debian/patches: Add patch from upstream to fix extraction from youtube.

[youtubedl] / test / test_download.py

diff --git a/test/test_download.py b/test/test_download.py
index f45c500c8b8348acfc2f0df1c38664a52deea68a..4639529897967ebc49883e488f5624a038c70c44 100644 (file)
--- a/test/test_download.py
+++ b/test/test_download.py
@@ -1,197 +1,235 @@
 #!/usr/bin/env python
 
 #!/usr/bin/env python
 

-# DO NOT EDIT THIS FILE BY HAND!
-# It is auto-generated from tests.json and gentests.py.
+from __future__ import unicode_literals
+
+# 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 (
+    assertGreaterEqual,
+    expect_warnings,
+    get_params,
+    gettestcases,
+    expect_info_dict,
+    try_rm,
+    report_warning,
+)
+

 
 import hashlib
 import io
 
 import hashlib
 import io

-import os

 import json
 import json

-import unittest
-import sys

 import socket
 
 import socket
 

-# Allow direct execution
-import os
-sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+import youtube_dl.YoutubeDL
+from youtube_dl.compat import (
+    compat_http_client,
+    compat_urllib_error,
+    compat_HTTPError,
+)
+from youtube_dl.utils import (
+    DownloadError,
+    ExtractorError,
+    format_bytes,
+    UnavailableVideoError,
+)
+from youtube_dl.extractor import get_info_extractor

 
 

-import youtube_dl.FileDownloader
-import youtube_dl.InfoExtractors
-from youtube_dl.utils import *
+RETRIES = 3

 
 

-# 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)

 
 

-class FileDownloader(youtube_dl.FileDownloader):
+class YoutubeDL(youtube_dl.YoutubeDL):

     def __init__(self, *args, **kwargs):
     def __init__(self, *args, **kwargs):

-        youtube_dl.FileDownloader.__init__(self, *args, **kwargs)

         self.to_stderr = self.to_screen
         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()
 
 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")

 
 

-    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')
+defs = gettestcases()
+
+
+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', [])]
+        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 (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('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', 'in_playlist')
+            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 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'],
+                        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):
+                        raise
+
+                    if try_num == RETRIES:
+                        report_warning('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.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)
+
+            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, 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)
+
+    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_%s_%d' % (test_case['name'], i)
+        i += 1
+    test_method.__name__ = str(tname)
+    setattr(TestDownload, test_method.__name__, test_method)
+    del test_method

 
 
 if __name__ == '__main__':
 
 
 if __name__ == '__main__':