From deee741fb145360576ceae9d69b1b43db082c404 Mon Sep 17 00:00:00 2001 From: dirkf Date: Tue, 9 Aug 2022 21:05:00 +0100 Subject: [PATCH 1/4] [test, etc] Improve download test logs; also clean up some new flake8 issues (#31153) * [test] Identify testcase errors better * [test] Identify download errors better * [extractor/minds] Linter * [extractor/aes] Linter --- test/test_download.py | 7 +++++-- youtube_dl/aes.py | 2 +- youtube_dl/extractor/minds.py | 2 +- 3 files changed, 7 insertions(+), 4 deletions(-) diff --git a/test/test_download.py b/test/test_download.py index 6a6673bc2..19936969f 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -33,6 +33,7 @@ from youtube_dl.compat import ( from youtube_dl.utils import ( DownloadError, ExtractorError, + error_to_compat_str, format_bytes, UnavailableVideoError, ) @@ -108,7 +109,7 @@ def generator(test_case, tname): for tc in test_cases: info_dict = tc.get('info_dict', {}) if not (info_dict.get('id') and info_dict.get('ext')): - raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?') + raise Exception('Test definition (%s) requires both \'id\' and \'ext\' keys present to define the output file' % (tname, )) if 'skip' in test_case: print_skipping(test_case['skip']) @@ -161,7 +162,9 @@ def generator(test_case, tname): except (DownloadError, ExtractorError) as err: # Check if the exception is not a network related one if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503): - raise + msg = getattr(err, 'msg', error_to_compat_str(err)) + err.msg = '%s (%s)' % (msg, tname, ) + raise err if try_num == RETRIES: report_warning('%s failed due to network errors, skipping...' % tname) diff --git a/youtube_dl/aes.py b/youtube_dl/aes.py index 461bb6d41..d0de2d93f 100644 --- a/youtube_dl/aes.py +++ b/youtube_dl/aes.py @@ -303,7 +303,7 @@ def xor(data1, data2): def rijndael_mul(a, b): - if(a == 0 or b == 0): + if (a == 0 or b == 0): return 0 return RIJNDAEL_EXP_TABLE[(RIJNDAEL_LOG_TABLE[a] + RIJNDAEL_LOG_TABLE[b]) % 0xFF] diff --git a/youtube_dl/extractor/minds.py b/youtube_dl/extractor/minds.py index 8e9f0f825..e8fd582aa 100644 --- a/youtube_dl/extractor/minds.py +++ b/youtube_dl/extractor/minds.py @@ -78,7 +78,7 @@ class MindsIE(MindsBaseIE): else: return self.url_result(entity['perma_url']) else: - assert(entity['subtype'] == 'video') + assert (entity['subtype'] == 'video') video_id = entity_id # 1080p and webm formats available only on the sources array video = self._call_api( From e6a836d54ca1d3cd02f3ee45ef707a46f23e8291 Mon Sep 17 00:00:00 2001 From: dirkf Date: Wed, 10 Aug 2022 15:37:59 +0100 Subject: [PATCH 2/4] [core] Make `--max-downloads ...` stop immediately on reaching the limit Based on and closes #26638. --- youtube_dl/YoutubeDL.py | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index 3895b408f..e77b8d50c 100755 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -1779,10 +1779,9 @@ class YoutubeDL(object): assert info_dict.get('_type', 'video') == 'video' - max_downloads = self.params.get('max_downloads') - if max_downloads is not None: - if self._num_downloads >= int(max_downloads): - raise MaxDownloadsReached() + max_downloads = int_or_none(self.params.get('max_downloads')) or float('inf') + if self._num_downloads >= max_downloads: + raise MaxDownloadsReached() # TODO: backward compatibility, to be removed info_dict['fulltitle'] = info_dict['title'] @@ -2062,6 +2061,9 @@ class YoutubeDL(object): self.report_error('postprocessing: %s' % str(err)) return self.record_download_archive(info_dict) + # avoid possible nugatory search for further items (PR #26638) + if self._num_downloads >= max_downloads: + raise MaxDownloadsReached() def download(self, url_list): """Download a given list of URLs.""" From d231b56717c73ee597d2e077d11b69ed48a1b02d Mon Sep 17 00:00:00 2001 From: dirkf Date: Sun, 14 Aug 2022 18:45:45 +0100 Subject: [PATCH 3/4] [jsinterp] Overhaul JSInterp to handle new YT players 4c3f79c5, 324f67b9 (#31170) * back-port from yt-dlp 8f53dc44a0cc1c2d98c35740b9293462c080f5d0, thanks pukkandan * also support void, improve <> precedence, improve expressions in comma-list * add more tests --- test/test_jsinterp.py | 49 ++- test/test_utils.py | 3 + test/test_youtube_signature.py | 13 + youtube_dl/compat.py | 54 ++- youtube_dl/jsinterp.py | 581 ++++++++++++++++++++------------- youtube_dl/utils.py | 47 ++- 6 files changed, 500 insertions(+), 247 deletions(-) diff --git a/test/test_jsinterp.py b/test/test_jsinterp.py index acdabffb1..c6c931743 100644 --- a/test/test_jsinterp.py +++ b/test/test_jsinterp.py @@ -19,6 +19,9 @@ class TestJSInterpreter(unittest.TestCase): jsi = JSInterpreter('function x3(){return 42;}') self.assertEqual(jsi.call_function('x3'), 42) + jsi = JSInterpreter('function x3(){42}') + self.assertEqual(jsi.call_function('x3'), None) + jsi = JSInterpreter('var x5 = function(){return 42;}') self.assertEqual(jsi.call_function('x5'), 42) @@ -51,8 +54,11 @@ class TestJSInterpreter(unittest.TestCase): jsi = JSInterpreter('function f(){return 11 >> 2;}') self.assertEqual(jsi.call_function('f'), 2) + jsi = JSInterpreter('function f(){return []? 