rh|5b%SrSSKJr SSKJr SSKrSSKJr SSKrSSK r SSK r SSK r SSK J r SSK Jr SSK Jr SS K Jr \R""S 5rS S SSS .SjrSrSSjrSrSSS.SjrSrSSjrSrS SjrS!Sjr/rS\S'\S:XaSSK r \ R@"5 gg)"az tools for segmenting -- that is, dividing up a score into small, possibly overlapping sections -- for searching across pieces for similarity. Speed notes: this module is definitely a case where running PyPy rather than cPython will give you a 3-5x speedup. If you really want to do lots of comparisons, the `scoreSimilarity` method will use python-Levenshtein if it is installed, unless forceDifflib is set to True. python-Levenshtein can be installed via **pip install python-Levenshtein**. The ratios are very slightly different, but the speedup is between 10 and 100x! (But then PyPy probably won't work.) ) annotations) OrderedDictN)partial)common) converter)corpus) environmentzsearch.segment )segmentLengthsoverlap algorithmjittercBSSKJn Uc URnUR5RR 5nU"USS9upx[ U5n [[R"XU- - 55n [U 5V s/sH oX- -PM n n /n /nU HznU[R"SU-U5- n[SU5n[XS- 5n[X-U 5nXUnXUUS- 4nU RU5 URU5 M| X4$s sn f)ap Translates a monophonic part with measures to a set of segments of length `segmentLengths` (measured in number of notes) with an overlap of `overlap` notes using a conversion algorithm of `algorithm` (default: search.translateStreamToStringNoRhythm). Returns two lists, a list of segments, and a list of tuples of measure start and end numbers that match the segments. If algorithm is None then a default algorithm of music21.search.translateStreamToStringNoRhythm is used >>> luca = corpus.parse('luca/gloria') >>> lucaCantus = luca.parts[0] >>> segments, measureLists = search.segment.translateMonophonicPartToSegments(lucaCantus) >>> segments[0:2] ['HJHEAAEHHCE@JHGECA@A>@A>@A>>> measureLists[0:2] [(1, 12), (7, 18)] >>> segments, measureLists = search.segment.translateMonophonicPartToSegments( ... lucaCantus, ... algorithm=search.translateDiatonicStreamToString) >>> segments[0:2] ['CRJOMTHCQNALRQPAGFEFDLFDCFEMOO', 'EFDLFDCFEMOOONPJDCBJSNTHLBOGFE'] >>> measureLists[0:2] [(1, 12), (7, 18)] r)searchT)returnMeasures)music21rtranslateStreamToStringNoRhythmrecursenotesstreamlenintmathceilrangerandomrandintmaxminappend) inputStreamr r rrrnStream outputStrmeasures totalLengthnumberOfSegmentsi segmentStarts segmentList measureList segmentStart segmentEndcurrentSegment measureTuples P/home/james-whalen/.local/lib/python3.13/site-packages/music21/search/segment.py!translateMonophonicPartToSegmentsr3.s1P:: !!#))002G#GDAIi.K499[W4L%MNO=BCS=TU=T.23=TMUKK% rF{F;; 1l+ *<(&  %%#VsDc xURn/nUH%n[U40UD6upVURUUS.5 M' U$)aQ Creates segment and measure lists for each part of a score Returns list of dictionaries of segment and measure lists >>> bach = corpus.parse('bwv66.6') >>> scoreList = search.segment.indexScoreParts(bach) >>> scoreList[1]['segmentList'][0] '@B@@@@ED@DBDA=BB@?==B@@EBBDBBA' >>> scoreList[1]['measureList'][0:3] [(0, 7), (4, 9), (8, 9)] )r,r-)partsr3r#) scoreFilekeywordsscoreFileParts indexedListpartr,r-s r2indexScorePartsr;tsU__NK#D $$ &&   Fc  [U[R5(d[R"U5nURn[ U40UD6nX4U4$![ a&nU(d[ SUSU35 SnSnAN-UeSnAff=f)z- Index one path in the context of multicore. zFailed on parse/index for, z: N) isinstancepathlibPathname indexOnePath Exceptionprint)filePathfailFastr7shortFp indexOutputes r2_indexSingleMulticorerKs h - -<<)mmG"88x8  ( ++  /zA3? @KG s A BA><A>>Bc 4[SUSSUSUS35 g)NzIndexed r (/))rE)numRuntotalRun latestOutputs r2_giveUpdatesMulticorerSs$ H\!