class ScramDLThread(threading.Thread): def __init__(self, urls, addendum_params=None): threading.Thread.__init__(self) override_params = dict() override_params['writeinfojson'] = False override_params['test'] = False override_params['format'] = 'mp4' if addendum_params is not None: override_params.update(addendum_params) params = get_params(override_params) self.scram_dl = YoutubeDL(params) self.scram_dl.add_progress_hook(self.hook_progress) self.scram_dl.add_default_info_extractors() self.is_finished = False self.urls = urls def hook_progress(self, status): if status['status'] == 'finished': self.is_finished = True else: self.is_finished = False def run(self): self.scram_dl.download(self.urls)
def test_template(self): ie = youtube_dl.extractor.get_info_extractor(test_case['name']) other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])] def print_skipping(reason): print('Skipping %s: %s' % (test_case['name'], reason)) if not ie.working(): print_skipping('IE marked as not _WORKING') return if 'playlist' not in test_case: info_dict = test_case.get('info_dict', {}) if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')): print_skipping('The output file cannot be know, the "file" ' 'key is missing or the info_dict is incomplete') return if 'skip' in test_case: print_skipping(test_case['skip']) return for other_ie in other_ies: if not other_ie.working(): print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) return params = get_params(test_case.get('params', {})) ydl = YoutubeDL(params) ydl.add_default_info_extractors() finished_hook_called = set() def _hook(status): if status['status'] == 'finished': finished_hook_called.add(status['filename']) ydl.add_progress_hook(_hook) def get_tc_filename(tc): return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {})) test_cases = test_case.get('playlist', [test_case]) def try_rm_tcs_files(): for tc in test_cases: tc_filename = get_tc_filename(tc) try_rm(tc_filename) try_rm(tc_filename + '.part') try_rm(os.path.splitext(tc_filename)[0] + '.info.json') try_rm_tcs_files() try: try_num = 1 while True: try: ydl.download([test_case['url']]) except (DownloadError, ExtractorError) as err: # Check if the exception is not a network related one if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503): raise if try_num == RETRIES: report_warning(u'Failed due to network errors, skipping...') return print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num)) try_num += 1 else: break for tc in test_cases: tc_filename = get_tc_filename(tc) if not test_case.get('params', {}).get('skip_download', False): self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) self.assertTrue(tc_filename in finished_hook_called) info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' self.assertTrue(os.path.exists(info_json_fn)) if 'md5' in tc: md5_for_file = _file_md5(tc_filename) self.assertEqual(md5_for_file, tc['md5']) with io.open(info_json_fn, encoding='utf-8') as infof: info_dict = json.load(infof) for (info_field, expected) in tc.get('info_dict', {}).items(): if isinstance(expected, compat_str) and expected.startswith('md5:'): got = 'md5:' + md5(info_dict.get(info_field)) else: got = info_dict.get(info_field) self.assertEqual(expected, got, u'invalid value for field %s, expected %r, got %r' % (info_field, expected, got)) # If checkable fields are missing from the test case, print the info_dict test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value)) for key, value in info_dict.items() if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location')) if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()): sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=4) + u'\n') # Check for the presence of mandatory fields for key in ('id', 'url', 'title', 'ext'): self.assertTrue(key in info_dict.keys() and info_dict[key]) # Check for mandatory fields that are automatically set by YoutubeDL for key in ['webpage_url', 'extractor', 'extractor_key']: self.assertTrue(info_dict.get(key), u'Missing field: %s' % key) finally: try_rm_tcs_files()
