monkee.ch/SickGear
1
0
Fork 0
mirror of https://github.com/SickGear/SickGear.git synced 2024-11-21 20:35:05 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'feature/UpdateTorrentParser' into dev

Browse source
This commit is contained in:
JackDandy 2023-02-09 14:23:20 +00:00
commit 9f0c04a374
2 changed files with 353 additions and 155 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
CHANGES.md
View file

@ -4,6 +4,7 @@
* Add Filelock 3.9.0 (ce3e891)
* Remove Lockfile no longer used by Cachecontrol
* Update Msgpack 1.0.0 (fa7d744) to 1.0.4 (b5acfd5)
* Update torrent_parser 0.3.0 (2a4eecb) to 0.4.0 (23b9e11)
* Update unidecode module 1.1.1 (632af82) to 1.3.6 (4141992)

493
lib/torrent_parser/torrent_parser.py
View file

@ -54,34 +54,39 @@ try:
# noinspection PyPackageRequirements
from chardet import detect as _detect
except ImportError:
def _detect(_):
warnings.warn("No chardet module installed, encoding will be utf-8")
return {'encoding': 'utf-8', 'confidence': 1}
return {"encoding": "utf-8", "confidence": 1}
try:
# noinspection PyUnresolvedReferences
# For Python 2
str_type = unicode
bytes_type = str
except NameError:
# For Python 3
str_type = str
bytes_type = bytes
__all__ = [
'InvalidTorrentDataException',
'BEncoder',
'BDecoder',
'encode',
'decode',
'TorrentFileParser',
'create_torrent_file',
'parse_torrent_file',
"InvalidTorrentDataException",
"BEncoder",
"BDecoder",
"encode",
"decode",
"TorrentFileParser",
"TorrentFileCreator",
"create_torrent_file",
"parse_torrent_file",
]
__version__ = '0.3.0'
__version__ = "0.4.1"
def detect(content):
return _detect(content)['encoding']
return _detect(content)["encoding"]
class InvalidTorrentDataException(Exception):
@ -99,32 +104,29 @@ _END = __EndCls()
def _check_hash_field_params(name, value):
return isinstance(name, str_type) \
and isinstance(value, tuple) and len(value) == 2 \
and isinstance(value[0], int) and isinstance(value[1], bool)
return (
isinstance(name, str_type)
and isinstance(value, tuple)
and len(value) == 2
and isinstance(value[0], int)
and isinstance(value[1], bool)
)
class TorrentFileParser(object):
class BDecoder(object):
TYPE_LIST = 'list'
TYPE_DICT = 'dict'
TYPE_INT = 'int'
TYPE_STRING = 'string'
TYPE_END = 'end'
TYPE_LIST = "list"
TYPE_DICT = "dict"
TYPE_INT = "int"
TYPE_STRING = "string"
TYPE_END = "end"
LIST_INDICATOR = b'l'
DICT_INDICATOR = b'd'
INT_INDICATOR = b'i'
END_INDICATOR = b'e'
STRING_INDICATOR = b''
STRING_DELIMITER = b':'
HASH_FIELD_PARAMS = {
# field length need_list
'pieces': (20, True),
'ed2k': (16, False),
'filehash': (20, False),
}
LIST_INDICATOR = b"l"
DICT_INDICATOR = b"d"
INT_INDICATOR = b"i"
END_INDICATOR = b"e"
STRING_INDICATOR = b""
STRING_DELIMITER = b":"
TYPES = [
(TYPE_LIST, LIST_INDICATOR),
@ -134,34 +136,55 @@ class TorrentFileParser(object):
(TYPE_STRING, STRING_INDICATOR),
]
# see https://docs.python.org/3/library/codecs.html#error-handlers
# for other usable error handler string
ERROR_HANDLER_USEBYTES = "usebytes"
def __init__(
self, fp, use_ordered_dict=False, encoding='utf-8', errors='strict',
hash_fields=None, hash_raw=False,
self,
data,
use_ordered_dict=False,
encoding="utf-8",
errors="strict",
hash_fields=None,
hash_raw=False,
):
"""
:param fp: a **binary** file-like object to parse,
:param bytes|file data: bytes or a **binary** file-like object to parse,
which means need 'b' mode when use built-in open function
:param bool use_ordered_dict: Use collections.OrderedDict as dict
container default False, which mean use built-in dict
:param str encoding: file content encoding, default utf-8, use 'auto'
to enable charset auto detection (need 'chardet' package installed)
:param str errors: how to deal with encoding error when try to parse
string from content with ``encoding``
string from content with ``encoding``.
