Exemple #1
0
    def findWorks(self):
        '''
        populate other information about the directory such as
        files and filenames.


        >>> di = corpus.work.DirectoryInformation('schoenberg',
        ...             corpusObject=corpus.corpora.CoreCorpus())
        >>> di.findWorks()
        OrderedDict([(...'opus19', CorpusWork(title='Opus 19',
                                       files=[CorpusFile(path='schoenberg/opus19/movement2.mxl',
                                                         title='Movement 2',
                                                         filename='movement2.mxl',
                                                         format='musicxml',
                                                         ext='.mxl'),
                                              CorpusFile(path='schoenberg/opus19/movement6.mxl',
                                                         title='Movement 6',
                                                         filename='movement6.mxl',
                                                         format='musicxml',
                                                         ext='.mxl')],
                                        virtual=False))])
        '''
        self.works.clear()
        works = self.corpusObject.getComposer(self.directoryName)
        # TODO: this should be renamed since not all are composers
        for path in works:
            # split by the composer dir to get relative path
            #environLocal.printDebug(['dir composer', composerDirectory, path])
            junk, fileStub = path.as_posix().split(self.directoryName)
            if fileStub.startswith(os.sep):
                fileStub = fileStub[len(os.sep):]
            # break into file components
            fileComponents = fileStub.split(os.sep)
            # the first is either a directory for containing components
            # or a top-level name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            if ext is None:
                #environLocal.printDebug([
                #    'file that does not seem to have an extension',
                #    ext, path])
                continue
            # if not a file w/ ext, we will get None for format
            if m21Format is None:
                workStub = fileComponents[0]
            else:  # remove the extension
                workStub = fileComponents[0].replace(ext, '')
            # create list location if not already added
            if workStub not in self.works:
                title = common.spaceCamelCase(workStub).title()
                self.works[workStub] = CorpusWork(title=title,
                                                  files=[],
                                                  virtual=False)
            # last component is name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            # all path parts after corpus
            corpusPath = os.path.join(str(self.directoryName), fileStub)
            corpusFileName = fileComponents[-1]  # all after
            title = None
            # this works but takes a long time!
            # title = converter.parse(path).metadata.title
            ## TODO: get from RichMetadataBundle!
            if title is None:
                title = common.spaceCamelCase(fileComponents[-1].replace(
                    ext, ''))
                title = title.title()
            fileTuple = CorpusFile(path=corpusPath,
                                   title=title,
                                   filename=corpusFileName,
                                   format=m21Format,
                                   ext=ext)
            self.works[workStub].files.append(fileTuple)
            # add this path
        return self.works
Exemple #2
0
    def findWorks(self):
        '''
        populate other information about the directory such as
        files and filenames.


        >>> di = corpus.work.DirectoryInformation('schoenberg',
        ...             corpusObject=corpus.corpora.CoreCorpus())
        >>> di.findWorks()
        OrderedDict([(...'opus19', CorpusWork(title='Opus 19',
                                       files=[CorpusFile(path='schoenberg/opus19/movement2.mxl',
                                                         title='Movement 2',
                                                         filename='movement2.mxl',
                                                         format='musicxml',
                                                         ext='.mxl'),
                                              CorpusFile(path='schoenberg/opus19/movement6.mxl',
                                                         title='Movement 6',
                                                         filename='movement6.mxl',
                                                         format='musicxml',
                                                         ext='.mxl')],
                                        virtual=False))])
        '''
        self.works.clear()
        works = self.corpusObject.getComposer(self.directoryName)
                # TODO: this should be renamed since not all are composers
        for path in works:
            # split by the composer dir to get relative path
            #environLocal.printDebug(['dir composer', composerDirectory, path])
            junk, fileStub = path.as_posix().split(self.directoryName)
            if fileStub.startswith(os.sep):
                fileStub = fileStub[len(os.sep):]
            # break into file components
            fileComponents = fileStub.split(os.sep)
            # the first is either a directory for containing components
            # or a top-level name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            if ext is None:
                #environLocal.printDebug([
                #    'file that does not seem to have an extension',
                #    ext, path])
                continue
            # if not a file w/ ext, we will get None for format
            if m21Format is None:
                workStub = fileComponents[0]
            else:  # remove the extension
                workStub = fileComponents[0].replace(ext, '')
            # create list location if not already added
            if workStub not in self.works:
                title = common.spaceCamelCase(workStub).title()
                self.works[workStub] = CorpusWork(title=title, files=[], virtual=False)
            # last component is name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            # all path parts after corpus
            corpusPath = os.path.join(str(self.directoryName), fileStub)
            corpusFileName = fileComponents[-1]  # all after
            title = None
            # this works but takes a long time!
            # title = converter.parse(path).metadata.title
            ## TODO: get from RichMetadataBundle!
            if title is None:
                title = common.spaceCamelCase(
                    fileComponents[-1].replace(ext, ''))
                title = title.title()
            fileTuple = CorpusFile(path=corpusPath, title=title, filename=corpusFileName,
                                   format=m21Format, ext=ext)
            self.works[workStub].files.append(fileTuple)
            # add this path
        return self.works
Exemple #3
0
def getWorkReferences(sort=True):
    '''
    Return a data dictionary for all works in the corpus and (optionally) the
    virtual corpus. Returns a list of reference dictionaries, each each
    dictionary for a each composer. A 'works' dictionary for each composer
    provides references to dictionaries for all associated works.