2+3: 4;}') + self.assertEqual(jsi.call_function('f'), 5) + def test_array_access(self): - jsi = JSInterpreter('function f(){var x = [1,2,3]; x[0] = 4; x[0] = 5; x[2] = 7; return x;}') + jsi = JSInterpreter('function f(){var x = [1,2,3]; x[0] = 4; x[0] = 5; x[2.0] = 7; return x;}') self.assertEqual(jsi.call_function('f'), [5, 2, 7]) def test_parens(self): @@ -62,6 +68,10 @@ class TestJSInterpreter(unittest.TestCase): jsi = JSInterpreter('function f(){return (1 + 2) * 3;}') self.assertEqual(jsi.call_function('f'), 9) + def test_quotes(self): + jsi = JSInterpreter(r'function f(){return "a\"\\("}') + self.assertEqual(jsi.call_function('f'), r'a"\(') + def test_assignments(self): jsi = JSInterpreter('function f(){var x = 20; x = 30 + 1; return x;}') self.assertEqual(jsi.call_function('f'), 31) @@ -104,18 +114,29 @@ class TestJSInterpreter(unittest.TestCase): }''') self.assertEqual(jsi.call_function('x'), [20, 20, 30, 40, 50]) + def test_builtins(self): + jsi = JSInterpreter(''' + function x() { return new Date('Wednesday 31 December 1969 18:01:26 MDT') - 0; } + ''') + self.assertEqual(jsi.call_function('x'), 86000) + jsi = JSInterpreter(''' + function x(dt) { return new Date(dt) - 0; } + ''') + self.assertEqual(jsi.call_function('x', 'Wednesday 31 December 1969 18:01:26 MDT'), 86000) + def test_call(self): jsi = JSInterpreter(''' function x() { return 2; } - function y(a) { return x() + a; } + function y(a) { return x() + (a?a:0); } function z() { return y(3); } ''') self.assertEqual(jsi.call_function('z'), 5) + self.assertEqual(jsi.call_function('y'), 2) def test_for_loop(self): # function x() { a=0; for (i=0; i-10; i++) {a++} a } jsi = JSInterpreter(''' - function x() { a=0; for (i=0; i-10; i = i + 1) {a++} a } + function x() { a=0; for (i=0; i-10; i++) {a++} return a } ''') self.assertEqual(jsi.call_function('x'), 10) @@ -156,19 +177,19 @@ class TestJSInterpreter(unittest.TestCase): def test_for_loop_continue(self): jsi = JSInterpreter(''' - function x() { a=0; for (i=0; i-10; i++) { continue; a++ } a } + function x() { a=0; for (i=0; i-10; i++) { continue; a++ } return a } ''') self.assertEqual(jsi.call_function('x'), 0) def test_for_loop_break(self): jsi = JSInterpreter(''' - function x() { a=0; for (i=0; i-10; i++) { break; a++ } a } + function x() { a=0; for (i=0; i-10; i++) { break; a++ } return a } ''') self.assertEqual(jsi.call_function('x'), 0) def test_literal_list(self): jsi = JSInterpreter(''' - function x() { [1, 2, "asdf", [5, 6, 7]][3] } + function x() { return [1, 2, "asdf", [5, 6, 7]][3] } ''') self.assertEqual(jsi.call_function('x'), [5, 6, 7]) @@ -177,6 +198,22 @@ class TestJSInterpreter(unittest.TestCase): function x() { a=5; a -= 1, a+=3; return a } ''') self.assertEqual(jsi.call_function('x'), 7) + jsi = JSInterpreter(''' + function x() { a=5; return (a -= 1, a+=3, a); } + ''') + self.assertEqual(jsi.call_function('x'), 7) + + def test_void(self): + jsi = JSInterpreter(''' + function x() { return void 42; } + ''') + self.assertEqual(jsi.call_function('x'), None) + + def test_return_function(self): + jsi = JSInterpreter(''' + function x() { return [1, function(){return 1}][1] } + ''') + self.assertEqual(jsi.call_function('x')([]), 1) if __name__ == '__main__': diff --git a/test/test_utils.py b/test/test_utils.py index 259c4763e..f1a748dde 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -370,6 +370,9 @@ class TestUtil(unittest.TestCase): self.assertEqual(unified_timestamp('Sep 11, 2013 | 5:49 AM'), 1378878540) self.assertEqual(unified_timestamp('December 15, 2017 at 7:49 am'), 1513324140) self.assertEqual(unified_timestamp('2018-03-14T08:32:43.1493874+00:00'), 1521016363) + self.assertEqual(unified_timestamp('December 31 1969 20:00:01 EDT'), 1) + self.assertEqual(unified_timestamp('Wednesday 31 December 1969 18:01:26 MDT'), 86) + self.assertEqual(unified_timestamp('12/31/1969 20:01:18 EDT', False), 78) def test_determine_ext(self): self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4') diff --git a/test/test_youtube_signature.py b/test/test_youtube_signature.py index fc5e9828e..6e955e0f0 100644 --- a/test/test_youtube_signature.py +++ b/test/test_youtube_signature.py @@ -90,12 +90,25 @@ _NSIG_TESTS = [ 'https://www.youtube.com/s/player/e06dea74/player_ias.vflset/en_US/base.js', 'AiuodmaDDYw8d3y4bf', 'ankd8eza2T6Qmw', ), + ( + 'https://www.youtube.com/s/player/5dd88d1d/player-plasma-ias-phone-en_US.vflset/base.js', + 'kSxKFLeqzv_ZyHSAt', 'n8gS8oRlHOxPFA', + ), + ( + 'https://www.youtube.com/s/player/324f67b9/player_ias.vflset/en_US/base.js', + 'xdftNy7dh9QGnhW', '22qLGxrmX8F1rA', + ), + ( + 'https://www.youtube.com/s/player/4c3f79c5/player_ias.vflset/en_US/base.js', + 'TDCstCG66tEAO5pR9o', 'dbxNtZ14c-yWyw', + ), ] class TestPlayerInfo(unittest.TestCase): def test_youtube_extract_player_info(self): PLAYER_URLS = ( + ('https://www.youtube.com/s/player/4c3f79c5/player_ias.vflset/en_US/base.js', '4c3f79c5'), ('https://www.youtube.com/s/player/64dddad9/player_ias.vflset/en_US/base.js', '64dddad9'), ('https://www.youtube.com/s/player/64dddad9/player_ias.vflset/fr_FR/base.js', '64dddad9'), ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-en_US.vflset/base.js', '64dddad9'), diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py index 9f5f85dae..6d2c31a61 100644 --- a/youtube_dl/compat.py +++ b/youtube_dl/compat.py @@ -2985,7 +2985,6 @@ except ImportError: except ImportError: compat_filter = filter - try: from future_builtins import zip as compat_zip except ImportError: # not 2.6+ or is 3.x @@ -2995,6 +2994,57 @@ except ImportError: # not 2.6+ or is 3.x compat_zip = zip +# method renamed between Py2/3 +try: + from itertools import zip_longest as compat_itertools_zip_longest +except ImportError: + from itertools import izip_longest as compat_itertools_zip_longest + + +# new class in collections +try: + from collections import ChainMap as compat_collections_chain_map +except ImportError: + # Py < 3.3 + class compat_collections_chain_map(compat_collections_abc.MutableMapping): + + maps = [{}] + + def __init__(self, *maps): + self.maps = list(maps) or [{}] + + def __getitem__(self, k): + for m in self.maps: + if k in m: + return m[k] + raise KeyError(k) + + def __setitem__(self, k, v): + self.maps[0].