# -*- coding: utf-8 -*- from ast import Str from asyncio import constants from . import GeneralMap as GM import re import string import unicodedata from . import PostProcess from . import ConvertFix as CF from aksharamukha.ScriptMap.EastIndic import PhagsPa, Burmese, Khmer from aksharamukha.ScriptMap.MainIndic import Tamil, Malayalam, Limbu, Chakma ### Use escape char in all functions def ItransLL(Strng): return Strng def RomanLoCChandrabindu(Strng): Strng = re.sub('n̐', 'm̐', Strng) return Strng def BalineseRomanLoCFix(Strng): Strng = Strng.replace('‘', 'ng̈') return Strng def JavaneseRomanLoCFix(Strng): Strng = Strng.replace('‘', 'ng̈') return Strng def HindiMarathiRomanLoCFix(Strng): Strng = Strng.replace('sh', 'ṣ') Strng = Strng.replace('ḷ', 'ḻ') Strng = Strng.replace('l̤', 'l̳') return Strng def RomanLoCLaUnderscoreDoubleDot(Strng): Strng = Strng.replace('ḻ', 'l̤') # reverse these in preprocess def RomanLoCVaWa(Strng): Strng = Strng.replace('w', 'v') return Strng def RomanLoCSasha(Strng): Strng = Strng.replace('sh', 'ṣ') return Strng def RomanLoCSLaDotLaUnderscore(Strng): Strng = Strng.replace('ḻ', 'l̤') Strng = Strng.replace( 'l̳', 'ḻ',) return Strng def RomanLoCLaUnderscoreDoubleDot(Strng): Strng = Strng.replace('l̤', 'ḻ') return Strng def MalayalamRomanLoCFix(Strng): Strng = Strng.replace('ṯṯ', 'ṟṟ') Strng = Strng.replace('ȧ', 'ŭ') return Strng def DevanagariRomanLoCFix(Strng): Strng = Strng.replace('g̳h̳', 'gḧ').replace('t̤', 'ṭ̈').replace('s̤', 's̈')\ .replace('h̤', 'ḧ') return Strng def IASTLDotRetroflex(Strng): Strng = Strng.replace('ḷ', 'l̤') return Strng def ArchaicJnaSimplifyRomanLOC(Strng): Strng = Strng.replace('ꦘ', 'ꦚ') return Strng def KhandaTaRomanLoC(Strng): Strng = Strng.replace('ৎ', 'ṯ') return Strng def TibetanLoCRomanLoCFix(Strng): Strng = re.sub('t(?![ʹhs])', 'tʹ', Strng) Strng = re.sub('n(?![ʹy])', 'nʹ', Strng) return Strng def BalineseJavaneseMoveRepha(Strng, tgt, reph): repha = '(' + reph + ')' cons = '(' + '|'.join(GM.CrunchSymbols(GM.Consonants, tgt)) + ')' vows = '(' + '|'.join(GM.CrunchSymbols(GM.VowelSignsNV, tgt)) + ')' vir = GM.CrunchSymbols(GM.virama, tgt)[0] candAnu = '[' + ''.join(GM.CrunchSymbols(GM.CombiningSigns, tgt)[:2]) + ']' Strng = re.sub('(' + candAnu + ')' + repha, r'\2\1', Strng) Strng = re.sub('('+cons+')'+'('+ vir + cons +')*'+'(' + vows + ')?'+ repha, r'\7\1\3\5', Strng) return Strng def DivesAkuruAlternateIndVowels(Strng): # use alt /y/ #Strng = Strng.replace("\U00011925", "\U00011926") # replace ind. vow with /y/ vow = "𑤁 𑤂 𑤃 𑤄 𑤅 𑤆 𑤆𑤵 𑤉 𑤀".split(" ") vowy = "𑤥𑤰 𑤥𑤱 𑤥𑤲 𑤥𑤳 𑤥𑤴 𑤥𑤵 𑤥𑤷 𑤥𑤸 𑤥".split(" ") for v, vy in zip(vow, vowy): Strng = Strng.replace(vy, v) return Strng def KawiMoveRepha(Strng): tgt = 'Kawi' repha = '\U00011F02' cons = '(' + '|'.join(GM.CrunchSymbols(GM.Consonants + GM.Vowels, tgt)) + ')' vows = '(' + '|'.join(GM.CrunchSymbols(GM.VowelSignsNV, tgt)) + ')' vir = '\U00011F42' Strng = re.sub(repha + '(('+cons+')' + '('+ vir + cons +')*'+'(' + vows + ')?)', r'\1' + repha, Strng) return Strng def JavaneseMoveRepha(Strng): return BalineseJavaneseMoveRepha(Strng, 'Javanese', 'ꦂ') def BalineseMoveRepha(Strng): return BalineseJavaneseMoveRepha(Strng, 'Balinese', 'ᬃ') def OriyaSubojinedVa(Strng): Strng = re.sub('(?? modifier letter comma consonants = '|'.join(GM.CrunchSymbols(GM.Consonants, 'IAST')) Strng = re.sub('(' + consonants + ')(‘)', r'\1' + 'ʻ', Strng) Strng = Strng.replace('o‘', 'oʻ') # reverse o' to au Strng = Strng.replace('oʻ', 'au') return Strng def paliTham(Strng): return Strng def segmentThamSyllabes(Strng): # segment text into syllables # https://github.com/ye-kyaw-thu/myWord/blob/main/syl_segment.py myConsonant = r"ᨠ-ᩌ" otherChar = r"ᩍ-ᩔ!-/:-@[-`{-~\s" ssSymbol = r'᩠' aThat = r'᩺' BreakPattern = re.compile(r"((? ui Strng = Strng.replace('\u102D\u102F', '\u102F\u102D') # mark Independent au -> o' Strng = Strng.replace('ဪ','ဩʻ') ## sort subjoined consonants yrvh = Burmese.ConsonantMap[25:27] + Burmese.ConsonantMap[28:29] + Burmese.ConsonantMap[32:33] yrvhsub = ['\u103B','\u103C','\u103D','\u103E'] vir = Burmese.ViramaMap[0] for x,y in zip(yrvh,yrvhsub): # Undo Replace subjoining forms: exp-virama + y/r/v/h <- subjoining y/r/v/h Strng = Strng.replace(y, vir + vir + x) # swap tone + virama to proper order Strng = Strng.replace("့်", "့်") # mark pure viramas aThat = r'်' Strng = Strng.replace(aThat, aThat + 'ʻ') # mark a as glottalstop Strng = Strng.replace('အ','’အ') # replace vowel + diac with vowels vowDep = 'အော် အော အိ အီ အု အူ အေ'.split(' ') vowIndep = 'ဪ ဩ ဣ ဤ ဥ ဦ ဧ'.split(' ') for x, y in zip(vowDep, vowIndep): Strng = Strng.replace(x, y) # danda