_%Rxq ! <=r<T) giveUpdates runMulticorec USLa[nOSn[[40UD6n[[ U55H@n[ X[ R5(aM&[ R"X5X'MB U(a[R"UUUS9nO[R"UUUS9n0nUH upn X4X'M /n UHn U RX5 M [U 5nU$)au Returns a dictionary of the lists from indexScoreParts for each score in scoreFilePaths >>> #_DOCS_SHOW searchResults = corpus.search('bwv190') >>> searchResults = corpus.corpora.CoreCorpus().search('bwv190') #_DOCS_HIDE >>> fpsNamesOnly = sorted([searchResult.sourcePath for searchResult in searchResults]) >>> len(fpsNamesOnly) 2 >>> scoreDict = search.segment.indexScoreFilePaths(fpsNamesOnly) >>> len(scoreDict['bwv190.7.mxl']) 4 >>> scoreDict['bwv190.7.mxl'][0]['measureList'] [(0, 9), (6, 15), (11, 20), (17, 25), (22, 31), (27, 32)] >>> scoreDict['bwv190.7.mxl'][0]['segmentList'][0] 'NNJLNOLLLJJIJLLLLNJJJIJLLJNNJL' TN)updateFunction) rSrrKrrr?r@rAr runParallelrunNonParallelr#r)scoreFilePathsrTrUr7rW indexFuncr*rpListUnOrderedrpDict outShortNameoutDataoriginalPathlibrpListp scoreDicts r2indexScoreFilePathsrds4d.-::I 3~& '.+W\\:: ' ^-> ?N ( ,,  )+ !//  )+ F2A. #/"93BF  fi F#I r<c  [U[R5(d[R"U5nUR5(d[R "U5nO[ R "U5n[U40UD6nU$)z0 Index a single path. Returns a scoreDictEntry )r?r@rA is_absoluterparserr;)rFr7scoreObjscoreDictEntrys r2rCrCsc h - -<<)    ! !<<)??8,$X::N r<cUc[RS5nO1[U[[45(a[ R "U5nURS5n[R"X5 SSS5 U$!,(df  U$=f)z Save the score dict from indexScoreFilePaths as a .json file for quickly reloading Returns the filepath (assumes you'll probably be using a temporary file) as a pathlib.Path() Nz.jsonwb) environLocal getTempFiler?strbytesr@rAopenjsondump)rcrFfs r2 saveScoreDictrtso++G4 HsEl + +<<) t  )  O   Os A== B c[U[R5(d[R"U5nURS5n[R "U5nSSS5 U$!,(df  W$=f)z) Load the scoreDictionary from filePath. rbN)r?r@rArprqload)rFrsrcs r2 loadScoreDictrxsX h - -<<) t IIaL      s A(( A7cUSLa[R"USU5nU$SSKJn URUSU5nU$![a [R"USU5nU$f=f)z Returns either a difflib.SequenceMatcher or pyLevenshtein StringMatcher.StringMatcher object depending on what is installed. If forceDifflib is True then use difflib even if pyLevenshtein is installed: Tr>r) StringMatcher)difflibSequenceMatcher Levenshteinrz ImportError)seq2junk forceDifflibsmObject pyLevenshteins r2getDifflibOrPyLevrstt**4T: O  ? B$224TBH O ?..tR>H O ?s<"A"!A"c ^^^^^ ^ ^ ^ ^ ^^^/m Sm [T5m[TR55m Sm Sm UUUU UU U U U UUU4 Sjn[T5HnT UmTTnT S- m USLa[ STST STS 35 [[U55HBm [ UT S 5H*um n[U5T:aMUT S T mU"U5 M, MD M T $) a Find the level of similarity between each pair of segments in a scoreDict. This takes twice as long as it should because it does not cache the pairwise similarity. >>> filePaths = [] >>> for p in ('bwv197.5.mxl', 'bwv190.7.mxl', 'bwv197.10.mxl'): ... #_DOCS_SHOW source = corpus.search(p)[0].sourcePath ... source = corpus.corpora.CoreCorpus().search(p)[0].sourcePath #_DOCS_HIDE ... filePaths.append(source) >>> scoreDict = search.segment.indexScoreFilePaths(filePaths) >>> scoreSim = search.segment.scoreSimilarity(scoreDict, forceDifflib=True) #_DOCS_HIDE >>> #_DOCS_SHOW scoreSim = search.segment.scoreSimilarity(scoreDict) >>> len(scoreSim) 496 Returns a list of tuples of first score name, first score voice number, first score measure number, second score name, second score voice number, second score measure number, and similarity score (0 to 1). >>> for result in scoreSim[133:137]: ... result ('bwv197.5.mxl', 1, 1, (4, 10), 'bwv190.7.mxl', 3, 4, (22, 30), 0.13...) ('bwv197.5.mxl', 1, 1, (4, 10), 'bwv197.10.mxl', 0, 0, (0, 8), 0.2) ('bwv197.5.mxl', 1, 1, (4, 10), 'bwv197.10.mxl', 1, 0, (0, 7), 0.266...) ('bwv197.5.mxl', 1, 1, (4, 10), 'bwv197.10.mxl', 1, 1, (4, 9), 0.307...) rNc > [UT S9n[TT5HnTUnTUn[[U55Hn[XES5Hupg[U5T:aMUR U5 UR 5nXESUn TTTTUUUU U4 n TR U 5 T (dMfUUUU TTTTU4 n TR U 5 M M M g)N)rr,r-)rrr enumerateset_seq1ratior#) thisSegmentdlthatScoreNumber thatScoreKey thatScorepNum2thatSegmentNumber thatSegmentrthatMeasureNumbersimilarityTuplesimilarityTupleReversedrincludeReverse minimumLengthpNumrc scoreDictKeys scoreIndex segmentNumbersimilarityScoresthisMeasureNumber thisScoreKey totalScoress r2 doOneSegment%scoreSimilarity..doOneSegmentTs  { F$Z=O(9L!,/Is9~.6?!(7792%;'-7 KK ,HHJE(1(8(GHY(Z%$%)$)) 'O%++O<) $))$%) /+%++,CDA79/ >r<rTz Comparing rMrNrOr,r-)rlistkeysrrEr)rcrrTrrrthisScoreNumber thisScorethisSegmentOuterrrrrrrrrs`` `` @@@@@@@@r2scoreSimilarityr)sHJi.K)*M DM'E'E'ER!-$_5 l+ a $  J|nBzl!K=J K#i.)D3rs ##  &&'78   B&L0,(>%*%);|  &   2 cP J z r<