X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/c4994c6aff3e03c7848b0a9d4ff9a1873412d31f..2646724d304a361bc9849ae5f3355176690e5494:/test/test_download.py diff --git a/test/test_download.py b/test/test_download.py index f45c500..7587a18 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -1,197 +1,178 @@ #!/usr/bin/env python -# DO NOT EDIT THIS FILE BY HAND! -# It is auto-generated from tests.json and gentests.py. - -import hashlib -import io +# Allow direct execution import os -import json -import unittest import sys -import socket +import unittest +sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -# Allow direct execution -import os -sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) +from test.helper import ( + get_params, + get_testcases, + try_rm, + md5, + report_warning +) -import youtube_dl.FileDownloader -import youtube_dl.InfoExtractors -from youtube_dl.utils import * -# General configuration (from __init__, not very elegant...) -jar = compat_cookiejar.CookieJar() -cookie_processor = compat_urllib_request.HTTPCookieProcessor(jar) -proxy_handler = compat_urllib_request.ProxyHandler() -opener = compat_urllib_request.build_opener(proxy_handler, cookie_processor, YoutubeDLHandler()) -compat_urllib_request.install_opener(opener) -socket.setdefaulttimeout(300) # 5 minutes should be enough (famous last words) +import hashlib +import io +import json +import socket -class FileDownloader(youtube_dl.FileDownloader): +import youtube_dl.YoutubeDL +from youtube_dl.utils import ( + compat_http_client, + compat_str, + compat_urllib_error, + compat_HTTPError, + DownloadError, + ExtractorError, + UnavailableVideoError, +) +from youtube_dl.extractor import get_info_extractor + +RETRIES = 3 + +class YoutubeDL(youtube_dl.YoutubeDL): def __init__(self, *args, **kwargs): - youtube_dl.FileDownloader.__init__(self, *args, **kwargs) self.to_stderr = self.to_screen + self.processed_info_dicts = [] + super(YoutubeDL, self).__init__(*args, **kwargs) + def report_warning(self, message): + # Don't accept warnings during tests + raise ExtractorError(message) + def process_info(self, info_dict): + self.processed_info_dicts.append(info_dict) + return super(YoutubeDL, self).process_info(info_dict) def _file_md5(fn): with open(fn, 'rb') as f: return hashlib.md5(f.read()).hexdigest() -try: - _skip_unless = unittest.skipUnless -except AttributeError: # Python 2.6 - def _skip_unless(cond, reason='No reason given'): - def resfunc(f): - # Start the function name with test to appease nosetests-2.6 - def test_wfunc(*args, **kwargs): - if cond: - return f(*args, **kwargs) - else: - print('Skipped test') - return - return test_wfunc - return resfunc -_skip = lambda *args, **kwargs: _skip_unless(False, *args, **kwargs) -class DownloadTest(unittest.TestCase): - PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json") +defs = get_testcases() - def setUp(self): - # Clear old files - self.tearDown() - - with io.open(self.PARAMETERS_FILE, encoding='utf-8') as pf: - self.parameters = json.load(pf) - - @_skip_unless(youtube_dl.InfoExtractors.YoutubeIE._WORKING, "IE marked as not _WORKING") - def test_Youtube(self): - filename = u'BaW_jenozKc.mp4' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.YoutubeIE()) - fd.download([u'http://www.youtube.com/watch?v=BaW_jenozKc']) - self.assertTrue(os.path.exists(filename)) - - @_skip_unless(youtube_dl.InfoExtractors.DailymotionIE._WORKING, "IE marked as not _WORKING") - def test_Dailymotion(self): - filename = u'x33vw9.mp4' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.DailymotionIE()) - fd.download([u'http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'392c4b85a60a90dc4792da41ce3144eb') - - @_skip_unless(youtube_dl.InfoExtractors.MetacafeIE._WORKING, "IE marked as not _WORKING") - def test_Metacafe(self): - filename = u'_aUehQsCQtM.flv' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.MetacafeIE()) - fd.add_info_extractor(youtube_dl.InfoExtractors.YoutubeIE()) - fd.download([u'http://www.metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/']) - self.assertTrue(os.path.exists(filename)) - - @_skip_unless(youtube_dl.InfoExtractors.BlipTVIE._WORKING, "IE