コード例 #1
0
def get_metadata(stream):
    """ Return metadata as a L{MetaInfo} object """
    title = 'Unknown'
    mi = MetaInformation(title, ['Unknown'])
    stream.seek(0)
    try:
        if stream.read(10) not in MAGIC:
            print >>sys.stderr, u'Couldn\'t read IMP header from file'
            return mi

        def cString(skip=0):
            result = ''
            while 1:
                data = stream.read(1)
                if data == '\x00':
                    if not skip: return result
                    skip -= 1
                    result, data = '', ''
                result += data

        stream.read(38) # skip past some uninteresting headers
        _, category, title, author = cString(), cString(), cString(1), cString(2)

        if title:
            mi.title = title
        if author:
            mi.authors = string_to_authors(author)
            mi.author = author
        if category:
            mi.category = category
    except Exception as err:
        msg = u'Couldn\'t read metadata from imp: %s with error %s'%(mi.title, unicode(err))
        print >>sys.stderr, msg.encode('utf8')
    return mi
コード例 #2
0
ファイル: meta.py プロジェクト: onyx-Sean/calibre
def get_metadata(stream):
    """
    Return basic meta-data about the LRF file in C{stream} as a
    L{MetaInformation} object.
    @param stream: A file like object or an instance of L{LRFMetaFile}
    """
    lrf = stream if isinstance(stream, LRFMetaFile) else LRFMetaFile(stream)
    authors = string_to_authors(lrf.author)
    mi = MetaInformation(lrf.title.strip(), authors)
    mi.author = lrf.author.strip()
    mi.comments = lrf.free_text.strip()
    mi.category = lrf.category.strip() + ', ' + lrf.classification.strip()
    tags = [x.strip() for x in mi.category.split(',') if x.strip()]
    if tags:
        mi.tags = tags
    if mi.category.strip() == ',':
        mi.category = None
    mi.publisher = lrf.publisher.strip()
    mi.cover_data = lrf.get_cover()
    try:
        mi.title_sort = lrf.title_reading.strip()
        if not mi.title_sort:
            mi.title_sort = None
    except:
        pass
    try:
        mi.author_sort = lrf.author_reading.strip()
        if not mi.author_sort:
            mi.author_sort = None
    except:
        pass
    if not mi.title or 'unknown' in mi.title.lower():
        mi.title = None
    if not mi.authors:
        mi.authors = None
    if not mi.author or 'unknown' in mi.author.lower():
        mi.author = None
    if not mi.category or 'unknown' in mi.category.lower():
        mi.category = None
    if not mi.publisher or 'unknown' in mi.publisher.lower() or \
            'some publisher' in mi.publisher.lower():
        mi.publisher = None

    return mi
コード例 #3
0
ファイル: meta.py プロジェクト: Farb/calibre
def get_metadata(stream):
    """
    Return basic meta-data about the LRF file in C{stream} as a
    L{MetaInformation} object.
    @param stream: A file like object or an instance of L{LRFMetaFile}
    """
    lrf = stream if isinstance(stream, LRFMetaFile) else LRFMetaFile(stream)
    authors = string_to_authors(lrf.author)
    mi = MetaInformation(lrf.title.strip(), authors)
    mi.author = lrf.author.strip()
    mi.comments = lrf.free_text.strip()
    mi.category = lrf.category.strip()+', '+lrf.classification.strip()
    tags = [x.strip() for x in mi.category.split(',') if x.strip()]
    if tags:
        mi.tags = tags
    if mi.category.strip() == ',':
        mi.category = None
    mi.publisher = lrf.publisher.strip()
    mi.cover_data = lrf.get_cover()
    try:
        mi.title_sort = lrf.title_reading.strip()
        if not mi.title_sort:
            mi.title_sort = None
    except:
        pass
    try:
        mi.author_sort = lrf.author_reading.strip()
        if not mi.author_sort:
            mi.author_sort = None
    except:
        pass
    if not mi.title or 'unknown' in mi.title.lower():
        mi.title = None
    if not mi.authors:
        mi.authors = None
    if not mi.author or 'unknown' in mi.author.lower():
        mi.author = None
    if not mi.category or 'unknown' in mi.category.lower():
        mi.category = None
    if not mi.publisher or 'unknown' in mi.publisher.lower() or \
            'some publisher' in mi.publisher.lower():
        mi.publisher = None

    return mi
コード例 #4
0
def get_metadata(stream, extract_cover=True):
    '''
    Return metadata as a L{MetaInfo} object
    '''
    mi = MetaInformation(_('Unknown'), [_('Unknown')])
    stream.seek(0)

    pheader = PdbHeaderReader(stream)
    section_data = None
    for i in range(1, pheader.num_sections):
        raw_data = pheader.section_data(i)
        section_header = SectionHeader(raw_data)
        if section_header.type == DATATYPE_METADATA:
            section_data = raw_data[8:]
            break

    if not section_data:
        return mi

    default_encoding = 'latin-1'
    record_count, = struct.unpack('>H', section_data[0:2])
    adv = 0
    title = None
    author = None
    pubdate = 0
    for i in range(record_count):
        try:
            type, length = struct.unpack_from('>HH', section_data, 2 + adv)
        except struct.error:
            break