    This is used in the generation of corpus documentation

    >>> post = corpus.getWorkReferences()

    '''
    # from music21 import corpus; corpus.getWorkReferences()
    # TODO: update this to use metadata
    results = []
    for composerDirectory, composer in corpora.CoreCorpus._composers:
        ref = {}
        ref['composer'] = composer
        ref['composerDir'] = composerDirectory
        ref['works'] = {}  # store by keys of name/dirname
        works = getComposer(composerDirectory)
        for path in works:
            # split by the composer dir to get relative path
            #environLocal.printDebug(['dir composer', composerDirectory, path])
            junk, fileStub = path.split(composerDirectory)
            if fileStub.startswith(os.sep):
                fileStub = fileStub[len(os.sep):]
            # break into file components
            fileComponents = fileStub.split(os.sep)
            # the first is either a directory for containing components
            # or a top-level name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            if ext is None:
                #environLocal.printDebug([
                #    'file that does not seem to have an extension',
                #    ext, path])
                continue
            # if not a file w/ ext, we will get None for format
            if m21Format is None:
                workStub = fileComponents[0]
            else:  # remove the extension
                workStub = fileComponents[0].replace(ext, '')
            # create list location if not already added
            if workStub not in ref['works']:
                ref['works'][workStub] = {}
                ref['works'][workStub]['files'] = []
                title = common.spaceCamelCase(workStub).title()
                ref['works'][workStub]['title'] = title
                ref['works'][workStub]['virtual'] = False
            # last component is name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            fileDict = {}
            fileDict['format'] = m21Format
            fileDict['ext'] = ext
            # all path parts after corpus
            fileDict['corpusPath'] = os.path.join(composerDirectory, fileStub)
            fileDict['fileName'] = fileComponents[-1]  # all after
            title = None
            # this works but takes a long time!
#             if format == 'musicxml':
#                 mxDocument = musicxml.Document()
#                 mxDocument.open(path)
#                 title = mxDocument.getBestTitle()
            if title is None:
                title = common.spaceCamelCase(
                    fileComponents[-1].replace(ext, ''))
                title = title.title()
            fileDict['title'] = title
            ref['works'][workStub]['files'].append(fileDict)
            # add this path
        results.append(ref)
    # get each VirtualWork object
    for vw in corpora.VirtualCorpus._virtual_works:
        composerDir = vw.corpusPath.split('/')[0]
        match = False
        for ref in results:
            # check composer reference or first part of corpusPath
            if (ref['composer'] == vw.composer or
                composerDir == ref['composerDir']):
                match = True
                break  # use this ref
        if not match:  # new composers, create a new ref
            ref = {}
            ref['composer'] = vw.composer
            ref['composerDir'] = composerDir
            ref['works'] = {}  # store by keys of name/dirname
        # work stub should be everything other than top-level
        workStub = vw.corpusPath.replace(composerDir + '/', '')
        ref['works'][workStub] = {}
        ref['works'][workStub]['virtual'] = True
        ref['works'][workStub]['files'] = []
        ref['works'][workStub]['title'] = vw.title
        for url in vw.urlList:
            m21Format, ext = common.findFormatExtURL(url)
            fileDict = {}
            fileDict['format'] = m21Format
            fileDict['ext'] = ext
            # all path parts after corpus
            fileDict['corpusPath'] = vw.corpusPath
            fileDict['title'] = vw.title
            fileDict['url'] = url
            ref['works'][workStub]['files'].append(fileDict)
        if not match:  # not found already, need to add
            results.append(ref)
    if sort:
        sortGroup = []
        for ref in results:
            sortGroupSub = []
            for workStub in ref['works']:
                # add title first for sorting
                sortGroupSub.append([
                    ref['works'][workStub]['title'],
                    workStub,
                    ])
            sortGroupSub.sort()
            ref['sortedWorkKeys'] = [y for unused_x, y in sortGroupSub]
            # prepare this sort group
            sortGroup.append([ref['composerDir'], ref])
        sortGroup.sort()
        results = [ref for junk, ref in sortGroup]
    return results
Exemple #4
0
def getWorkReferences(sort=True):
    '''
    Return a data dictionary for all works in the corpus and (optionally) the
    virtual corpus. Returns a list of reference dictionaries, each each
    dictionary for a each composer. A 'works' dictionary for each composer
    provides references to dictionaries for all associated works.

