Class Text

src/summarycode/copyright_tallies.py:157–187

# keep track of an original text value and the corresponding clustering "key"
@attr.attributes(slots=True)
class Text(object):
    # cleaned, normalized, clustering text for a copyright holder
    key = attr.attrib()
    # original text for a copyright holder
    original = attr.attrib()
    # count of occurences of a text
    count = attr.attrib(default=1)

    def normalize(self):
        if TRACE_TEXT:
            logger_debug('Text.normalize:', self)
        key = self.key.lower()
        key = ' '.join(key.split())
        key = key.strip('.,').strip()
        key = clean(key)
        self.key = key.strip('.,').strip()

    def transliterate(self):
        self.key = toascii(self.key, translit=True)

    def fingerprint(self):
        key = self.key
        if not isinstance(key, str):
            key = unidecode(key)
        fp = fingerprints.generate(key)

        if TRACE_TEXT or TRACE_FP:
            logger_debug('Text.fingerprint:key: ', repr(self.key))
            logger_debug('Text.fingerprint:fp :    ', fingerprints.generate(unidecode(self.key)))

        self.key = fp


def tally_copyrights(texts, _detector=CopyrightDetector()):