X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2FFileDownloader.py;h=574863e7c9da066b1fc29741a83a58f2a2baaed4;hb=086d7b4500dad720e548233e425b4c08bfe99e3e;hp=5087b4cc824db2d7857407f3215cd0baaede9838;hpb=46b9d8295d43bb4a5954da95293ef2490eb28b5e;p=youtube-dl diff --git a/youtube_dl/FileDownloader.py b/youtube_dl/FileDownloader.py index 5087b4cc8..574863e7c 100644 --- a/youtube_dl/FileDownloader.py +++ b/youtube_dl/FileDownloader.py @@ -7,6 +7,7 @@ import math import io import os import re +import shutil import socket import subprocess import sys @@ -17,6 +18,7 @@ if os.name == 'nt': import ctypes from .utils import * +from .InfoExtractors import get_info_extractor class FileDownloader(object): @@ -78,6 +80,7 @@ class FileDownloader(object): updatetime: Use the Last-modified header to set output file timestamps. writedescription: Write the video description to a .description file writeinfojson: Write the video description to a .info.json file + writethumbnail: Write the thumbnail image to a file writesubtitles: Write the video subtitles to a file onlysubtitles: Downloads only the subtitles of the video allsubtitles: Downloads all the subtitles of the video @@ -88,6 +91,7 @@ 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 +124,7 @@ 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 +235,21 @@ 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): @@ -243,7 +257,7 @@ 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:' @@ -255,7 +269,7 @@ class FileDownloader(object): Do the same as trouble, but prefixes the message with 'ERROR:', colored in red if stderr is a tty file. ''' - if sys.stderr.isatty(): + if sys.stderr.isatty() and os.name != 'nt': _msg_header = u'\033[0;31mERROR:\033[0m' else: _msg_header = u'ERROR:' @@ -333,12 +347,13 @@ class FileDownloader(object): """Report download progress.""" if self.params.get('noprogress', False): return + clear_line = (u'\x1b[K' if sys.stderr.isatty() and os.name != 'nt' else u'') if self.params.get('progress_with_newline', False): self.to_screen(u'[download] %s of %s at %s ETA %s' % (percent_str, data_len_str, speed_str, eta_str)) else: - self.to_screen(u'\r[download] %s of %s at %s ETA %s' % - (percent_str, data_len_str, speed_str, eta_str), skip_eol=True) + self.to_screen(u'\r%s[download] %s of %s at %s ETA %s' % + (clear_line, percent_str, data_len_str, speed_str, eta_str), skip_eol=True) self.to_cons_title(u'youtube-dl - %s of %s at %s ETA %s' % (percent_str.strip(), data_len_str.strip(), speed_str.strip(), eta_str.strip())) @@ -378,7 +393,13 @@ 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), @@ -389,10 +410,10 @@ 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): @@ -407,11 +428,132 @@ 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, 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 + + #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.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 + suitable_found = True + + # Extract information from URL and process it + try: + ie_results = ie.extract(url) + 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.report_error(compat_str(de), de.format_traceback()) + break + except Exception as e: + if self.params.get('ignoreerrors', False): + self.report_error(compat_str(e), tb=compat_str(traceback.format_exc())) + break + else: + raise + if not suitable_found: + self.report_error(u'no suitable InfoExtractor: %s' % url) + + def process_ie_result(self, ie_result, download = True): + """ + 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'. + """ + 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.""" + #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'] @@ -475,14 +617,17 @@ 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.report_error(u'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 @@ -491,14 +636,17 @@ 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 @@ -512,6 +660,20 @@ class FileDownloader(object): 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' + 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)) + if not self.params.get('skip_download', False): if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)): success = True @@ -540,53 +702,14 @@ class FileDownloader(object): raise SameFileError(self.params['outtmpl']) for url in url_list: - suitable_found = False - for ie in self._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.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 - suitable_found = True - - # Extract information from URL and process it - try: - videos = ie.extract(url) - except ExtractorError as de: # An error we somewhat expected - self.trouble(u'ERROR: ' + compat_str(de), de.format_traceback()) - break - except MaxDownloadsReached: - self.to_screen(u'[info] Maximum number of downloaded files reached.') - raise - except Exception as e: - if self.params.get('ignoreerrors', False): - self.report_error(u'' + compat_str(e), tb=compat_str(traceback.format_exc())) - break - else: - raise - - if len(videos or []) > 1 and self.fixed_template(): - raise SameFileError(self.params['outtmpl']) - - for video in videos or []: - video['extractor'] = ie.IE_NAME - try: - self.increment_downloads() - self.process_info(video) - except UnavailableVideoError: - self.to_stderr(u"\n") - self.report_error(u'unable to download video') - - # Suitable InfoExtractor had been found; go to next URL - break - - if not suitable_found: - self.report_error(u'no suitable InfoExtractor: %s' % url) + 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 @@ -613,7 +736,7 @@ 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) @@ -632,6 +755,8 @@ 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: @@ -686,7 +811,8 @@ 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