X-Git-Url: http://git.cielonegro.org/gitweb.cgi?a=blobdiff_plain;ds=inline;f=youtube-dl;h=fe8a591606a38c243cb1214372d90b505f41e207;hb=bd3cdf6dc42d48d010fad03d8d601b3920c7a8a7;hp=b2dc7952d9c3a0a1c7ed41bde7be98d3e887f3a2;hpb=b74c859d0ffe2250c7730f18f24e9515084e838a;p=youtube-dl.git diff --git a/youtube-dl b/youtube-dl index b2dc7952d..fe8a59160 100755 --- a/youtube-dl +++ b/youtube-dl @@ -2,6 +2,7 @@ # -*- coding: utf-8 -*- # Author: Ricardo Garcia Gonzalez # Author: Danny Colligan +# Author: Benjamin Johnson # License: Public domain code import htmlentitydefs import httplib @@ -13,13 +14,20 @@ import os.path import re import socket import string +import subprocess import sys import time import urllib import urllib2 +# parse_qs was moved from the cgi module to the urlparse module recently. +try: + from urlparse import parse_qs +except ImportError: + from cgi import parse_qs + std_headers = { - 'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.0.8) Gecko/2009032609 Firefox/3.0.8', + 'User-Agent': 'Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6', 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', 'Accept': 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5', 'Accept-Language': 'en-us,en;q=0.5', @@ -27,6 +35,75 @@ std_headers = { simple_title_chars = string.ascii_letters.decode('ascii') + string.digits.decode('ascii') +def preferredencoding(): + """Get preferred encoding. + + Returns the best encoding scheme for the system, based on + locale.getpreferredencoding() and some further tweaks. + """ + def yield_preferredencoding(): + try: + pref = locale.getpreferredencoding() + u'TEST'.encode(pref) + except: + pref = 'UTF-8' + while True: + yield pref + return yield_preferredencoding().next() + +def htmlentity_transform(matchobj): + """Transforms an HTML entity to a Unicode character. + + This function receives a match object and is intended to be used with + the re.sub() function. + """ + entity = matchobj.group(1) + + # Known non-numeric HTML entity + if entity in htmlentitydefs.name2codepoint: + return unichr(htmlentitydefs.name2codepoint[entity]) + + # Unicode character + mobj = re.match(ur'(?u)#(x?\d+)', entity) + if mobj is not None: + numstr = mobj.group(1) + if numstr.startswith(u'x'): + base = 16 + numstr = u'0%s' % numstr + else: + base = 10 + return unichr(long(numstr, base)) + + # Unknown entity in name, return its literal representation + return (u'&%s;' % entity) + +def sanitize_title(utitle): + """Sanitizes a video title so it could be used as part of a filename.""" + utitle = re.sub(ur'(?u)&(.+?);', htmlentity_transform, utitle) + return utitle.replace(unicode(os.sep), u'%') + +def sanitize_open(filename, open_mode): + """Try to open the given filename, and slightly tweak it if this fails. + + Attempts to open the given filename. If this fails, it tries to change + the filename slightly, step by step, until it's either able to open it + or it fails and raises a final exception, like the standard open() + function. + + It returns the tuple (stream, definitive_file_name). + """ + try: + stream = open(filename, open_mode) + return (stream, filename) + except (IOError, OSError), err: + # In case of error, try to remove win32 forbidden chars + filename = re.sub(ur'[<>:"\|\?\*]', u'#', filename) + + # An exception here should be caught in the caller + stream = open(filename, open_mode) + return (stream, filename) + + class DownloadError(Exception): """Download Error exception. @@ -58,6 +135,22 @@ class UnavailableFormatError(Exception): This exception will be thrown when a video is requested in a format that is not available for that video. """ + pass + +class ContentTooShortError(Exception): + """Content Too Short exception. + + This exception may be raised by FileDownloader objects when a file they + download is too small for what the server announced first, indicating + the connection was probably interrupted. + """ + # Both in bytes + downloaded = None + expected = None + + def __init__(self, downloaded, expected): + self.downloaded = downloaded + self.expected = expected class FileDownloader(object): """File Downloader class. @@ -98,6 +191,7 @@ class FileDownloader(object): ignoreerrors: Do not stop on download errors. ratelimit: Download speed limit, in bytes/sec. nooverwrites: Prevent overwriting files. + continuedl: Try to continue downloads if possible. """ params = None @@ -126,10 +220,12 @@ class FileDownloader(object): def format_bytes(bytes): if bytes is None: return 'N/A' - if bytes == 0: + if type(bytes) is str: + bytes = float(bytes) + if bytes == 0.0: exponent = 0 else: - exponent = long(math.log(float(bytes), 1024.0)) + exponent = long(math.log(bytes, 1024.0)) suffix = 'bkMGTPEZY'[exponent] converted = float(bytes) / float(1024**exponent) return '%.2f%s' % (converted, suffix) @@ -166,13 +262,13 @@ class FileDownloader(object): new_min = max(bytes / 2.0, 1.0) new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB if elapsed_time < 0.001: - return int(new_max) + return long(new_max) rate = bytes / elapsed_time if rate > new_max: - return int(new_max) + return long(new_max) if rate < new_min: - return int(new_min) - return int(rate) + return long(new_min) + return long(rate) @staticmethod def parse_bytes(bytestr): @@ -184,6 +280,16 @@ class FileDownloader(object): multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower()) return long(round(number * multiplier)) + @staticmethod + def verify_url(url): + """Verify a URL is valid and data could be downloaded. Return real data URL.""" + request = urllib2.Request(url, None, std_headers) + data = urllib2.urlopen(request) + data.read(1) + url = data.geturl() + data.close() + return url + def add_info_extractor(self, ie): """Add an InfoExtractor object to the end of the list.""" self._ies.append(ie) @@ -197,12 +303,12 @@ class FileDownloader(object): def to_stdout(self, message, skip_eol=False): """Print message to stdout if not in quiet mode.""" if not self.params.get('quiet', False): - print (u'%s%s' % (message, [u'\n', u''][skip_eol])).encode(locale.getpreferredencoding()), + print (u'%s%s' % (message, [u'\n', u''][skip_eol])).encode(preferredencoding()), sys.stdout.flush() def to_stderr(self, message): """Print message to stderr.""" - print >>sys.stderr, message + print >>sys.stderr, message.encode(preferredencoding()) def fixed_template(self): """Checks if the output template is fixed.""" @@ -242,6 +348,18 @@ class FileDownloader(object): """Report download progress.""" self.to_stdout(u'\r[download] %s of %s at %s ETA %s' % (percent_str, data_len_str, speed_str, eta_str), skip_eol=True) + + def report_resuming_byte(self, resume_len): + """Report attemtp to resume at given byte.""" + self.to_stdout(u'[download] Resuming download at byte %s' % resume_len) + + def report_file_already_downloaded(self, file_name): + """Report file has already been fully downloaded.""" + self.to_stdout(u'[download] %s has already been downloaded' % file_name) + + def report_unable_to_resume(self): + """Report it was impossible to resume download.""" + self.to_stdout(u'[download] Unable to resume') def report_finish(self): """Report download finished.""" @@ -249,25 +367,31 @@ class FileDownloader(object): def process_info(self, info_dict): """Process a single dictionary returned by an InfoExtractor.""" - # Forced printings - if self.params.get('forcetitle', False): - print info_dict['title'].encode(locale.getpreferredencoding()) - if self.params.get('forceurl', False): - print info_dict['url'].encode(locale.getpreferredencoding()) - # Do nothing else if in simulate mode if self.params.get('simulate', False): - return + # Verify URL if it's an HTTP one + if info_dict['url'].startswith('http'): + try: + info_dict['url'] = self.verify_url(info_dict['url'].encode('utf-8')).decode('utf-8') + except (OSError, IOError, urllib2.URLError, httplib.HTTPException, socket.error), err: + raise UnavailableFormatError + + # Forced printings + if self.params.get('forcetitle', False): + print info_dict['title'].encode(preferredencoding(), 'xmlcharrefreplace') + if self.params.get('forceurl', False): + print