def test_template(self): ie = youtube_dl.extractor.get_info_extractor(test_case['name']) other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])] is_playlist = any(k.startswith('playlist') for k in test_case) test_cases = test_case.get( 'playlist', [] if is_playlist else [test_case]) def print_skipping(reason): print('Skipping %s: %s' % (test_case['name'], reason)) if not ie.working(): print_skipping('IE marked as not _WORKING') return for tc in test_cases: info_dict = tc.get('info_dict', {}) if not tc.get('file') and not (info_dict.get('id') and info_dict.get('ext')): raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?') if 'skip' in test_case: print_skipping(test_case['skip']) return for other_ie in other_ies: if not other_ie.working(): print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) return params = get_params(test_case.get('params', {})) if is_playlist and 'playlist' not in test_case: params.setdefault('extract_flat', True) params.setdefault('skip_download', True) ydl = YoutubeDL(params, auto_init=False) ydl.add_default_info_extractors() finished_hook_called = set() def _hook(status): if status['status'] == 'finished': finished_hook_called.add(status['filename']) ydl.add_progress_hook(_hook) expect_warnings(ydl, test_case.get('expected_warnings', [])) def get_tc_filename(tc): return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {})) res_dict = None def try_rm_tcs_files(tcs=None): if tcs is None: tcs = test_cases for tc in tcs: tc_filename = get_tc_filename(tc) try_rm(tc_filename) try_rm(tc_filename + '.part') try_rm(os.path.splitext(tc_filename)[0] + '.info.json') try_rm_tcs_files() try: try_num = 1 while True: try: # We're not using .download here sine that is just a shim # for outside error handling, and returns the exit code # instead of the result dict. res_dict = ydl.extract_info(test_case['url']) except (DownloadError, ExtractorError) as err: # Check if the exception is not a network related one if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503): raise if try_num == RETRIES: report_warning(u'Failed due to network errors, skipping...') return print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num)) try_num += 1 else: break if is_playlist: self.assertEqual(res_dict['_type'], 'playlist') self.assertTrue('entries' in res_dict) expect_info_dict(self, test_case.get('info_dict', {}), res_dict) if 'playlist_mincount' in test_case: assertGreaterEqual( self, len(res_dict['entries']), test_case['playlist_mincount'], 'Expected at least %d in playlist %s, but got only %d' % ( test_case['playlist_mincount'], test_case['url'], len(res_dict['entries']))) if 'playlist_count' in test_case: self.assertEqual( len(res_dict['entries']), test_case['playlist_count'], 'Expected %d entries in playlist %s, but got %d.' % ( test_case['playlist_count'], test_case['url'], len(res_dict['entries']), )) if 'playlist_duration_sum' in test_case: got_duration = sum(e['duration'] for e in res_dict['entries']) self.assertEqual( test_case['playlist_duration_sum'], got_duration) for tc in test_cases: tc_filename = get_tc_filename(tc) if not test_case.get('params', {}).get('skip_download', False): self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) self.assertTrue(tc_filename in finished_hook_called) expected_minsize = tc.get('file_minsize', 10000) if expected_minsize is not None: if params.get('test'): expected_minsize = max(expected_minsize, 10000) got_fsize = os.path.getsize(tc_filename) assertGreaterEqual( self, got_fsize, expected_minsize, 'Expected %s to be at least %s, but it\'s only %s ' % (tc_filename, format_bytes(expected_minsize), format_bytes(got_fsize))) if 'md5' in tc: md5_for_file = _file_md5(tc_filename) self.assertEqual(md5_for_file, tc['md5']) info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' self.assertTrue( os.path.exists(info_json_fn), 'Missing info file %s' % info_json_fn) with io.open(info_json_fn, encoding='utf-8') as infof: info_dict = json.load(infof) expect_info_dict(self, tc.get('info_dict', {}), info_dict) finally: try_rm_tcs_files() if is_playlist and res_dict is not None and res_dict.get('entries'): # Remove all other files that may have been extracted if the # extractor returns full results even with extract_flat res_tcs = [{'info_dict': e} for e in res_dict['entries']] try_rm_tcs_files(res_tcs)
def test_template(self): ie = youtube_dl.extractor.get_info_extractor(test_case['name'])() other_ies = [ get_info_extractor(ie_key)() for ie_key in test_case.get('add_ie', []) ] is_playlist = any(k.startswith('playlist') for k in test_case) test_cases = test_case.get('playlist', [] if is_playlist else [test_case]) def print_skipping(reason): print('Skipping %s: %s' % (test_case['name'], reason)) self.skipTest(reason) if not ie.working(): print_skipping('IE marked as not _WORKING') for tc in test_cases: info_dict = tc.get('info_dict', {}) if not (info_dict.get('id') and info_dict.get('ext')): raise Exception( 'Test definition (%s) requires both \'id\' and \'ext\' keys present to define the output