to comma and double danda to full stop Strng = Strng.replace('၊', ',').replace('။', '.') #print(Strng) return Strng def insertViramaSyriac(Strng): Strng += "\uF001" return Strng def BengaliSwitchYaYYa(Strng): Strng = re.sub('(? Aleph approximation AyinAlephInitial = [('ʾa', 'ʾ'),('ʾā', 'ā̂'), ('ʾi', 'î'), ('ʾī', 'ī̂'), ('ʾu', 'û'), ('ʾū', 'ū̂'), ('ʾe', 'ê'), ('ʾē', 'ē̂'),\ ('ʾo', 'ô'), ('ʾō', 'ō̂')] for x, y in AyinAlephInitial: Strng = Strng.replace(x, y) ## Also make changes in PostOptionsLatn if Source == 'Arab': Strng = PostProcess.arabizeLatn(Strng, target="indic") #print('after2', Strng) elif Source == 'Arab-Ur' or Source == 'Arab-Pa' or Source == 'Arab-Fa': Strng = PostProcess.urduizeLatn(Strng, target="indic") elif Source == 'Syrn': Strng = PostProcess.syricizeLatn(Strng, target="indic") elif Source == 'Syrj' or Source == 'Hebr': Strng = PostProcess.hebraizeLatn(Strng, target="indic") # remove sin/shin dot Strng = Strng.replace('\u032A', '').replace('\u032E', '') # duplicate consonants ## move do chashmee ha + gemination sign ## tʰ꞉ -> t꞉ʰ Strng = re.sub('([ʰ])(꞉)', r'\2\1', Strng) Strng = re.sub('([aiuāīū])(꞉)', r'\2\1', Strng) Strng = re.sub('(.)(꞉)', r'\1' + vir + r'\1', Strng) # avoid fusion of Ayn Strng = Strng.replace("ʿ" + vir, "ʿ" + vir + "\u200B") cons_prev = '|'.join(GM.SemiticConsonants) if 'Syr' in Source: ## converting from Syriac through Latin consSyrc = "|".join(["ʾ", "b", "v", "g", "ġ", "d", "ḏ", "h", "w", "z", "ḥ", "ṭ", "y", "k", "ḫ", "l", "m", "n", "s", "ʿ", "p", "f", "ṣ", "q", "r", "š", "t", "ṯ", "č", "ž", "j"]) vowelSyrc = ["a", "ā", "e", "ē", "ū", "ō", "ī", "â", "ā̂", "ê", "ē̂"] vowelsDepSyrc = "|".join(["a", "ā", "e", "ē", "u", "i", "o"]) vowelsInDepSyrc1 = ["i", "u", "o"] vowelsInDepSyrc2 = ["ī̂", "û", "ô"] if any([vow in Strng for vow in vowelSyrc]): #print(Strng) #print([(vow, vow in Strng) for vow in vowelSyrc]) Strng = Strng.replace('ī', 'i').replace('ū', 'u').replace('ō', 'o') for vow1, vow2 in zip(vowelsInDepSyrc1, vowelsInDepSyrc2): Strng = re.sub('(? i^ if "\uF001" in Strng: Strng = re.sub('(' + consSyrc + ')' + '(?!' + vowelsDepSyrc + ')', r'\1' + vir, Strng) # print(Strng) Strng = re.sub('(?<=' + cons_prev + ')' + 'a(?!\u0304)', '', Strng) Strng = Strng.replace("\uF001", "") ## Add scripts with vowels here if "Arab" in Source or Source == 'Latn' or Source == "Hebr" or Source == "Thaa" or Source == 'Type': basic_vowels = '(' + '|'.join(['a', 'ā', 'i', 'ī', 'u', 'ū', 'ē', 'ō', 'e', 'o', '#', vir]) + ')' Strng = re.sub('(ŵ)(?=' + basic_vowels + ')', "w", Strng) Strng = re.sub('(ŷ)(?=' + basic_vowels + ')', "y", Strng) Strng = re.sub('(?<=' + cons_prev + ')' + 'a(?!(ŵ|ŷ|\u0304|\u032E))', '', Strng) Strng = re.sub('(?<=ḧ)' + 'a(?!(ŵ|ŷ|\u0304|\u032E))', '', Strng) ## Hamza with vowels if 'Arab' in Source: simp_vow = 'a ā i ī u ū'.split(' ') init_vow = 'â ā̂ î ī̂ û ū̂'.split(' ') # Hamsa + vow -> indep vow for x, y in zip(simp_vow, init_vow): Strng = re.sub('ʔ' + x, y, Strng) # remove leftover hamza if nativizing if "\u05CC" in Strng: Strng = Strng.replace("ʔ", "") ## Semitic to Indic equivalents SemiticIndic=[('ṣ', 'sQ'), ('ʿ', 'ʾQ'), ('ṭ', 'tQ'), ('ḥ', 'hQ'), \ ('ḍ', 'dQ'), ('p̣', 'pQ'), ('ž', 'šQ'), ('ẓ', 'jʰQ'), ('ḏ', 'dʰQ'), ('ṯ', 'tʰQ'),\ ('w', 'vQ'), ('ḵ', 'k'), ('\u032A', ''), ('\u032E', ''),('a̮', "ā"),('\u0308', ""),\ ('ĕ\u0302', 'ê'), ('ă\u0302', 'â'), ('ŏ\u0302','ô'), ('ĕ', 'e'), ('ă', ''), ('ŏ','o'), ('ḵ', 'k'),\ ('ʾQā', 'ā̂Q'), ('ʾQi', 'îQ'), ('ʾQī', 'ī̂Q'), ('ʾQu', 'ûQ'), ('ʾQū', 'ū̂Q'), ('ʾQe', 'êQ'), ('ʾQē', 'ē̂Q'),\ ('ʾQo', 'ôQ'), ('ʾQō', 'ō̂Q'), ('ⁿ', 'n\u033D'), ('ʾā', 'ā̂')] for s, i in SemiticIndic: Strng = Strng.replace(s, i) if 'Arab' in Source: Strng = re.sub('(\u033D)([iuā])', r'\2', Strng) Strng = re.sub('(\u033D)([a])', '', Strng) ## Lone Aliph to Nukta Strng = Strng.replace('ʾ', 'â') return Strng def perisanizeArab(Strng): arabKafYe = 'ك ي'.split(' ') persKafYe = 'ک ی'.split(' ') for x, y in zip(arabKafYe, persKafYe): Strng = Strng.replace(x, y) return Strng def ArabizePersian(Strng): ## how to deal with ga arabKafYe = 'ك ي'.split(' ') persKafYe = 'ک ی'.split(' ') for x, y in zip(arabKafYe, persKafYe): Strng = Strng.replace(y, x) return Strng def semiticizeUrdu(Strng): urduSpecific = 'ے ڈ ٹ ہ'.split(' ') semitic = 'ي د ت ه'.split(' ') for x, y in zip(urduSpecific, semitic): Strng = Strng.replace(x, y) ## remove Do chashme he Strng = Strng.replace('ھ', '') return Strng def ShowChillus(Strng): return PostProcess.MalayalamChillu(Strng, True, True) def ShowKhandaTa(Strng): Strng = Strng.replace('ৎ', 'ত্ˍ') return Strng def eiaudipthongs(Strng): return Strng def