2 # -*- coding: utf-8 -*-
20 import xml.etree.ElementTree
24 import urllib.request as compat_urllib_request
25 except ImportError: # Python 2
26 import urllib2 as compat_urllib_request
29 import urllib.error as compat_urllib_error
30 except ImportError: # Python 2
31 import urllib2 as compat_urllib_error
34 import urllib.parse as compat_urllib_parse
35 except ImportError: # Python 2
36 import urllib as compat_urllib_parse
39 from urllib.parse import urlparse as compat_urllib_parse_urlparse
40 except ImportError: # Python 2
41 from urlparse import urlparse as compat_urllib_parse_urlparse
44 import urllib.parse as compat_urlparse
45 except ImportError: # Python 2
46 import urlparse as compat_urlparse
49 import http.cookiejar as compat_cookiejar
50 except ImportError: # Python 2
51 import cookielib as compat_cookiejar
54 import html.entities as compat_html_entities
55 except ImportError: # Python 2
56 import htmlentitydefs as compat_html_entities
59 import html.parser as compat_html_parser
60 except ImportError: # Python 2
61 import HTMLParser as compat_html_parser
64 import http.client as compat_http_client
65 except ImportError: # Python 2
66 import httplib as compat_http_client
69 from urllib.error import HTTPError as compat_HTTPError
70 except ImportError: # Python 2
71 from urllib2 import HTTPError as compat_HTTPError
74 from urllib.request import urlretrieve as compat_urlretrieve
75 except ImportError: # Python 2
76 from urllib import urlretrieve as compat_urlretrieve
80 from subprocess import DEVNULL
81 compat_subprocess_get_DEVNULL = lambda: DEVNULL
83 compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
86 from urllib.parse import parse_qs as compat_parse_qs
87 except ImportError: # Python 2
88 # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib.
89 # Python 2's version is apparently totally broken
90 def _unquote(string, encoding='utf-8', errors='replace'):
93 res = string.split('%')
100 # pct_sequence: contiguous sequence of percent-encoded bytes, decoded
107 pct_sequence += item[:2].decode('hex')
110 # This segment was just a single percent-encoded character.
111 # May be part of a sequence of code units, so delay decoding.
112 # (Stored in pct_sequence).
116 # Encountered non-percent-encoded characters. Flush the current
118 string += pct_sequence.decode(encoding, errors) + rest
121 # Flush the final pct_sequence
122 string += pct_sequence.decode(encoding, errors)
125 def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False,
126 encoding='utf-8', errors='replace'):
127 qs, _coerce_result = qs, unicode
128 pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')]
130 for name_value in pairs:
131 if not name_value and not strict_parsing:
133 nv = name_value.split('=', 1)
136 raise ValueError("bad query field: %r" % (name_value,))
137 # Handle case of a control-name with no equal sign
138 if keep_blank_values:
142 if len(nv[1]) or keep_blank_values:
143 name = nv[0].replace('+', ' ')
144 name = _unquote(name, encoding=encoding, errors=errors)
145 name = _coerce_result(name)
146 value = nv[1].replace('+', ' ')
147 value = _unquote(value, encoding=encoding, errors=errors)
148 value = _coerce_result(value)
149 r.append((name, value))
152 def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False,
153 encoding='utf-8', errors='replace'):
155 pairs = _parse_qsl(qs, keep_blank_values, strict_parsing,
156 encoding=encoding, errors=errors)
157 for name, value in pairs:
158 if name in parsed_result:
159 parsed_result[name].append(value)
161 parsed_result[name] = [value]
165 compat_str = unicode # Python 2
170 compat_chr = unichr # Python 2
175 if type(c) is int: return c
178 # This is not clearly defined otherwise
179 compiled_regex_type = type(re.compile(''))
182 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome)',
183 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
184 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
185 'Accept-Encoding': 'gzip, deflate',
186 'Accept-Language': 'en-us,en;q=0.5',
189 def preferredencoding():
190 """Get preferred encoding.
192 Returns the best encoding scheme for the system, based on
193 locale.getpreferredencoding() and some further tweaks.
196 pref = locale.getpreferredencoding()
203 if sys.version_info < (3,0):
205 print(s.encode(preferredencoding(), 'xmlcharrefreplace'))
208 assert type(s) == type(u'')
211 # In Python 2.x, json.dump expects a bytestream.