    This is used in the generation of corpus documentation

    ::

        >>> from music21 import corpus
        >>> post = corpus.getWorkReferences()

    '''
    # from music21 import corpus; corpus.getWorkReferences()
    # TODO: update this to use metadata
    results = []
    for composerDirectory, composer in corpora.CoreCorpus._composers:
        ref = {}
        ref['composer'] = composer
        ref['composerDir'] = composerDirectory
        ref['works'] = {}  # store by keys of name/dirname
        works = getComposer(composerDirectory)
        for path in works:
            # split by the composer dir to get relative path
            #environLocal.printDebug(['dir composer', composerDirectory, path])
            junk, fileStub = path.split(composerDirectory)
            if fileStub.startswith(os.sep):
                fileStub = fileStub[len(os.sep):]
            # break into file components
            fileComponents = fileStub.split(os.sep)
            # the first is either a directory for containing components
            # or a top-level name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            if ext is None:
                #environLocal.printDebug([
                #    'file that does not seem to have an extension',
                #    ext, path])
                continue
            # if not a file w/ ext, we will get None for format
            if m21Format is None:
                workStub = fileComponents[0]
            else:  # remove the extension
                workStub = fileComponents[0].replace(ext, '')
            # create list location if not already added
            if workStub not in ref['works']:
                ref['works'][workStub] = {}
                ref['works'][workStub]['files'] = []
                title = common.spaceCamelCase(workStub).title()
                ref['works'][workStub]['title'] = title
                ref['works'][workStub]['virtual'] = False
            # last component is name
            m21Format, ext = common.findFormatExtFile(fileComponents[-1])
            fileDict = {}
            fileDict['format'] = m21Format
            fileDict['ext'] = ext
            # all path parts after corpus
            fileDict['corpusPath'] = os.path.join(composerDirectory, fileStub)
            fileDict['fileName'] = fileComponents[-1]  # all after
            title = None
            # this works but takes a long time!
            #             if format == 'musicxml':
            #                 mxDocument = musicxml.Document()
            #                 mxDocument.open(path)
            #                 title = mxDocument.getBestTitle()
            if title is None:
                title = common.spaceCamelCase(fileComponents[-1].replace(
                    ext, ''))
                title = title.title()
            fileDict['title'] = title
            ref['works'][workStub]['files'].append(fileDict)
            # add this path
        results.append(ref)
    # get each VirtualWork object
    for vw in corpora.VirtualCorpus._virtual_works:
        composerDir = vw.corpusPath.split('/')[0]
        match = False
        for ref in results:
            # check composer reference or first part of corpusPath
            if (ref['composer'] == vw.composer
                    or composerDir == ref['composerDir']):
                match = True
                break  # use this ref
        if not match:  # new composers, create a new ref
            ref = {}
            ref['composer'] = vw.composer
            ref['composerDir'] = composerDir
            ref['works'] = {}  # store by keys of name/dirname
        # work stub should be everything other than top-level
        workStub = vw.corpusPath.replace(composerDir + '/', '')
        ref['works'][workStub] = {}
        ref['works'][workStub]['virtual'] = True
        ref['works'][workStub]['files'] = []
        ref['works'][workStub]['title'] = vw.title
        for url in vw.urlList:
            m21Format, ext = common.findFormatExtURL(url)
            fileDict = {}
            fileDict['format'] = m21Format
            fileDict['ext'] = ext
            # all path parts after corpus
            fileDict['corpusPath'] = vw.corpusPath
            fileDict['title'] = vw.title
            fileDict['url'] = url
            ref['works'][workStub]['files'].append(fileDict)
        if not match:  # not found already, need to add
            results.append(ref)
    if sort:
        sortGroup = []
        for ref in results:
            sortGroupSub = []
            for workStub in ref['works']:
                # add title first for sorting
                sortGroupSub.append([
                    ref['works'][workStub]['title'],
                    workStub,
                ])
            sortGroupSub.sort()
            ref['sortedWorkKeys'] = [y for unused_x, y in sortGroupSub]
            # prepare this sort group
            sortGroup.append([ref['composerDir'], ref])
        sortGroup.sort()
        results = [ref for junk, ref in sortGroup]
    return results