11i<"SrSSKrSSKrSSKJrJrJrJr SSKJ r SSK J r Sr \R"S/SQ5r"S S 5rg) z(Implementation of punctuation processingN)ListUnionTuplePattern)str2list) Separatoru!;:,.!?¡¿—…"«»“”(){}[] _mark_index)indexmarkpositionc l\rSrSrSr\4S\\\44Sjjr \ S5r \ S5r \ RS\\\44Sj5r S \\\\4S \\\\44S jrS \\\\4S \\\\\\44S jrS \S\S \\\\\44Sjr\S \\\\4S\\S\S\S \\4 Sj5rSrg) Punctuationa2Preserve or remove the punctuation during phonemization Backends behave differently with punctuation: festival and espeak ignore it and remove it silently whereas segments will raise an error. The Punctuation class solves that issue by "hiding" the punctuation to the phonemization backend and restoring it afterwards. Parameters ---------- marks (str or re.Pattern) : The punctuation marks to consider for processing (either removal or preservation). If a string, each mark must be made of a single character. Default to Punctuation.default_marks(). marksc,SUlSUlXlgN_marks _marks_rer)selfrs P/home/james-whalen/.local/lib/python3.13/site-packages/phonemizer/punctuation.py__init__Punctuation.__init__/s '+ c[$)z1Returns the default punctuation marks as a string)_DEFAULT_MARKSrr default_marksPunctuation.default_marks4s rcRUR(a UR$[S5e)z!The punctuation marks as a stringzCpunctuation initialized from regex, cannot access marks as a string)r ValueError)rs rrPunctuation.marks9s! ;;;; ^__rvaluec[U[5(a3[R"SUR-S-5UlSUlg[U[5(a]SR[U55Ul[R"S[R"UR 5S35Ulg[S5e)Nz((z)|\s)+z(\s*[z]+\s*)+z;punctuation marks must be defined as a string or re.Pattern) isinstancerrecompilepatternrrstrjoinsetescaper!)rr#s rrr"@s eW % %ZZ (= (IJDNDK s # #''#e*-DK ZZ54;;1G0H(PQDNZ[ [rtextreturnc^S[S[4U4Sjjn[U[5(aU"U5$UVs/sH o2"U5PM sn$s snf)zReturns the `text` with all punctuation marks replaced by spaces The input `text` can be a string or a list and is returned with the same type and punctuation removed. r.r/cd>[R"TRSU5R5$)N )r'subrstrip)r.rs rauxPunctuation.remove..auxWs#66$..#t4::< ['hello', 'my world'], [',', '!'] )r enumerate_preserve_line)rr.preserved_textpreserved_marksnumr7rs rpreservePunctuation.preserve^sj#4."4IC--d8KD  "N  $O)"0848/II8s  AAr7r?c[[R"URU55nU(dU//4$[ U5S:Xa&USR 5U:Xa/[ X!S5/4$/nUHnSnXSS:Xa'URUR 55(aSnO.XSS:Xa&URUR 55(aSnUR[ X%R 5U55 M /nUHSnURUR5n U SURRU SS5pURU 5 U nMU Xq/-U4$) z+Auxiliary method for Punctuation.preserve()rAIBEN) listr'finditerrlengroup _MarkIndex startswithendswithappendsplitr r+) rr7r?matchesrmatchr preserved_liner rQprefixsuffixs rr<Punctuation._preserve_lineqs>r{{4>>48962:  w<1 !1!1!3t!; 3c233 3EH "tu{{}'E'E"+%$-- *F*F LLCA BDJJtyy)E"1Xtyy~~eABi'@F  ! !& )D &--rsepr4c [U5n/nSnU(dU(GaU(ddUHZnU(d?UR(a.URUR5(dXsR-nURU5 M\ /nGO"U(dKUR[R "SURSR SU5555 /nGOUSnURU:XGaUSn USSn[R "SURU R5n UR(aBUSRUR5(aUSS[UR5*US'URS:Xa XS-US'GOURS:XaXURUSU -U(d U RUR5(aSO UR-5 USSnUS-nOURS :XaLURX(d U RUR5(aSO UR-5 US-nOQ[U5S:Xa USU -US'O6USn USSnX-US-US'OURUS5 USSnUS-nU(aGMU(aGMU$) aJRestore punctuation in a text. This is the reverse operation of Punctuation.preserve(). It takes a list of punctuated chunks and a list of punctuation marks, as well as the separator and strip parameters used by phonemize. It returns the punctuated text as a list: ['hello', 'my world'], [',', '!'] -> ['hello, my world!'] rr2r%c38# UHoRv M g7fr)r ).0ms r &Punctuation.restore..sD[UZPQVVUZsrCNrFrHrD) rwordrOrPr'r3r+r r rKr ) clsr.rrXr4punctuated_textposr7 current_markr first_words rrestorePunctuation.restoresr~e D SXXdmmCHH6M6M#hh#**40 !  &&rvvc388RWWD[UZD[=['\] %Qx %%,!8D!!"IE66#sxx;DxxDG$4$4SXX$>$>"&q'/CM>":Q#,,3"&a.Q%..#5'..tAw~uPTP]P]^a^f^fPgPgmpmumu/vw#ABx!Ag%..#5'..tUdmmTWT\T\F]F]rcfckck/lm!Agt9>'+1gnDG)-aJ#'8D&0&7$q'&ADG$**4738D'CideenrrN)__name__ __module__ __qualname____firstlineno____doc__rrr*rr staticmethodrpropertyrsetterrr8rrMr@intr< classmethodrboolre__static_attributes__rrrrrsi 5CeCL1 ``  \\ \5g. \ \ ,5d3i0 ,U3S >5J ,JU49c>2JuT$s)_dS]N^=^7_J&.3.S.U49d:FV;V5W.BI5d3i0IJ'III!%S IIrr)rk collectionsr'typingrrrrphonemizer.utilsrphonemizer.separatorrr namedtuplerMrrrrrxsE/ ..%*5  # #02 }}r