X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2FFileDownloader.py;h=80c2546ea9b571005088716a63dcf3291ca384be;hb=767e00277f379410288476ecd05622968eec1d1b;hp=d2b9be9ef88cb93e3e54fb0d3650880e35c36b43;hpb=7eab8dc7504cf1f5f1dd03eb62e266ce24948b93;p=youtube-dl diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py index d2b9be9ef..80c2546ea 100644 --- a/youtube_dl/FileDownloader.py +++ b/youtube_dl/FileDownloader.py @@ -17,6 +17,7 @@ if os.name == 'nt': import ctypes from .utils import * +from .InfoExtractors import get_info_extractor class FileDownloader(object): @@ -88,6 +89,7 @@ class FileDownloader(object): keepvideo: Keep the video file after post-processing min_filesize: Skip files smaller than this size max_filesize: Skip files larger than this size + daterange: A DateRange object, download only if the upload_date is in the range. """ params = None @@ -120,7 +122,7 @@ class FileDownloader(object): exponent = 0 else: exponent = int(math.log(bytes, 1024.0)) - suffix = 'bkMGTPEZY'[exponent] + suffix = ['B','KiB','MiB','GiB','TiB','PiB','EiB','ZiB','YiB'][exponent] converted = float(bytes) / float(1024 ** exponent) return '%.2f%s' % (converted, suffix) @@ -231,11 +233,21 @@ class FileDownloader(object): self.to_stderr(message) if self.params.get('verbose'): if tb is None: - tb_data = traceback.format_list(traceback.extract_stack()) - tb = u''.join(tb_data) + if sys.exc_info()[0]: # if .trouble has been called from an except block + tb = u'' + if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: + tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) + tb += compat_str(traceback.format_exc()) + else: + tb_data = traceback.format_list(traceback.extract_stack()) + tb = u''.join(tb_data) self.to_stderr(tb) if not self.params.get('ignoreerrors', False): - raise DownloadError(message) + if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: + exc_info = sys.exc_info()[1].exc_info + else: + exc_info = sys.exc_info() + raise DownloadError(message, exc_info) self._download_retcode = 1 def report_warning(self, message): @@ -243,13 +255,25 @@ class FileDownloader(object): Print the message to stderr, it will be prefixed with 'WARNING:' If stderr is a tty file the 'WARNING:' will be colored ''' - if sys.stderr.isatty(): + if sys.stderr.isatty() and os.name != 'nt': _msg_header=u'\033[0;33mWARNING:\033[0m' else: _msg_header=u'WARNING:' warning_message=u'%s %s' % (_msg_header,message) self.to_stderr(warning_message) + def report_error(self, message, tb=None): + ''' + Do the same as trouble, but prefixes the message with 'ERROR:', colored + in red if stderr is a tty file. + ''' + if sys.stderr.isatty() and os.name != 'nt': + _msg_header = u'\033[0;31mERROR:\033[0m' + else: + _msg_header = u'ERROR:' + 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) @@ -281,7 +305,7 @@ class FileDownloader(object): return os.rename(encodeFilename(old_filename), encodeFilename(new_filename)) except (IOError, OSError) as err: - self.trouble(u'ERROR: unable to rename file') + self.report_error(u'unable to rename file') def try_utime(self, filename, last_modified_hdr): """Try to set the last-modified time of the given file.""" @@ -366,7 +390,13 @@ class FileDownloader(object): template_dict = dict(info_dict) template_dict['epoch'] = int(time.time()) - template_dict['autonumber'] = u'%05d' % self._num_downloads + autonumber_size = self.params.get('autonumber_size') + if autonumber_size is None: + autonumber_size = 5 + autonumber_templ = u'%0' + str(autonumber_size) + u'd' + template_dict['autonumber'] = autonumber_templ % self._num_downloads + if template_dict['playlist_index'] is not None: + template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index'] sanitize = lambda k,v: sanitize_filename( u'NA' if v is None else compat_str(v), @@ -377,10 +407,10 @@ class FileDownloader(object): filename = self.params['outtmpl'] % template_dict return filename except KeyError as err: - self.trouble(u'ERROR: Erroneous output template') + self.report_error(u'Erroneous output template') return None except ValueError as err: - self.trouble(u'ERROR: Insufficient system charset ' + repr(preferredencoding())) + self.report_error(u'Insufficient system charset ' + repr(preferredencoding())) return None def _match_entry(self, info_dict): @@ -395,21 +425,37 @@ class FileDownloader(object): if rejecttitle: if re.search(rejecttitle, title, re.IGNORECASE): return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' + date = info_dict.get('upload_date', None) + if date is not None: + 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) return None - def extract_info(self, url): + def extract_info(self, url, download = True, ie_name = None): ''' Returns a list with a dictionary for each video we find. + If 'download', also downloads the videos. ''' suitable_found = False - for ie in self._ies: + + #We copy the original list + ies = list(self._ies) + + if ie_name is not None: + #We put in the first place the given info extractor + first_ie = get_info_extractor(ie_name)() + first_ie.set_downloader(self) + ies.insert(0, first_ie) + + for ie in ies: # Go to next InfoExtractor if not suitable if not ie.suitable(url): continue # Warn if the _WORKING attribute is False if not ie.working(): - self.to_stderr(u'WARNING: the program functionality for this site has been marked as broken, ' + self.report_warning(u'the