212 # In Python 3.x, it writes to a character stream
213 if sys.version_info < (3,0):
214 def write_json_file(obj, fn):
215 with open(fn, 'wb') as f:
218 def write_json_file(obj, fn):
219 with open(fn, 'w', encoding='utf-8') as f:
222 if sys.version_info >= (2,7):
223 def find_xpath_attr(node, xpath, key, val):
224 """ Find the xpath xpath[@key=val] """
225 assert re.match(r'^[a-zA-Z]+$', key)
226 assert re.match(r'^[a-zA-Z0-9@\s]*$', val)
227 expr = xpath + u"[@%s='%s']" % (key, val)
228 return node.find(expr)
230 def find_xpath_attr(node, xpath, key, val):
231 for f in node.findall(xpath):
232 if f.attrib.get(key) == val:
236 # On python2.6 the xml.etree.ElementTree.Element methods don't support
237 # the namespace parameter
238 def xpath_with_ns(path, ns_map):
239 components = [c.split(':') for c in path.split('/')]
243 replaced.append(c[0])
246 replaced.append('{%s}%s' % (ns_map[ns], tag))
247 return '/'.join(replaced)
249 def htmlentity_transform(matchobj):
250 """Transforms an HTML entity to a character.
252 This function receives a match object and is intended to be used with
253 the re.sub() function.
255 entity = matchobj.group(1)
257 # Known non-numeric HTML entity
258 if entity in compat_html_entities.name2codepoint:
259 return compat_chr(compat_html_entities.name2codepoint[entity])
261 mobj = re.match(u'(?u)#(x?\\d+)', entity)
263 numstr = mobj.group(1)
264 if numstr.startswith(u'x'):
266 numstr = u'0%s' % numstr
269 return compat_chr(int(numstr, base))
271 # Unknown entity in name, return its literal representation
272 return (u'&%s;' % entity)
274 compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
275 class BaseHTMLParser(compat_html_parser.HTMLParser):
277 compat_html_parser.HTMLParser.__init__(self)
280 def loads(self, html):
285 class AttrParser(BaseHTMLParser):
286 """Modified HTMLParser that isolates a tag with the specified attribute"""
287 def __init__(self, attribute, value):
288 self.attribute = attribute
293 self.watch_startpos = False
295 BaseHTMLParser.__init__(self)
297 def error(self, message):
298 if self.error_count > 10 or self.started:
299 raise compat_html_parser.HTMLParseError(message, self.getpos())
300 self.rawdata = '\n'.join(self.html.split('\n')[self.getpos()[0]:]) # skip one line
301 self.error_count += 1
304 def handle_starttag(self, tag, attrs):
307 self.find_startpos(None)
308 if self.attribute in attrs and attrs[self.attribute] == self.value:
311 self.watch_startpos = True
313 if not tag in self.depth: self.depth[tag] = 0
316 def handle_endtag(self, tag):
318 if tag in self.depth: self.depth[tag] -= 1
319 if self.depth[self.result[0]] == 0:
321 self.result.append(self.getpos())
323 def find_startpos(self, x):
324 """Needed to put the start position of the result (self.result[1])
325 after the opening tag with the requested id"""
326 if self.watch_startpos:
327 self.watch_startpos = False
328 self.result.append(self.getpos())
329 handle_entityref = handle_charref = handle_data = handle_comment = \
330 handle_decl = handle_pi = unknown_decl = find_startpos
332 def get_result(self):
333 if self.result is None:
335 if len(self.result) != 3:
337 lines = self.html.split('\n')
338 lines = lines[self.result[1][0]-1:self.result[2][0]]
339 lines[0] = lines[0][self.result[1][1]:]
341 lines[-1] = lines[-1][:self.result[2][1]-self.result[1][1]]
342 lines[-1] = lines[-1][:self.result[2][1]]
343 return '\n'.join(lines).strip()
344 # Hack for https://github.com/rg3/youtube-dl/issues/662
345 if sys.version_info < (2, 7, 3):
346 AttrParser.parse_endtag = (lambda self, i:
347 i + len("</scr'+'ipt>")
348 if self.rawdata[i:].startswith("</scr'+'ipt>")
349 else compat_html_parser.HTMLParser.parse_endtag(self, i))
351 def get_element_by_id(id, html):
352 """Return the content of the tag with the specified ID in the passed HTML document"""
353 return get_element_by_attribute("id", id, html)
355 def get_element_by_attribute(attribute, value, html):
356 """Return the content of the tag with the specified attribute in the passed HTML document"""
357 parser = AttrParser(attribute, value)
360 except compat_html_parser.HTMLParseError:
362 return parser.get_result()
364 class MetaParser(BaseHTMLParser):
366 Modified HTMLParser that isolates a meta tag with the specified name
369 def __init__(self, name):
370 BaseHTMLParser.__init__(self)
375 def handle_starttag(self, tag, attrs):
379 if attrs.get('name') == self.name:
380 self.result = attrs.get('content')
382 def get_result(self):
385 def get_meta_content(name, html):