wasvnukta(Strng): return Strng def default(Strng): return Strng def SogdReshAyin(Strng): Strng = Strng.replace('𐽀', '[\uEA01-\uEA02]') # resh yin return Strng def SogoReshAyinDaleth(Strng): Strng = Strng.replace('𐼘', '[\uEA01-\uEA02-\uEA06]') # resh yin daleth return Strng def PhlpMemQoph(Strng): Strng = Strng.replace('𐮋', '[\uEA03-\uEA04]') # mem qoph return Strng def PhlpWawAyinResh(Strng): Strng = Strng.replace('𐮅', '[\uEA05-\uEA02-\uEA01]') # resh yin return Strng def PhliWawAyinResh(Strng): Strng = Strng.replace('𐭥', '[\uEA05-\uEA02-\uEA01]') # resh yin return Strng def HatrDalethResh(Strng): Strng = Strng.replace('𐣣', '[\uEA06-\uEA01]') # resh yin return Strng def MalayalamHalfu(Strng): consu = '[കചടതറനണരലവഴളറ]' vir = GM.CrunchSymbols(GM.VowelSigns, 'Malayalam')[0] consAll = "(" + '|'.join(GM.CrunchSymbols(GM.Consonants, 'Malayalam')) + ")" Strng = re.sub('(?<=' + consu + ')' + '(' + vir + ')' + '(?!' + consAll + ')', r'\2' + 'ു്', Strng) return Strng def MalayalamTranscribe(Strng): Strng = MalayalamHalfu(Strng) script = "Malayalam" ListC = GM.CrunchList('ConsonantMap',script) ListSC = GM.CrunchList('SouthConsonantMap',script) vir = GM.CrunchSymbols(GM.VowelSigns, script)[0] ConUnVoiced = [ListC[x] for x in [0,5,10,15,20]] ConVoicedJ = [ListC[x] for x in [2,7,12,17,22]] ConVoicedS = [ListC[x] for x in [2,5,12,17,22]] ConNasalsAll = '|'.join([ListC[x] for x in [4, 9, 14, 19, 24]]) conNasalCa = '|'.join([ListC[x] for x in [9]]) ConNasalsGroup = [ConNasalsAll, conNasalCa, ConNasalsAll, ConNasalsAll, ConNasalsAll] ConMedials = '|'.join(ListC[25:28]+ListSC[0:2]+ListSC[3:4]) Vowels = '|'.join(GM.CrunchSymbols(GM.Vowels+GM.VowelSignsNV, script)) Aytham = GM.CrunchList('Aytham',script)[0] Consonants = '|'.join(GM.CrunchSymbols(GM.Consonants,script)) NRA =ListSC[3] + vir + ListSC[2] NDRA = ListC[14] + vir + ListC[12] + vir + ListC[26] ### Check Siva Siva Mails ### Do something about Eyelash ra in Transliterated text for i in range(len(ConUnVoiced)): pass #Strng = re.sub('('+Vowels+Consonants+')'+ConUnVoiced[i]+'('+Vowels+Consonants+')',r'\1'+ConVoicedS[i]+r'\2',Strng) Strng = re.sub('('+Vowels+'|'+Consonants+'|'+Aytham+')'+ConUnVoiced[i]+'(?!'+vir+')',r'\1'+ConVoicedS[i],Strng) Strng = re.sub('('+ConVoicedS[i]+')'+ConUnVoiced[i]+'(?!'+vir+')',r'\1'+ConVoicedS[i],Strng) # Nasals + Unvoiced -> voiced # Rule applied even if spaced -> ulakaJ cuRRu -> ulakaJ juRRu; cenJu -> senju # Strng = re.sub('('+ConNasalsSpace+')'+'('+vir+')'+'( ?)'+ConUnVoiced[i],r'\1\2\3'+ConVoicedJ[i],Strng) # Only without space : vampu -> vambu; varim panam -> varim panam Strng = re.sub('('+ConNasalsGroup[i]+')'+'('+vir+')'+ConUnVoiced[i],r'\1\2'+ConVoicedJ[i],Strng) # Medials + Unvoiced -> Voiced Strng = re.sub('('+ConMedials+')'+'('+vir+')'+ConUnVoiced[i]+'(?!'+vir+')',r'\1\2'+ConVoicedS[i],Strng) #RRA NRA Strng = Strng.replace('റ്റ', 'ട്ട').replace('ന്റ', 'ണ്ഡ') #anusvara with m Strng = Strng.replace('ം', 'മ്') return Strng def retainLatin(Strng, reverse=False): latn_basic_lower = 'a b c d e f g h i j k l m n o p q r s t u v w x y z ḥ ṭ ṣ ʾ ʿ š ā ī ū ē ō' latn_basic_upper = latn_basic_lower.upper() latn_all = latn_basic_lower + latn_basic_upper latn_all = latn_all.split(' ') if not reverse: for i, c in enumerate(latn_all): Strng = Strng.replace(c, chr(60929+i)) # for disambiguating preprocessed consonants Strng = Strng.replace('\uEA01', 'r').replace('\uEA02', 'ʿ').replace('\uEA03', 'm').replace('\uEA04', 'q').replace('\uEA05', 'w').replace('\uEA06', 'd') else: # print('I am trying to reverse Lating') for i, c in enumerate(latn_all): Strng = Strng.replace(chr(60929+i), c) return Strng def JapanesePreProcess(src, txt, preoptions): import pykakasi from . import Convert if src == 'Hiragana' or src == 'Katakana': # preserve roman words kks = pykakasi.kakasi() txt = Convert.convertScript(txt.lower(), "ISO", "Devanagari") #print(txt) cv = kks.convert(txt) txt = '' for item in cv: txt = txt + ' ' + item['hepburn'] if 'eiaudipthongs' in preoptions: txt = txt.replace('ou', 'o\u02BDu').replace('ei', 'e\u02BDi') txt = re.sub('(r)([aiueo])(\u309A\u309A)', 'l' + r'\2\2', txt) txt = re.sub('(r)([aāiīuūeēoō])(\u309A)', 'l' + r'\2', txt) txt = re.sub('(k)([aiueo])(\u309A\u309A)', 'ṅ' + r'\2\2', txt) txt = re.sub('(k)([aāiīuūeēoō])(\u309A)', 'ṅ' + r'\2', txt) txt = txt.replace('aa', 'ā').replace('ii', 'ī').replace('ee', 'ē').replace('oo', 'ō').replace('uu','ū') txt = txt.replace('a-', 'ā').replace('i-', 'ī').replace('e-', 'ē').replace('o-', 'ō').replace('u-','ū') txt = txt.replace("n'", 'n_').replace('ch', 'c').replace('sh', 'ṣ').replace('sṣ', 'ṣṣ').replace('ai', 