__setitem__(k, v) + return + + def __delitem__(self, k): + if k in self.maps[0]: + del self.maps[0][k] + return + raise KeyError(k) + + def __iter__(self): + return itertools.chain(*reversed(self.maps)) + + def __len__(self): + return len(iter(self)) + + def new_child(self, m=None, **kwargs): + m = m or {} + m.update(kwargs) + return compat_collections_chain_map(m, *self.maps) + + @property + def parents(self): + return compat_collections_chain_map(*(self.maps[1:])) + + if sys.version_info < (3, 3): def compat_b64decode(s, *args, **kwargs): if isinstance(s, compat_str): @@ -3031,6 +3081,7 @@ __all__ = [ 'compat_basestring', 'compat_chr', 'compat_collections_abc', + 'compat_collections_chain_map', 'compat_cookiejar', 'compat_cookiejar_Cookie', 'compat_cookies', @@ -3051,6 +3102,7 @@ __all__ = [ 'compat_input', 'compat_integer_types', 'compat_itertools_count', + 'compat_itertools_zip_longest', 'compat_kwargs', 'compat_map', 'compat_numeric_types', diff --git a/youtube_dl/jsinterp.py b/youtube_dl/jsinterp.py index 8eaa911cd..c60a9b3c2 100644 --- a/youtube_dl/jsinterp.py +++ b/youtube_dl/jsinterp.py @@ -1,42 +1,87 @@ from __future__ import unicode_literals +import itertools import json +import math import operator import re from .utils import ( + NO_DEFAULT, ExtractorError, + js_to_json, remove_quotes, + unified_timestamp, ) from .compat import ( - compat_collections_abc, + compat_collections_chain_map as ChainMap, + compat_itertools_zip_longest as zip_longest, compat_str, ) -MutableMapping = compat_collections_abc.MutableMapping +_NAME_RE = r'[a-zA-Z_$][\w$]*' -class Nonlocal: - pass +# (op, definition) in order of binding priority, tightest first +# avoid dict to maintain order +# definition None => Defined in JSInterpreter._operator +_DOT_OPERATORS = ( + ('.', None), + # TODO: ('?.', None), +) - -_OPERATORS = [ +_OPERATORS = ( ('|', operator.or_), ('^', operator.xor), ('&', operator.and_), ('>>', operator.rshift), ('<<', operator.lshift), - ('-', operator.sub), ('+', operator.add), - ('%', operator.mod), - ('/', operator.truediv), + ('-', operator.sub), ('*', operator.mul), -] -_ASSIGN_OPERATORS = [(op + '=', opfunc) for op, opfunc in _OPERATORS] -_ASSIGN_OPERATORS.append(('=', (lambda cur, right: right))) + ('/', operator.truediv), + ('%', operator.mod), +) -_NAME_RE = r'[a-zA-Z_$][a-zA-Z_$0-9]*' +_COMP_OPERATORS = ( + ('===', operator.is_), + ('==', operator.eq), + ('!==', operator.is_not), + ('!=', operator.ne), + ('<=', operator.le), + ('>=', operator.ge), + ('<', operator.lt), + ('>', operator.gt), +) + +_LOG_OPERATORS = ( + ('&', operator.and_), + ('|', operator.or_), + ('^', operator.xor), +) + +_SC_OPERATORS = ( + ('?', None), + ('||', None), + ('&&', None), + # TODO: ('??', None), +) + +_OPERATOR_RE = '|'.join(map(lambda x: re.escape(x[0]), _OPERATORS + _LOG_OPERATORS)) _MATCHING_PARENS = dict(zip(*zip('()', '{}', '[]'))) +_QUOTES = '\'"' + + +def _ternary(cndn, if_true=True, if_false=False): + """Simulate JS's ternary operator (cndn?if_true:if_false)""" + if cndn in (False, None, 0, ''): + return if_false + try: + if math.isnan(cndn): # NB: NaN cannot be checked by membership + return if_false + except TypeError: + pass + return if_true class JS_Break(ExtractorError): @@ -49,70 +94,77 @@ class JS_Continue(ExtractorError): ExtractorError.__init__(self, 'Invalid continue') -class LocalNameSpace(MutableMapping): - def __init__(self, *stack): - self.stack = tuple(stack) - - def __getitem__(self, key): - for scope in self.stack: - if key in scope: - return scope[key] - raise KeyError(key) - +class LocalNameSpace(ChainMap): def __setitem__(self, key, value): - for scope in self.stack: + for scope in self.maps: if key in scope: scope[key] = value - break - else: - self.stack[0][key] = value - return value + return + self.maps[0][key] = value def __delitem__(self, key): raise NotImplementedError('Deleting is not supported') - def __iter__(self): - for scope in self.stack: - for scope_item in iter(scope): - yield scope_item - - def __len__(self, key): - return len(iter(self)) - def __repr__(self): - return 'LocalNameSpace%s' % (self.stack, ) + return 'LocalNameSpace%s' % (self.maps, ) class JSInterpreter(object): + __named_object_counter = 0 + def __init__(self, code, objects=None): - if objects is None: - objects = {} - self.code = code - self._functions = {} - self._objects = objects - self.__named_object_counter = 0 + self.code, self._functions = code, {} + self._objects = {} if objects is None else objects + + class Exception(ExtractorError): + def __init__(self, msg, *args, **kwargs): + expr = kwargs.pop('expr', None) + if expr is not None: + msg = '{0} in: {1!r}'.format(msg.rstrip(), expr[:100]) + super(JSInterpreter.Exception, self).__init__(msg, *args, **kwargs) def _named_object(self, namespace, obj): self.__named_object_counter += 1 - name = '__youtube_dl_jsinterp_obj%s' % (self.__named_object_counter, ) + name = '__youtube_dl_jsinterp_obj%d' % (self.