v5`n^@ddlmZddlmZddlmZmZ ddl Z n #e $rdZ YnwxYwgdZ GddeZ Gd d eZ Gd d eZGd deZGddeZGddeZGddeZGddeZGddeZGddeZe Ze ZexZZeZeZeZeZeZeZeZ dS)) defaultdict) zip_longest)BaseBaseSimilarityN)HammingMLIPNS LevenshteinDamerauLevenshteinJaro JaroWinklerStrCmp95NeedlemanWunschGotoh SmithWatermanhammingmlipns levenshteindamerau_levenshteinjaro jaro_winklerstrcmp95needleman_wunschgotohsmith_watermanc eZdZdZddZdZdS) rz Compute the Hamming distance between the two or more sequences. The Hamming distance is the number of differing items in ordered sequences. https://en.wikipedia.org/wiki/Hamming_distance rNFTcL||_|p|j|_||_||_dSN)qval_ident test_functruncateexternal)selfrr!r"r#s Blib/python3.11/site-packages/textdistance/algorithms/edit_based.py__init__zHamming.__init__#s) "1dk    cj|}j|}||Sjrtnt}t fd||DS)Nc&g|] }j| S)r!).0esr$s r% z$Hamming.__call__..1s%GGG++GGGr')_get_sequences quick_answerr"ziprsum)r$ sequencesresult_zips` r%__call__zHamming.__call__)sf'D'3 ""I.  Mm4ssGGGGddI6FGGGHHHr')rNFT)__name__ __module__ __qualname____doc__r&r5r*r'r%rrsF !!!! IIIIIr'rc,eZdZdZd dZdZdZdZdS) r a Compute the absolute Levenshtein distance between the two sequences. The Levenshtein distance is the minimum number of edit operations necessary for transforming one sequence into the other. The edit operations allowed are: * deletion: ABC -> BC, AC, AB * insertion: ABC -> ABCD, EABC, AEBC.. * substitution: ABC -> ABE, ADC, FBC.. https://en.wikipedia.org/wiki/Levenshtein_distance TODO: https://gist.github.com/kylebgorman/1081951/9b38b7743a3cb5167ab2c6608ac8eea7fc629dca rNTc>||_|p|j|_||_dSrrr r!r#r$rr!r#s r%r&zLevenshtein.__init__A" "1dk  r'c |r|st|t|zS||d|dr||dd|ddSt||dd||||dd}||dd|dd}t||dzS)Nr)lenr!min)r$s1s2dss r% _recursivezLevenshtein._recursiveFs % %r77SWW$ $ >>"R&"R& ) ) *43B3CRC)) )  DCRC"   DRW     DCRC"SbS' " "1ayy1}r'ct|dz}t|dz}d}trtj|}nt|}td|D]}||gdg|dz zz}}td|D]f}||dz} ||dz dz} |||dz ||dz } ||dz | z} t | | | ||<g|dS)zp source: https://github.com/jamesturk/jellyfish/blob/master/jellyfish/_jellyfish.py#L18 rNrr@)rAnumpyarangeranger!rB) r$rCrDrowscolsprevcurrcdeletion insertiondistedits r%_cicledzLevenshtein._cicledWs 2ww{2ww{  ,t$$CC++Cq$ 8 8AaSA3$(#33#D1d^^ 8 87Q;AJN ~~bQiAE;;AE{$h/T8Y77A  8 2wr'c|||\}}|||}||S|||Sr)r.r/rVr$rCrDr3s r%r5zLevenshtein.__call__nsO$$R,,B""2r**  M||B###r'rNT)r6r7r8r9r&rGrVr5r*r'r%r r 4s_  !!!! ".$$$$$r'r c,eZdZdZd dZdZdZdZdS) r a Compute the absolute Damerau-Levenshtein distance between the two sequences. The Damerau-Levenshtein distance is the minimum number of edit operations necessary for transforming one sequence into the other. The edit operations allowed are: * deletion: ABC -> BC, AC, AB * insertion: ABC -> ABCD, EABC, AEBC.. * substitution: ABC -> ABE, ADC, FBC.. * transposition: ABC -> ACB, BAC https://en.wikipedia.org/wiki/Damerau%E2%80%93Levenshtein_distance rNTc>||_|p|j|_||_dSrr<r=s r%r&zDamerauLevenshtein.