file' % (tname, )) if 'skip' in test_case: print_skipping(test_case['skip']) for other_ie in other_ies: if not other_ie.working(): print_skipping('test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) params = get_params(test_case.get('params', {})) params['outtmpl'] = tname + '_' + params['outtmpl'] if is_playlist and 'playlist' not in test_case: params.setdefault('extract_flat', 'in_playlist') params.setdefault('playlistend', test_case.get('playlist_mincount')) params.setdefault('skip_download', True) ydl = YoutubeDL(params, auto_init=False) ydl.add_default_info_extractors() finished_hook_called = set() def _hook(status): if status['status'] == 'finished': finished_hook_called.add(status['filename']) ydl.add_progress_hook(_hook) expect_warnings(ydl, test_case.get('expected_warnings', [])) def get_tc_filename(tc): return ydl.prepare_filename(tc.get('info_dict', {})) res_dict = None def try_rm_tcs_files(tcs=None): if tcs is None: tcs = test_cases for tc in tcs: tc_filename = get_tc_filename(tc) try_rm(tc_filename) try_rm(tc_filename + '.part') try_rm(os.path.splitext(tc_filename)[0] + '.info.json') try_rm_tcs_files() try: try_num = 1 while True: try: # We're not using .download here since that is just a shim # for outside error handling, and returns the exit code # instead of the result dict. res_dict = ydl.extract_info( test_case['url'], force_generic_extractor=params.get( 'force_generic_extractor', False)) except (DownloadError, ExtractorError) as err: # Check if the exception is not a network related one if not err.exc_info[0] in ( compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or ( err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503): msg = getattr(err, 'msg', error_to_compat_str(err)) err.msg = '%s (%s)' % ( msg, tname, ) raise err if try_num == RETRIES: report_warning( '%s failed due to network errors, skipping...' % tname) return print( 'Retrying: {0} failed tries\n\n##########\n\n'.format( try_num)) try_num += 1 else: break if is_playlist: self.assertTrue( res_dict['_type'] in ['playlist', 'multi_video']) self.assertTrue('entries' in res_dict) expect_info_dict(self, res_dict, test_case.get('info_dict', {})) if 'playlist_mincount' in test_case: assertGreaterEqual( self, len(res_dict['entries']), test_case['playlist_mincount'], 'Expected at least %d in playlist %s, but got only %d' % (test_case['playlist_mincount'], test_case['url'], len(res_dict['entries']))) if 'playlist_count' in test_case: self.assertEqual( len(res_dict['entries']), test_case['playlist_count'], 'Expected %d entries in playlist %s, but got %d.' % ( test_case['playlist_count'], test_case['url'], len(res_dict['entries']), )) if 'playlist_duration_sum' in test_case: got_duration = sum(e['duration'] for e in res_dict['entries']) self.assertEqual(test_case['playlist_duration_sum'], got_duration) # Generalize both playlists and single videos to unified format for # simplicity if 'entries' not in res_dict: res_dict['entries'] = [res_dict] for tc_num, tc in enumerate(test_cases): tc_res_dict = res_dict['entries'][tc_num] # First, check test cases' data against extracted data alone expect_info_dict(self, tc_res_dict, tc.get('info_dict', {})) # Now, check downloaded file consistency tc_filename = get_tc_filename(tc) if not test_case.get('params', {}).get('skip_download', False): self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) self.assertTrue(tc_filename in finished_hook_called) expected_minsize = tc.get('file_minsize', 10000) if expected_minsize is not None: if params.get('test'): expected_minsize = max(expected_minsize, 10000) got_fsize = os.path.getsize(tc_filename) assertGreaterEqual( self, got_fsize, expected_minsize, 'Expected %s to be at least %s, but it\'s only %s ' % (tc_filename, format_bytes(expected_minsize), format_bytes(got_fsize))) if 'md5' in tc: md5_for_file = _file_md5(tc_filename) self.assertEqual(tc['md5'], md5_for_file) # Finally, check test cases' data again but this time against # extracted data from info JSON file written during processing info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' self.assertTrue(os.path.exists(info_json_fn), 'Missing info file %s' % info_json_fn) with io.open(info_json_fn, encoding='utf-8') as infof: info_dict = json.load(infof) expect_info_dict(self, info_dict, tc.get('info_dict', {})) finally: try_rm_tcs_files() if is_playlist and res_dict is not None and res_dict.get( 'entries'): # Remove all other files that may have been extracted if the # extractor returns full results even with extract_flat res_tcs = [{'info_dict': e} for e in res_dict['entries']] try_rm_tcs_files(res_tcs)