__named_object_counter, ) namespace[name] = obj return name @staticmethod - def _separate(expr, delim=',', max_split=None): + def _separate(expr, delim=',', max_split=None, skip_delims=None): if not expr: return counters = {k: 0 for k in _MATCHING_PARENS.values()} - start, splits, pos, delim_len = 0, 0, 0, len(delim) - 1 + start, splits, pos, skipping, delim_len = 0, 0, 0, 0, len(delim) - 1 + in_quote, escaping = None, False for idx, char in enumerate(expr): - if char in _MATCHING_PARENS: - counters[_MATCHING_PARENS[char]] += 1 - elif char in counters: - counters[char] -= 1 - if char != delim[pos] or any(counters.values()): - pos = 0 + if not in_quote: + if char in _MATCHING_PARENS: + counters[_MATCHING_PARENS[char]] += 1 + elif char in counters: + counters[char] -= 1 + if not escaping: + if char in _QUOTES and in_quote in (char, None): + in_quote = None if in_quote else char + else: + escaping = in_quote and char == '\\' + else: + escaping = False + + if char != delim[pos] or any(counters.values()) or in_quote: + pos = skipping = 0 continue - elif pos != delim_len: + elif skipping > 0: + skipping -= 1 + continue + elif pos == 0 and skip_delims: + here = expr[idx:] + for s in skip_delims if isinstance(skip_delims, (list, tuple)) else [skip_delims]: + if here.startswith(s) and s: + skipping = len(s) - 1 + break + if skipping > 0: + continue + if pos < delim_len: pos += 1 continue yield expr[start: idx - delim_len] @@ -122,61 +174,108 @@ class JSInterpreter(object): break yield expr[start:] - @staticmethod - def _separate_at_paren(expr, delim): - separated = list(JSInterpreter._separate(expr, delim, 1)) + @classmethod + def _separate_at_paren(cls, expr, delim): + separated = list(cls._separate(expr, delim, 1)) + if len(separated) < 2: - raise ExtractorError('No terminating paren {0} in {1}'.format(delim, expr)) + raise cls.Exception('No terminating paren {delim} in {expr}'.format(**locals())) return separated[0][1:].strip(), separated[1].strip() + @staticmethod + def _all_operators(): + return itertools.chain( + _SC_OPERATORS, _LOG_OPERATORS, _COMP_OPERATORS, _OPERATORS) + + def _operator(self, op, left_val, right_expr, expr, local_vars, allow_recursion): + if op in ('||', '&&'): + if (op == '&&') ^ _ternary(left_val): + return left_val # short circuiting + elif op == '?': + right_expr = _ternary(left_val, *self._separate(right_expr, ':', 1)) + + right_val = self.interpret_expression(right_expr, local_vars, allow_recursion) + opfunc = op and next((v for k, v in self._all_operators() if k == op), None) + if not opfunc: + return right_val + + try: + return opfunc(left_val, right_val) + except Exception as e: + raise self.Exception('Failed to evaluate {left_val!r} {op} {right_val!r}'.format(**locals()), expr, cause=e) + + def _index(self, obj, idx): + if idx == 'length': + return len(obj) + try: + return obj[int(idx)] if isinstance(obj, list) else obj[idx] + except Exception as e: + raise self.Exception('Cannot get index {idx}'.format(**locals()), expr=repr(obj), cause=e) + + def _dump(self, obj, namespace): + try: + return json.dumps(obj) + except TypeError: + return self._named_object(namespace, obj) + def interpret_statement(self, stmt, local_vars, allow_recursion=100): if allow_recursion < 0: - raise ExtractorError('Recursion limit reached') + raise self.Exception('Recursion limit reached') + allow_recursion -= 1 - sub_statements = list(self._separate(stmt, ';')) - stmt = (sub_statements or ['']).pop() + should_return = False + sub_statements = list(self._separate(stmt, ';')) or [''] + expr = stmt = sub_statements.pop().strip() for sub_stmt in sub_statements: - ret, should_abort = self.interpret_statement(sub_stmt, local_vars, allow_recursion - 1) - if should_abort: - return ret + ret, should_return = self.interpret_statement(sub_stmt, local_vars, allow_recursion) + if should_return: + return ret, should_return - should_abort = False - stmt = stmt.lstrip() - stmt_m = re.match(r'var\s', stmt) - if stmt_m: - expr = stmt[len(stmt_m.group(0)):] - else: - return_m = re.match(r'return(?:\s+|$)', stmt) - if return_m: - expr = stmt[len(return_m.group(0)):] - should_abort = True + m = re.match(r'(?P(?:var|const|let)\s)|return(?:\s+|$)', stmt) + if m: + expr = stmt[len(m.group(0)):].strip() + should_return = not m.group('var') + if not expr: + return None, should_return + + if expr[0] in _QUOTES: + inner, outer = self._separate(expr, expr[0], 1) + inner = json.loads(js_to_json(inner + expr[0])) # , strict=True)) + if not outer: + return inner, should_return + expr = self._named_object(local_vars, inner) + outer + + if expr.startswith('new '): + obj = expr[4:] + if obj.startswith('Date('): + left, right = self._separate_at_paren(obj[4:], ')') + left = self.interpret_expression(left, local_vars, allow_recursion) + expr = unified_timestamp(left, False) + if not expr: + raise self.Exception('Failed to parse date {left!r}'.format(**locals()), expr=expr) + expr = self._dump(int(expr * 1000), local_vars) + right else: - # Try interpreting it as an expression - expr = stmt + raise self.Exception('Unsupported object {obj}'.format(**locals()), expr=expr) - v = self.interpret_expression(expr, local_vars, allow_recursion) - return v, should_abort - - def interpret_expression(self, expr, local_vars, allow_recursion): - expr = expr.strip() - if expr == '': # Empty expression - return None + if expr.startswith('void '): + left = self.interpret_expression(expr[5:], local_vars, allow_recursion) + return None, should_return if expr.startswith('{'): inner, outer = self._separate_at_paren(expr, '}') - inner, should_abort = self.interpret_statement(inner, local_vars, allow_recursion - 1) + inner, should_abort = self.interpret_statement(inner, local_vars, allow_recursion) if not outer or should_abort: - return inner + return inner, should_abort or should_return else: - expr = json.dumps(inner) + outer + expr = self._dump(inner, local_vars) + outer if expr.startswith('('): inner, outer = self._separate_at_paren(expr, ')') - inner = self.interpret_expression(inner, local_vars, allow_recursion) - if not outer: - return inner + inner, should_abort = self.interpret_statement(inner, local_vars, allow_recursion) + if not outer or should_abort: + return inner, should_abort or should_return else: - expr = json.dumps(inner) + outer + expr = self._dump(inner, local_vars) + outer if expr.startswith('['): inner, outer = self._separate_at_paren(expr, ']') @@ -185,57 +284,53 @@ class JSInterpreter(object): for item in self._separate(inner)]) expr = name + outer - m = re.match(r'try\s*', expr) - if m: + m = re.match(r'(?Ptry|finally)\s*|(?:(?Pcatch)|(?Pfor)|(?Pswitch))\s*\(', expr) + md = m.groupdict() if m else {} + if md.get('try'): if expr[m.end()] == '{': try_expr, expr = self._separate_at_paren(expr[m.end():], '}') else: try_expr, expr = expr[m.end() - 1:], '' - ret, should_abort = self.interpret_statement(try_expr, local_vars, allow_recursion - 1) + ret, should_abort = self.interpret_statement(try_expr, local_vars, allow_recursion) if should_abort: - return ret - return self.interpret_statement(expr, local_vars, allow_recursion - 1)[0] + return ret, True + ret, should_abort = self.interpret_statement(expr, local_vars, allow_recursion) + return ret, should_abort or should_return - m = re.match(r'(?:(?Pcatch)|(?Pfor)|(?Pswitch))\s*\(', expr) - md = m.groupdict() if m else {} - if md.get('catch'): + elif md.get('catch'): # We ignore the catch block _, expr = self._separate_at_paren(expr, '}') - return self.interpret_statement(expr, local_vars, allow_recursion - 1)[0] + ret, should_abort = self.interpret_statement(expr, local_vars, allow_recursion) + return ret, should_abort or should_return elif md.get('for'): - def raise_constructor_error(c): - raise ExtractorError( - 'Premature return in the initialization of a for loop in {0!r}'.format(c)) - constructor, remaining = self._separate_at_paren(expr[m.end() - 1:], ')') if remaining.startswith('{'): body, expr = self._separate_at_paren(remaining, '}') else: - m = re.match(r'switch\s*\(', remaining) # FIXME - if m: - switch_val, remaining = self._separate_at_paren(remaining[m.end() - 1:], ')') + switch_m = re.match(r'switch\s*\(', remaining) # FIXME + if switch_m: + switch_val, remaining = self._separate_at_paren(remaining[switch_m.end() - 1:], ')') body, expr = self._separate_at_paren(remaining, '}') body = 'switch(%s){%s}' % (switch_val, body) else: body, expr = remaining, '' start, cndn, increment = self._separate(constructor, ';') - if self.interpret_statement(start, local_vars, allow_recursion - 1)[1]: - raise_constructor_error(constructor) + self.interpret_expression(start, local_vars, allow_recursion) while True: - if not self.interpret_expression(cndn, local_vars, allow_recursion): + if not _ternary(self.interpret_expression(cndn, local_vars, allow_recursion)): break try: - ret, should_abort = self.interpret_statement(body, local_vars, allow_recursion - 1) + ret, should_abort = self.interpret_statement(body, local_vars, allow_recursion) if should_abort: - return ret + return ret, True except JS_Break: break except JS_Continue: pass - if self.interpret_statement(increment, local_vars, allow_recursion - 1)[1]: - raise_constructor_error(constructor) - return self.interpret_statement(expr, local_vars, allow_recursion - 1)[0] + self.interpret_expression(increment, local_vars, allow_recursion) + ret, should_abort = self.interpret_statement(expr, local_vars, allow_recursion) + return ret, should_abort or should_return elif md.get('switch'): switch_val, remaining = self._separate_at_paren(expr[m.end() - 1:], ')') @@ -245,7 +340,7 @@ class JSInterpreter(object): for default in (False, True): matched = False for item in items: - case, stmt = [i.strip() for i in self._separate(item, ':', 1)] + case, stmt = (i.strip() for i in self._separate(item, ':', 1)) if default: matched = matched or case == 'default' elif not matched: @@ -254,24 +349,28 @@ class JSInterpreter(object): if not matched: continue try: - ret, should_abort = self.interpret_statement(stmt, local_vars, allow_recursion - 1) + ret, should_abort = self.interpret_statement(stmt, local_vars, allow_recursion) if should_abort: return ret except JS_Break: break if matched: break - return self.interpret_statement(expr, local_vars, allow_recursion - 1)[0] + ret, should_abort = self.interpret_statement(expr, local_vars, allow_recursion) + return ret, should_abort or should_return # Comma separated statements sub_expressions = list(self._separate(expr)) - expr = sub_expressions.pop().strip() if sub_expressions else '' - for sub_expr in sub_expressions: - self.interpret_expression(sub_expr, local_vars, allow_recursion) + if len(sub_expressions) > 1: + for sub_expr in sub_expressions: + ret, should_abort = self.interpret_statement(sub_expr, local_vars, allow_recursion) + if should_abort: + return ret, True + return ret, False for m in re.finditer(r'''(?x) - (?P\+\+|--)(?P%(_NAME_RE)s)| - (?P%(_NAME_RE)s)(?P\+\+|--)''' % globals(), expr): + (?P\+\+|--)(?P{_NAME_RE})| + (?P{_NAME_RE})(?P\+\+|--)'''.format(**globals()), expr): var = m.group('var1') or m.group('var2') start, end = m.span() sign = m.group('pre_sign') or m.group('post_sign') @@ -279,85 +378,87 @@ class JSInterpreter(object): local_vars[var] += 1 if sign[0] == '+' else -1 if m.group('pre_sign'): ret = local_vars[var] - expr = expr[:start] + json.dumps(ret) + expr[end:] + expr = expr[:start] + self._dump(ret, local_vars) + expr[end:] - for op, opfunc in _ASSIGN_OPERATORS: - m = re.match(r'''(?x) - (?P%s)(?:\[(?P[^\]]+?)