info_dict['url'].encode(preferredencoding(), 'xmlcharrefreplace') + return + try: template_dict = dict(info_dict) template_dict['epoch'] = unicode(long(time.time())) filename = self.params['outtmpl'] % template_dict - self.report_destination(filename) except (ValueError, KeyError), err: self.trouble('ERROR: invalid output template or system charset: %s' % str(err)) - if self.params['nooverwrites'] and os.path.exists(filename): - self.to_stderr('WARNING: file exists: %s; skipping' % filename) + if self.params.get('nooverwrites', False) and os.path.exists(filename): + self.to_stderr(u'WARNING: file exists: %s; skipping' % filename) return try: @@ -277,27 +401,23 @@ class FileDownloader(object): return try: - outstream = open(filename, 'wb') + success = self._do_download(filename, info_dict['url'].encode('utf-8')) except (OSError, IOError), err: - self.trouble('ERROR: unable to open for writing: %s' % str(err)) - return - - try: - self._do_download(outstream, info_dict['url']) - outstream.close() - except (OSError, IOError), err: - os.remove(filename) raise UnavailableFormatError except (urllib2.URLError, httplib.HTTPException, socket.error), err: self.trouble('ERROR: unable to download video data: %s' % str(err)) return - - try: - self.post_process(filename, info_dict) - except (PostProcessingError), err: - self.trouble('ERROR: postprocessing: %s' % str(err)) + except (ContentTooShortError, ), err: + self.trouble('ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) return + if success: + try: + self.post_process(filename, info_dict) + except (PostProcessingError), err: + self.trouble('ERROR: postprocessing: %s' % str(err)) + return + def download(self, url_list): """Download a given list of URLs.""" if len(url_list) > 1 and self.fixed_template(): @@ -333,21 +453,79 @@ class FileDownloader(object): if info is None: break - def _do_download(self, stream, url): + def _download_with_rtmpdump(self, filename, url): + self.report_destination(filename) + + # Check for rtmpdump first + try: + subprocess.call(['rtmpdump', '-h'], stdout=(file(os.path.devnull, 'w')), stderr=subprocess.STDOUT) + except (OSError, IOError): + self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run') + return False + + # Download using rtmpdump. rtmpdump returns exit code 2 when + # the connection was interrumpted and resuming appears to be + # possible. This is part of rtmpdump's normal usage, AFAIK. + basic_args = ['rtmpdump', '-q', '-r', url, '-o', filename] + retval = subprocess.call(basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)]) + while retval == 2 or retval == 1: + self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename), skip_eol=True) + time.sleep(2.0) # This seems to be needed + retval = subprocess.call(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1]) + if retval == 0: + self.to_stdout(u'\r[rtmpdump] %s bytes' % os.path.getsize(filename)) + return True + else: + self.trouble('ERROR: rtmpdump exited with code %d' % retval) + return False + + def _do_download(self, filename, url): + # Attempt to download using rtmpdump + if url.startswith('rtmp'): + return self._download_with_rtmpdump(filename, url) + + stream = None + open_mode = 'wb' + basic_request = urllib2.Request(url, None, std_headers) request = urllib2.Request(url, None, std_headers) - data = urllib2.urlopen(request) + + # Establish possible resume length + if os.path.isfile(filename): + resume_len = os.path.getsize(filename) + else: + resume_len = 0 + + # Request parameters in case of being able to resume + if self.params.get('continuedl', False) and resume_len != 0: + self.report_resuming_byte(resume_len) + request.add_header('Range','bytes=%d-' % resume_len) + open_mode = 'ab' + + # Establish connection + try: + data = urllib2.urlopen(request) + except (urllib2.HTTPError, ), err: + if err.code != 416: # 416 is 'Requested range not satisfiable' + raise + # Unable to resume + data = urllib2.urlopen(basic_request) + content_length = data.info()['Content-Length'] + + if content_length is not None and long(content_length) == resume_len: + # Because the file had already been fully downloaded + self.report_file_already_downloaded(filename) + return True + else: + # Because the server didn't let us + self.report_unable_to_resume() + open_mode = 'wb' + data_len = data.info().get('Content-length', None) data_len_str = self.format_bytes(data_len) byte_counter = 0 block_size = 1024 start = time.time() while True: - # Progress message - percent_str = self.calc_percent(byte_counter, data_len) - eta_str = self.calc_eta(start, time.time(), data_len, byte_counter) - speed_str = self.calc_speed(start, time.time(), byte_counter) - self.report_progress(percent_str, data_len_str, speed_str, eta_str) - # Download and write before = time.time() data_block = data.read(block_size) @@ -356,15 +534,31 @@ class FileDownloader(object): if data_block_len == 0: break byte_counter += data_block_len + + # Open file just in time + if stream is None: + try: + (stream, filename) = sanitize_open(filename, open_mode) + self.report_destination(filename) + except (OSError, IOError), err: + self.trouble('ERROR: unable to open for writing: %s' % str(err)) + return False stream.write(data_block) block_size = self.best_block_size(after - before, data_block_len) + # Progress message + percent_str = self.calc_percent(byte_counter, data_len) + eta_str = self.calc_eta(start, time.time(), data_len, byte_counter) + speed_str = self.calc_speed(start, time.time(), byte_counter) + self.report_progress(percent_str, data_len_str, speed_str, eta_str) + # Apply rate limit self.slow_down(start, byte_counter) self.report_finish() if data_len is not None and str(byte_counter) != data_len: - raise ValueError('Content too short: %s/%s bytes' % (byte_counter, data_len)) + raise ContentTooShortError(byte_counter, long(data_len)) + return True class InfoExtractor(object): """Information Extractor class. @@ -430,46 +624,24 @@ class InfoExtractor(object): class YoutubeIE(InfoExtractor): """Information extractor for youtube.com.""" - _VALID_URL = r'^((?:http://)?(?:\w+\.)?youtube\.com/(?:(?:v/)|(?:(?:watch(?:\.php)?)?\?(?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$' + _VALID_URL = r'^((?:http://)?(?:\w+\.)?youtube\.com/(?:(?:v/)|(?:(?:watch(?:\.php)?)?[\?#](?:.+&)?v=)))?([0-9A-Za-z_-]+)(?(1).+)?$' _LANG_URL = r'http://uk.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1' _LOGIN_URL = 'http://www.youtube.com/signup?next=/&gl=US&hl=en' _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en' _NETRC_MACHINE = 'youtube' - _available_formats = ['22', '35', '18', '17', '13'] # listed in order of priority for -b flag + _available_formats = ['37', '22', '35', '18', '5', '17', '13', None] # listed in order of priority for -b flag _video_extensions = { '13': '3gp', '17': 'mp4', '18': 'mp4', '22': 'mp4', + '37': 'mp4', } @staticmethod def suitable(url): return (re.match(YoutubeIE._VALID_URL, url) is not None) - @staticmethod - def htmlentity_transform(matchobj): - """Transforms an HTML entity to a Unicode character.""" - entity = matchobj.group(1) - - # Known non-numeric HTML entity - if entity in htmlentitydefs.name2codepoint: - return unichr(htmlentitydefs.name2codepoint[entity]) - - # Unicode character - mobj = re.match(ur'(?u)#(x?\d+)', entity) - if mobj is not None: - numstr = mobj.group(1) - if numstr.startswith(u'x'): - base = 16 - numstr = u'0%s' % numstr - else: - base = 10 - return unichr(long(numstr, base)) - - # Unknown entity in name, return its literal representation - return (u'&%s;' % entity) - def report_lang(self): """Report attempt to set language.""" self._downloader.to_stdout(u'[youtube] Setting language') @@ -482,22 +654,22 @@ class YoutubeIE(InfoExtractor): """Report attempt to confirm age.""" self._downloader.to_stdout(u'[youtube] Confirming age') - def report_webpage_download(self, video_id): - """Report attempt to download webpage.""" - self._downloader.to_stdout(u'[youtube] %s: Downloading video webpage' % video_id) + def report_video_info_webpage_download(self, video_id): + """Report attempt to download video info webpage.""" + self._downloader.to_stdout(u'[youtube] %s: Downloading video info webpage' % video_id) def report_information_extraction(self, video_id): """Report attempt to extract video information.""" self._downloader.to_stdout(u'[youtube] %s: Extracting video information' % video_id) - def report_video_url(self, video_id, video_real_url): - """Report extracted video URL.""" - self._downloader.to_stdout(u'[youtube] %s: URL: %s' % (video_id, video_real_url)) - def report_unavailable_format(self, video_id, format): """Report extracted video URL.""" self._downloader.to_stdout(u'[youtube] %s: Format %s not available' % (video_id, format)) + def report_rtmp_download(self): + """Indicate the download will use the RTMP protocol.""" + self._downloader.to_stdout(u'[youtube] RTMP download detected') + def _real_initialize(self): if self._downloader is None: return @@ -587,53 +759,62 @@ class YoutubeIE(InfoExtractor): best_quality = True while True: + # Extension + video_extension = self._video_extensions.get(format_param, 'flv') + + # Get video info + video_info_url = 'http://www.youtube.com/get_video_info?&video_id=%s&el=detailpage&ps=default&eurl=&gl=US&hl=en' % video_id + request = urllib2.Request(video_info_url, None, std_headers) try: - # Extension - video_extension = self._video_extensions.get(format_param, 'flv') - - # Normalize URL, including format - normalized_url = 'http://www.youtube.com/watch?v=%s&gl=US&hl=en' % video_id - if format_param is not None: - normalized_url = '%s&fmt=%s' % (normalized_url, format_param) - request = urllib2.Request(normalized_url, None, std_headers) - try: - self.report_webpage_download(video_id) - video_webpage = urllib2.urlopen(request).read() - except (urllib2.URLError, httplib.HTTPException, socket.error), err: - self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % str(err)) - return - self.report_information_extraction(video_id) - - # "t" param - mobj = re.search(r', "t": "([^"]+)"', video_webpage) - if mobj is None: - self._downloader.trouble(u'ERROR: unable to extract "t" parameter') - return - video_real_url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&el=detailpage&ps=' % (video_id, mobj.group(1)) - if format_param is not None: - video_real_url = '%s&fmt=%s' % (video_real_url, format_param) - self.report_video_url(video_id, video_real_url) - - # uploader - mobj = re.search(r"var watchUsername = '([^']+)';", video_webpage) - if mobj is None: - self._downloader.trouble(u'ERROR: unable to extract uploader nickname') - return - video_uploader = mobj.group(1) + self.report_video_info_webpage_download(video_id) + video_info_webpage = urllib2.urlopen(request).read() + video_info = parse_qs(video_info_webpage) + except (urllib2.URLError, httplib.HTTPException, socket.error), err: + self._downloader.trouble(u'ERROR: unable to download video info webpage: %s' % str(err)) + return + self.report_information_extraction(video_id) + + # "t" param + if 'token' not in video_info: + # Attempt to see if YouTube has issued an error message + if 'reason' not in video_info: + self._downloader.trouble(u'ERROR: unable to extract "t" parameter for unknown reason') + stream = open('reportme-ydl-%s.dat' % time.time(), 'wb') + stream.write(video_info_webpage) + stream.close() + else: + reason = urllib.unquote_plus(video_info['reason'][0]) + self._downloader.trouble(u'ERROR: YouTube said: %s' % reason.decode('utf-8')) + return + token = urllib.unquote_plus(video_info['token'][0]) + video_real_url = 'http://www.youtube.com/get_video?video_id=%s&t=%s&eurl=&el=detailpage&ps=default&gl=US&hl=en' % (video_id, token) + if format_param is not None: + video_real_url = '%s&fmt=%s' % (video_real_url, format_param) + + # Check possible RTMP download + if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): + self.report_rtmp_download() + video_real_url = video_info['conn'][0] + + # uploader + if 'author' not in video_info: + self._downloader.trouble(u'ERROR: unable to extract uploader nickname') + return + video_uploader = urllib.unquote_plus(video_info['author'][0]) - # title - mobj = re.search(r'(?im)