__init__r>r'c tjt|dzt|dzgtj}t dt|dzD]}|dz||d<t dt|dzD]}|dz|d|<t |D]\}}t |D]\}}t||| }t||dz |dz|||dz dz||dz |dz |z|||<|r|s||||dz st|||||dz |dz |z|||<܌|t|dz t|dz S)Nrdtyper@)rIzerosrAintrK enumerater!rB r$rCrDrEijcs1cs2costs r%_numpyzDamerauLevenshtein._numpys KR1c"ggk2%) D D Dr3r77Q;''  A1uAaDHHr3r77Q;''  A1uAbE!HHmm  FAs#B--  3t~~c377788a!eHQK!OaDQK!Oa!eHQUOd*!Q~~c2a!e955aDGa!eHQUOd*!Q (R1~c"ggk**r'c i}tdt|dzD] }|dz||df< tdt|dzD] }|dz|d|f< t|D]\}}t|D]\}}t||| }t ||dz |fdz|||dz fdz||dz |dz f|z|||f<|r|sq||||dz s|||dz |st |||f||dz |dz f|z|||f<|t|dz t|dz fS)za https://www.guyrutenberg.com/2008/12/15/damerau-levenshtein-distance-in-python/ r@rr_)rKrArbrar!rBrcs r% _pure_pythonzDamerauLevenshtein._pure_pythons r3r77Q;''  A1uAaeHHr3r77Q;''  A1uAb!eHHmm  FAs#B--  3t~~c377788a!eQhK!OaQhK!Oa!eQUlOd*!Q$~~c2a!e955~~bQi55adGa!eQUlOd*!Q$# ,R1c"ggk)**r'c|||\}}|||}||S|||Sr)r.r/rkrXs r%r5zDamerauLevenshtein.__call__sS$$R,,B""2r**  M   R(((r'rY)r6r7r8r9r&rirkr5r*r'r%r r xsa  !!!! +++B#+#+#+J ) ) ) ) )r'r c(eZdZdZd dZdZd dZd S) r a Computes the Jaro-Winkler measure between two strings. The Jaro-Winkler measure is designed to capture cases where two strings have a low Jaro score, but share a prefix. and thus are likely to match. https://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance https://github.com/Yomguithereal/talisman/blob/master/src/metrics/distance/jaro.js https://github.com/Yomguithereal/talisman/blob/master/src/metrics/distance/jaro-winkler.js FTrc>||_||_||_||_dSr)rlong_tolerance winklerizer#)r$rorprr#s r%r&zJaroWinkler.__init__s# ,$  r'cdSNrr*r$r2s r%maximumzJaroWinkler.maximumqr'皙?c|||\}}|||}||St|}t|}|r|sdSt||}|dzdz }|dkrd}dg|z} dg|z} d} t |D]i\} } td| |z }t | |z|dz }t ||dzD]'}| |s||| krdx| | <| |<| dz } n(j| sdSdx}}t | D]@\} }|r9t ||D]}| |r|dz}n|| ||kr|dz }A|dz}| |z | |z z}|| |z | z z }|dz}|js|S|dks |dks|dkr|St |d }d} | |kr?|| || kr-|| r%| dz } | |kr|| || kr|| %| r|| |zd |z zz }|jr|d kr|S| | dzks d| z|| zkr|S| | z dz ||z| dzz dzz }|d |z |zz }|S) Nr_rrFTffffff??) r.r/rAmaxrbrBrKrpro)r$rCrD prefix_weightr3s1_lens2_lenmin_len search_ranges1_flagss2_flags common_charsrds1_chlowhirek trans_counts1_fweighttmps r%r5zJaroWinkler.__call__sN$$R,,B""2r**  MRR V 3ff%%1 ) !  L7V#7V# !"   HAua\)**CQ%vz22B3Q''  {r!u~~044HQK(1+ A%LE 3K ** % %GAt %q&))A{Ea5BqE>>1$K &)>><+-==!  M S==FaKK6Q;;M OO !ee1A2a5 FA!ee1A2a5  9 a-'3<8 8F " gllM 1q5 A $4w{$B$BMa!#!a%(?!(CD3<3&& r'N)FTrT)rvr6r7r8r9r&rtr5r*r'r%r r s[  !!!! MMMMMMr'r c eZdZdfd ZxZS)r FrTcRt|d||dS)NF)rorprr#)superr&)r$rorr# __class__s r%r&z Jaro.