\])? - \s*%s - (?P.*)$''' % (_NAME_RE, re.escape(op)), expr) - if not m: - continue - right_val = self.interpret_expression(m.group('expr'), local_vars, allow_recursion) + if not expr: + return None, should_return - if m.groupdict().get('index'): - lvar = local_vars[m.group('out')] - idx = self.interpret_expression(m.group('index'), local_vars, allow_recursion) - if not isinstance(idx, int): - raise ExtractorError('List indices must be integers: %s' % (idx, )) - cur = lvar[idx] - val = opfunc(cur, right_val) - lvar[idx] = val - return val - else: - cur = local_vars.get(m.group('out')) - val = opfunc(cur, right_val) - local_vars[m.group('out')] = val - return val + m = re.match(r'''(?x) + (?P + (?P{_NAME_RE})(?:\[(?P[^\]]+?)\])?\s* + (?P{_OPERATOR_RE})? + =(?P.*)$ + )|(?P + (?!if|return|true|false|null|undefined)(?P{_NAME_RE})$ + )|(?P + (?P{_NAME_RE})\[(?P.+)\]$ + )|(?P + (?P{_NAME_RE})(?:\.(?P[^(]+)|\[(?P[^\]]+)\])\s* + )|(?P + (?P{_NAME_RE})\((?P.*)\)$ + )'''.format(**globals()), expr) + md = m.groupdict() if m else {} + if md.get('assign'): + left_val = local_vars.get(m.group('out')) - if expr.isdigit(): - return int(expr) + if not m.group('index'): + local_vars[m.group('out')] = self._operator( + m.group('op'), left_val, m.group('expr'), expr, local_vars, allow_recursion) + return local_vars[m.group('out')], should_return + elif left_val is None: + raise self.Exception('Cannot index undefined variable ' + m.group('out'), expr=expr) - if expr == 'break': + idx = self.interpret_expression(m.group('index'), local_vars, allow_recursion) + if not isinstance(idx, (int, float)): + raise self.Exception('List index %s must be integer' % (idx, ), expr=expr) + idx = int(idx) + left_val[idx] = self._operator( + m.group('op'), left_val[idx], m.group('expr'), expr, local_vars, allow_recursion) + return left_val[idx], should_return + + elif expr.isdigit(): + return int(expr), should_return + + elif expr == 'break': raise JS_Break() elif expr == 'continue': raise JS_Continue() - var_m = re.match( - r'(?!if|return|true|false|null)(?P%s)$' % _NAME_RE, - expr) - if var_m: - return local_vars[var_m.group('name')] + elif md.get('return'): + return local_vars[m.group('name')], should_return try: - return json.loads(expr) + ret = json.loads(js_to_json(expr)) # strict=True) + if not md.get('attribute'): + return ret, should_return except ValueError: pass - m = re.match( - r'(?P%s)\[(?P.+)\]$' % _NAME_RE, expr) - if m: + if md.get('indexing'): val = local_vars[m.group('in')] idx = self.interpret_expression(m.group('idx'), local_vars, allow_recursion) - return val[idx] + return self._index(val, idx), should_return - def raise_expr_error(where, op, exp): - raise ExtractorError('Premature {0} return of {1} in {2!r}'.format(where, op, exp)) - - for op, opfunc in _OPERATORS: - separated = list(self._separate(expr, op)) + for op, _ in self._all_operators(): + # hackety: have higher priority than <>, but don't confuse them + skip_delim = (op + op) if op in ('<', '>') else None + separated = list(self._separate(expr, op, skip_delims=skip_delim)) if len(separated) < 2: continue - right_val = separated.pop() - left_val = op.join(separated) - left_val, should_abort = self.interpret_statement( - left_val, local_vars, allow_recursion - 1) - if should_abort: - raise_expr_error('left-side', op, expr) - right_val, should_abort = self.interpret_statement( - right_val, local_vars, allow_recursion - 1) - if should_abort: - raise_expr_error('right-side', op, expr) - return opfunc(left_val or 0, right_val) - m = re.match( - r'(?P%s)(?:\.(?P[^(]+)|\[(?P[^]]+)\])\s*' % _NAME_RE, - expr) - if m: + right_expr = separated.pop() + while op == '-' and len(separated) > 1 and not separated[-1].strip(): + right_expr = '-' + right_expr + separated.pop() + left_val = self.interpret_expression(op.join(separated), local_vars, allow_recursion) + return self._operator(op, 0 if left_val is None else left_val, + right_expr, expr, local_vars, allow_recursion), should_return + + if md.get('attribute'): variable = m.group('var') - nl = Nonlocal() - - nl.member = remove_quotes(m.group('member') or m.group('member2')) + member = m.group('member') + if not member: + member = self.interpret_expression(m.group('member2'), local_vars, allow_recursion) arg_str = expr[m.end():] if arg_str.startswith('('): arg_str, remaining = self._separate_at_paren(arg_str, ')') @@ -367,25 +468,24 @@ class JSInterpreter(object): def assertion(cndn, msg): """ assert, but without risk of getting optimized out """ if not cndn: - raise ExtractorError('{0} {1}: {2}'.format(nl.member, msg, expr)) + raise ExtractorError('{member} {msg}'.format(**locals()), expr=expr) def eval_method(): - # nonlocal member - member = nl.member - if variable == 'String': - obj = compat_str - elif variable in local_vars: - obj = local_vars[variable] - else: + if (variable, member) == ('console', 'debug'): + return + types = { + 'String': compat_str, + 'Math': float, + } + obj = local_vars.get(variable, types.get(variable, NO_DEFAULT)) + if