387 Return the content attribute from the meta tag with the given name attribute.
389 parser = MetaParser(name)
392 except compat_html_parser.HTMLParseError:
394 return parser.get_result()
397 def clean_html(html):
398 """Clean an HTML snippet into a readable string"""
400 html = html.replace('\n', ' ')
401 html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html)
402 html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html)
404 html = re.sub('<.*?>', '', html)
405 # Replace html entities
406 html = unescapeHTML(html)
410 def sanitize_open(filename, open_mode):
411 """Try to open the given filename, and slightly tweak it if this fails.
413 Attempts to open the given filename. If this fails, it tries to change
414 the filename slightly, step by step, until it's either able to open it
415 or it fails and raises a final exception, like the standard open()
418 It returns the tuple (stream, definitive_file_name).
422 if sys.platform == 'win32':
424 msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
425 return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename)
426 stream = open(encodeFilename(filename), open_mode)
427 return (stream, filename)
428 except (IOError, OSError) as err:
429 if err.errno in (errno.EACCES,):
432 # In case of error, try to remove win32 forbidden chars
433 alt_filename = os.path.join(
434 re.sub(u'[/<>:"\\|\\\\?\\*]', u'#', path_part)
435 for path_part in os.path.split(filename)
437 if alt_filename == filename:
440 # An exception here should be caught in the caller
441 stream = open(encodeFilename(filename), open_mode)
442 return (stream, alt_filename)
445 def timeconvert(timestr):
446 """Convert RFC 2822 defined time string into system timestamp"""
448 timetuple = email.utils.parsedate_tz(timestr)
449 if timetuple is not None:
450 timestamp = email.utils.mktime_tz(timetuple)
453 def sanitize_filename(s, restricted=False, is_id=False):
454 """Sanitizes a string so it could be used as part of a filename.
455 If restricted is set, use a stricter subset of allowed characters.
456 Set is_id if this is not an arbitrary string, but an ID that should be kept if possible
458 def replace_insane(char):
459 if char == '?' or ord(char) < 32 or ord(char) == 127:
462 return '' if restricted else '\''
464 return '_-' if restricted else ' -'
465 elif char in '\\/|*<>':
467 if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()):
469 if restricted and ord(char) > 127:
473 result = u''.join(map(replace_insane, s))
475 while '__' in result:
476 result = result.replace('__', '_')
477 result = result.strip('_')
478 # Common case of "Foreign band name - English song title"
479 if restricted and result.startswith('-_'):
485 def orderedSet(iterable):
486 """ Remove all duplicates from the input iterable """
497 assert type(s) == type(u'')
499 result = re.sub(u'(?u)&(.+?);', htmlentity_transform, s)
502 def encodeFilename(s):
504 @param s The name of the file
507 assert type(s) == type(u'')
509 # Python 3 has a Unicode API
510 if sys.version_info >= (3, 0):
513 if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5:
514 # Pass u'' directly to use Unicode APIs on Windows 2000 and up
515 # (Detecting Windows NT 4 is tricky because 'major >= 4' would
516 # match Windows 9x series as well. Besides, NT 4 is obsolete.)