see https://docs.python.org/3/library/codecs.html#error-handlers
for usable error handler string.
in particular, you can use "usebytes" to use "strict" decode mode
and let it return raw bytes if error happened.
:param Dict[str, Tuple[int, bool]] hash_fields: extra fields should
be treated as hash value. dict key is the field name, value is a
two-element tuple of (hash_block_length, as_a_list).
See :any:`hash_field` for detail
"""
if getattr(fp, 'read', ) is None \
or getattr(fp, 'seek') is None:
raise ValueError('Parameter fp needs a file like object')
if isinstance(data, bytes_type):
data = io.BytesIO(data)
elif getattr(data, "read") is not None and getattr(data, "seek") is not None:
pass
else:
raise ValueError("Parameter data must be bytes or file like object")
self._pos = 0
self._encoding = encoding
self._content = fp
self._content = data
self._use_ordered_dict = use_ordered_dict
self._error_handler = errors
self._hash_fields = dict(TorrentFileParser.HASH_FIELD_PARAMS)
self._error_use_bytes = False
if self._error_handler == BDecoder.ERROR_HANDLER_USEBYTES:
self._error_handler = "strict"
self._error_use_bytes = True
self._hash_fields = {}
if hash_fields is not None:
for k, v in hash_fields.items():
if _check_hash_field_params(k, v):
@ -192,9 +215,9 @@ class TorrentFileParser(object):
raise ValueError("Invalid hash field parameter")
return self
def parse(self):
def decode(self):
"""
:rtype: dict|list|int|str|bytes
:rtype: dict|list|int|str|unicode|bytes
:raise: :any:`InvalidTorrentDataException` when parse failed or error
happened when decode string using specified encoding
"""
@ -204,7 +227,7 @@ class TorrentFileParser(object):
try:
c = self._read_byte(1, True)
raise InvalidTorrentDataException(
0, 'Expect EOF, but get [{}] at pos {}'.format(c, self._pos)
0, "Expect EOF, but get [{}] at pos {}".format(c, self._pos)
)
except EOFError: # expect EOF
pass
@ -218,8 +241,7 @@ class TorrentFileParser(object):
if raise_eof:
raise EOFError()
raise InvalidTorrentDataException(
self._pos,
'Unexpected EOF when reading torrent file'
self._pos, "Unexpected EOF when reading torrent file"
)
self._pos += count
return gotten
@ -237,7 +259,7 @@ class TorrentFileParser(object):
k = self._next_element()
if k is _END:
return
if not isinstance(k, str_type):
if not isinstance(k, str_type) and not isinstance(k, bytes_type):
raise InvalidTorrentDataException(
self._pos, "Type of dict key can't be " + type(k).__name__
)
@ -245,7 +267,7 @@ class TorrentFileParser(object):
v = self._next_hash(*self._hash_fields[k])
else:
v = self._next_element(k)
if k == 'encoding':
if k == "encoding":
self._encoding = v
yield k, v
@ -270,12 +292,12 @@ class TorrentFileParser(object):
char = self._read_byte(1)
neg = False
while char != end:
if not neg and char == b'-':
if not neg and char == b"-":
neg = True
elif not b'0' <= char <= b'9':
elif not b"0" <= char <= b"9":
raise InvalidTorrentDataException(self._pos - 1)
else:
value = value * 10 + int(char) - int(b'0')
value = value * 10 + int(char) - int(b"0")
char = self._read_byte(1)
return -value if neg else value
@ -284,26 +306,33 @@ class TorrentFileParser(object):
raw = self._read_byte(length)
if need_decode:
encoding = self._encoding
if encoding == 'auto':
if encoding == "auto":
self.encoding = encoding = detect(raw)
try:
string = raw.decode(encoding, self._error_handler)
except UnicodeDecodeError as e:
if self._error_use_bytes:
return raw
else:
msg = [
"Fail to decode string at pos {pos} using encoding ",
e.encoding
e.encoding,
]
if field:
msg.extend([
' when parser field "', field, '"'