'a_i').replace('au', 'a_u') txt = txt.replace('w', 'v') txt = txt.replace('ou', 'ō').replace('ei', 'ē') txt = txt.replace('、', ',').replace('。', '.') # Nasalization of n txt= txt.replace('ng', 'ṅg').replace('nk', 'ṅk').replace('nk', 'ṅk').replace('np', 'mp').replace('nb', 'mb').replace('nm', 'mm') if 'wasvnukta' in preoptions: txt = txt.replace('v', 'v̈') #print(txt) txt = txt.replace('、', ',').replace('。', '.') return txt def holamlong(Strng): Strng = Strng.replace('ֹּ','ֹּ') Strng = re.sub('(? bhArtIy Strng = re.sub(ISyl+Cons+Syl+SylAny+"(?!" + Char + ")", r'\1\8' + vir + r'\9\15', Strng) # bhAratIya --> bhArtIy Strng = re.sub(ISyl+Cons+Syl+"(?!" + Char + ")", r'\1\8' + vir + r'\9', Strng) # namakIn -> namkIn Strng = re.sub(ISyl+Cons+"(?!" + Char + ")", r'\1\8' + vir, Strng) # kama -> kam Cons_sss = '((' + Cons + vir + ')' +'([शषस]))' # delete if double ends with s, sh, s Strng = re.sub(ISyl+Cons_sss+"(?!" + Char + ")", r'\1\8' + vir, Strng) ## remove from geminated consonants Target = 'Devanagari' ConUnAsp = [GM.CrunchList('ConsonantMap', Target)[x] for x in [0,2,5,7,10,12,15,17,20,22,4,9,14,19,24,25,26,27,28,29,30,31,32]] ConUnAsp = ConUnAsp + GM.CrunchList('SouthConsonantMap',Target) + GM.CrunchList('NuktaConsonantMap',Target) ConAsp = [GM.CrunchList('ConsonantMap', Target)[x] for x in [1,3,6,8,11,13,16,18,21,23]] Strng = re.sub(ISyl + '('+'|'.join(ConUnAsp)+')'+'('+vir+')('+r'\8'+")(?!" + Char + ")", r'\1\8\9\10' + vir,Strng) for i in range(len(ConAsp)): Strng = re.sub(ISyl +'('+ConUnAsp[i]+')'+'('+vir+')'+'('+ConAsp[i]+')'+'(?!" + Char + ")', r'\1\8\9\10' + vir,Strng) #Strng = re.sub(VowI + Nas + Cons2+"(?!" + Char + ")", r'\1\2\3' + vir, Strng) cons_pyramid = ['[यरलव]', '[नमण]', '[शषस]', '[कखपफगघबभ]', '[टठतथडढदध]', '[चछजझज़]'] for c1, cons1 in enumerate(cons_pyramid): for c2, cons2 in enumerate(cons_pyramid): if c1 < c2: # delete if ascending order Cons_pyr = '((' + cons1 + vir + ')' +'('+cons2+'))' Strng = re.sub(ISyl+Cons_pyr+"(?!" + Char + ")", r'\1\8' + vir, Strng) Strng = Strng.replace(vir, vir2) return Strng # consider adding an optional NUkta to the post consonantal position def RemoveFinal(Strng, Target): if Target == 'Bengali': Strng = PostProcess.KhandaTa(Strng, Target, True) VowI = "(" + '|'.join(GM.CrunchSymbols(GM.Vowels,Target)) + ")" VowS = "(" + '|'.join(GM.CrunchSymbols(GM.VowelSignsNV, Target)) + ")" Cons = "(" + '|'.join(GM.CrunchSymbols(GM.Consonants, Target)) + ")" Char = "(" + '|'.join(GM.CrunchSymbols(GM.Characters, Target)) + ")" Nas = "([" + '|'.join(GM.CrunchList('AyogavahaMap',Target)) + "]?)" ISyl = "((" + VowI + "|" + "(" + Cons + VowS + "?" + ")" + Nas + '))' Syl = "((" + Cons + VowS + ')' + Nas + ")" SylAny = "((" + Cons + VowS + "?" + ')' + Nas + ")" vir = GM.CrunchList("ViramaMap", Target)[0] if Target != 'Bengali': Cons2 = '((' + Cons + vir + ')?' + Cons + ')' else: Cons2 = '(()?' + Cons + ')' if Target == 'Khmer' or Target == 'KhmerLoC': ListC ='|'.join(GM.CrunchSymbols(sorted(GM.Consonants, key=len, reverse=True),Target)) ra = Khmer.ConsonantMap[26] Strng = re.sub('('+ListC+')'+'\u17CC',ra+'\u17D2'+r'\1',Strng) Strng = re.sub(ISyl + '('+ListC +')' + '(((\u17D2)' + '('+ListC +'))*)([៍៎៏]?)(?=[\s\n])', r'\1\8\9' + vir + r'\13', Strng) # kama -> kam Strng = re.sub(ISyl + '('+ListC +')' + '(((\u17D2)' + '('+ListC +'))*)([៍៎៏]?)$', r'\1\8\9' + vir + r'\13', Strng) # kama -> kam else: Strng = re.sub(ISyl + Cons2+"(?!" + Char + ")", r'\1\8' + vir, Strng) # kama -> kam Strng = re.sub(ISyl + Cons2+"(?!" + Char + ")", r'\1\8' + vir, Strng) # kama -> kam #Strng = re.sub(VowI + Nas + Cons2+"(?!" + Char + ")", r'\1\2\3' + vir, Strng) return Strng def SchwaFinalGurmukhi(Strng): Strng = RemoveFinal(Strng, 'Gurmukhi') return Strng def SchwaFinalGujarati(Strng): Strng = RemoveFinal(Strng, 'Gujarati') return Strng def SchwaFinalBengali(Strng): Strng = RemoveFinal(Strng, 'Bengali') return Strng def SchwaFinalKhmerLoC(Strng): Strng = RemoveFinal(Strng, 'KhmerLoC') return Strng def SchwaFinalKhmer(Strng): Strng = RemoveFinal(Strng, 'Khmer') return Strng def SchwaFinalWarangCiti(Strng): Target = "WarangCiti" VowI = "(" + '|'.join(GM.CrunchSymbols(GM.Vowels,Target)) + ")" VowS = "(" + '|'.join(GM.CrunchSymbols(GM.VowelSignsNV, Target)) + ")" Cons = "(" + '|'.join(GM.CrunchSymbols(GM.Consonants, Target)) + ")" Char = "(" + '|'.join(GM.CrunchSymbols(GM.Characters, Target)) + ")" Nas = "([" + '|'.join(GM.CrunchList('AyogavahaMap',Target)) + "]?)" ISyl = "((" + VowI + "|" + "(" + Cons + VowS + "?" + ")" + Nas + '))' Syl = "((" + Cons + VowS + ')' + Nas + ")" SylAny = "((" + Cons + VowS + "?" + ')' + Nas + ")" vir = '\u02BB' Cons2 = '((' + Cons + vir + ')?' + Cons + ')' Strng = re.sub(ISyl+Cons2+"(?!" + Char + ")", r'\1\8' + vir, Strng) # kama -> kam return Strng def siddhamUnicode(Strng): return Strng def ThaiOrthography(Strng): Strng += "\u02BB\u02BB" return Strng def LaoTranscription(Strng): Strng += "\u02BB\u02BB" return Strng def LimbuDevanagariConvention(Strng): Strng = Strng.replace('ए़', 'ऎ') Strng = Strng.replace('ओ़', 'ऒ') Strng = Strng.replace('े़', 'ॆ') Strng = Strng.replace('ो़', 'ॊ') Strng = Strng.replace('ः', '꞉') return Strng def LimbuSpellingSaI(Strng): ## Fix this not for all vir = Limbu.ViramaMap[0] FCons = [x+vir for x in [Limbu.ConsonantMap[x] for x in[0,4,15,19,20,24,26,27]]] FinalCons = ['\u1930','\u1931','\u1933','\u1934','\u1935','\u1936','\u1937','\u1938'] for x, y in zip(FCons, FinalCons): Strng = Strng.replace(x, '\u193A' + y) return Strng def removeChillus(Strng): Chillus=['\u0D7A','\u0D7B','\u0D7C','\u0D7D','\u0D7E'] vir = Malayalam.ViramaMap[0] ConVir =[ Malayalam.ConsonantMap[14]+vir, Malayalam.ConsonantMap[19]+vir, Malayalam.ConsonantMap[26]+vir, Malayalam.ConsonantMap[27]+vir, Malayalam.SouthConsonantMap[0]+vir, ] for x,y in zip(Chillus, ConVir): Strng = Strng.replace(x, y) return Strng def SinhalaPali(Strng): Strng = PostProcess.SinhalaPali(Strng, reverse = True) return Strng def IASTPali(Strng): Strng = Strng.replace('ḷ', 'l̤') return Strng def CyrillicPali(Strng): Strng = Strng.replace('л̣', 'л̤',) return Strng def MalayalamPrakrit(Strng): ## Reverse Gemination Strng = PostProcess.ReverseGeminationSign(Strng, 'Malayalam') Strng = Strng.replace("ഀ", "ം") return Strng def GranthaPrakrit(Strng): ## Reverse Gemination Strng = PostProcess.ReverseGeminationSign(Strng, 'Grantha') Strng = Strng.replace("𑌀", "𑌂") return Strng def RomanPreFix(Strng,Source): DepV = '\u1E7F' Asp = '\u02B0' Vir = GM.CrunchList("ViramaMap", Source)[0] Nuk = GM.CrunchList("NuktaMap", Source)[0] VowelA = GM.CrunchSymbols(['VowelMap'],Source)[0] ListV = '|'.join(GM.CrunchSymbols(GM.VowelSigns,Source)) ListC = '|'.join(GM.CrunchSymbols(GM.Consonants,Source)) Strng = re.sub('('+ListC+')'+'(?!'+ListV+'|'+VowelA+')',r'\1'+DepV+Vir,Strng) Strng = re.sub('('+ListC+'|'+Nuk+')'+'('+ListV+')',r'\1'+DepV+r'\2',Strng) #print Strng Strng = re.sub('(?<='+ListC+')'+'('+VowelA+')',r'',Strng) #print Strng #Fix Nukta त़् त़्अ त़्ा --> त़् त़ त़ा Strng = Strng.replace(DepV+Vir+Nuk+VowelA,Nuk) Strng = re.sub(DepV+Vir+Nuk+'(?=['+DepV+'])',Nuk,Strng) Strng = Strng.replace(DepV+Vir+Nuk,Nuk+DepV+Vir) #print Strng return Strng def joinVowelCons(Strng, script): consonantsAll = '(' + '|'.join(sorted(GM.CrunchSymbols(GM.Consonants, script), key = len, reverse=True)) + ')' vowelsAll = '(' + '|'.join(sorted(GM.CrunchSymbols(GM.Vowels, script), key = len, reverse=True)) + ')' Strng = re.sub(consonantsAll + ' ' + vowelsAll, r'\1\2', Strng) Strng = re.sub(consonantsAll + ' ' + consonantsAll, r'\1\2', Strng) return Strng def joinVowelConsIAST(Strng): return joinVowelCons(Strng, 'IAST') def joinVowelConsISO(Strng): return joinVowelCons(Strng, 'ISO') def PreProcess(Strng,Source,Target,postoptions,preoptions): if Source in GM.RomanDiacritic or Source == 'Latn': Strng = Strng.lower() if Source in GM.pipeScripts: if '|' in Strng: Strng = Strng.replace(".", "●") Strng = Strng.replace("।", "|").replace("॥", "||") Strng = Strng.replace("|", ".").replace("||", "..") if Source == "Itrans": Strng = re.sub("●([acCDN])", "."+r'\1', Strng) elif Source == "Velthuis": Strng = re.sub("●([rRlLmhtdnTDsyg])", "."+r'\1', Strng) Strng = re.sub("●●([lr])", "."+r'\1', Strng) if 'Arab' in Source: Strng = re.sub('([وي])(?=[\u064E\u0650\u064F\u0651\u064B\u064C\u064D])', '\u02DE' + r'\1', Strng) #Strng = re.sub('((\u064E|\u0650|\u0652|\u064F))(\u0651)', r'\2', Strng) if Source in ["Syrj", "Syrn"]: Strng = Strng.replace('\u0323', '\u0742') if Source == 'Itrans': sOm = 'OM' tOm = 'oM' punc = '(' + '|'.join(["\u005C"+x for x in list(string.punctuation)]+ ['\s'] + [x.replace('.', '\.') for x in GM.CrunchSymbols(GM.Signs,Source)[1:3]]) + ')' Strng = re.sub(punc + sOm + punc, r'\1' + tOm + r'\2', Strng) Strng = re.sub('^' + sOm + punc, tOm + r'\1', Strng) Strng = re.sub(punc + sOm + '$', r'\1' + tOm, Strng) Strng = re.sub('^' + sOm + '$', tOm, Strng) punc = '(\s)' Strng = re.sub(punc + sOm + punc, r'\1' + tOm + r'\2', Strng) Strng = re.sub('^' + sOm + punc, tOm + r'\1', Strng) Strng = re.sub(punc + sOm + '$', r'\1' + tOm, Strng) Strng = re.sub('^' + sOm + '$', tOm, Strng) # kaRRi -> katri etc. not kaVoc.R Strng = re.sub('([aAiuUeoEO])([R]){2}([iI])', r'\1'+r'\2'+'_'+r'\2'+r'\3', Strng) Strng = re.sub('([aAiuUeoEO])([L]){2}([iI])', r'\1'+r'\2'+'_'+r'\2'+r'\3', Strng) AltForm = ['O', 'aa','ii','uu','RRi','RRI','LLi','LLI','N^','JN','chh','shh','x','GY','.n','.m','.h', 'AUM', 'E', 'J', 'c.o', 'c.e'] NormForm = ['^o', 'A','I','U','R^i','R^I','L^i','L^I','~N','~n','Ch','Sh','kSh','j~n','M','M','','oM', '^e', 'z', 'A.c', 'e.c'] for x,y in zip(AltForm,NormForm): Strng = Strng.replace(x,y) AltForms = [('ee', 'I'), ('dny', 'j~n'), ('oo', 'U'), ('kS', 'kSh'), ('w', 'v'), ('|', '.'), ('kShh', 'kSh')] for x,y in AltForms: Strng = Strng.replace(x,y) Strng = Strng.replace('OM', 'oM') if Source == 'BarahaNorth' or Source == 'BarahaSouth': # alternate representations alt_baraha = [('A', 'aa'), ('I', 'ee'), ('U', 'oo'), ('~loo', '~lU'), ('Roo', 'RU'), ('ou', 'au'), ('K', 'kh'), ('G','gh'), ('ch', 'c'), ('Ch', 'C'), ('J','jh'), ('P', 'ph'), ('B', 'bh'), ('w', 'v'), ('sh', 'S'), ('~h', '_h'), ('Y', 'yx'), ('^^', '{}'), ('^', '()'), ('tx', 'rx'), ('zh', 'Lx'), ('~e', '~a'), ('q', '\_'), ('#', "\\'"), ('$', '\\"')] for alt, norm in alt_baraha: Strng = Strng.replace(alt, norm) if Target == 'Tamil': Strng = Strng.replace('n', 'nx').replace('~nx', 'n') print(Strng) if 'IAST' in Source: Strng = Strng.replace("aï", "a_i") Strng = Strng.replace("aü", "a_u") Strng = Strng.replace('\u0303', 'ṃ') if "ISO" in Source: Strng = Strng.replace('a:i', 'a_i') Strng = Strng.replace('a:u', 'a_u') Strng = Strng.replace('\u0303', 'ṁ') ## People use the wrong convention sometimes Strng = Strng.replace('ṃ', 'ṁ') if Source == "Titus": Strng = Strng if Source == "ISO" or Source == "IAST" or Source == "Titus" or "RussianCyrillic": Strng = CF.VedicSvarasNonDiacritic(Strng) if Source == "Latn" and 'Syr' in Target: Strng = Strng.replace('ḇ', 'v').replace('ḡ', 'ḡ').replace('ḵ', 'ḫ').replace('p̄', 'f') if ('↓' in Strng or '↑' in Strng) and Target in GM.IndicScripts : Strng = Strng.replace('↓', '॒') Strng = Strng.replace('↑↑', '᳚') Strng = Strng.replace('↑', '॑') if ('↓' in Strng or '↑' in Strng) and Target in GM.LatinScripts : Strng = Strng.replace('↓', '\\_') Strng = Strng.replace('↑↑', '\\"') Strng = Strng.replace('↑', '\\\'') if Source == "WarangCiti": Strng = Strng.replace('\u200D', '\u00D7') if Source == "Hebr-Ar": # accept variants with dotes dot_var = [('עׄ','ג'), ('תׄ','ת֒'), ('ת','ת̈'), ('ק','ק̈')] for char, char_var in dot_var: Strng = Strng.replace(char_var, char) ## Normalize Input Strings Strng = normalize(Strng,Source) ## Remove unsupported letters and replace with supported ones #Fix Udata + Avagraha combination # नमो॑ऽयम् --> namo̍​'yam if Source in GM.IndicScripts and Target in GM.Transliteration: udatta = '\u0951' avagrahaSrc = GM.CrunchList('SignMap', Source)[0] avagrahaTgt = GM.CrunchList('SignMap', Target)[0] if avagrahaTgt == "'": Strng = Strng.replace(udatta + avagrahaSrc, udatta + '\u200B' + avagrahaSrc) return Strng def ISO259Target(Strng): Strng = Strng.replace('א', 'ʾ').replace('׳', '’') return Strng def ISO233Target(Strng): replacements = [('أ', 'ˈʾ'), ('ء', '¦'), ('إ', 'ˌʾ')] for x, y in replacements: Strng = Strng.replace(x, y) return Strng def PersianDMGTarget(Strng): replacements = [('ا', 'ʾ')] for x, y in replacements: Strng = Strng.replace(x, y) return Strng def ISO233Source(Strng): replacements = [('أ', 'ˈʾ'), ('ء', '¦'), ('إ', 'ˌʾ')] for x, y in replacements: Strng = Strng.replace(y, x) replacements = [('j', 'ǧ'), ('g', 'ǧ'), ('ḧ', 'ẗ'), ('ḫ', 'ẖ'), ('a̮', 'ỳ'), \ ('aⁿ', 'á'), ('iⁿ', 'í'), ('uⁿ', 'ú'), ('ā̂', 'ʾâ'), ('ˀ', 'ˈ')] for x, y in replacements: Strng = Strng.replace(y, x) return Strng def HebrewSBLTarget(Strng): Strng = Strng.replace('א', 'ʾ').replace('׳', '’') return Strng def HebrewSBLSource(Strng): Strng = Strng.replace('ʾ', 'א',).replace('’', '׳') Strng = Strng.replace('\u0307\u00B0', '\u00B0\u0307') replacements = [('v', 'ḇ'), ('f', 'p̄'), ('d꞉', 'd'), ('d', 'ḏ'), \ ('g꞉', 'g'), ('g', 'ḡ'), \ ('t꞉', 't'), ('t', 'ṯ'), ('š̮', 'š'), ('š̪', 'ś'), ('o', 'ō'), ('ō', 'ô'), ('ū', 'û'), ('\u033D', 'ĕ') ] for x, y in replacements: Strng = Strng.replace(y, x) return Strng def ISO259Source(Strng): Strng = Strng.replace('ʾ', 'א',).replace('’', '׳') Strng = Strng.replace('\u0307\u00B0', '\u00B0\u0307') replacements = [('ḵ', 'k'), ('v', 'b'), ('f', 'p'), ('b', 'ḃ'), ('p', 'ṗ'), ('k', 'k̇'), ('꞉', '\u0307'),\ ('š̮', 'š'), ('š', 's̀'), ('š̪', 'ś'), ('ā', 'å'), ('e', 'ȩ'), ('ō', 'ŵ'), ('ū', 'ẇ'), ('\u033D', '°'), ('ĕ', 'ḝ') ] for x, y in replacements: Strng = Strng.replace(y, x) Strng = unicodedata.normalize('NFD', Strng) Strng = Strng.replace('\u0307', '꞉') Strng = unicodedata.normalize('NFC', Strng) return Strng def