        # CharSet
        if type == 1:
            val, = struct.unpack('>H', section_data[6+adv:8+adv])
            default_encoding = MIBNUM_TO_NAME.get(val, 'latin-1')
        # Author
        elif type == 4:
            author = section_data[6+adv+(2*length)]
        # Title
        elif type == 5:
            title = section_data[6+adv+(2*length)]
        # Publication Date
        elif type == 6:
            pubdate, = struct.unpack('>I', section_data[6+adv:6+adv+4])

        adv += 2*length

    if title:
        mi.title = title.replace('\0', '').decode(default_encoding, 'replace')
    if author:
        author = author.replace('\0', '').decode(default_encoding, 'replace')
        mi.author = author.split(',')
    mi.pubdate = datetime.fromtimestamp(pubdate)

    return mi
コード例 #5
0
ファイル: plucker.py プロジェクト: MarioJC/calibre
def get_metadata(stream, extract_cover=True):
    '''
    Return metadata as a L{MetaInfo} object
    '''
    mi = MetaInformation(_('Unknown'), [_('Unknown')])
    stream.seek(0)

    pheader = PdbHeaderReader(stream)
    section_data = None
    for i in range(1, pheader.num_sections):
        raw_data = pheader.section_data(i)
        section_header = SectionHeader(raw_data)
        if section_header.type == DATATYPE_METADATA:
            section_data = raw_data[8:]
            break

    if not section_data:
        return mi

    default_encoding = 'latin-1'
    record_count, = struct.unpack('>H', section_data[0:2])
    adv = 0
    title = None
    author = None
    pubdate = 0
    for i in xrange(record_count):
        try:
            type, length = struct.unpack_from('>HH', section_data, 2 + adv)
        except struct.error:
            break

        # CharSet
        if type == 1:
            val, = struct.unpack('>H', section_data[6+adv:8+adv])
            default_encoding = MIBNUM_TO_NAME.get(val, 'latin-1')
        # Author
        elif type == 4:
            author = section_data[6+adv+(2*length)]
        # Title
        elif type == 5:
            title = section_data[6+adv+(2*length)]
        # Publication Date
        elif type == 6:
            pubdate, = struct.unpack('>I', section_data[6+adv:6+adv+4])

        adv += 2*length

    if title:
        mi.title = title.replace('\0', '').decode(default_encoding, 'replace')
    if author:
        author = author.replace('\0', '').decode(default_encoding, 'replace')
        mi.author = author.split(',')
    mi.pubdate = datetime.fromtimestamp(pubdate)

    return mi
コード例 #6
0
ファイル: rb.py プロジェクト: wynick27/calibre
def get_metadata(stream):
    """ Return metadata as a L{MetaInfo} object """
    title = 'Unknown'
    mi = MetaInformation(title, ['Unknown'])
    stream.seek(0)
    try:
        if not stream.read(14) == MAGIC:
            print >> sys.stderr, u'Couldn\'t read RB header from file'
            return mi
        stream.read(10)

        read_i32 = lambda: struct.unpack('<I', stream.read(4))[0]

        stream.seek(read_i32())
        toc_count = read_i32()

        for i in range(toc_count):
            stream.read(32)
            length, offset, flag = read_i32(), read_i32(), read_i32()
            if flag == 2:
                break
        else:
            print >> sys.stderr, u'Couldn\'t find INFO from RB file'
            return mi

        stream.seek(offset)
        info = stream.read(length).splitlines()
        for line in info:
            if '=' not in line:
                continue
            key, value = line.split('=')
            if key.strip() == 'TITLE':
                mi.title = value.strip()
            elif key.strip() == 'AUTHOR':
                mi.author = value
                mi.authors = string_to_authors(value)
    except Exception as err:
        msg = u'Couldn\'t read metadata from rb: %s with error %s' % (
            mi.title, unicode(err))
        print >> sys.stderr, msg.encode('utf8')
        raise
    return mi
コード例 #7
0
ファイル: rb.py プロジェクト: artbycrunk/calibre
def get_metadata(stream):
    """ Return metadata as a L{MetaInfo} object """
    title = 'Unknown'
    mi = MetaInformation(title, ['Unknown'])
    stream.seek(0)
    try:
        if not stream.read(14) == MAGIC:
            print(u'Couldn\'t read RB header from file', file=sys.stderr)
            return mi
        stream.read(10)

        read_i32 = lambda: struct.unpack('<I', stream.read(4))[0]

        stream.seek(read_i32())
        toc_count = read_i32()

        for i in range(toc_count):
            stream.read(32)
            length, offset, flag = read_i32(), read_i32(), read_i32()
            if flag == 2:
                break
        else:
            print(u'Couldn\'t find INFO from RB file', file=sys.stderr)
            return mi

        stream.seek(offset)
        info = stream.read(length).splitlines()
        for line in info:
            if '=' not in line:
                continue
            key, value = line.split('=')
            if key.strip() == 'TITLE':
                mi.title = value.strip()
            elif key.strip() == 'AUTHOR':
                mi.author = value
                mi.authors = string_to_authors(value)
    except Exception as err:
        msg = u'Couldn\'t read metadata from rb: %s with error %s'%(mi.title, unicode(err))
        print(msg.encode('utf8'), file=sys.stderr)
        raise
    return mi