-#!/usr/bin/env python2
+#!/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 test.helper import (
+ get_params,
+ get_testcases,
+ global_setup,
+ try_rm,
+ md5,
+ report_warning
+)
+global_setup()
+
+
import hashlib
-import os
+import io
import json
-
-from youtube_dl.FileDownloader import FileDownloader
-from youtube_dl.InfoExtractors import YoutubeIE, DailymotionIE
-from youtube_dl.InfoExtractors import MetacafeIE, BlipTVIE
-from youtube_dl.InfoExtractors import XVideosIE, VimeoIE
-from youtube_dl.InfoExtractors import SoundcloudIE, StanfordOpenClassroomIE
-from youtube_dl.InfoExtractors import CollegeHumorIE, XNXXIE
-
-
-class DownloadTest(unittest.TestCase):
- PARAMETERS_FILE = "test/parameters.json"
- #calculated with md5sum:
- #md5sum (GNU coreutils) 8.19
-
- YOUTUBE_SIZE = 1993883
- YOUTUBE_URL = "http://www.youtube.com/watch?v=BaW_jenozKc"
- YOUTUBE_FILE = "BaW_jenozKc.mp4"
-
- DAILYMOTION_MD5 = "d363a50e9eb4f22ce90d08d15695bb47"
- DAILYMOTION_URL = "http://www.dailymotion.com/video/x33vw9_tutoriel-de-youtubeur-dl-des-video_tech"
- DAILYMOTION_FILE = "x33vw9.mp4"
-
- METACAFE_SIZE = 5754305
- METACAFE_URL = "http://www.metacafe.com/watch/yt-_aUehQsCQtM/the_electric_company_short_i_pbs_kids_go/"
- METACAFE_FILE = "aUehQsCQtM.flv"
-
- BLIP_MD5 = "4962f94441605832eb1008eb820ef47a"
- BLIP_URL = "http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352"
- BLIP_FILE = "5779306.m4v"
-
- XVIDEO_MD5 = "aecab2ea59b7996110a7e409f0c55da3"
- XVIDEO_URL = "http://www.xvideos.com/video939581/funny_porns_by_s_-1"
- XVIDEO_FILE = "939581.flv"
-
- VIMEO_MD5 = "1ab4dedc01f771cb2a65e91caa801aaf"
- VIMEO_URL = "http://vimeo.com/14160053"
- VIMEO_FILE = ""
-
- VIMEO2_MD5 = ""
- VIMEO2_URL = "http://player.vimeo.com/video/47019590"
- VIMEO2_FILE = ""
-
- SOUNDCLOUD_MD5 = "c1b9b9ea8bfd620b96b2628664576e1c"
- SOUNDCLOUD_URL = "http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy"
- SOUNDCLOUD_FILE = "n6FLbx6ZzMiu.mp3"
-
- STANDFORD_MD5 = "8aac7873a07dcfaed66b1559ab128514"
- STANDFORD_URL = "http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100"
- STANDFORD_FILE = "PracticalUnix_intro-environment.mp4"
-
- COLLEGEHUMOR_MD5 = ""
- COLLEGEHUMOR_URL = "http://www.collegehumor.com/video/6830834/mitt-romney-style-gangnam-style-parody"
- COLLEGEHUMOR_FILE = ""
-
- XNXX_MD5 = "c5c67df477eb0d9b058200351448ba4c"
- XNXX_URL = "http://video.xnxx.com/video1135332/lida_naked_funny_actress_5_"
- XNXX_FILE = "1135332.flv"
-
- def test_youtube(self):
- #let's download a file from youtube
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(YoutubeIE())
- if not YoutubeIE._WORKING: return
- fd.download([DownloadTest.YOUTUBE_URL])
- self.assertTrue(os.path.exists(DownloadTest.YOUTUBE_FILE))
- self.assertEqual(os.path.getsize(DownloadTest.YOUTUBE_FILE), DownloadTest.YOUTUBE_SIZE)
-
- def test_dailymotion(self):
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(DailymotionIE())
- if not DailymotionIE._WORKING: return
- 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):
- #this emulate a skip,to be 2.6 compatible
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(MetacafeIE())
- if not MetacafeIE._WORKING: return
- fd.add_info_extractor(YoutubeIE())
- if not YoutubeIE._WORKING: return
- fd.download([DownloadTest.METACAFE_URL])
- self.assertTrue(os.path.exists(DownloadTest.METACAFE_FILE))
- self.assertEqual(os.path.getsize(DownloadTest.METACAFE_FILE), DownloadTest.METACAFE_SIZE)
-