UnSupThaana(Strng): return Strng def RemoveJoiners(Strng): Strng = Strng.replace("\u200D", "") Strng = Strng.replace("\u200C", "") return Strng def ArabicGimelJa(Strng): Strng = Strng.replace("ج", "ڨ") return Strng # Normalize Input # ka + -> qa etc def normalize(Strng,Source): # Replace Decomposed Nukta Characters by pre-composed Nukta consonants #nuktaDecom = [u"क़",u"ख़",u"ग़",u"ज़",u"ड़",u"ढ़",u"फ़",u"य़",u"ਲ਼",u"ਸ਼",u"ਖ਼",u"ਗ਼",u"ਜ਼",u"ਫ਼",u"ড়",u"ঢ়",u"য়",u"ଡ଼",u"ଢ଼"] #nuktaPrecom = [u"क़",u"ख़",u"ग़",u"ज़",u"ड़",u"ढ़",u"फ़",u"य़",u"ਲ਼",u"ਸ਼",u"ਖ਼",u"ਗ਼",u"ਜ਼",u"ਫ਼",u"ড়",u"ঢ়",u"য়",u"ଡ଼",u"ଢ଼"] nuktaDecom = ["\u0915\u093C","\u0916\u093C","\u0917\u093C","\u091C\u093C","\u0921\u093C","\u0922\u093C","\u092B\u093C","\u092F\u093C","\u0A32\u0A3C","\u0A38\u0A3C","\u0A16\u0A3C","\u0A17\u0A3C","\u0A1C\u0A3C","\u0A2B\u0A3C","\u09A1\u09BC","\u09A2\u09BC","\u09AF\u09BC","\u0B21\u0B3C","\u0B22\u0B3C"] nuktaPrecom = ["\u0958","\u0959","\u095A","\u095B","\u095C","\u095D","\u095E","\u095F","\u0A33","\u0A36","\u0A59","\u0A5A","\u0A5B","\u0A5E","\u09DC","\u09DD","\u09DF","\u0B5C","\u0B5D"] if Source not in ["Grantha","TamilGrantha"]: for x,y in zip(nuktaDecom,nuktaPrecom): Strng = Strng.replace(x,y) if Source in ['IAST', 'ISO', 'ISOPali', 'Titus', 'IASTPali']: Strng = Strng.replace("ü", "uʼ").replace("ǖ", "ūʼ").replace( 'ö', 'aʼ',).replace('ȫ', 'āʼ') if Source in ['IAST', 'ISO', 'ISOPali', 'IASTPali', 'Titus'] or 'RomanLoC' in Source: Strng = unicodedata.normalize('NFC', Strng) Strng = Strng.replace("ẗ", "ẗ").replace("ÿ", "ÿ").replace("ḧ", "ḧ") if 'RomanLoC' in Source: Strng = Strng.replace('\u02D9', '\u02DA') if Source == 'Arab-Ur' or Source == 'Arab-Pa': Strng = Strng.replace('ك', 'ک') Strng = Strng.replace('ي', 'ی') if Source == "Hebr": vowels = ['ְ','ֱ','ֲ','ֳ','ִ','ֵ','ֶ','ַ','ָ','ֹ','ֺ','ֻ','ׇ'] vowelsR = '(' + '|'.join(vowels + ['וֹ', 'וּ']) + ')' # Swap the order of diacritics Strng = re.sub(vowelsR + '([ּׁׂ])', r'\2\1', Strng) Strng = Strng.replace('\u05BC\u05B0\u05C1', '\u05C1\u05BC\u05B0') # Sindhi C+Anudatta <- Sindhi Underscore characters # For easy Transliteration # Malayalam Chillu Characters chilluZwj=["ണ്‍","ന്‍","ര്‍","ല്‍","ള്‍","ക്‍"] chilluAtom=["ൺ","ൻ","ർ","ൽ","ൾ","ൿ"] for x,y in zip(chilluZwj,chilluAtom): Strng = Strng.replace(x,y) Strng = Strng.replace('ൌ', 'ൗ') Strng = Strng.replace('ൟ', 'ഈ') Strng = Strng.replace('ൎ', 'ര്') # Malayalam chill + virama to just letter + virama Strng = Strng.replace('ൻ്റ', 'ന്റ') # Telugu/Kananda Nakaarapollu Strng = Strng.replace('ೝ', 'ನ್') # Kannada Strng = Strng.replace('ౝ', 'న్') # Telugu # Bengali Khanda Ta # Strng = Strng.replace("ৎ","ত্‍") # Tamil tamAlt = ["ஸ்ரீ","க்‌ஷ","ர‌ி","ர‌ீ"] tamNorm = ["ஶ்ரீ","க்ஷ","ரி","ரீ"] # Burmese Strng = Strng.replace("ဿ", "သ္သ") # Tamil Strng.replace("ஸ²", "ஶ") ## The following is a reversal of custom options check !!!! subNum = ["¹","₁","₂","₃","₄"] supNum = ["","","²","³","⁴"] for x,y in zip(tamAlt+subNum,tamNorm+supNum): Strng = Strng.replace(x,y) # Tibetan Vowels oldVow = ["ྲྀ","ཷ","ླྀ","ཹ","ཱི","ཱུ","ཀྵ","ྐྵ"] newVow= ["ྲྀ","ྲཱྀ","ླྀ","ླཱྀ","ཱི","ཱུ","ཀྵ","ྐྵ"] for x,y in zip(oldVow,newVow): Strng = Strng.replace(x,y) # Decomposed to Precomposed Roman Characters latinDecom= ["ā","ī","ū","ē","ō","ṃ","ṁ","ḥ","ś","ṣ","ṇ","ṛ","ṝ","ḷ","ḹ","ḻ","ṉ","ṟ"] latinPrecom = ["ā","ī","ū","ē","ō","ṃ","ṁ","ḥ","ś","ṣ","ṇ","ṛ","ṝ","ḷ","ḹ","ḻ","ṉ","ṟ"] for x,y in zip(latinDecom,latinPrecom): Strng = Strng.replace(x,y) # Thai Decomposed AM to Precomposed AM Strng = Strng.replace("ํา","ำ") # For Lao Strng = Strng.replace("ໍາ", 'ຳ') ## Two Single Danda to Double Danda Strng = Strng.replace("।।","॥") ## Decomposed Limbu to precomposed Strng = Strng.replace("ᤠ᤺ᤣ", "ᤠᤣ᤺") Strng = Strng.replace("᤺ᤣ", "ᤣ᤺") Strng = Strng.replace("ᤠᤣ", "ᤥ") ## Replace Thai ฎ Strng = Strng.replace("ฎ", "ฏ") ## Replace Grantha old au with new au Strng = Strng.replace('𑍌', '𑍗') ## Replace Tibetan Chandra with Nada to the normal one Strng = Strng.replace('\u0F82', '\u0F83') ## Replace Candra A iwht regualr AE Strng = Strng.replace('ॲ', 'ऍ') #Strng = Strng.replace('', "ຣ\uE00A") ## Normalization for Bengali, Tamil, Malayalam and Grantha # Bengali o/au Strng = Strng.replace('ো', 'ো') Strng = Strng.replace('াে', 'ো') Strng = Strng.replace('ৌ', 'ৌ') Strng = Strng.replace('ৗে', 'ৌ') # Tamil o, O, au Strng = Strng.replace('ொ', 'ொ') Strng = Strng.replace('ாெ', 'ொ') Strng = Strng.replace('ோ', 'ோ') Strng = Strng.replace('ாே', 'ோ') Strng = Strng.replace('ௌ', 'ௌ') Strng = Strng.replace('ௗெ', 'ௌ') # Malayalam Strng = Strng.replace('ൊ', 'ൊ') Strng = Strng.replace('ാെ', 'ൊ') Strng = Strng.replace('ോ', 