def _trydecode(self, data, codec: Optional[str], width: int,
linecount: int) -> str:
remaining = linecount
result = []
if codec is None:
from refinery.units.encoding.esc import esc
decoded = data[:abs(width * linecount)]
decoded = str(decoded | -esc(bare=True))
limit = min(abs(linecount) * width, len(decoded))
for k in range(0, limit, width):
result.append(decoded[k:k + width])
return result
try:
import unicodedata
unprintable = {'Cc', 'Cf', 'Co', 'Cs'}
self.log_info(F'trying to decode as {codec}.')
decoded = codecs.decode(data, codec, errors='strict')
count = sum(
unicodedata.category(c) not in unprintable for c in decoded)
ratio = count / len(decoded)
except UnicodeDecodeError as DE:
self.log_info('decoding failed:', DE.reason)
return None
except ValueError as V:
self.log_info('decoding failed:', V)
return None
if ratio < 0.8:
self.log_info(
F'data contains {ratio * 100:.2f}% printable characters, this is too low.'
)
return None
for paragraph in decoded.splitlines(False):
if not remaining:
break
wrapped = [
line for chunk in textwrap.wrap(
paragraph,
width,
break_long_words=True,
break_on_hyphens=False,
drop_whitespace=False,
expand_tabs=True,
max_lines=abs(remaining + 1),
replace_whitespace=False,
tabsize=4,
) for line in chunk.splitlines(keepends=False)
]
remaining -= len(wrapped)
result.extend(wrapped)
return result[:abs(linecount)]
def test_quoted_string_01(self):
unit = esc(quoted=True)
self.assertEqual(unit.process(RB'"r\x65\x66\x69\x6ee\x72\x79"'),
B'refinery')
def test_quoted_string_04(self):
unit = esc(quoted=True)
with self.assertRaises(ValueError):
unit(RB'"r\x65\x66\x69\x6ee\x72\x79')
def test_quoted_string_03(self):
unit = esc(quoted=True, hex=False)
result = unit.reverse(B'binary\n\a\t.."refinery"!')
self.assertEqual(result, BR'"binary\n\a\t..\"refinery\"!"')
def test_quoted_string_02(self):
unit = esc(quoted=True, hex=True)
result = unit.reverse(RB'refinery')
self.assertEqual(result, BR'"\x72\x65\x66\x69\x6e\x65\x72\x79"')
def REG_SZ(data: str) -> bytes:
return data.encode(self.codec) | esc(quoted=True) | bytes