program functionality for this site has been marked as broken, ' u'and will probably not work. If you want to go on, use the -i option.') # Suitable InfoExtractor found @@ -418,63 +464,93 @@ class FileDownloader(object): # Extract information from URL and process it try: ie_results = ie.extract(url) - results = self.process_ie_results(ie_results, ie) + if ie_results is None: # Finished already (backwards compatibility; listformats and friends should be moved here) + break + results = [] + for ie_result in ie_results: + if not 'extractor' in ie_result: + #The extractor has already been set somewhere else + ie_result['extractor'] = ie.IE_NAME + results.append(self.process_ie_result(ie_result, download)) return results except ExtractorError as de: # An error we somewhat expected - self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback()) + self.report_error(compat_str(de), de.format_traceback()) break except Exception as e: if self.params.get('ignoreerrors', False): - self.trouble(u'ERROR: ' + compat_str(e), tb=compat_str(traceback.format_exc())) + self.report_error(compat_str(e), tb=compat_str(traceback.format_exc())) break else: raise if not suitable_found: - self.trouble(u'ERROR: no suitable InfoExtractor: %s' % url) - def extract_info_iterable(self, urls): - ''' - Return the videos founded for the urls - ''' - results = [] - for url in urls: - results.extend(self.extract_info(url)) - return results + self.report_error(u'no suitable InfoExtractor: %s' % url) - def process_ie_results(self, ie_results, ie): + def process_ie_result(self, ie_result, download = True): """ - Take the results of the ie and return a list of videos. - For url elements it will seartch the suitable ie and get the videos + Take the result of the ie and return a list of videos. + For url elements it will search the suitable ie and get the videos For playlist elements it will process each of the elements of the 'entries' key + + It will also download the videos if 'download'. """ - results = [] - for result in ie_results or []: - result_type = result.get('_type', 'video') #If not given we suppose it's a video, support the dafault old system - if result_type == 'video': - if not 'extractor' in result: - #The extractor has already been set somewhere else - result['extractor'] = ie.IE_NAME - results.append(result) - elif result_type == 'url': - #We get the videos pointed by the url - results.extend(self.extract_info(result['url'])) - elif result_type == 'playlist': - #We process each entry in the playlist - entries_result = self.process_ie_results(result['entries'], ie) - result['entries'] = entries_result - results.extend([result]) - return results + result_type = ie_result.get('_type', 'video') #If not given we suppose it's a video, support the dafault 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: + #Do the download: + self.process_info(ie_result) + return ie_result + elif result_type == 'url': + #We get the video pointed by the url + result = self.extract_info(ie_result['url'], download, ie_name = ie_result['ie_key'])[0] + return result + 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) + + playlist_results = [] + + n_all_entries = len(ie_result['entries']) + playliststart = self.params.get('playliststart', 1) - 1 + playlistend = self.params.get('playlistend', -1) + + if playlistend == -1: + entries = ie_result['entries'][playliststart:] + else: + entries = ie_result['entries'][playliststart:playlistend] + + n_entries = len(entries) + + 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)) + entry_result = self.process_ie_result(entry, False) + entry_result['playlist'] = playlist + entry_result['playlist_index'] = i + playliststart + #We must do the download here to correctly set the 'playlist' key + if download: + self.process_info(entry_result) + playlist_results.append(entry_result) + result = ie_result.copy() + result['entries'] = playlist_results + return result def process_info(self, info_dict): """Process a single dictionary returned by an InfoExtractor.""" - if info_dict.get('_type','video') == 'playlist': - playlist = info_dict.get('title', None) or info_dict.get('id', None) - self.to_screen(u'[download] Downloading playlist: %s' % playlist) - for video in info_dict['entries']: - video['playlist'] = playlist - self.process_info(video) - return + #We increment the download the download count here to match the previous behaviour. + self.increment_downloads() + info_dict['fulltitle'] = info_dict['title'] + if len(info_dict['title']) > 200: + info_dict['title'] = info_dict['title'][:197] + u'...' + # Keep for backwards compatibility info_dict['stitle'] = info_dict['title'] @@ -519,7 +595,7 @@ class FileDownloader(object): if dn != '' and not os.path.exists(dn): # dn is already encoded os.makedirs(dn) except (OSError, IOError) as err: - self.trouble(u'ERROR: unable to create directory ' + compat_str(err)) + self.report_error(u'unable to create directory ' + compat_str(err)) return if self.params.get('writedescription', False): @@ -529,7 +605,7 @@ class FileDownloader(object): with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: descfile.write(info_dict['description']) except (OSError, IOError): - self.trouble(u'ERROR: Cannot write description file ' + descfn) + self.report_error(u'Cannot write description file ' + descfn) return if