__init__Bs: )      r')FrT)r6r7r8r& __classcell__)rs@r%r r As=          r'r cBeZdZdZdZddZdZd Zd Zd Z d Z d Z dS)ra0 Computes the Needleman-Wunsch measure between two strings. The Needleman-Wunsch generalizes the Levenshtein distance and considers global alignment between two strings. Specifically, it is computed by assigning a score to each alignment between two input strings and choosing the score of the best alignment, that is, the maximal score. An alignment between two strings is a set of correspondences between the characters of between them, allowing for gaps. https://en.wikipedia.org/wiki/Needleman%E2%80%93Wunsch_algorithm Fr|NrTc\||_||_|r||_n |j|_||_dSr)rgap_costsim_funcr r#r$rrrr#s r%r&zNeedlemanWunsch.__init__Ys5    ($DMM KDM  r'cXttt| |jzSr)r}maprArrss r%minimumzNeedlemanWunsch.minimumbs$CY''(((4=88r'cFttt|Sr)r}rrArss r%rtzNeedlemanWunsch.maximume3sI&&'''r'cd|j|zS)z'Get distance between sequences r@) similarityrss r%distancezNeedlemanWunsch.distancehsODOY///r'c`|j|}|j|}|dkrdS|j||z ||z z S)!Get distance from 0 to 1 r)rrtrr$r2rrts r%normalized_distancez#NeedlemanWunsch.normalized_distancemsL$, *$, * a<<1 y)G3'8IJJr'c`|j|}|j|}|dkrdS|j||z |dzz S)rrrr_)rrtrrs r%normalized_similarityz%NeedlemanWunsch.normalized_similarityvsK$, *$, * a<<1+g5'A+FFr'ctstd|||\}}tjt |dzt |dzftj}t t |dzD]}||jz ||df<t t |dzD]}||jz |d|f<t|dD]\}}t|dD]l\}}||dz |dz f| ||z}||dz |f|jz } |||dz f|jz } t|| | |||f<m||j ddz |j ddz fS)Nz2Please, install numpy for Needleman-Wunsch measurerr]r) rI ImportErrorr.r`rAfloatrKrrbrr}shape) r$rCrDdist_matrdrec1c2matchdeleteinserts r%r5zNeedlemanWunsch.__call__s TRSS S$$R,,B ; WWq[#b''A+ &+    s2ww{## 2 2A 4=01HQTNNs2ww{## 2 2A 4=01HQTNNr1%% < !>A ?q)A-x~a/@1/DDEEr'rrr*r'r%rrsU  !!!! (((FFFFFr'rc,eZdZdZd dZdZdZd ZdS) rzGotoh score Gotoh's algorithm is essentially Needleman-Wunsch with affine gap penalties: https://www.cs.umd.edu/class/spring2003/cmsc838t/papers/gotoh1982.pdf r皙?NTcj||_||_||_|r||_n |j|_||_dSr)rgap_opengap_extrr r#)r$rrrrr#s r%r&zGotoh.__init__s<     ($DMM KDM  r'cHttt| Srrrss r%rz Gotoh.minimumsCY''((((r'cFttt|Srrrss r%rtz Gotoh.maximumrr'c tstd|||\}}t|}t|}tj|dz|dzftj}tj|dz|dzftj}tj|dz|dzftj}d|d<t d|d<t d|d<t d|dzD]R}t d||df<|j |j|dz zz ||df<t d||df<|j ||df<St d|dzD]R} t d|d| f<t d|d| f<|j |d| f<|j |j| dz zz |d| f<St|dD]\}} t|dD]\} } | | | } t||dz | dz f| z||dz | dz f| z||dz | dz f| z||| f<t||dz | f|jz ||dz | f|jz ||| f<t||| dz f|jz ||| dz f|jz ||| f<یd|j D\}} t||| f||| f||| fS) Nz'Please, install numpy for Gotoh measurerr]r)rrz-infrc3 K|] }|dz V dS)rNr*)r+ns r% z!Gotoh.__call__..s&++!A++++++r') rIrr.rAr`rrKrrrbrr}r) r$rCrDlen_s1len_s2d_matp_matq_matrdrerrsim_vals r%r5zGotoh.