'ോ') Strng = Strng.replace('ാേ', 'ോ') # Grantha Strng = Strng.replace('𑍋', '𑍋') Strng = Strng.replace('𑌾𑍇', '𑍋') return Strng # Remove ZWJ/ZWNJ characters def removeZW(Strng): Strng = Strng.replace("\u200C").replace("\u200D") return Strng def PhagsPaArrange(Strng,Source): if Source in GM.IndicScripts: ListC = "|".join(sorted(GM.CrunchSymbols(GM.Consonants, Source),key=len,reverse=True)) #Do this for all ListV = "|".join(sorted(GM.CrunchSymbols(GM.Vowels, Source),key=len,reverse=True)) ListVS = "|".join(sorted(GM.CrunchSymbols(GM.VowelSignsNV, Source),key=len,reverse=True)) ListCS = "|".join(sorted(GM.CrunchSymbols(GM.CombiningSigns, Source),key=len,reverse=True)) vir = GM.CrunchSymbols(GM.VowelSigns,Source)[0] yrv = "|".join([GM.CrunchSymbols(GM.Consonants, Source)[i] for i in [25,26,28]]) Strng = re.sub("("+ListC+")"+"("+vir+")"+"("+yrv+")"+"("+"("+ListVS+")?"+"("+ListCS+")?"+")",r' \1\2\3\4',Strng) Strng = re.sub("("+ListC+ListV+")"+"("+"("+ListVS+")?"+"("+ListCS+")?"+")"+"("+ListC+")"+"("+vir+")"+"(?!\s)",r"\1\2\5\6 ",Strng) Strng = re.sub("("+ListC+ListV+")"+"("+"("+ListVS+")?"+"("+ListCS+")?"+")"+"("+ListC+")"+'(?!'+vir+')',r"\1\2 \5",Strng) Strng = re.sub("("+ListC+ListV+")"+"("+"("+ListVS+")?"+"("+ListCS+")?"+")"+"("+ListC+")"+'(?!'+vir+')',r"\1\2 \5",Strng) elif Source in GM.LatinScripts: pass return Strng def TamilTranscribeCommon(Strng, c = 31): script = "Tamil" ListC = GM.CrunchList('ConsonantMap',script) ListSC = GM.CrunchList('SouthConsonantMap',script) vir = GM.CrunchSymbols(GM.VowelSigns, script)[0] ConUnVoiced = [ListC[x] for x in [0,5,10,15,20]] ConVoicedJ = [ListC[x] for x in [2,7,12,17,22]] ConVoicedS = [ListC[x] for x in [2,31,12,17,22]] ConNasalsAll = '|'.join([ListC[x] for x in [4, 9, 14, 19, 24]]) conNasalCa = '|'.join([ListC[x] for x in [9]]) ConNasalsGroup = [ConNasalsAll, conNasalCa, ConNasalsAll, ConNasalsAll, ConNasalsAll] ConMedials = '|'.join(ListC[25:28]+ListSC[0:2]+ListSC[3:4]) Vowels = '|'.join(GM.CrunchSymbols(GM.Vowels+GM.VowelSignsNV, script)) Aytham = GM.CrunchList('Aytham',script)[0] Consonants = '|'.join(GM.CrunchSymbols(GM.Consonants,script)) NRA =ListSC[3] + vir + ListSC[2] NDRA = ListC[14] + vir + ListC[12] + vir + ListC[26] ### Check Siva Siva Mails ### Do something about Eyelash ra in Transliterated text for i in range(len(ConUnVoiced)): pass #Strng = re.sub('('+Vowels+Consonants+')'+ConUnVoiced[i]+'('+Vowels+Consonants+')',r'\1'+ConVoicedS[i]+r'\2',Strng) Strng = re.sub('('+Vowels+'|'+Consonants+'|'+Aytham+')'+ConUnVoiced[i]+'(?!'+vir+')',r'\1'+ConVoicedS[i],Strng) Strng = re.sub('([³])'+ConUnVoiced[i]+'(?!'+vir+')',r'\1'+ConVoicedS[i],Strng) Strng = re.sub('³+', '³', Strng) # Nasals + Unvoiced -> voiced # Rule applied even if spaced -> ulakaJ cuRRu -> ulakaJ juRRu; cenJu -> senju # Strng = re.sub('('+ConNasalsSpace+')'+'('+vir+')'+'( ?)'+ConUnVoiced[i],r'\1\2\3'+ConVoicedJ[i],Strng) # Only without space : vampu -> vambu; varim panam -> varim panam Strng = re.sub('('+ConNasalsGroup[i]+')'+'('+vir+')'+ConUnVoiced[i],r'\1\2'+ConVoicedJ[i],Strng) # Medials + Unvoiced -> Voiced Strng = re.sub('('+ConMedials+')'+'('+vir+')'+ConUnVoiced[i]+'(?!'+vir+')',r'\1\2'+ConVoicedS[i],Strng) Strng = Strng.replace(NRA,NDRA) # Intervocalic S Strng = re.sub('(? CCA Strng = re.sub('('+ListC[5]+vir+')'+'(('+punct+')+)'+'('+ListC[c]+')',r'\1\2'+ListC[5],Strng) # JA + Spac | Punct + SA -> JCA Strng = re.sub('('+ListC[9]+vir+')'+'(('+punct+')+)'+'('+ListC[c]+')',r'\1\2'+ListC[7],Strng) # GA + Spac | Punct + KA -> Strng = re.sub('('+ListC[4]+vir+')'+'(('+punct+')+)'+'('+ListC[0]+')',r'\1\2'+ListC[2],Strng) # NA + Spac | Punct + TTA -> Strng = re.sub('('+ListC[14]+vir+')'+'(('+punct+')+)'+'('+ListC[10]+')',r'\1\2'+ListC[12],Strng) # NA + Spac | Punct + TA -> Strng = re.sub('('+ListC[19]+vir+')'+'(('+punct+')+)'+'('+ListC[15]+')',r'\1\2'+ListC[17],Strng) # K + k -> hh Strng = Strng.replace(Tamil.Aytham[0]+ListC[0],ListC[32]+vir+ListC[32]) # K -> H Strng = Strng.replace(Tamil.Aytham[0],ListC[32]+vir) # RR + RRA -> TT+RA Strng = re.sub(ListSC[2]+vir+ListSC[2],ListC[10]+vir+ListC[26],Strng) # RR | TT + /s + SA -> RR + /s + CA Strng = re.sub("("+'['+ListC[10]+ListSC[2]+']'+vir+')'+'(\s)'+'('+ListC[c]+')',r'\1\2'+ListC[5],Strng) ## NNN to N, RR to R Strng = Strng.replace(ListSC[3],ListC[19]) #Strng = Strng.replace(ListSC[2],ListC[26]) return Strng def TamilTranscribe(Strng): Strng = TamilTranscribeCommon(Strng) return Strng def TamilTranscribeDialect(Strng): Strng = TamilTranscribeCommon(Strng, c=29) return Strng def IPAIndic(Strng): Strng = Strng.replace("ʊ","u") Strng = Strng.replace("ɛ","e") return