marked as not _WORKING") - def test_BlipTV(self): - filename = u'5779306.m4v' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.BlipTVIE()) - fd.download([u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'b2d849efcf7ee18917e4b4d9ff37cafe') - - @_skip_unless(youtube_dl.InfoExtractors.XVideosIE._WORKING, "IE marked as not _WORKING") - def test_XVideos(self): - filename = u'939581.flv' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.XVideosIE()) - fd.download([u'http://www.xvideos.com/video939581/funny_porns_by_s_-1']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'1d0c835822f0a71a7bf011855db929d0') - - @_skip_unless(youtube_dl.InfoExtractors.VimeoIE._WORKING, "IE marked as not _WORKING") - def test_Vimeo(self): - filename = u'14160053.mp4' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.VimeoIE()) - fd.download([u'http://vimeo.com/14160053']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'60540a4ec7cc378ec84b919c0aed5023') - - @_skip_unless(youtube_dl.InfoExtractors.SoundcloudIE._WORKING, "IE marked as not _WORKING") - def test_Soundcloud(self): - filename = u'62986583.mp3' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.SoundcloudIE()) - fd.download([u'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'ebef0a451b909710ed1d7787dddbf0d7') - - @_skip_unless(youtube_dl.InfoExtractors.StanfordOpenClassroomIE._WORKING, "IE marked as not _WORKING") - def test_StanfordOpenClassroom(self): - filename = u'PracticalUnix_intro-environment.mp4' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.StanfordOpenClassroomIE()) - fd.download([u'http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'544a9468546059d4e80d76265b0443b8') - - @_skip_unless(youtube_dl.InfoExtractors.XNXXIE._WORKING, "IE marked as not _WORKING") - def test_XNXX(self): - filename = u'1135332.flv' - params = self.parameters - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.XNXXIE()) - fd.download([u'http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'0831677e2b4761795f68d417e0b7b445') - - @_skip_unless(youtube_dl.InfoExtractors.YoukuIE._WORKING, "IE marked as not _WORKING") - def test_Youku(self): - filename = u'XNDgyMDQ2NTQw_part00.flv' - params = self.parameters - params["test"] = False - fd = FileDownloader(params) - fd.add_info_extractor(youtube_dl.InfoExtractors.YoukuIE()) - fd.download([u'http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html']) - self.assertTrue(os.path.exists(filename)) - md5_for_file = _file_md5(filename) - self.assertEqual(md5_for_file, u'ffe3f2e435663dc2d1eea34faeff5b5b') - - - def tearDown(self): - if os.path.exists(u'BaW_jenozKc.mp4'): - os.remove(u'BaW_jenozKc.mp4') - if os.path.exists(u'x33vw9.mp4'): - os.remove(u'x33vw9.mp4') - if os.path.exists(u'_aUehQsCQtM.flv'): - os.remove(u'_aUehQsCQtM.flv') - if os.path.exists(u'5779306.m4v'): - os.remove(u'5779306.m4v') - if os.path.exists(u'939581.flv'): - os.remove(u'939581.flv') - if os.path.exists(u'14160053.mp4'): - os.remove(u'14160053.mp4') - if os.path.exists(u'62986583.mp3'): - os.remove(u'62986583.mp3') - if os.path.exists(u'PracticalUnix_intro-environment.mp4'): - os.remove(u'PracticalUnix_intro-environment.mp4') - if os.path.exists(u'1135332.flv'): - os.remove(u'1135332.flv') - if os.path.exists(u'XNDgyMDQ2NTQw_part00.flv'): - os.remove(u'XNDgyMDQ2NTQw_part00.flv') +class TestDownload(unittest.TestCase): + maxDiff = None + def setUp(self): + self.defs = defs + +### Dynamically generate tests +def generator(test_case): + + 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() + + return test_template + +### And add them to TestDownload +for n, test_case in enumerate(defs): + test_method = generator(test_case) + tname = 'test_' + str(test_case['name']) + i = 1 + while hasattr(TestDownload, tname): + tname = 'test_' + str(test_case['name']) + '_' + str(i) + i += 1 + test_method.__name__ = tname + setattr(TestDownload, test_method.__name__, test_method) + del test_method if __name__ == '__main__':