obj is NO_DEFAULT: if variable not in self._objects: self._objects[variable] = self.extract_object(variable) obj = self._objects[variable] + # Member access if arg_str is None: - # Member access - if member == 'length': - return len(obj) - return obj[member] + return self._index(obj, member) # Function call argvals = [ @@ -396,12 +496,17 @@ class JSInterpreter(object): if member == 'fromCharCode': assertion(argvals, 'takes one or more arguments') return ''.join(map(chr, argvals)) - raise ExtractorError('Unsupported string method %s' % (member, )) + raise self.Exception('Unsupported string method ' + member, expr=expr) + elif obj == float: + if member == 'pow': + assertion(len(argvals) == 2, 'takes two arguments') + return argvals[0] ** argvals[1] + raise self.Exception('Unsupported Math method ' + member, expr=expr) if member == 'split': assertion(argvals, 'takes one or more arguments') - assertion(argvals == [''], 'with arguments is not implemented') - return list(obj) + assertion(len(argvals) == 1, 'with limit argument is not implemented') + return obj.split(argvals[0]) if argvals[0] else list(obj) elif member == 'join': assertion(isinstance(obj, list), 'must be applied on a list') assertion(len(argvals) == 1, 'takes exactly one argument') @@ -447,7 +552,7 @@ class JSInterpreter(object): assertion(argvals, 'takes one or more arguments') assertion(len(argvals) <= 2, 'takes at-most 2 arguments') f, this = (argvals + [''])[:2] - return [f((item, idx, obj), this=this) for idx, item in enumerate(obj)] + return [f((item, idx, obj), {'this': this}, allow_recursion) for idx, item in enumerate(obj)] elif member == 'indexOf': assertion(argvals, 'takes one or more arguments') assertion(len(argvals) <= 2, 'takes at-most 2 arguments') @@ -457,32 +562,35 @@ class JSInterpreter(object): except ValueError: return -1 - if isinstance(obj, list): - member = int(member) - nl.member = member - return obj[member](argvals) + idx = int(member) if isinstance(obj, list) else member + return obj[idx](argvals, allow_recursion=allow_recursion) if remaining: - return self.interpret_expression( + ret, should_abort = self.interpret_statement( self._named_object(local_vars, eval_method()) + remaining, local_vars, allow_recursion) + return ret, should_return or should_abort else: - return eval_method() + return eval_method(), should_return - m = re.match(r'^(?P%s)\((?P[a-zA-Z0-9_$,]*)\)$' % _NAME_RE, expr) - if m: - fname = m.group('func') - argvals = tuple([ - int(v) if v.isdigit() else local_vars[v] - for v in self._separate(m.group('args'))]) + elif md.get('function'): + fname = m.group('fname') + argvals = [self.interpret_expression(v, local_vars, allow_recursion) + for v in self._separate(m.group('args'))] if fname in local_vars: - return local_vars[fname](argvals) + return local_vars[fname](argvals, allow_recursion=allow_recursion), should_return elif fname not in self._functions: self._functions[fname] = self.extract_function(fname) - return self._functions[fname](argvals) + return self._functions[fname](argvals, allow_recursion=allow_recursion), should_return - if expr: - raise ExtractorError('Unsupported JS expression %r' % expr) + raise self.Exception( + 'Unsupported JS expression ' + (expr[:40] if expr != stmt else ''), expr=stmt) + + def interpret_expression(self, expr, local_vars, allow_recursion): + ret, should_return = self.interpret_statement(expr, local_vars, allow_recursion) + if should_return: + raise self.Exception('Cannot return from an expression', expr) + return ret def extract_object(self, objname): _FUNC_NAME_RE = r'''(?:[a-zA-Z$0-9]+|"[a-zA-Z$0-9]+"|'[a-zA-Z$0-9]+')''' @@ -494,15 +602,17 @@ class JSInterpreter(object): }\s*; ''' % (re.escape(objname), _FUNC_NAME_RE), self.code) + if not obj_m: + raise self.Exception('Could not find object ' + objname) fields = obj_m.group('fields') # Currently, it only supports function definitions fields_m = re.finditer( r'''(?x) - (?P%s)\s*:\s*function\s*\((?P[a-z,]+)\){(?P[^}]+)} - ''' % _FUNC_NAME_RE, + (?P%s)\s*:\s*function\s*\((?P(?:%s|,)*)\){(?P[^}]+)} + ''' % (_FUNC_NAME_RE, _NAME_RE), fields) for f in fields_m: - argnames = f.group('args').split(',') + argnames = self.build_arglist(f.group('args')) obj[remove_quotes(f.group('key'))] = self.build_function(argnames, f.group('code')) return obj @@ -510,15 +620,19 @@ class JSInterpreter(object): def extract_function_code(self, funcname): """ @returns argnames, code """ func_m = re.search( - r'''(?x) - (?:function\s+%(f_n)s|[{;,]\s*%(f_n)s\s*=\s*function|var\s+%(f_n)s\s*=\s*function)\s* + r'''(?xs) + (?: + function\s+%(name)s| + [{;,]\s*%(name)s\s*=\s*function| + (?:var|const|let)\s+%(name)s\s*=\s*function + )\s* \((?P[^)]*)\)\s* - (?P\{(?:(?!};)[^"]|"([^"]|\\")*")+\})''' % {'f_n': re.escape(funcname), }, + (?P{.