', maybe it is an hash field. ',
'You can use self.hash_field("', field, '") ',
'to let it be treated as hash value, ',
'so this error may disappear'
])
msg.extend(
[
' when parser field "',
field,
'"' ", maybe it is an hash field. ",
'You can use self.hash_field("',
field,
'") ',
"to let it be treated as hash value, ",
"so this error may disappear",
]
)
raise InvalidTorrentDataException(
self._pos - length + e.start,
''.join(msg)
self._pos - length + e.start, "".join(msg)
)
return string
return raw
@ -317,11 +346,11 @@ class TorrentFileParser(object):
if self._hash_raw:
return raw
res = [
binascii.hexlify(chunk).decode('ascii')
binascii.hexlify(chunk).decode("ascii")
for chunk in (raw[x : x + p_len] for x in range(0, len(raw), p_len))
]
if len(res) == 0 and not need_list:
return ''
return ""
if len(res) == 1 and not need_list:
return res[0]
return res
@ -340,11 +369,11 @@ class TorrentFileParser(object):
raise InvalidTorrentDataException(self._pos)
def _type_to_func(self, t):
return getattr(self, '_next_' + t)
return getattr(self, "_next_" + t)
def _next_element(self, field=None):
element_type = self._next_type()
if element_type is TorrentFileParser.TYPE_STRING and field is not None:
if element_type is BDecoder.TYPE_STRING and field is not None:
element = self._type_to_func(element_type)(field=field)
else:
element = self._type_to_func(element_type)()
@ -354,28 +383,28 @@ class TorrentFileParser(object):
class BEncoder(object):
TYPES = {
(dict,): TorrentFileParser.TYPE_DICT,
(list,): TorrentFileParser.TYPE_LIST,
(int,): TorrentFileParser.TYPE_INT,
(str_type, bytes): TorrentFileParser.TYPE_STRING,
(dict,): BDecoder.TYPE_DICT,
(list,): BDecoder.TYPE_LIST,
(int,): BDecoder.TYPE_INT,
(str_type, bytes_type): BDecoder.TYPE_STRING,
}
def __init__(self, data, encoding='utf-8', hash_fields=None):
def __init__(self, data, encoding="utf-8", hash_fields=None):
"""
:param dict|list|int|str data: data will be encoded
:param str encoding: string field output encoding
:param List[str] hash_fields: see
:any:`TorrentFileParser.__init__`
:any:`BDecoder.__init__`
"""
self._data = data
self._encoding = encoding
self._hash_fields = list(TorrentFileParser.HASH_FIELD_PARAMS.keys())
self._hash_fields = []
if hash_fields is not None:
self._hash_fields.extend(str_type(hash_fields))
self._hash_fields = hash_fields
def hash_field(self, name):
"""
see :any:`TorrentFileParser.hash_field`
see :any:`BDecoder.hash_field`
:param str name:
:return: return self, so you can chained call
@ -388,7 +417,7 @@ class BEncoder(object):
:rtype: bytes
"""
return b''.join(self._output_element(self._data))
return b"".join(self._output_element(self._data))
def encode_to_filelike(self):
"""
@ -401,15 +430,15 @@ class BEncoder(object):
def _output_string(self, data):
if isinstance(data, str_type):
data = data.encode(self._encoding)
yield str(len(data)).encode('ascii')
yield TorrentFileParser.STRING_DELIMITER
yield str(len(data)).encode("ascii")
yield BDecoder.STRING_DELIMITER
yield data
@staticmethod
def _output_int(data):
yield TorrentFileParser.INT_INDICATOR
yield str(data).encode('ascii')
yield TorrentFileParser.END_INDICATOR
yield BDecoder.INT_INDICATOR
yield str(data).encode("ascii")
yield BDecoder.END_INDICATOR
def _output_decode_hash(self, data):
if isinstance(data, str_type):
@ -419,31 +448,41 @@ class BEncoder(object):
if not isinstance(hash_line, str_type):
raise InvalidTorrentDataException(
None,
"Hash must be " + str_type.__name__ + " not " +
type(hash_line).__name__,
"Hash must be "
+ str_type.__name__
+ " not "