self.params.get('writesubtitles', False) and 'subtitles' in info_dict and info_dict['subtitles']: @@ -538,14 +614,17 @@ class FileDownloader(object): subtitle = info_dict['subtitles'][0] (sub_error, sub_lang, sub) = subtitle sub_format = self.params.get('subtitlesformat') - 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.trouble(u'ERROR: Cannot write subtitles file ' + descfn) - return + 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('onlysubtitles', False): return @@ -554,14 +633,17 @@ class FileDownloader(object): sub_format = self.params.get('subtitlesformat') for subtitle in subtitles: (sub_error, sub_lang, sub) = subtitle - 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.trouble(u'ERROR: Cannot write subtitles file ' + descfn) - return + 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('onlysubtitles', False): return @@ -572,7 +654,7 @@ class FileDownloader(object): 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.trouble(u'ERROR: Cannot write metadata to JSON file ' + infofn) + self.report_error(u'Cannot write metadata to JSON file ' + infofn) return if not self.params.get('skip_download', False): @@ -584,17 +666,17 @@ class FileDownloader(object): except (OSError, IOError) as err: raise UnavailableVideoError() except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: - self.trouble(u'ERROR: unable to download video data: %s' % str(err)) + self.report_error(u'unable to download video data: %s' % str(err)) return except (ContentTooShortError, ) as err: - self.trouble(u'ERROR: content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) + self.report_error(u'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) as err: - self.trouble(u'ERROR: postprocessing: %s' % str(err)) + self.report_error(u'postprocessing: %s' % str(err)) return def download(self, url_list): @@ -603,14 +685,14 @@ class FileDownloader(object): raise SameFileError(self.params['outtmpl']) for url in url_list: - videos = self.extract_info(url) - - for video in videos or []: - try: - self.increment_downloads() - self.process_info(video) - except UnavailableVideoError: - self.trouble(u'\nERROR: unable to download video') + try: + #It also downloads the videos + videos = self.extract_info(url) + except UnavailableVideoError: + self.report_error(u'unable to download video') + except MaxDownloadsReached: + self.to_screen(u'[info] Maximum number of downloaded files reached.') + raise return self._download_retcode @@ -637,7 +719,7 @@ class FileDownloader(object): except (IOError, OSError): self.report_warning(u'Unable to remove downloaded video file') - def _download_with_rtmpdump(self, filename, url, player_url, page_url): + def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path): self.report_destination(filename) tmpfilename = self.temp_name(filename) @@ -645,7 +727,7 @@ class FileDownloader(object): try: subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT) except (OSError, IOError): - self.trouble(u'ERROR: RTMP download detected but "rtmpdump" could not be run') + self.report_error(u'RTMP download detected but "rtmpdump" could not be run') return False # Download using rtmpdump. rtmpdump returns exit code 2 when @@ -656,6 +738,8 @@ class FileDownloader(object): basic_args += ['-W', player_url] if page_url is not None: basic_args += ['--pageUrl', page_url] + if play_path is not None: + basic_args += ['-y', play_path] args = basic_args + [[], ['-e', '-k', '1']][self.params.get('continuedl', False)] if self.params.get('verbose', False): try: @@ -690,7 +774,8 @@ class FileDownloader(object): }) return True else: - self.trouble(u'\nERROR: rtmpdump exited with code %d' % retval) + self.to_stderr(u"\n") + self.report_error(u'rtmpdump exited with code %d' % retval) return False def _do_download(self, filename, info_dict): @@ -709,7 +794,8 @@ class FileDownloader(object): if url.startswith('rtmp'): return self._download_with_rtmpdump(filename, url, info_dict.get('player_url', None), - info_dict.get('page_url', None)) + info_dict.get('page_url', None), + info_dict.get('play_path', None)) tmpfilename = self.temp_name(filename) stream = None @@ -790,7 +876,7 @@ class FileDownloader(object): self.report_retry(count, retries) if count > retries: - self.trouble(u'ERROR: giving up after %s retries' % retries) + self.report_error(u'giving up after %s retries' % retries) return False data_len = data.info().get('Content-length', None) @@ -826,12 +912,13 @@ class FileDownloader(object): filename = self.undo_temp_name(tmpfilename) self.report_destination(filename) except (OSError, IOError) as err: - self.trouble(u'ERROR: unable to open for writing: %s' % str(err)) + self.report_error(u'unable to open for writing: %s' % str(err)) return False try: stream.write(data_block) except (IOError, OSError) as err: - self.trouble(u'\nERROR: unable to write data: %s' % str(err)) + self.to_stderr(u"\n") + self.report_error(u'unable to write data: %s' % str(err)) return False if not self.params.get('noresizebuffer', False): block_size = self.best_block_size(after - before, len(data_block)) @@ -857,7 +944,8 @@ class FileDownloader(object): self.slow_down(start, byte_counter - resume_len) if stream is None: - self.trouble(u'\nERROR: Did not get any data blocks') + self.to_stderr(u"\n") + self.report_error(u'Did not get any data blocks') return False stream.close() self.report_finish()