__call__sV IGHH H$$R,,B RR VaZ!4EKHHH VaZ!4EKHHH VaZ!4EKHHHd Fmmd Fmmd q&1*%% ) )A--E!Q$K=.4<1q5+AAE!Q$K--E!Q$K=.E!Q$KKq&1*%% B BA--E!Q$K--E!Q$K=.E!Q$K=.4<1q5+AAE!Q$KK!,,,  FAs#Ba000  3--S11!!a%Q,''1!a%Q,''1!a%Q,''1ad "!a%(Odm3!a%(Odl2ad "!QU(Odm3!QU(Odl2ad   ,+u{+++15A;ad U1a4[999r')rrNrT)r6r7r8r9r&rrtr5r*r'r%rrs_ !!!!)))(((0:0:0:0:0:r'rc@eZdZdZdZd dZdZedZdZ d S) rz`strcmp95 similarity http://cpansearch.perl.org/src/SCW/Text-JaroWinkler-0.1/strcmp95.c )$)AE)rI)rO)rU)BV)rr)rr)rr)rr)rr)rr)rY)rr)CG)rF)Wr)rr)XK)SZ)rr)Qr)rr)MN)Lr)rr)PR)rJ)2r)5r)8r)1r)rr)0r)rr)rr)rrFTc"||_||_dSr) long_stringsr#)r$rr#s r%r&zStrCmp95.__init__#s(  r'cdSrrr*rss r%rtzStrCmp95.maximum'rur'c<dt|cxkodkncS)Nr[)ord)chars r% _in_rangezStrCmp95._in_range*s*3t99!!!!r!!!!!r'c|}|}|||}||St|}t|}t t }|jD]\}}d|||f<d|||f<||kr|} |} n|} |} dg| z} dg| z} td| dzdz } d} |dz }t|D]j\}}t|| z d}t|| z|}t||dzD]+}| |dkr|||krd| |<d| |<| dz } n,k| dkrdSdx}}t|D]E\}}| |st||D]}| |dkr|dz}n|||kr|dz }F|dz}d}| | krt|D]}| |dkr| ||s+t|D]\}| |dkr| ||s+||||f|vr>||||||fz }d| |<|dz | z}||z ||z z}|| |z | z z }|dz }|dkr|St| d }d}t||D]0\}}||krn$||krn| rn|dz }1|r||d zd |z zz }|js|S| d kr|S| |dzks d| z| |zkr|S|d r|S| |z dz ||z|dzz dzz }|d |z |zz }|S) Nryrr_rrxg$@g@rzr{rvr|)stripupperr/rArrasp_mxr}rbrBrKrr0isdigitr)r$rCrDr3rradjwtrrrminvs1_flags2_flagnum_comyl1rdrlowlimhilimrern_transn_siminum_simrrress r%r5zStrCmp95.__call__.s XXZZ     XXZZ    ""2r**  MRRC   j  FBE"b&ME"b&MM F??!LDD!LD# $# $1la/!344 qjmm  FAs\)1--FL(#..E6519--  1:??r!u||!"GAJ!"GAJqLGE a<<3Gmm  FAs1: 1f%%  1:??AAE#be||1 Q, '>>6]]  1:??~~be,,v  AqzQ >>"Q%00! 1r!u~U22 eBqE2a5L11F!"GAJ4-')6!Gf$447W$//# S==M aLL B   HCAvvczz{{}}  FAA  / a#gv. .F   M 199M a!e  q7{TAX55M a5==?? M{Q6F?QU#:Q#>?3<3&& r'N)FT) r6r7r8r9rr&rt staticmethodrr5r*r'r%rrsw E!!!!""\"xxxxxr'rc&eZdZdZd dZdZdZd S) r a* Compute the Hamming distance between the two or more sequences. The Hamming distance is the number of differing items in ordered sequences. http://www.sial.iias.spb.su/files/386-386-1-PB.pdf https://github.com/Yomguithereal/talisman/blob/master/src/metrics/distance/mlipns.js ?r_rTc>||_||_||_||_dSr)r threshold maxmismatchesr#)r$rrrr#s r%r&zMLIPNS.__init__s# "*  r'cdSrrr*rss r%rtzMLIPNS.maximumrur'cb|j|}|j|}||Sd}t|}tt t |}t |rN||jkrC|sdSd||z |z z |jkrdS|dz }|dz}|dz}t |r ||jkC|sdSdS)Nrr) r.r/rr}rrAallrr)r$r2r3 mismatcheshammaxlens r%r5zMLIPNS.__call__s'D'3 ""I.  M gii#Si(()))nn t/A!A!A qFSLF**dn<<q !OJ 1HC aKF)nn t/A!A!A 1qr'N)rr_rTrr*r'r%r r sP!!!! r'r )! collectionsr itertoolsrbaser_Baser_BaseSimilarityrIr__all__rr r r r rrrrr rrdameraurrrrrrrrr*r'r%rs######!!!!!!CBBBBBBBLLLL EEE   IIIIIeIII0A$A$A$A$A$%A$A$A$Hb)b)b)b)b)b)b)b)Jaaaaa/aaaH     ;   OFOFOFOFOFoOFOFOFd,F,F,F,F,FO,F,F,F^F:F:F:F:F:OF:F:F:RQQQQQQQQh&&&&&_&&&R '))kmm 2 2 4 44  tvv{}} "?$$  8:: s %%