def _parse_dictionary()

in torchaudio/datasets/cmudict.py [0:0]

• 20 lines of code
• 12 McCabe index (conditional complexity)

def _parse_dictionary(lines: Iterable[str], exclude_punctuations: bool) -> List[str]:
    _alt_re = re.compile(r"\([0-9]+\)")
    cmudict: List[Tuple[str, List[str]]] = list()
    for line in lines:
        if not line or line.startswith(";;;"):  # ignore comments
            continue

        word, phones = line.strip().split("  ")
        if word in _PUNCTUATIONS:
            if exclude_punctuations:
                continue
            # !EXCLAMATION-POINT -> !
            # --DASH -> --
            # ...ELLIPSIS -> ...
            if word.startswith("..."):
                word = "..."
            elif word.startswith("--"):
                word = "--"
            else:
                word = word[0]

        # if a word have multiple pronunciations, there will be (number) appended to it
        # for example, DATAPOINTS and DATAPOINTS(1),
        # the regular expression `_alt_re` removes the '(1)' and change the word DATAPOINTS(1) to DATAPOINTS
        word = re.sub(_alt_re, "", word)
        phones = phones.split(" ")
        cmudict.append((word, phones))

    return cmudict