519 encoding = sys.getfilesystemencoding()
522 return s.encode(encoding, 'ignore')
524 def decodeOption(optval):
527 if isinstance(optval, bytes):
528 optval = optval.decode(preferredencoding())
530 assert isinstance(optval, compat_str)
533 def formatSeconds(secs):
535 return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60)
537 return '%d:%02d' % (secs // 60, secs % 60)
542 def make_HTTPS_handler(opts):
543 if sys.version_info < (3, 2):
546 class HTTPSConnectionV3(httplib.HTTPSConnection):
547 def __init__(self, *args, **kwargs):
548 httplib.HTTPSConnection.__init__(self, *args, **kwargs)
551 sock = socket.create_connection((self.host, self.port), self.timeout)
552 if self._tunnel_host:
556 self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv3)
557 except ssl.SSLError as e:
558 self.sock = ssl.wrap_socket(sock, self.key_file, self.cert_file, ssl_version=ssl.PROTOCOL_SSLv23)
560 class HTTPSHandlerV3(compat_urllib_request.HTTPSHandler):
561 def https_open(self, req):
562 return self.do_open(HTTPSConnectionV3, req)
563 return HTTPSHandlerV3()
565 context = ssl.SSLContext(ssl.PROTOCOL_SSLv3)
566 context.set_default_verify_paths()
568 context.verify_mode = (ssl.CERT_NONE
569 if opts.no_check_certificate
570 else ssl.CERT_REQUIRED)
571 return compat_urllib_request.HTTPSHandler(context=context)
573 class ExtractorError(Exception):
574 """Error during info extraction."""
575 def __init__(self, msg, tb=None, expected=False, cause=None):
576 """ tb, if given, is the original traceback (so that it can be printed out).
577 If expected is set, this is a normal error message and most likely not a bug in youtube-dl.
580 if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
583 msg = msg + u'; please report this issue on https://yt-dl.org/bug . Be sure to call youtube-dl with the --verbose flag and include its complete output. Make sure you are using the latest version; type youtube-dl -U to update.'
584 super(ExtractorError, self).__init__(msg)
587 self.exc_info = sys.exc_info() # preserve original exception
590 def format_traceback(self):
591 if self.traceback is None:
593 return u''.join(traceback.format_tb(self.traceback))
596 class RegexNotFoundError(ExtractorError):
597 """Error when a regex didn't match"""
601 class DownloadError(Exception):
602 """Download Error exception.
604 This exception may be thrown by FileDownloader objects if they are not
605 configured to continue on errors. They will contain the appropriate
608 def __init__(self, msg, exc_info=None):
609 """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """
610 super(DownloadError, self).__init__(msg)
611 self.exc_info = exc_info
614 class SameFileError(Exception):
615 """Same File exception.
617 This exception will be thrown by FileDownloader objects if they detect
618 multiple files would have to be downloaded to the same file on disk.
623 class PostProcessingError(Exception):
624 """Post Processing exception.
626 This exception may be raised by PostProcessor's .run() method to
627 indicate an error in the postprocessing task.
629 def __init__(self, msg):
632 class MaxDownloadsReached(Exception):
633 """ --max-downloads limit has been reached. """
637 class UnavailableVideoError(Exception):
638 """Unavailable Format exception.
640 This exception will be thrown when a video is requested
641 in a format that is not available for that video.
646 class ContentTooShortError(Exception):
647 """Content Too Short exception.
649 This exception may be raised by FileDownloader objects when a file they
650 download is too small for what the server announced first, indicating
651 the connection was probably interrupted.
657 def __init__(self, downloaded, expected):
658 self.downloaded = downloaded
659 self.expected = expected
661 class YoutubeDLHandler(compat_urllib_request.HTTPHandler):
662 """Handler for HTTP requests and responses.
664 This class, when installed with an OpenerDirector, automatically adds
665 the standard headers to every HTTP request and handles gzipped and
666 deflated responses from web servers. If compression is to be avoided in
667 a particular request, the original request in the program code only has
668 to include the HTTP header "Youtubedl-No-Compression", which will be
669 removed before making the real request.
