X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=test%2Ftest_download.py;h=d1d6b119b9290fb60f1ce56fa68ff1e67ce034d9;hb=f6cc16f5d821a50df173b865164e4fa9cbe854af;hp=a2f5abcb2f6dcefc1ac5b087f9d11f6263f81dca;hpb=4c9afeca34ba3d981fa51db18fe9522212ffad4c;p=youtube-dl diff --git a/test/test_download.py b/test/test_download.py index a2f5abcb2..dfb04d010 100644 --- a/test/test_download.py +++ b/test/test_download.py @@ -1,148 +1,170 @@ -#!/usr/bin/env python2 -import unittest -import hashlib +#!/usr/bin/env python + +# Allow direct execution import os +import sys +import unittest +sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) -from youtube_dl.FileDownloader import FileDownloader -from youtube_dl.InfoExtractors import YoutubeIE, DailymotionIE -from youtube_dl.InfoExtractors import MetacafeIE, PhotobucketIE -from youtube_dl.InfoExtractors import FacebookIE, BlipTVIE -from youtube_dl.InfoExtractors import VimeoIE, XVideosIE +from test.helper import ( + get_params, + get_testcases, + global_setup, + try_rm, + md5, + report_warning +) +global_setup() -class DownloadTest(unittest.TestCase): - #calculated with md5sum: - #md5sum (GNU coreutils) 8.19 - YOUTUBE_MD5 = "8547978241cb87dd6782b10b8e90acc3" - YOUTUBE_URL = "http://www.youtube.com/watch?v=BaW_jenozKc" - YOUTUBE_FILE = "BaW_jenozKc.flv" - - - DAILYMOTION_MD5 = "" - DAILYMOTION_URL = "http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech" - DAILYMOTION_FILE = "" - - - METACAFE_MD5 = "" - METACAFE_URL = "http://www.metacafe.com/watch/yt-bV9L5Ht9LgY/download_youtube_playlist_with_youtube_dl/" - METACAFE_FILE = "" - - - PHOTOBUCKET_MD5 = "" - PHOTOBUCKET_URL = "http://www.metacafe.com/watch/yt-bV9L5Ht9LgY/download_youtube_playlist_with_youtube_dl/" - PHOTOBUCKET_FILE = "" - - - FACEBOOK_MD5 = "" - FACEBOOK_URL = "https://www.facebook.com/video/video.php?v=207446242657384" - FACEBOOK_FILE = "" - - - BLIP_MD5 = "" - BLIP_URL = "https://www.facebook.com/video/video.php?v=207446242657384" - BLIP_FILE = "" - - VIMEO_MD5 = "" - VIMEO_URL = "https://www.facebook.com/video/video.php?v=207446242657384" - VIMEO_FILE = "" - - XVIDEO_MD5 = "" - XVIDEO_URL = "https://www.facebook.com/video/video.php?v=207446242657384" - XVIDEO_FILE = "" - - - def test_youtube(self): - #let's download a file from youtube - fd = FileDownloader({}) - fd.add_info_extractor(YoutubeIE()) - fd.download([DownloadTest.YOUTUBE_URL]) - self.assertTrue(os.path.exists(DownloadTest.YOUTUBE_FILE)) - md5_down_file = md5_for_file(DownloadTest.YOUTUBE_FILE) - self.assertEqual(md5_down_file, DownloadTest.YOUTUBE_MD5) - - def test_dailymotion(self): - fd = FileDownloader({}) - fd.add_info_extractor(DailymotionIE()) - fd.download([DownloadTest.DAILYMOTION_URL]) - self.assertTrue(os.path.exists(DownloadTest.DAILYMOTION_FILE)) - md5_down_file = md5_for_file(DownloadTest.DAILYMOTION_FILE) - self.assertEqual(md5_down_file, DownloadTest.DAILYMOTION_MD5) - - - def test_metacafe(self): - fd = FileDownloader({}) - fd.add_info_extractor(MetacafeIE()) - fd.download([DownloadTest.METACAFE_URL]) - self.assertTrue(os.path.exists(DownloadTest.METACAFE_FILE)) - md5_down_file = md5_for_file(DownloadTest.METACAFE_FILE) - self.assertEqual(md5_down_file, DownloadTest.METACAFE_MD5) - - def test_photobucket(self): - fd = FileDownloader({}) - fd.add_info_extractor(PhotobucketIE()) - fd.download([DownloadTest.PHOTOBUCKET_URL]) - self.assertTrue(os.path.exists(DownloadTest.PHOTOBUCKET_FILE)) - md5_down_file = md5_for_file(DownloadTest.PHOTOBUCKET_FILE) - self.assertEqual(md5_down_file, DownloadTest.PHOTOBUCKET_MD5) - - - def test_facebook(self): - fd = FileDownloader({}) - fd.add_info_extractor(FacebookIE()) - fd.download([DownloadTest.FACEBOOK_URL]) - self.assertTrue(os.path.exists(DownloadTest.FACEBOOK_FILE)) - md5_down_file = md5_for_file(DownloadTest.FACEBOOK_FILE) - self.assertEqual(md5_down_file, DownloadTest.FACEBOOK_MD5) - - def test_blip(self): - fd = FileDownloader({}) - fd.add_info_extractor(BlipTVIE()) - fd.download([DownloadTest.BLIP_URL]) - self.assertTrue(os.path.exists(DownloadTest.BLIP_FILE)) - md5_down_file = md5_for_file(DownloadTest.BLIP_FILE) - self.assertEqual(md5_down_file, DownloadTest.BLIP_MD5) - - - def test_vimeo(self): - fd = FileDownloader({}) - fd.add_info_extractor(VimeoIE()) - fd.download([DownloadTest.VIMEO_URL]) - self.assertTrue(os.path.exists(DownloadTest.VIMEO_FILE)) - md5_down_file = md5_for_file(DownloadTest.VIMEO_FILE) - self.assertEqual(md5_down_file, DownloadTest.VIMEO_MD5) - - - def test_xvideo(self): - fd = FileDownloader({}) - fd.add_info_extractor(XVideosIE()) - fd.download([DownloadTest.XVIDEO_URL]) - self.assertTrue(os.path.exists(DownloadTest.XVIDEO_FILE)) - md5_down_file = md5_for_file(DownloadTest.XVIDEO_FILE) - self.assertEqual(md5_down_file, DownloadTest.XVIDEO_MD5) - - def cleanUp(self): - if os.path.exists(DownloadTest.YOUTUBE_FILE): - os.remove(DownloadTest.YOUTUBE_FILE) - if os.path.exists(DownloadTest.DAILYMOTION_FILE): - os.remove(DownloadTest.DAILYMOTION_FILE) - if os.path.exists(DownloadTest.METACAFE_FILE): - os.remove(DownloadTest.METACAFE_FILE) - if os.path.exists(DownloadTest.PHOTOBUCKET_FILE): - os.remove(DownloadTest.PHOTOBUCKET_FILE) - if os.path.exists(DownloadTest.FACEBOOK_FILE): - os.remove(DownloadTest.FACEBOOK_FILE) - if os.path.exists(DownloadTest.BLIP_FILE): - os.remove(DownloadTest.BLIP_FILE) - if os.path.exists(DownloadTest.VIMEO_FILE): - os.remove(DownloadTest.VIMEO_FILE) - if os.path.exists(DownloadTest.XVIDEO_FILE): - os.remove(DownloadTest.XVIDEO_FILE) - -def md5_for_file(f, block_size=2**20): - md5 = hashlib.md5() - while True: - data = f.read(block_size) - if not data: - break - md5.update(data) - return md5.digest() +import hashlib +import io +import json +import socket + +import youtube_dl.YoutubeDL +from youtube_dl.utils import ( + compat_str, + compat_urllib_error, + compat_HTTPError, + DownloadError, + ExtractorError, + UnavailableVideoError, +) + +RETRIES = 3 + +class YoutubeDL(youtube_dl.YoutubeDL): + def __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() + +defs = get_testcases() + + +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']) + 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 + + 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.fd.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(tc_filename + '.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) 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) + self.assertTrue(os.path.exists(tc_filename + '.info.json')) + if 'md5' in tc: + md5_for_file = _file_md5(tc_filename) + self.assertEqual(md5_for_file, tc['md5']) + with io.open(tc_filename + '.info.json', 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=2) + 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]) + 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__': + unittest.main()