number, heading)) |
| for match in _LATIN_NUM_LIST_RE.finditer(text): |
| number = ( |
| match.group("nparen") |
| or match.group("nclose") |
| or match.group("nord") |
| or match.group("ndot") |
| or match.group("ncolon") |
| or match.group("ndash") |
| or "" |
| ) |
| dbl = match.group("dbl") |
| if dbl: |
| number = str(ord(dbl[0]) - 96) |
| if not number: |
| continue |
| line_end = text.find("\n", match.start()) |
| if line_end < 0: |
| line_end = len(text) |
| heading = re.sub(r"\s+", " ", text[match.start() : line_end]).strip()[:240] |
| hits.append(_Hit(match.start(), "section", number, heading)) |
| for i, match in enumerate(_BULLET_RE.finditer(text), start=1): |
| line_end = text.find("\n", match.start()) |
| if line_end < 0: |
| line_end = len(text) |
| heading = re.sub(r"\s+", " ", text[match.start() : line_end]).strip()[:240] |
| hits.append(_Hit(match.start(), "section", str(i), heading)) |
| for match in _DASH_NUM_RE.finditer(text): |
| number = (match.group("bn") or "").strip() |
| if not number: |
| continue |
| line_end = text.find("\n", match.start()) |
| if line_end < 0: |
| line_end = len(text) |
| heading = re.sub(r"\s+", " ", text[match.start() : line_end]).strip()[:240] |
| hits.append(_Hit(match.start(), "section", number, heading)) |
| for regex in _detected_script_regexes(text, language): |
| for match in regex.finditer(text): |
| if _in_range_boilerplate(text, match.start()): |
| continue |
| if _SCRIPT_CITE_TAIL_RE.match(text[match.end() : match.end() + 48]): |
| continue |
| raw = match.group(0) |
| number = _fold_digits_in_number(_script_article_number(raw)) |
| heading = re.sub(r"\s+", " ", raw).strip()[:240] |
| hits.append(_Hit(match.start(), _script_hit_kind(regex), number, heading)) |
| if latin: |
| from .profiles import iso_lang |
|
|
| lang = iso_lang(language) |
| keyword_hits = [h for h in hits if h.kind in {"article", "section"}] |
| if len(keyword_hits) < MIN_SPLIT_HEADINGS and lang in _CW_NUM_LANGS: |
| for match in _CW_NUM_RE.finditer(text): |
| number = (match.group("cw_n") or "").strip() |
| if not _plausible_cw_number(number): |
| continue |
| if _MONTH_START_RE.match(text[match.end() : match.end() + 16]): |
| continue |
| hits.append( |
| _Hit( |
| match.start(), |
| "section", |
| number, |
| match.group(0).strip()[:240], |
| ) |
| ) |
| hits.sort(key=lambda h: (h.start, -len(h.heading))) |
| merged: list[_Hit] = [] |
| seen_starts: set[int] = set() |
| for hit in hits: |
| if hit.start in seen_starts: |
| continue |
| seen_starts.add(hit.start) |
| merged.append(hit) |
| return merged |
|
|
|
|
| def _units_from_hits(text: str, hits: list[_Hit]) -> list[StructureUnit]: |
| if not hits: |
| return [ |
| StructureUnit(kind="preamble", number="", heading="", body=text, char_start=0, char_end=len(text)) |
| ] |
| cursor = {"title": "", "chapter": "", "part": "", "article": "", "section": ""} |
| units: list[StructureUnit] = [] |
| if hits[0].start > 0: |
| units.append( |
| StructureUnit( |
| kind="preamble", |
| number="", |
| heading="", |
| body=text[0 : hits[0].start], |
| char_start=0, |
| char_end=hits[0].start, |
| ) |
| ) |
| for i, hit in enumerate(hits): |
| start = hit.start |
| end = hits[i + 1].start if i + 1 < len(hits) else len(text) |
| body = text[start:end] |
| cursor[hit.kind] = hit.number |
| for lower, rank in _KIND_RANK.items(): |
| if rank > _KIND_RANK.get(hit.kind, 99): |
| cursor[lower] = "" |
| units.append( |
| StructureUnit( |
| kind=hit.kind, |
| number=hit.number, |
| heading=hit.heading[:240], |
| body=body, |
| title_number=cursor["title"], |
| chapter_number=cursor["chapter"], |
| part_number=cursor["part"], |
| article_number=cursor["article"], |
| section_number=cursor["section"], |
| subsections=_subsection_tokens(body), |
| hierarchy_path=_cursor_path(cursor), |
| char_start=start, |
| char_end=end, |
| ) |
| ) |
| return units |
|
|
|
|
| def segment_exclusive(text: str, *, language: str = "") -> list[StructureUnit]: |
| """Internal exclusive-span segmentation of *text* (includes preamble).""" |
| from .profiles import iso_lang, latin_split_allowed |
|
|
| if not text: |
| return [] |
| latin = not language or latin_split_allowed(language) |
| hits = _collect_hits(text, language=language, latin=latin) |
| retrieval_like = [h for h in hits if h.kind in {"article", "section"}] |
| if ( |
| not latin |
| and len(retrieval_like) < MIN_SPLIT_HEADINGS |
| and iso_lang(language) in _LATIN_FALLBACK_LANGS |
| ): |
| hits = _collect_hits(text, language=language, latin=True) |
| return _units_from_hits(text, hits) |
|
|
|
|
| def split_script_units(text: str, language: str) -> list[StructureUnit]: |
| """Retrieval projection of script-specific exclusive spans.""" |
| internal = segment_exclusive(text, language=language) |
| retrieval = [u for u in internal if u.kind == "article" and len(u.body.strip()) >= 16] |
| return retrieval if len(retrieval) >= MIN_SPLIT_HEADINGS else [] |
|
|
|
|
| def split_structured_units(text: str, *, language: str = "") -> list[StructureUnit]: |
| """Retrieval units. Empty list means keep the whole instrument. |
| |
| Prefers article/section. When those are scarce, numbered chapter then title |
| headings (Cameroon CHAPITRE, Bolivia TÍTULO, Cape Verde CAPÍTULO). |
| """ |
| if not text or len(text) < 16: |
| return [] |
| from .profiles import latin_split_allowed |
|
|
| internal = segment_exclusive(text, language=language) |
|
|
| def _take(kinds: set[str]) -> list[StructureUnit]: |
| out: list[StructureUnit] = [] |
| for u in internal: |
| if u.kind not in kinds: |
| continue |
| min_len = 16 if u.number else MIN_UNIT_CHARS |
| if language and not latin_split_allowed(language): |
| min_len = 16 |
| if len(u.body.strip()) >= min_len: |
| out.append(u) |
| return out |
|
|
| for kinds in ( |
| {"article", "section"}, |
| {"chapter"}, |
| {"part"}, |
| {"title"}, |
| ): |
| retrieval = _take(kinds) |
| if len(retrieval) >= MIN_SPLIT_HEADINGS: |
| return retrieval |
| return [] |
|
|
