def _to_textio(fp: IO, mode: str, read_codec: str) -> TextIO:
if 'b' in mode:
fp = cast(TextIO, fp)
# TODO: FIx me
fp.decoder = read_codec
fp.native_reader = fp.read
fp.read = lambda *args: _auto_decode(fp, *args)
if getattr(fp, 'native_closer', None):
fp.native_closer = fp.close
fp.close = lambda *a: _wrapped_close(fp)
return fp
def _auto_decode(fp: IO, nbytes: Optional[int] = None) -> str:
# We have a file opened in binary mode and haven't formally established a decoder.
# We may need as many as four bytes to determine the encoding.
# TODO: should the BOM (if present) count in nbytes?
if not fp.decoder:
if nbytes and nbytes < 4:
# Nothing we can do if they're asking for less than 4 bytes
data = fp.native_reader(nbytes)
return data.decode() if data else data
# Use the BOM to figure out what is toing on. If no BOM, we always go to UTF-8
bom = fp.native_reader(4)
if len(bom) < 4:
return bom.decode(fp.decoder if fp.decoder else 'utf-8')
fp.decoder = json.detect_encoding(bom)
if nbytes is None:
return (bom + fp.native_reader()).decode(fp.decoder)
else:
return (bom + fp.native_reader(nbytes - 4)).decode(fp.decoder)
else:
return fp.native_reader(nbytes).decode(fp.decoder)