671 Part of this code was copied from:
673 http://techknack.net/python-urllib2-handlers/
675 Andrew Rowls, the author of that code, agreed to release it to the
682 return zlib.decompress(data, -zlib.MAX_WBITS)
684 return zlib.decompress(data)
687 def addinfourl_wrapper(stream, headers, url, code):
688 if hasattr(compat_urllib_request.addinfourl, 'getcode'):
689 return compat_urllib_request.addinfourl(stream, headers, url, code)
690 ret = compat_urllib_request.addinfourl(stream, headers, url)
694 def http_request(self, req):
695 for h,v in std_headers.items():
699 if 'Youtubedl-no-compression' in req.headers:
700 if 'Accept-encoding' in req.headers:
701 del req.headers['Accept-encoding']
702 del req.headers['Youtubedl-no-compression']
703 if 'Youtubedl-user-agent' in req.headers:
704 if 'User-agent' in req.headers:
705 del req.headers['User-agent']
706 req.headers['User-agent'] = req.headers['Youtubedl-user-agent']
707 del req.headers['Youtubedl-user-agent']
710 def http_response(self, req, resp):
713 if resp.headers.get('Content-encoding', '') == 'gzip':
714 content = resp.read()
715 gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb')
717 uncompressed = io.BytesIO(gz.read())
718 except IOError as original_ioerror:
719 # There may be junk add the end of the file
720 # See http://stackoverflow.com/q/4928560/35070 for details
721 for i in range(1, 1024):
723 gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb')
724 uncompressed = io.BytesIO(gz.read())
729 raise original_ioerror
730 resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
731 resp.msg = old_resp.msg
733 if resp.headers.get('Content-encoding', '') == 'deflate':
734 gz = io.BytesIO(self.deflate(resp.read()))
735 resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code)
736 resp.msg = old_resp.msg
739 https_request = http_request
740 https_response = http_response
742 def unified_strdate(date_str):
743 """Return a string with the date in the format YYYYMMDD"""
746 date_str = date_str.replace(',',' ')
747 # %z (UTC offset) is only supported in python>=3.2
748 date_str = re.sub(r' (\+|-)[\d]*$', '', date_str)
749 format_expressions = [
757 '%Y-%m-%dT%H:%M:%SZ',
758 '%Y-%m-%dT%H:%M:%S.%fZ',
759 '%Y-%m-%dT%H:%M:%S.%f0Z',
762 for expression in format_expressions:
764 upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d')
769 def determine_ext(url, default_ext=u'unknown_video'):
770 guess = url.partition(u'?')[0].rpartition(u'.')[2]
771 if re.match(r'^[A-Za-z0-9]+$', guess):
776 def subtitles_filename(filename, sub_lang, sub_format):
777 return filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format
779 def date_from_str(date_str):
781 Return a datetime object from a string in the format YYYYMMDD or
782 (now|today)[+-][0-9](day|week|month|year)(s)?"""
783 today = datetime.date.today()
784 if date_str == 'now'or date_str == 'today':
786 match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
787 if match is not None:
788 sign = match.group('sign')
789 time = int(match.group('time'))
792 unit = match.group('unit')
801 delta = datetime.timedelta(**{unit: time})
803 return datetime.datetime.strptime(date_str, "%Y%m%d").date()
805 class DateRange(object):
806 """Represents a time interval between two dates"""
807 def __init__(self, start=None, end=None):
808 """start and end must be strings in the format accepted by date"""
809 if start is not None:
810 self.start = date_from_str(start)
812 self.start = datetime.datetime.min.date()
814 self.end = date_from_str(end)
816 self.end = datetime.datetime.max.date()
817 if self.start > self.end:
818 raise ValueError('Date range: "%s" , the start date must be before the end date' % self)
821 """Returns a range that only contains the given day"""
823 def __contains__(self, date):
824 """Check if the date is in the range"""
825 if not isinstance(date, datetime.date):
826 date = date_from_str(date)
827 return self.start <= date <= self.end
829 return '%s - %s' % ( self.start.isoformat(), self.end.isoformat())
833 """ Returns the platform name as a compat_str """
834 res = platform.platform()
835 if isinstance(res, bytes):
836 res = res.decode(preferredencoding())
838 assert isinstance(res, compat_str)
842 def write_string(s, out=None):
845 assert type(s) == type(u'')
847 if ('b' in getattr(out, 'mode', '') or
