X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=a03f7184ddf5b8303be799e9857d1416bd7d4aec;hb=3b9323d96e2be7341d2e309124075134ec8a50ca;hp=c8308ba3a3b7e3473e53423b45c7950b57eb5aef;hpb=329ca3bef695bff011ed9b2d5f03e1331bf5bf0f;p=youtube-dl diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index c8308ba3a..a03f7184d 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -47,6 +47,7 @@ from .compat import ( compat_socket_create_connection, compat_str, compat_struct_pack, + compat_struct_unpack, compat_urllib_error, compat_urllib_parse, compat_urllib_parse_urlencode, @@ -110,6 +111,50 @@ ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐ،٠itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'], 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy'))) +DATE_FORMATS = ( + '%d %B %Y', + '%d %b %Y', + '%B %d %Y', + '%b %d %Y', + '%b %dst %Y %I:%M', + '%b %dnd %Y %I:%M', + '%b %dth %Y %I:%M', + '%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:%S', + '%Y-%m-%d %H:%M:%S.%f', + '%d.%m.%Y %H:%M', + '%d.%m.%Y %H.%M', + '%Y-%m-%dT%H:%M:%SZ', + '%Y-%m-%dT%H:%M:%S.%fZ', + '%Y-%m-%dT%H:%M:%S.%f0Z', + '%Y-%m-%dT%H:%M:%S', + '%Y-%m-%dT%H:%M:%S.%f', + '%Y-%m-%dT%H:%M', +) + +DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) +DATE_FORMATS_DAY_FIRST.extend([ + '%d-%m-%Y', + '%d.%m.%Y', + '%d.%m.%y', + '%d/%m/%Y', + '%d/%m/%y', + '%d/%m/%Y %H:%M:%S', +]) + +DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) +DATE_FORMATS_MONTH_FIRST.extend([ + '%m-%d-%Y', + '%m.%d.%Y', + '%m/%d/%Y', + '%m/%d/%y', + '%m/%d/%Y %H:%M:%S', +]) + def preferredencoding(): """Get preferred encoding. @@ -267,9 +312,17 @@ def get_element_by_id(id, html): return get_element_by_attribute('id', id, html) -def get_element_by_attribute(attribute, value, html): +def get_element_by_class(class_name, html): + return get_element_by_attribute( + 'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name), + html, escape_value=False) + + +def get_element_by_attribute(attribute, value, html, escape_value=True): """Return the content of the tag with the specified attribute in the passed HTML document""" + value = re.escape(value) if escape_value else value + m = re.search(r'''(?xs) <([a-zA-Z0-9:._-]+) (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*? @@ -278,7 +331,7 @@ def get_element_by_attribute(attribute, value, html): \s*> (?P.*?) - ''' % (re.escape(attribute), re.escape(value)), html) + ''' % (re.escape(attribute), value), html) if not m: return None @@ -975,6 +1028,24 @@ class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor): https_response = http_response +def extract_timezone(date_str): + m = re.search( + r'^.{8,}?(?PZ$| ?(?P\+|-)(?P[0-9]{2}):?(?P[0-9]{2})$)', + date_str) + if not m: + timezone = datetime.timedelta() + else: + date_str = date_str[:-len(m.group('tz'))] + if not m.group('sign'): + timezone = datetime.timedelta() + else: + sign = 1 if m.group('sign') == '+' else -1 + timezone = datetime.timedelta( + hours=sign * int(m.group('hours')), + minutes=sign * int(m.group('minutes'))) + return timezone, date_str + + def parse_iso8601(date_str, delimiter='T', timezone=None): """ Return a UNIX timestamp from the given date """ @@ -984,20 +1055,8 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): date_str = re.sub(r'\.[0-9]+', '', date_str) if timezone is None: - m = re.search( - r'(?:Z$| ?(?P\+|-)(?P[0-9]{2}):?(?P[0-9]{2})$)', - date_str) - if not m: - timezone = datetime.timedelta() - else: - date_str = date_str[:-len(m.group(0))] - if not m.group('sign'): - timezone = datetime.timedelta() - else: - sign = 1 if m.group('sign') == '+' else -1 - timezone = datetime.timedelta( - hours=sign * int(m.group('hours')), - minutes=sign * int(m.group('minutes'))) + timezone, date_str = extract_timezone(date_str) + try: date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) dt = datetime.datetime.strptime(date_str, date_format) - timezone @@ -1006,6 +1065,10 @@ def parse_iso8601(date_str, delimiter='T', timezone=None): pass +def date_formats(day_first=True): + return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST + + def unified_strdate(date_str, day_first=True): """Return a string with the date in the format YYYYMMDD""" @@ -1014,53 +1077,11 @@ def unified_strdate(date_str, day_first=True): upload_date = None # Replace commas date_str = date_str.replace(',', ' ') - # %z (UTC offset) is only supported in python>=3.2 - if not re.match(r'^[0-9]{1,2}-[0-9]{1,2}-[0-9]{4}$', date_str): - date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str) # Remove AM/PM + timezone date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) + _, date_str = extract_timezone(date_str) - format_expressions = [ - '%d %B %Y', - '%d %b %Y', - '%B %d %Y', - '%b %d %Y', - '%b %dst %Y %I:%M', - '%b %dnd %Y %I:%M', - '%b %dth %Y %I:%M', - '%Y %m %d', - '%Y-%m-%d', - '%Y/%m/%d', - '%Y/%m/%d %H:%M:%S', - '%Y-%m-%d %H:%M:%S', - '%Y-%m-%d %H:%M:%S.%f', - '%d.%m.%Y %H:%M', - '%d.%m.%Y %H.%M', - '%Y-%m-%dT%H:%M:%SZ', - '%Y-%m-%dT%H:%M:%S.%fZ', - '%Y-%m-%dT%H:%M:%S.%f0Z', - '%Y-%m-%dT%H:%M:%S', - '%Y-%m-%dT%H:%M:%S.%f', - '%Y-%m-%dT%H:%M', - ] - if day_first: - format_expressions.extend([ - '%d-%m-%Y', - '%d.%m.%Y', - '%d.%m.%y', - '%d/%m/%Y', - '%d/%m/%y', - '%d/%m/%Y %H:%M:%S', - ]) - else: - format_expressions.extend([ - '%m-%d-%Y', - '%m.%d.%Y', - '%m/%d/%Y', - '%m/%d/%y', - '%m/%d/%Y %H:%M:%S', - ]) - for expression in format_expressions: + for expression in date_formats(day_first): try: upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') except ValueError: @@ -1076,6 +1097,29 @@ def unified_strdate(date_str, day_first=True): return compat_str(upload_date) +def unified_timestamp(date_str, day_first=True): + if date_str is None: + return None + + date_str = date_str.replace(',', ' ') + + pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0 + timezone, date_str = extract_timezone(date_str) + + # Remove AM/PM + timezone + date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) + + for expression in date_formats(day_first): + try: + dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta) + return calendar.timegm(dt.timetuple()) + except ValueError: + pass + timetuple = email.utils.parsedate_tz(date_str) + if timetuple: + return calendar.timegm(timetuple) + pm_delta * 3600 + + def determine_ext(url, default_ext='unknown_video'): if url is None: return default_ext @@ -1410,6 +1454,8 @@ def shell_quote(args): def smuggle_url(url, data): """ Pass additional data in a URL for internal use. """ + url, idata = unsmuggle_url(url, {}) + data.update(idata) sdata = compat_urllib_parse_urlencode( {'__youtubedl_smuggle': json.dumps(data)}) return url + '#' + sdata @@ -1591,6 +1637,11 @@ class HEADRequest(compat_urllib_request.Request): return 'HEAD' +class PUTRequest(compat_urllib_request.Request): + def get_method(self): + return 'PUT' + + def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): if get_attr: if v is not None: @@ -1626,6 +1677,10 @@ def float_or_none(v, scale=1, invscale=1, default=None): return default +def strip_or_none(v): + return None if v is None else v.strip() + + def parse_duration(s): if not isinstance(s, compat_basestring): return None @@ -1882,7 +1937,13 @@ def update_Request(req, url=None, data=None, headers={}, query={}): req_headers.update(headers) req_data = data or req.data req_url = update_url_query(url or req.get_full_url(), query) - req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request + req_get_method = req.get_method() + if req_get_method == 'HEAD': + req_type = HEADRequest + elif req_get_method == 'PUT': + req_type = PUTRequest + else: + req_type = compat_urllib_request.Request new_req = req_type( req_url, data=req_data, headers=req_headers, origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) @@ -1924,11 +1985,27 @@ US_RATINGS = { } +TV_PARENTAL_GUIDELINES = { + 'TV-Y': 0, + 'TV-Y7': 7, + 'TV-G': 0, + 'TV-PG': 0, + 'TV-14': 14, + 'TV-MA': 17, +} + + def parse_age_limit(s): - if s is None: + if type(s) == int: + return s if 0 <= s <= 21 else None + if not isinstance(s, compat_basestring): return None m = re.match(r'^(?P\d{1,2})\+?$', s) - return int(m.group('age')) if m else US_RATINGS.get(s) + if m: + return int(m.group('age')) + if s in US_RATINGS: + return US_RATINGS[s] + return TV_PARENTAL_GUIDELINES.get(s) def strip_jsonp(code): @@ -1970,7 +2047,7 @@ def js_to_json(code): '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| /\*.*?\*/|,(?