+ type(hash_line).__name__,
)
if len(hash_line) % 2 != 0:
raise InvalidTorrentDataException(
None,
"Hash(" + hash_line + ") length(" + str(len(hash_line)) +
") is a not even number",
"Hash("
+ hash_line
+ ") length("
+ str(len(hash_line))
+ ") is a not even number",
)
try:
raw = binascii.unhexlify(hash_line)
except binascii.Error as e:
raise InvalidTorrentDataException(
None, str(e),
None,
str(e),
)
result.append(raw)
for x in self._output_string(b''.join(result)):
for x in self._output_string(b"".join(result)):
yield x
def _output_dict(self, data):
yield TorrentFileParser.DICT_INDICATOR
yield BDecoder.DICT_INDICATOR
for k, v in data.items():
if not isinstance(k, str_type):
if not isinstance(k, str_type) and not isinstance(k, bytes_type):
raise InvalidTorrentDataException(
None, "Dict key must be " + str_type.__name__,
None,
"Dict key must be "
+ str_type.__name__
+ " or "
+ bytes_type.__name__,
)
for x in self._output_element(k):
yield x
@ -453,17 +492,17 @@ class BEncoder(object):
else:
for x in self._output_element(v):
yield x
yield TorrentFileParser.END_INDICATOR
yield BDecoder.END_INDICATOR
def _output_list(self, data):
yield TorrentFileParser.LIST_INDICATOR
yield BDecoder.LIST_INDICATOR
for v in data:
for x in self._output_element(v):
yield x
yield TorrentFileParser.END_INDICATOR
yield BDecoder.END_INDICATOR
def _type_to_func(self, t):
return getattr(self, '_output_' + t)
return getattr(self, "_output_" + t)
def _output_element(self, data):
for types, t in self.TYPES.items():
@ -476,47 +515,119 @@ class BEncoder(object):
)
class BDecoder(object):
class TorrentFileParser(object):
HASH_FIELD_DEFAULT_PARAMS = {
# field length need_list
"pieces": (20, True),
"ed2k": (16, False),
"filehash": (20, False),
"pieces root": (32, False),
}
def __init__(
self, data, use_ordered_dict=False, encoding='utf-8', errors='strict',
hash_fields=None, hash_raw=False,
self,
fp,
use_ordered_dict=False,
encoding="utf-8",
errors=BDecoder.ERROR_HANDLER_USEBYTES,
hash_fields=None,
hash_raw=False,
):
"""
See :any:`TorrentFileParser.__init__` for parameter description.
See :any:`BDecoder.__init__` for parameter description.
This class will use some default ``hash_fields`` values, and use "usebytes" as error handler
compare to use :any:`BDecoder` directly.
:param bytes data: raw data to be decoded
:param file fp: file to be parse
:param bool use_ordered_dict:
:param str encoding:
:param str errors:
:param Dict[str, Tuple[int, bool]] hash_fields:
:param bool hash_raw:
"""
self._parser = TorrentFileParser(
io.BytesIO(bytes(data)),
torrent_hash_fields = dict(TorrentFileParser.HASH_FIELD_DEFAULT_PARAMS)
if hash_fields is not None:
torrent_hash_fields.update(hash_fields)
self._decoder = BDecoder(
fp,
use_ordered_dict,
encoding,
errors,
hash_fields,
torrent_hash_fields,
hash_raw,
)
def hash_field(self, name, block_length=20, need_dict=False):
"""
See :any:`TorrentFileParser.hash_field` for parameter description
See :any:`BDecoder.hash_field` for parameter description
:param name:
:param block_length:
:param need_dict:
:return: return self, so you can chained call
"""
self._parser.hash_field(name, block_length, need_dict)
self._decoder.hash_field(name, block_length, need_dict)
return self
def decode(self):
return self._parser.parse()
def parse(self):
"""
Parse provided file
"""
return self._decoder.decode()
def encode(data, encoding='utf-8', hash_fields=None):
class TorrentFileCreator(object):
def __init__(self, data, encoding="utf-8", hash_fields=None):
"""
See :any:`BEncoder.__init__` for parameter description.