+})''' % {'name': re.escape(funcname)}, self.code) code, _ = self._separate_at_paren(func_m.group('code'), '}') # refine the match if func_m is None: - raise ExtractorError('Could not find JS function %r' % funcname) - return func_m.group('args').split(','), code + raise self.Exception('Could not find JS function "{funcname}"'.format(**locals())) + return self.build_arglist(func_m.group('args')), code def extract_function(self, funcname): return self.extract_function_from_code(*self.extract_function_code(funcname)) @@ -534,7 +648,7 @@ class JSInterpreter(object): name = self._named_object( local_vars, self.extract_function_from_code( - [x.strip() for x in mobj.group('args').split(',')], + self.build_arglist(mobj.group('args')), body, local_vars, *global_stack)) code = code[:start] + name + remaining return self.build_function(argnames, code, local_vars, *global_stack) @@ -542,17 +656,22 @@ class JSInterpreter(object): def call_function(self, funcname, *args): return self.extract_function(funcname)(args) + @classmethod + def build_arglist(cls, arg_text): + if not arg_text: + return [] + return list(filter(None, (x.strip() or None for x in cls._separate(arg_text)))) + def build_function(self, argnames, code, *global_stack): global_stack = list(global_stack) or [{}] - local_vars = global_stack.pop(0) + argnames = tuple(argnames) - def resf(args, **kwargs): - local_vars.update(dict(zip(argnames, args))) - local_vars.update(kwargs) - var_stack = LocalNameSpace(local_vars, *global_stack) - for stmt in self._separate(code.replace('\n', ''), ';'): - ret, should_abort = self.interpret_statement(stmt, var_stack) - if should_abort: - break - return ret + def resf(args, kwargs={}, allow_recursion=100): + global_stack[0].update( + zip_longest(argnames, args, fillvalue=None)) + global_stack[0].update(kwargs) + var_stack = LocalNameSpace(*global_stack) + ret, should_abort = self.interpret_statement(code.replace('\n', ''), var_stack, allow_recursion - 1) + if should_abort: + return ret return resf diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 4e00317f1..a5f584ec5 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1696,6 +1696,17 @@ MONTH_NAMES = { 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'], } +# Timezone names for RFC2822 obs-zone +# From https://github.com/python/cpython/blob/3.11/Lib/email/_parseaddr.py#L36-L42 +TIMEZONE_NAMES = { + 'UT': 0, 'UTC': 0, 'GMT': 0, 'Z': 0, + 'AST': -4, 'ADT': -3, # Atlantic (used in Canada) + 'EST': -5, 'EDT': -4, # Eastern + 'CST': -6, 'CDT': -5, # Central + 'MST': -7, 'MDT': -6, # Mountain + 'PST': -8, 'PDT': -7 # Pacific +} + KNOWN_EXTENSIONS = ( 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac', 'flv', 'f4v', 'f4a', 'f4b', @@ -1735,12 +1746,17 @@ DATE_FORMATS = ( '%b %dth %Y %I:%M', '%Y %m %d', '%Y-%m-%d', + '%Y.%m.%d.', '%Y/%m/%d', '%Y/%m/%d %H:%M', '%Y/%m/%d %H:%M:%S', + '%Y%m%d%H%M', + '%Y%m%d%H%M%S', + '%Y%m%d', '%Y-%m-%d %H:%M', '%Y-%m-%d %H:%M:%S', '%Y-%m-%d %H:%M:%S.%f', + '%Y-%m-%d %H:%M:%S:%f', '%d.%m.%Y %H:%M', '%d.%m.%Y %H.%M', '%Y-%m-%dT%H:%M:%SZ', @@ -1753,6 +1769,7 @@ DATE_FORMATS = ( '%b %d %Y at %H:%M:%S', '%B %d %Y at %H:%M', '%B %d %Y at %H:%M:%S', + '%H:%M %d-%b-%Y', ) DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) @@ -1763,6 +1780,7 @@ DATE_FORMATS_DAY_FIRST.extend([ '%d/%m/%Y', '%d/%m/%y', '%d/%m/%Y %H:%M:%S', + '%d-%m-%Y %H:%M', ]) DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) @@ -2966,10 +2984,22 @@ class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler): def extract_timezone(date_str): m = re.search( - r'^.{8,}?(?PZ$| ?(?P\+|-)(?P[0-9]{2}):?(?P[0-9]{2})$)', - date_str) + r'''(?x) + ^.{8,}? # >=8 char non-TZ prefix, if present + (?PZ| # just the UTC Z, or + (?:(?<=.\b\d{4}|\b\d{2}:\d\d)| # preceded by 4 digits or hh:mm or + (?= 4 alpha or 2 digits + [ ]? # optional space + (?P\+|-) # +/- + (?P[0-9]{2}):?(?P[0-9]{2}) # hh[:]mm + $) + ''', date_str) if not m: - timezone = datetime.timedelta() + m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P\s*[A-Z]+)$', date_str) + timezone = TIMEZONE_NAMES.get(m and m.group('tz').strip()) + if timezone is not None: + date_str = date_str[:-len(m.group('tz'))] + timezone = datetime.timedelta(hours=timezone or 0) else: date_str = date_str[:-len(m.group('tz'))] if not m.group('sign'): @@ -3037,7 +3067,8 @@ def unified_timestamp(date_str, day_first=True): if date_str is None: return None - date_str = re.sub(r'[,|]', '', date_str) + date_str = re.sub(r'\s+', ' ', re.sub( + r'(?i)[,|]|(mon|tues?|wed(nes)?|thu(rs)?|fri|sat(ur)?)(day)?', '', date_str)) pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0 timezone, date_str = extract_timezone(date_str) @@ -3063,7 +3094,7 @@ def unified_timestamp(date_str, day_first=True): pass timetuple = email.utils.parsedate_tz(date_str) if timetuple: - return calendar.timegm(timetuple) + pm_delta * 3600 + return calendar.timegm(timetuple) + pm_delta * 3600 - timezone.total_seconds() def determine_ext(url, default_ext='unknown_video'): @@ -3673,13 +3704,11 @@ def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): if get_attr: if v is not None: v = getattr(v, get_attr, None) - if v == '': - v = None - if v is None: + if v in (None, ''): return default try: return int(v) * invscale // scale - except (ValueError, TypeError): + except (ValueError, TypeError, OverflowError): return default From e52e8b8111cf7ca27daef184bacd926865e951b1 Mon Sep 17 00:00:00 2001 From: dirkf Date: Mon, 15 Aug 2022 16:45:04 +0100 Subject: [PATCH 4/4] [postprocessor] Don't replace existing value with null metadata parsed from title --- youtube_dl/postprocessor/metadatafromtitle.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/youtube_dl/postprocessor/metadatafromtitle.py b/youtube_dl/postprocessor/metadatafromtitle.py index f5c14d974..6cd5bb70f 100644 --- a/youtube_dl/postprocessor/metadatafromtitle.py +++ b/youtube_dl/postprocessor/metadatafromtitle.py @@ -40,6 +40,8 @@ class MetadataFromTitlePP(PostProcessor): % self._titleformat) return [], info for attribute, value in match.groupdict().items(): + if value is None: + continue info[attribute] = value self._downloader.to_screen( '[fromtitle] parsed %s: %s'