=\s*[\]}])| [a-zA-Z_][.a-zA-Z_0-9]*| - (?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:\s*:)?| + \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:\s*:)?| [0-9]+(?=\s*:) ''', fix_kv, code) @@ -2046,6 +2123,7 @@ def mimetype2ext(mt): return ext _, _, res = mt.rpartition('/') + res = res.lower() return { '3gpp': '3gp', @@ -2057,9 +2135,53 @@ def mimetype2ext(mt): 'x-flv': 'flv', 'x-mp4-fragmented': 'mp4', 'x-ms-wmv': 'wmv', + 'mpegurl': 'm3u8', + 'x-mpegurl': 'm3u8', + 'vnd.apple.mpegurl': 'm3u8', + 'dash+xml': 'mpd', + 'f4m': 'f4m', + 'f4m+xml': 'f4m', + 'hds+xml': 'f4m', + 'vnd.ms-sstr+xml': 'ism', }.get(res, res) +def parse_codecs(codecs_str): + # http://tools.ietf.org/html/rfc6381 + if not codecs_str: + return {} + splited_codecs = list(filter(None, map( + lambda str: str.strip(), codecs_str.strip().strip(',').split(',')))) + vcodec, acodec = None, None + for full_codec in splited_codecs: + codec = full_codec.split('.')[0] + if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v'): + if not vcodec: + vcodec = full_codec + elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac'): + if not acodec: + acodec = full_codec + else: + write_string('WARNING: Unknown codec %s' % full_codec, sys.stderr) + if not vcodec and not acodec: + if len(splited_codecs) == 2: + return { + 'vcodec': vcodec, + 'acodec': acodec, + } + elif len(splited_codecs) == 1: + return { + 'vcodec': 'none', + 'acodec': vcodec, + } + else: + return { + 'vcodec': vcodec or 'none', + 'acodec': acodec or 'none', + } + return {} + + def urlhandle_detect_ext(url_handle): getheader = url_handle.headers.get @@ -2852,3 +2974,123 @@ def decode_packed_codes(code): return re.sub( r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], obfucasted_code) + + +def parse_m3u8_attributes(attrib): + info = {} + for (key, val) in re.findall(r'(?P[A-Z0-9-]+)=(?P"[^"]+"|[^",]+)(?:,|$)', attrib): + if val.startswith('"'): + val = val[1:-1] + info[key] = val + return info + + +def urshift(val, n): + return val >> n if val >= 0 else (val + 0x100000000) >> n + + +# Based on png2str() written by @gdkchan and improved by @yokrysty +# Originally posted at https://github.com/rg3/youtube-dl/issues/9706 +def decode_png(png_data): + # Reference: https://www.w3.org/TR/PNG/ + header = png_data[8:] + + if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR': + raise IOError('Not a valid PNG file.') + + int_map = {1: '>B', 2: '>H', 4: '>I'} + unpack_integer = lambda x: compat_struct_unpack(int_map[len(x)], x)[0] + + chunks = [] + + while header: + length = unpack_integer(header[:4]) + header = header[4:] + + chunk_type = header[:4] + header = header[4:] + + chunk_data = header[:length] + header = header[length:] + + header = header[4:] # Skip CRC + + chunks.append({ + 'type': chunk_type, + 'length': length, + 'data': chunk_data + }) + + ihdr = chunks[0]['data'] + + width = unpack_integer(ihdr[:4]) + height = unpack_integer(ihdr[4:8]) + + idat = b'' + + for chunk in chunks: + if chunk['type'] == b'IDAT': + idat += chunk['data'] + + if not idat: + raise IOError('Unable to read PNG data.') + + decompressed_data = bytearray(zlib.decompress(idat)) + + stride = width * 3 + pixels = [] + + def _get_pixel(idx): + x = idx % stride + y = idx // stride + return pixels[y][x] + + for y in range(height): + basePos = y * (1 + stride) + filter_type = decompressed_data[basePos] + + current_row = [] + + pixels.append(current_row) + + for x in range(stride): + color = decompressed_data[1 + basePos + x] + basex = y * stride + x + left = 0 + up = 0 + + if x > 2: + left = _get_pixel(basex - 3) + if y > 0: + up = _get_pixel(basex - stride) + + if filter_type == 1: # Sub + color = (color + left) & 0xff + elif filter_type == 2: # Up + color = (color + up) & 0xff + elif filter_type == 3: # Average + color = (color + ((left + up) >> 1)) & 0xff + elif filter_type == 4: # Paeth + a = left + b = up + c = 0 + + if x > 2 and y > 0: + c = _get_pixel(basex - stride - 3) + + p = a + b - c + + pa = abs(p - a) + pb = abs(p - b) + pc = abs(p - c) + + if pa <= pb and pa <= pc: + color = (color + a) & 0xff + elif pb <= pc: + color = (color + b) & 0xff + else: + color = (color + c) & 0xff + + current_row.append(color) + + return width, height, pixels