This class will use some default ``hash_fields`` values,
compare to use ``BEncoder`` directly.
:param dict|list|int|str data:
:param str encoding:
:param List[str] hash_fields:
"""
torrent_hash_fields = list(TorrentFileParser.HASH_FIELD_DEFAULT_PARAMS.keys())
if hash_fields is not None:
torrent_hash_fields.extend(hash_fields)
self._encoder = BEncoder(
data,
encoding,
torrent_hash_fields,
)
def hash_field(self, name):
"""
See :any:`BEncoder.hash_field` for parameter description
:param name:
:return: return self, so you can chained call
"""
self._encoder.hash_field(name)
return self
def create_filelike(self):
"""
Create a file-like(BytesIO) object according to provided data
:rtype: BytesIO
"""
return self._encoder.encode_to_filelike()
def create(self, filename):
"""
Create torrent file according to provided data
:param filename: output filename
:return:
"""
with open(filename, "wb") as f:
f.write(self._encoder.encode())
def encode(data, encoding="utf-8", hash_fields=None):
"""
Shortcut function for encode python object to torrent file format(bencode)
@ -531,8 +642,12 @@ def encode(data, encoding='utf-8', hash_fields=None):
def decode(
data, use_ordered_dict=False, encoding='utf-8', errors='strict',
hash_fields=None, hash_raw=False,
data,
use_ordered_dict=False,
encoding="utf-8",
errors="strict",
hash_fields=None,
hash_raw=False,
):
"""
Shortcut function for decode bytes as torrent file format(bencode) to python
@ -540,7 +655,7 @@ def decode(
See :any:`BDecoder.__init__` for parameter description
:param bytes data: raw data to be decoded
:param bytes|file data: data or file object to be decoded
:param bool use_ordered_dict:
:param str encoding:
:param str errors:
@ -549,13 +664,22 @@ def decode(
:rtype: dict|list|int|str|bytes|bytes
"""
return BDecoder(
data, use_ordered_dict, encoding, errors, hash_fields, hash_raw,
data,
use_ordered_dict,
encoding,
errors,
hash_fields,
hash_raw,
).decode()
def parse_torrent_file(
filename, use_ordered_dict=False, encoding='utf-8', errors='strict',
hash_fields=None, hash_raw=False,
filename,
use_ordered_dict=False,
encoding="utf-8",
errors="usebytes",
hash_fields=None,
hash_raw=False,
):
"""
Shortcut function for parse torrent object using TorrentFileParser
@ -570,13 +694,18 @@ def parse_torrent_file(
:param bool hash_raw:
:rtype: dict|list|int|str|bytes
"""
with open(filename, 'rb') as f:
with open(filename, "rb") as f:
return TorrentFileParser(
f, use_ordered_dict, encoding, errors, hash_fields, hash_raw,
f,
use_ordered_dict,
encoding,
errors,
hash_fields,
hash_raw,
).parse()
def create_torrent_file(filename, data, encoding='utf-8', hash_fields=None):
def create_torrent_file(filename, data, encoding="utf-8", hash_fields=None):
"""
Shortcut function for create a torrent file using BEncoder
@ -587,30 +716,93 @@ def create_torrent_file(filename, data, encoding='utf-8', hash_fields=None):
:param str encoding:
:param List[str] hash_fields:
"""
with open(filename, 'wb') as f:
f.write(BEncoder(data, encoding, hash_fields).encode())
TorrentFileCreator(data, encoding, hash_fields).create(filename)
class DataWrapper:
def __init__(self, data):
self.data = data
class JSONEncoderDataWrapperBytesToString(json.JSONEncoder):
def process(self, o):
if isinstance(o, bytes_type):
return binascii.hexlify(o).decode("ascii")
if isinstance(o, collections.OrderedDict):
output = collections.OrderedDict()
for k, v in o.items():
output[self.process(k)] = self.process(v)