def test_template(self): ie = youtube_dl.extractor.get_info_extractor(test_case['name']) other_ies = [ get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', []) ] def print_skipping(reason): print('Skipping %s: %s' % (test_case['name'], reason)) if not ie.working(): print_skipping('IE marked as not _WORKING') return if 'playlist' not in test_case: info_dict = test_case.get('info_dict', {}) if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')): raise Exception( 'Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?' ) if 'skip' in test_case: print_skipping(test_case['skip']) return for other_ie in other_ies: if not other_ie.working(): print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) return params = get_params(test_case.get('params', {})) ydl = YoutubeDL(params) ydl.add_default_info_extractors() finished_hook_called = set() def _hook(status): if status['status'] == 'finished': finished_hook_called.add(status['filename']) ydl.add_progress_hook(_hook) def get_tc_filename(tc): return tc.get('file') or ydl.prepare_filename( tc.get('info_dict', {})) test_cases = test_case.get('playlist', [test_case]) def try_rm_tcs_files(): for tc in test_cases: tc_filename = get_tc_filename(tc) try_rm(tc_filename) try_rm(tc_filename + '.part') try_rm(os.path.splitext(tc_filename)[0] + '.info.json') try_rm_tcs_files() try: try_num = 1 while True: try: ydl.download([test_case['url']]) except (DownloadError, ExtractorError) as err: # Check if the exception is not a network related one if not err.exc_info[0] in ( compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or ( err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503): raise if try_num == RETRIES: report_warning( u'Failed due to network errors, skipping...') return print( 'Retrying: {0} failed tries\n\n##########\n\n'.format( try_num)) try_num += 1 else: break for tc in test_cases: tc_filename = get_tc_filename(tc) if not test_case.get('params', {}).get('skip_download', False): self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) self.assertTrue(tc_filename in finished_hook_called) info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' self.assertTrue(os.path.exists(info_json_fn)) if 'md5' in tc: md5_for_file = _file_md5(tc_filename) self.assertEqual(md5_for_file, tc['md5']) with io.open(info_json_fn, encoding='utf-8') as infof: info_dict = json.load(infof) expect_info_dict(self, tc.get('info_dict', {}), info_dict) finally: try_rm_tcs_files()
def test_template(self): ie = youtube_dl.extractor.get_info_extractor(test_case['name']) other_ies = [get_info_extractor(ie_key) for ie_key in test_case.get('add_ie', [])] def print_skipping(reason): print('Skipping %s: %s' % (test_case['name'], reason)) if not ie.working(): print_skipping('IE marked as not _WORKING') return if 'playlist' not in test_case: info_dict = test_case.get('info_dict', {}) if not test_case.get('file') and not (info_dict.get('id') and info_dict.get('ext')): raise Exception('Test definition incorrect. The output file cannot be known. Are both \'id\' and \'ext\' keys present?') if 'skip' in test_case: print_skipping(test_case['skip']) return for other_ie in other_ies: if not other_ie.working(): print_skipping(u'test depends on %sIE, marked as not WORKING' % other_ie.ie_key()) return params = get_params(test_case.get('params', {})) ydl = YoutubeDL(params) ydl.add_default_info_extractors() finished_hook_called = set() def _hook(status): if status['status'] == 'finished': finished_hook_called.add(status['filename']) ydl.add_progress_hook(_hook) def get_tc_filename(tc): return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {})) test_cases = test_case.get('playlist', [test_case]) def try_rm_tcs_files(): for tc in test_cases: tc_filename = get_tc_filename(tc) try_rm(tc_filename) try_rm(tc_filename + '.part') try_rm(os.path.splitext(tc_filename)[0] + '.info.json') try_rm_tcs_files() try: try_num = 1 while True: try: ydl.download([test_case['url']]) except (DownloadError, ExtractorError) as err: # Check if the exception is not a network related one if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503): raise if try_num == RETRIES: report_warning(u'Failed due to network errors, skipping...') return print('Retrying: {0} failed tries\n\n##########\n\n'.format(try_num)) try_num += 1 else: break for tc in test_cases: tc_filename = get_tc_filename(tc) if not test_case.get('params', {}).get('skip_download', False): self.assertTrue(os.path.exists(tc_filename), msg='Missing file ' + tc_filename) self.assertTrue(tc_filename in finished_hook_called) info_json_fn = os.path.splitext(tc_filename)[0] + '.info.json' self.assertTrue(os.path.exists(info_json_fn)) if 'md5' in tc: md5_for_file = _file_md5(tc_filename) self.assertEqual(md5_for_file, tc['md5']) with io.open(info_json_fn, encoding='utf-8') as infof: info_dict = json.load(infof) expect_info_dict(self, tc.get('info_dict', {}), info_dict) finally: try_rm_tcs_files()