848 sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr
849 s = s.encode(preferredencoding(), 'ignore')
854 def bytes_to_intlist(bs):
857 if isinstance(bs[0], int): # Python 3
860 return [ord(c) for c in bs]
863 def intlist_to_bytes(xs):
866 if isinstance(chr(0), bytes): # Python 2
867 return ''.join([chr(x) for x in xs])
872 def get_cachedir(params={}):
873 cache_root = os.environ.get('XDG_CACHE_HOME',
874 os.path.expanduser('~/.cache'))
875 return params.get('cachedir', os.path.join(cache_root, 'youtube-dl'))
878 # Cross-platform file locking
879 if sys.platform == 'win32':
880 import ctypes.wintypes
883 class OVERLAPPED(ctypes.Structure):
885 ('Internal', ctypes.wintypes.LPVOID),
886 ('InternalHigh', ctypes.wintypes.LPVOID),
887 ('Offset', ctypes.wintypes.DWORD),
888 ('OffsetHigh', ctypes.wintypes.DWORD),
889 ('hEvent', ctypes.wintypes.HANDLE),
892 kernel32 = ctypes.windll.kernel32
893 LockFileEx = kernel32.LockFileEx
894 LockFileEx.argtypes = [
895 ctypes.wintypes.HANDLE, # hFile
896 ctypes.wintypes.DWORD, # dwFlags
897 ctypes.wintypes.DWORD, # dwReserved
898 ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
899 ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
900 ctypes.POINTER(OVERLAPPED) # Overlapped
902 LockFileEx.restype = ctypes.wintypes.BOOL
903 UnlockFileEx = kernel32.UnlockFileEx
904 UnlockFileEx.argtypes = [
905 ctypes.wintypes.HANDLE, # hFile
906 ctypes.wintypes.DWORD, # dwReserved
907 ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow
908 ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh
909 ctypes.POINTER(OVERLAPPED) # Overlapped
911 UnlockFileEx.restype = ctypes.wintypes.BOOL
912 whole_low = 0xffffffff
913 whole_high = 0x7fffffff
915 def _lock_file(f, exclusive):
916 overlapped = OVERLAPPED()
917 overlapped.Offset = 0
918 overlapped.OffsetHigh = 0
919 overlapped.hEvent = 0
920 f._lock_file_overlapped_p = ctypes.pointer(overlapped)
921 handle = msvcrt.get_osfhandle(f.fileno())
922 if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0,
923 whole_low, whole_high, f._lock_file_overlapped_p):
924 raise OSError('Locking file failed: %r' % ctypes.FormatError())
927 assert f._lock_file_overlapped_p
928 handle = msvcrt.get_osfhandle(f.fileno())
929 if not UnlockFileEx(handle, 0,
930 whole_low, whole_high, f._lock_file_overlapped_p):
931 raise OSError('Unlocking file failed: %r' % ctypes.FormatError())
936 def _lock_file(f, exclusive):
937 fcntl.lockf(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH)
940 fcntl.lockf(f, fcntl.LOCK_UN)
943 class locked_file(object):
944 def __init__(self, filename, mode, encoding=None):
945 assert mode in ['r', 'a', 'w']
946 self.f = io.open(filename, mode, encoding=encoding)
950 exclusive = self.mode != 'r'
952 _lock_file(self.f, exclusive)
958 def __exit__(self, etype, value, traceback):
967 def write(self, *args):
968 return self.f.write(*args)
970 def read(self, *args):
971 return self.f.read(*args)
974 def shell_quote(args):
976 encoding = sys.getfilesystemencoding()
980 if isinstance(a, bytes):
981 # We may get a filename encoded with 'encodeFilename'
982 a = a.decode(encoding)
983 quoted_args.append(pipes.quote(a))
984 return u' '.join(quoted_args)
987 def takewhile_inclusive(pred, seq):
988 """ Like itertools.takewhile, but include the latest evaluated element
989 (the first element so that Not pred(e)) """
996 def smuggle_url(url, data):
997 """ Pass additional data in a URL for internal use. """
999 sdata = compat_urllib_parse.urlencode(
1000 {u'__youtubedl_smuggle': json.dumps(data)})
1001 return url + u'#' + sdata
1004 def unsmuggle_url(smug_url):
1005 if not '#__youtubedl_smuggle' in smug_url:
1006 return smug_url, None
1007 url, _, sdata = smug_url.rpartition(u'#')
1008 jsond = compat_parse_qs(sdata)[u'__youtubedl_smuggle'][0]
1009 data = json.loads(jsond)
1013 def parse_xml_doc(s):
1014 assert isinstance(s, type(u''))
1015 return xml.etree.ElementTree.fromstring(s.encode('utf-8'))
1018 def format_bytes(bytes):
1021 if type(bytes) is str:
1022 bytes = float(bytes)
1026 exponent = int(math.log(bytes, 1024.0))
1027 suffix = [u'B', u'KiB', u'MiB', u'GiB', u'TiB', u'PiB', u'EiB', u'ZiB', u'YiB'][exponent]
1028 converted = float(bytes) / float(1024 ** exponent)
1029 return u'%.2f%s' % (converted, suffix)