X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/feb5020b37d7d3ba4005a8bac6f4efece4ce4b8c..80f96043cb349c4235e242e323556f561d147911:/youtube_dl/YoutubeDL.py diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py index d3281fe..e2332f9 100644 --- a/youtube_dl/YoutubeDL.py +++ b/youtube_dl/YoutubeDL.py @@ -3,6 +3,7 @@ from __future__ import absolute_import +import errno import io import os import re @@ -70,17 +71,27 @@ class YoutubeDL(object): logtostderr: Log messages to stderr instead of stdout. writedescription: Write the video description to a .description file writeinfojson: Write the video description to a .info.json file + writeannotations: Write the video annotations to a .annotations.xml file writethumbnail: Write the thumbnail image to a file writesubtitles: Write the video subtitles to a file writeautomaticsub: Write the automatic subtitles to a file allsubtitles: Downloads all the subtitles of the video + (requires writesubtitles or writeautomaticsub) listsubtitles: Lists all available subtitles for the video subtitlesformat: Subtitle format [srt/sbv/vtt] (default=srt) - subtitleslang: Language of the subtitles to download + subtitleslangs: List of languages of the subtitles to download keepvideo: Keep the video file after post-processing daterange: A DateRange object, download only if the upload_date is in the range. skip_download: Skip the actual download of the video file - + cachedir: Location of the cache files in the filesystem. + None to disable filesystem cache. + noplaylist: Download single video instead of a playlist if in doubt. + age_limit: An integer representing the user's age in years. + Unsuitable videos for the given age are skipped. + downloadarchive: File name of a file where all downloads are recorded. + Videos already present in the file are not downloaded + again. + The following parameters are not used by YoutubeDL itself, they are used by the FileDownloader: nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, @@ -97,11 +108,23 @@ class YoutubeDL(object): def __init__(self, params): """Create a FileDownloader object with the given options.""" self._ies = [] + self._ies_instances = {} self._pps = [] self._progress_hooks = [] self._download_retcode = 0 self._num_downloads = 0 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] + + if (sys.version_info >= (3,) and sys.platform != 'win32' and + sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] + and not params['restrictfilenames']): + # On Python 3, the Unicode filesystem API will throw errors (#1474) + self.report_warning( + u'Assuming --restrict-filenames since file system encoding ' + u'cannot encode all charactes. ' + u'Set the LC_ALL environment variable to fix this.') + params['restrictfilenames'] = True + self.params = params self.fd = FileDownloader(self, self.params) @@ -111,8 +134,21 @@ class YoutubeDL(object): def add_info_extractor(self, ie): """Add an InfoExtractor object to the end of the list.""" self._ies.append(ie) + self._ies_instances[ie.ie_key()] = ie ie.set_downloader(self) + def get_info_extractor(self, ie_key): + """ + Get an instance of an IE with name ie_key, it will try to get one from + the _ies list, if there's no instance it will create a new one and add + it to the extractor list. + """ + ie = self._ies_instances.get(ie_key) + if ie is None: + ie = get_info_extractor(ie_key)() + self.add_info_extractor(ie) + return ie + def add_default_info_extractors(self): """ Add the InfoExtractors returned by gen_extractors to the end of the list @@ -127,14 +163,10 @@ class YoutubeDL(object): def to_screen(self, message, skip_eol=False): """Print message to stdout if not in quiet mode.""" - assert type(message) == type(u'') if not self.params.get('quiet', False): terminator = [u'\n', u''][skip_eol] output = message + terminator - if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr - output = output.encode(preferredencoding(), 'ignore') - self._screen_file.write(output) - self._screen_file.flush() + write_string(output, self._screen_file) def to_stderr(self, message): """Print message to stderr.""" @@ -184,10 +216,10 @@ class YoutubeDL(object): If stderr is a tty file the 'WARNING:' will be colored ''' if sys.stderr.isatty() and os.name != 'nt': - _msg_header=u'\033[0;33mWARNING:\033[0m' + _msg_header = u'\033[0;33mWARNING:\033[0m' else: - _msg_header=u'WARNING:' - warning_message=u'%s %s' % (_msg_header,message) + _msg_header = u'WARNING:' + warning_message = u'%s %s' % (_msg_header, message) self.to_stderr(warning_message) def report_error(self, message, tb=None): @@ -202,19 +234,6 @@ class YoutubeDL(object): error_message = u'%s %s' % (_msg_header, message) self.trouble(error_message, tb) - def slow_down(self, start_time, byte_counter): - """Sleep if the download speed is over the rate limit.""" - rate_limit = self.params.get('ratelimit', None) - if rate_limit is None or byte_counter == 0: - return - now = time.time() - elapsed = now - start_time - if elapsed <= 0.0: - return - speed = float(byte_counter) / elapsed - if speed > rate_limit: - time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit) - def report_writedescription(self, descfn): """ Report that the description file is being written """ self.to_screen(u'[info] Writing video description to: ' + descfn) @@ -227,6 +246,10 @@ class YoutubeDL(object): """ Report that the metadata file has been written """ self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn) + def report_writeannotations(self, annofn): + """ Report that the annotations file has been written. """ + self.to_screen(u'[info] Writing video annotations to: ' + annofn) + def report_file_already_downloaded(self, file_name): """Report file has already been fully downloaded.""" try: @@ -252,19 +275,21 @@ class YoutubeDL(object): if template_dict['playlist_index'] is not None: template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index'] - sanitize = lambda k,v: sanitize_filename( + sanitize = lambda k, v: sanitize_filename( u'NA' if v is None else compat_str(v), restricted=self.params.get('restrictfilenames'), - is_id=(k==u'id')) - template_dict = dict((k, sanitize(k, v)) for k,v in template_dict.items()) + is_id=(k == u'id')) + template_dict = dict((k, sanitize(k, v)) + for k, v in template_dict.items()) - filename = self.params['outtmpl'] % template_dict + tmpl = os.path.expanduser(self.params['outtmpl']) + filename = tmpl % template_dict return filename except KeyError as err: self.report_error(u'Erroneous output template') return None except ValueError as err: - self.report_error(u'Insufficient system charset ' + repr(preferredencoding())) + self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')') return None def _match_entry(self, info_dict): @@ -284,19 +309,24 @@ class YoutubeDL(object): dateRange = self.params.get('daterange', DateRange()) if date not in dateRange: return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) + age_limit = self.params.get('age_limit') + if age_limit is not None: + if age_limit < info_dict.get('age_limit', 0): + return u'Skipping "' + title + '" because it is age restricted' + if self.in_download_archive(info_dict): + return (u'%(title)s has already been recorded in archive' + % info_dict) return None - + def extract_info(self, url, download=True, ie_key=None, extra_info={}): ''' Returns a list with a dictionary for each video we find. If 'download', also downloads the videos. extra_info is a dict containing the extra values to add to each result ''' - + if ie_key: - ie = get_info_extractor(ie_key)() - ie.set_downloader(self) - ies = [ie] + ies = [self.get_info_extractor(ie_key)] else: ies = self._ies @@ -336,7 +366,7 @@ class YoutubeDL(object): raise else: self.report_error(u'no suitable InfoExtractor: %s' % url) - + def process_ie_result(self, ie_result, download=True, extra_info={}): """ Take the result of the ie(may be modified) and resolve all unresolved @@ -348,13 +378,8 @@ class YoutubeDL(object): result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system if result_type == 'video': - if 'playlist' not in ie_result: - # It isn't part of a playlist - ie_result['playlist'] = None - ie_result['playlist_index'] = None - if download: - self.process_info(ie_result) - return ie_result + ie_result.update(extra_info) + return self.process_video_result(ie_result) elif result_type == 'url': # We have to add extra_info to the results because it may be # contained in a playlist @@ -365,7 +390,7 @@ class YoutubeDL(object): elif result_type == 'playlist': # We process each entry in the playlist playlist = ie_result.get('title', None) or ie_result.get('id', None) - self.to_screen(u'[download] Downloading playlist: %s' % playlist) + self.to_screen(u'[download] Downloading playlist: %s' % playlist) playlist_results = [] @@ -383,12 +408,12 @@ class YoutubeDL(object): self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" % (ie_result['extractor'], playlist, n_all_entries, n_entries)) - for i,entry in enumerate(entries,1): - self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_entries)) + for i, entry in enumerate(entries, 1): + self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries)) extra = { - 'playlist': playlist, - 'playlist_index': i + playliststart, - } + 'playlist': playlist, + 'playlist_index': i + playliststart, + } if not 'extractor' in entry: # We set the extractor, if it's an url it will be set then to # the new extractor, but if it's already a video we must make @@ -412,6 +437,103 @@ class YoutubeDL(object): else: raise Exception('Invalid result type: %s' % result_type) + def select_format(self, format_spec, available_formats): + if format_spec == 'best' or format_spec is None: + return available_formats[-1] + elif format_spec == 'worst': + return available_formats[0] + else: + extensions = [u'mp4', u'flv', u'webm', u'3gp'] + if format_spec in extensions: + filter_f = lambda f: f['ext'] == format_spec + else: + filter_f = lambda f: f['format_id'] == format_spec + matches = list(filter(filter_f, available_formats)) + if matches: + return matches[-1] + return None + + def process_video_result(self, info_dict, download=True): + assert info_dict.get('_type', 'video') == 'video' + + if 'playlist' not in info_dict: + # It isn't part of a playlist + info_dict['playlist'] = None + info_dict['playlist_index'] = None + + # This extractors handle format selection themselves + if info_dict['extractor'] in [u'youtube', u'Youku', u'YouPorn', u'mixcloud']: + if download: + self.process_info(info_dict) + return info_dict + + # We now pick which formats have to be downloaded + if info_dict.get('formats') is None: + # There's only one format available + formats = [info_dict] + else: + formats = info_dict['formats'] + + # We check that all the formats have the format and format_id fields + for (i, format) in enumerate(formats): + if format.get('format_id') is None: + format['format_id'] = compat_str(i) + if format.get('format') is None: + format['format'] = u'{id} - {res}{note}'.format( + id=format['format_id'], + res=self.format_resolution(format), + note=u' ({})'.format(format['format_note']) if format.get('format_note') is not None else '', + ) + + if self.params.get('listformats', None): + self.list_formats(info_dict) + return + + format_limit = self.params.get('format_limit', None) + if format_limit: + formats = list(takewhile_inclusive( + lambda f: f['format_id'] != format_limit, formats + )) + if self.params.get('prefer_free_formats'): + def _free_formats_key(f): + try: + ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext']) + except ValueError: + ext_ord = -1 + # We only compare the extension if they have the same height and width + return (f.get('height'), f.get('width'), ext_ord) + formats = sorted(formats, key=_free_formats_key) + + req_format = self.params.get('format', 'best') + if req_format is None: + req_format = 'best' + formats_to_download = [] + # The -1 is for supporting YoutubeIE + if req_format in ('-1', 'all'): + formats_to_download = formats + else: + # We can accept formats requestd in the format: 34/5/best, we pick + # the first that is available, starting from left + req_formats = req_format.split('/') + for rf in req_formats: + selected_format = self.select_format(rf, formats) + if selected_format is not None: + formats_to_download = [selected_format] + break + if not formats_to_download: + raise ExtractorError(u'requested format not available') + + if download: + if len(formats_to_download) > 1: + self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) + for format in formats_to_download: + new_info = dict(info_dict) + new_info.update(format) + self.process_info(new_info) + # We update the info dict with the best quality format (backwards compatibility) + info_dict.update(formats_to_download[-1]) + return info_dict + def process_info(self, info_dict): """Process a single resolved IE result.""" @@ -447,7 +569,8 @@ class YoutubeDL(object): if self.params.get('forceid', False): compat_print(info_dict['id']) if self.params.get('forceurl', False): - compat_print(info_dict['url']) + # For RTMP URLs, also include the playpath + compat_print(info_dict['url'] + info_dict.get('play_path', u'')) if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict: compat_print(info_dict['thumbnail']) if self.params.get('forcedescription', False) and 'description' in info_dict: @@ -478,68 +601,70 @@ class YoutubeDL(object): self.report_writedescription(descfn) with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: descfile.write(info_dict['description']) + except (KeyError, TypeError): + self.report_warning(u'There\'s no description to write.') except (OSError, IOError): self.report_error(u'Cannot write description file ' + descfn) return - if (self.params.get('writesubtitles', False) or self.params.get('writeautomaticsub')) and 'subtitles' in info_dict and info_dict['subtitles']: + if self.params.get('writeannotations', False): + try: + annofn = filename + u'.annotations.xml' + self.report_writeannotations(annofn) + with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: + annofile.write(info_dict['annotations']) + except (KeyError, TypeError): + self.report_warning(u'There are no annotations to write.') + except (OSError, IOError): + self.report_error(u'Cannot write annotations file: ' + annofn) + return + + subtitles_are_requested = any([self.params.get('writesubtitles', False), + self.params.get('writeautomaticsub')]) + + if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']: # subtitles download errors are already managed as troubles in relevant IE # that way it will silently go on when used with unsupporting IE - subtitle = info_dict['subtitles'][0] - (sub_error, sub_lang, sub) = subtitle + subtitles = info_dict['subtitles'] sub_format = self.params.get('subtitlesformat') - if sub_error: - self.report_warning("Some error while getting the subtitles") - else: + for sub_lang in subtitles.keys(): + sub = subtitles[sub_lang] + if sub is None: + continue try: - sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format + sub_filename = subtitles_filename(filename, sub_lang, sub_format) self.report_writesubtitles(sub_filename) with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: - subfile.write(sub) + subfile.write(sub) except (OSError, IOError): self.report_error(u'Cannot write subtitles file ' + descfn) return - if self.params.get('allsubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']: - subtitles = info_dict['subtitles'] - sub_format = self.params.get('subtitlesformat') - for subtitle in subtitles: - (sub_error, sub_lang, sub) = subtitle - if sub_error: - self.report_warning("Some error while getting the subtitles") - else: - try: - sub_filename = filename.rsplit('.', 1)[0] + u'.' + sub_lang + u'.' + sub_format - self.report_writesubtitles(sub_filename) - with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: - subfile.write(sub) - except (OSError, IOError): - self.report_error(u'Cannot write subtitles file ' + descfn) - return - if self.params.get('writeinfojson', False): infofn = filename + u'.info.json' self.report_writeinfojson(infofn) try: - json_info_dict = dict((k, v) for k,v in info_dict.items() if not k in ['urlhandle']) + json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle']) write_json_file(json_info_dict, encodeFilename(infofn)) except (OSError, IOError): self.report_error(u'Cannot write metadata to JSON file ' + infofn) return if self.params.get('writethumbnail', False): - if 'thumbnail' in info_dict: - thumb_format = info_dict['thumbnail'].rpartition(u'/')[2].rpartition(u'.')[2] - if not thumb_format: - thumb_format = 'jpg' + if info_dict.get('thumbnail') is not None: + thumb_format = determine_ext(info_dict['thumbnail'], u'jpg') thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format self.to_screen(u'[%s] %s: Downloading thumbnail ...' % (info_dict['extractor'], info_dict['id'])) - uf = compat_urllib_request.urlopen(info_dict['thumbnail']) - with open(thumb_filename, 'wb') as thumbf: - shutil.copyfileobj(uf, thumbf) - self.to_screen(u'[%s] %s: Writing thumbnail to: %s' % - (info_dict['extractor'], info_dict['id'], thumb_filename)) + try: + uf = compat_urllib_request.urlopen(info_dict['thumbnail']) + with open(thumb_filename, 'wb') as thumbf: + shutil.copyfileobj(uf, thumbf) + self.to_screen(u'[%s] %s: Writing thumbnail to: %s' % + (info_dict['extractor'], info_dict['id'], thumb_filename)) + except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: + self.report_warning(u'Unable to download thumbnail "%s": %s' % + (info_dict['thumbnail'], compat_str(err))) if not self.params.get('skip_download', False): if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)): @@ -547,11 +672,11 @@ class YoutubeDL(object): else: try: success = self.fd._do_download(filename, info_dict) - except (OSError, IOError) as err: - raise UnavailableVideoError() except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: self.report_error(u'unable to download video data: %s' % str(err)) return + except (OSError, IOError) as err: + raise UnavailableVideoError(err) except (ContentTooShortError, ) as err: self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) return @@ -563,6 +688,8 @@ class YoutubeDL(object): self.report_error(u'postprocessing: %s' % str(err)) return + self.record_download_archive(info_dict) + def download(self, url_list): """Download a given list of URLs.""" if len(url_list) > 1 and self.fixed_template(): @@ -587,7 +714,7 @@ class YoutubeDL(object): keep_video = None for pp in self._pps: try: - keep_video_wish,new_info = pp.run(info) + keep_video_wish, new_info = pp.run(info) if keep_video_wish is not None: if keep_video_wish: keep_video = keep_video_wish @@ -595,10 +722,62 @@ class YoutubeDL(object): # No clear decision yet, let IE decide keep_video = keep_video_wish except PostProcessingError as e: - self.to_stderr(u'ERROR: ' + e.msg) + self.report_error(e.msg) if keep_video is False and not self.params.get('keepvideo', False): try: self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename) os.remove(encodeFilename(filename)) except (IOError, OSError): self.report_warning(u'Unable to remove downloaded video file') + + def in_download_archive(self, info_dict): + fn = self.params.get('download_archive') + if fn is None: + return False + vid_id = info_dict['extractor'] + u' ' + info_dict['id'] + try: + with locked_file(fn, 'r', encoding='utf-8') as archive_file: + for line in archive_file: + if line.strip() == vid_id: + return True + except IOError as ioe: + if ioe.errno != errno.ENOENT: + raise + return False + + def record_download_archive(self, info_dict): + fn = self.params.get('download_archive') + if fn is None: + return + vid_id = info_dict['extractor'] + u' ' + info_dict['id'] + with locked_file(fn, 'a', encoding='utf-8') as archive_file: + archive_file.write(vid_id + u'\n') + + @staticmethod + def format_resolution(format): + if format.get('height') is not None: + if format.get('width') is not None: + res = u'%sx%s' % (format['width'], format['height']) + else: + res = u'%sp' % format['height'] + else: + res = '???' + return res + + def list_formats(self, info_dict): + formats_s = [] + for format in info_dict.get('formats', [info_dict]): + formats_s.append(u'%-15s: %-5s %-15s[%s]' % ( + format['format_id'], + format['ext'], + format.get('format_note') or '-', + self.format_resolution(format), + ) + ) + if len(formats_s) != 1: + formats_s[0] += ' (worst)' + formats_s[-1] += ' (best)' + formats_s = "\n".join(formats_s) + self.to_screen(u'[info] Available formats for %s:\n' + u'format code extension note resolution\n%s' % ( + info_dict['id'], formats_s))