class DownloadManager(QtCore.QAbstractTableModel): playVid = QtCore.Signal(str) def __init__(self): super(DownloadManager, self).__init__() self.initialized_ = True self.downloader_ = YoutubeDL() self.downloader_.add_progress_hook(self.progress) self.workerThreads_ = [] self.vids_ = [] self._max_simulateous_downloads = 1 self.download_location_ = tempfile.gettempdir()+'/maxitube' self.downloader_.params['outtmpl'] = self.download_location_+'/%(autonumber)s.%(ext)s' self.downloader_.params['noprogress'] = False #self.setHorizontalHeaderLabels(['name', "status", "progress", "file"]) #self.requestUpdateLine.connect(self.updateLine) def __del__(self): try: shutil.rmtree(self.download_location_) except FileNotFoundError: pass @QtCore.Slot(str) def add(self, vid): for vid2 in self.vids_: if vid['url'] == vid2['url']: return vid['status'] = 'queued' index = len(self.vids_)-1 self.beginInsertRows(QtCore.QModelIndex(), index, index) self.vids_.append(vid) self.endInsertRows() #self.insertRows(self.rowCount(), 1) self.dataChanged.emit(self.index(0, 0), self.index(self.rowCount()-1, self.columnCount()-1)) #url = vid['url'] #item = QStandardItem(url) #self.appendRow([item, QStandardItem('queued'), QStandardItem(''), QStandardItem('')]) self.update() def columnCount(self, parent=QtCore.QModelIndex()): return 5 def rowCount(self, parent=QtCore.QModelIndex()): return len(self.vids_) def data(self, index, role): data = None if role == QtCore.Qt.DisplayRole: row = index.row() vid = self.vids_[row] section = index.column() if section == 0: data = vid['url'] elif section == 1 and 'status' in vid: data = vid['status'] elif section == 2 and '_percent_str' in vid: data = vid['_percent_str'] elif section == 3 and 'title' in vid: data = vid['title'] elif section == 4 and 'filename' in vid: data = vid['filename'] return data def headerData(self, section, orientation, role=QtCore.Qt.DisplayRole): data = None if role == QtCore.Qt.DisplayRole and orientation == QtCore.Qt.Horizontal: if section == 0: data = 'Url' elif section == 1: data = 'Status' elif section == 2: data = 'Progress' elif section == 3: data = 'Title' elif section == 4: data = 'File' return data def update(self): simulateous_downloads = 0 first_queued = None first_index = None # FIXME iterate over self.vids_ for index in range(self.rowCount()): vid = self.vids_[index] status = vid['status'] if status == 'downloading': simulateous_downloads += 1 elif status == 'queued': if not first_queued: first_queued = vid first_index = index if first_queued and (simulateous_downloads < self._max_simulateous_downloads): self.vids_[first_index]['status']='downloading' self.dataChanged.emit(self.index(0, 0), self.index(self.rowCount()-1, self.columnCount()-1)) self.worker_ = DownloadWorker(self, first_queued) self.worker_.finished.connect(self.update) thread = QtCore.QThread() self.worker_.moveToThread(thread) thread.started.connect(self.worker_.doWork) self.worker_.finished.connect(thread.quit) thread.finished.connect(self.worker_.deleteLater) #self.workerThread_.finished.connect(self.workerThread_.deleteLater) thread.start() self.workerThreads_.append(thread) def progress(self, dl_infos): filename = None if 'tmpfilename' in dl_infos: filename = dl_infos['tmpfilename'] elif 'filename' in dl_infos: filename = dl_infos['filename'] if filename: basename_we = os.path.basename(filename).split('.')[0] index = int(basename_we)-1 self.vids_[index]['filename'] = filename self.vids_[index]['status'] = dl_infos['status'] if '_percent_str' in dl_infos: self.vids_[index]['_percent_str']= dl_infos['_percent_str'] self.dataChanged.emit(self.index(0, 0), self.index(self.rowCount()-1, self.columnCount()-1)) def start_download(self, vid): self.downloader_.add_progress_hook(self.progress) url = vid['url'] print('-- downloading:',url) try: infos = self.downloader_.extract_info(url) except DownloadError: for index in range(len(self.vids_)): if url == self.vids_[index]['url']: self.vids_[index]['status'] = 'error' self.dataChanged.emit(self.index(0, 0), self.index(self.rowCount()-1, self.columnCount()-1))