- def test_blip(self):
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(BlipTVIE())
- if not BlipTVIE._WORKING: return
- 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_xvideo(self):
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(XVideosIE())
- if not XVideosIE._WORKING: return
- 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 test_vimeo(self):
- #skipped for the moment produce an error
- return
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- 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_vimeo2(self):
- #skipped for the moment produce an error
- return
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(VimeoIE())
- fd.download([DownloadTest.VIMEO2_URL])
- self.assertTrue(os.path.exists(DownloadTest.VIMEO2_FILE))
- md5_down_file = md5_for_file(DownloadTest.VIMEO2_FILE)
- self.assertEqual(md5_down_file, DownloadTest.VIMEO2_MD5)
-
- def test_soundcloud(self):
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(SoundcloudIE())
- if not SoundcloudIE._WORKING: return
- fd.download([DownloadTest.SOUNDCLOUD_URL])
- self.assertTrue(os.path.exists(DownloadTest.SOUNDCLOUD_FILE))
- md5_down_file = md5_for_file(DownloadTest.SOUNDCLOUD_FILE)
- self.assertEqual(md5_down_file, DownloadTest.SOUNDCLOUD_MD5)
-
- def test_standford(self):
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(StanfordOpenClassroomIE())
- if not StanfordOpenClassroomIE._WORKING: return
- fd.download([DownloadTest.STANDFORD_URL])
- self.assertTrue(os.path.exists(DownloadTest.STANDFORD_FILE))
- md5_down_file = md5_for_file(DownloadTest.STANDFORD_FILE)
- self.assertEqual(md5_down_file, DownloadTest.STANDFORD_MD5)
-
- def test_collegehumor(self):
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(CollegeHumorIE())
- if not CollegeHumorIE._WORKING: return
- fd.download([DownloadTest.COLLEGEHUMOR_URL])
- self.assertTrue(os.path.exists(DownloadTest.COLLEGEHUMOR_FILE))
- md5_down_file = md5_for_file(DownloadTest.COLLEGEHUMOR_FILE)
- self.assertEqual(md5_down_file, DownloadTest.COLLEGEHUMOR_MD5)
-
- def test_xnxx(self):
- with open(DownloadTest.PARAMETERS_FILE) as f:
- fd = FileDownloader(json.load(f))
- fd.add_info_extractor(XNXXIE())
- if not XNXXIE._WORKING: return
- fd.download([DownloadTest.XNXX_URL])
- self.assertTrue(os.path.exists(DownloadTest.XNXX_FILE))
- md5_down_file = md5_for_file(DownloadTest.XNXX_FILE)
- self.assertEqual(md5_down_file, DownloadTest.XNXX_MD5)
-
- def tearDown(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.BLIP_FILE):
- os.remove(DownloadTest.BLIP_FILE)
- if os.path.exists(DownloadTest.XVIDEO_FILE):
- os.remove(DownloadTest.XVIDEO_FILE)
- if os.path.exists(DownloadTest.VIMEO_FILE):
- os.remove(DownloadTest.VIMEO_FILE)
- if os.path.exists(DownloadTest.SOUNDCLOUD_FILE):
- os.remove(DownloadTest.SOUNDCLOUD_FILE)
- if os.path.exists(DownloadTest.STANDFORD_FILE):
- os.remove(DownloadTest.STANDFORD_FILE)
- if os.path.exists(DownloadTest.COLLEGEHUMOR_FILE):
- os.remove(DownloadTest.COLLEGEHUMOR_FILE)
- if os.path.exists(DownloadTest.XNXX_FILE):
- os.remove(DownloadTest.XNXX_FILE)
-
-def md5_for_file(filename, block_size=2**20):
- with open(filename) as f:
- md5 = hashlib.md5()
- while True:
- data = f.read(block_size)
- if not data:
- break
- md5.update(data)
- return md5.hexdigest()
+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])
+ # 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__':
+ unittest.main()
projects / youtube-dl / blobdiff