return output
if isinstance(o, dict):
return {self.process(k): self.process(v) for k, v in o.items()}
if isinstance(o, list):
return [self.process(v) for v in o]
return o
def default(self, o):
if isinstance(o, DataWrapper):
return self.process(o.data)
return json.JSONEncoder.default(self, o)
def __main():
parser = argparse.ArgumentParser()
parser.add_argument('file', nargs='?', default='',
help='input file, will read form stdin if empty')
parser.add_argument('--dict', '-d', action='store_true', default=False,
help='use built-in dict, default will be OrderedDict')
parser.add_argument('--sort', '-s', action='store_true', default=False,
help='sort output json item by key')
parser.add_argument('--indent', '-i', type=int, default=None,
help='json output indent for every inner level')
parser.add_argument('--ascii', '-a', action='store_true', default=False,
help='ensure output json use ascii char, '
'escape other char use \\u')
parser.add_argument('--coding', '-c', default='utf-8',
help='string encoding, default "utf-8"')
parser.add_argument('--errors', '-e', default='strict',
help='decoding error handler, default "strict", you can'
' use "ignore" or "replace" to avoid exception')
parser.add_argument('--version', '-v', action='store_true', default=False,
help='print version and exit')
parser.add_argument(
"file", nargs="?", default="", help="input file, will read form stdin if empty"
)
parser.add_argument(
"--dict",
"-d",
action="store_true",
default=False,
help="use built-in dict, default will be OrderedDict",
)
parser.add_argument(
"--sort",
"-s",
action="store_true",
default=False,
help="sort output json item by key",
)
parser.add_argument(
"--indent",
"-i",
type=int,
default=None,
help="json output indent for every inner level",
)
parser.add_argument(
"--ascii",
"-a",
action="store_true",
default=False,
help="ensure output json use ascii char, " "escape other char use \\u",
)
parser.add_argument(
"--coding", "-c", default="utf-8", help='string encoding, default "utf-8"'
)
parser.add_argument(
"--errors",
"-e",
default=BDecoder.ERROR_HANDLER_USEBYTES,
help='decoding error handler, default "'
+ BDecoder.ERROR_HANDLER_USEBYTES
+ '"',
)
parser.add_argument(
"--hash-raw",
"-r",
action="store_true",
default=False,
help="do not group hash field by block, keeps it as raw bytes",
)
parser.add_argument(
"--version",
"-v",
action="store_true",
default=False,
help="print version and exit",
)
args = parser.parse_args()
if args.version:
@ -618,28 +810,33 @@ def __main():
exit(0)
try:
if args.file == '':
target_file = io.BytesIO(
getattr(sys.stdin, 'buffer', sys.stdin).read()
)
if args.file == "":
target_file = io.BytesIO(getattr(sys.stdin, "buffer", sys.stdin).read())
else:
target_file = open(args.file, 'rb')
target_file = open(args.file, "rb")
except FileNotFoundError:
sys.stderr.write('File "{}" not exist\n'.format(args.file))
exit(1)
# noinspection PyUnboundLocalVariable
data = TorrentFileParser(
target_file, not args.dict, args.coding, args.errors
target_file,
use_ordered_dict=not args.dict,
encoding=args.coding,
errors=args.errors,
hash_raw=args.hash_raw,
).parse()
data = json.dumps(
data, ensure_ascii=args.ascii,
sort_keys=args.sort, indent=args.indent
text = json.dumps(
DataWrapper(data),
ensure_ascii=args.ascii,
sort_keys=args.sort,
indent=args.indent,
cls=JSONEncoderDataWrapperBytesToString,
)
print(data)
print(text)
if __name__ == '__main__':
if __name__ == "__main__":
__main()