debian/NEWS: Write news about new behavior of youtube-dl.

[youtubedl] / test / test_download.py

diff --git a/test/test_download.py b/test/test_download.py
index 8d869847821877042a3a73cc4939f833f17cd37b..7587a18aa18fee1cdf61a735246c62dbf91e691d 100644 (file)
--- a/test/test_download.py
+++ b/test/test_download.py
@@ -1,71 +1,61 @@
 #!/usr/bin/env python
 
 #!/usr/bin/env python
 

-import errno
-import hashlib
-import io
+# Allow direct execution

 import os
 import os

-import json
-import unittest

 import sys
 import sys

-import socket
-import binascii
+import unittest
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

 
 

-# Allow direct execution
-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.extractor
-from youtube_dl.utils import *

 
 

-DEF_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'tests.json')
-PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)), "parameters.json")
+import hashlib
+import io
+import json
+import socket
+
+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
 
 
 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(10)
-
-def _try_rm(filename):
-    """ Remove a file if it exists """
-    try:
-        os.remove(filename)
-    except OSError as ose:
-        if ose.errno != errno.ENOENT:
-            raise
-
-md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
-
-class FileDownloader(youtube_dl.FileDownloader):
+class YoutubeDL(youtube_dl.YoutubeDL):

     def __init__(self, *args, **kwargs):
         self.to_stderr = self.to_screen
         self.processed_info_dicts = []
     def __init__(self, *args, **kwargs):
         self.to_stderr = self.to_screen
         self.processed_info_dicts = []

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

-with io.open(DEF_FILE, encoding='utf-8') as deff:
-    defs = json.load(deff)
-with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
-    parameters = json.load(pf)
+defs = get_testcases()

 
 
 class TestDownload(unittest.TestCase):
     maxDiff = None
     def setUp(self):
 
 
 class TestDownload(unittest.TestCase):
     maxDiff = None
     def setUp(self):

-        self.parameters = parameters

         self.defs = defs
 
 ### Dynamically generate tests
         self.defs = defs
 
 ### Dynamically generate tests


@@ -73,89 +63,114 @@ def generator(test_case):
 
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])
 
     def test_template(self):
         ie = youtube_dl.extractor.get_info_extractor(test_case['name'])

-        if not ie._WORKING:
-            print('Skipping: IE marked as not _WORKING')
-            return
-        if 'playlist' not in test_case and not test_case['file']:
-            print('Skipping: No output file specified')
+        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
             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:
         if 'skip' in test_case:

-            print('Skipping: {0}'.format(test_case['skip']))
+            print_skipping(test_case['skip'])

             return
             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 = self.parameters.copy()
-        params.update(test_case.get('params', {}))
+        params = get_params(test_case.get('params', {}))

 
 

-        fd = FileDownloader(params)
-        for ie in youtube_dl.extractor.gen_extractors():
-            fd.add_info_extractor(ie)
+        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'])
         finished_hook_called = set()
         def _hook(status):
             if status['status'] == 'finished':
                 finished_hook_called.add(status['filename'])

-        fd.add_progress_hook(_hook)
+        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])
 
         test_cases = test_case.get('playlist', [test_case])

-        for tc in test_cases:
-            _try_rm(tc['file'])
-            _try_rm(tc['file'] + '.part')
-            _try_rm(tc['file'] + '.info.json')
+        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:

-            for retry in range(1, RETRIES + 1):
+            try_num = 1
+            while True:

                 try:
                 try:

-                    fd.download([test_case['url']])
+                    ydl.download([test_case['url']])

                 except (DownloadError, ExtractorError) as err:
                 except (DownloadError, ExtractorError) as err:

-                    if retry == RETRIES: raise
-

                     # Check if the exception is not a network related one
                     # Check if the exception is not a network related one

-                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError):
+                    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
 
                         raise
 

-                    print('Retrying: {0} failed tries\n\n##########\n\n'.format(retry))
+                    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:
                 else:
                     break
 
             for tc in test_cases:

+                tc_filename = get_tc_filename(tc)

                 if not test_case.get('params', {}).get('skip_download', False):
                 if not test_case.get('params', {}).get('skip_download', False):

-                    self.assertTrue(os.path.exists(tc['file']), msg='Missing file ' + tc['file'])
-                    self.assertTrue(tc['file'] in finished_hook_called)
-                self.assertTrue(os.path.exists(tc['file'] + '.info.json'))
+                    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:
                 if 'md5' in tc:

-                    md5_for_file = _file_md5(tc['file'])
+                    md5_for_file = _file_md5(tc_filename)

                     self.assertEqual(md5_for_file, tc['md5'])
                     self.assertEqual(md5_for_file, tc['md5'])

-                with io.open(tc['file'] + '.info.json', encoding='utf-8') as infof:
+                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:'):
                     info_dict = json.load(infof)
                 for (info_field, expected) in tc.get('info_dict', {}).items():
                     if isinstance(expected, compat_str) and expected.startswith('md5:'):

-                        self.assertEqual(expected, 'md5:' + md5(info_dict.get(info_field)))
+                        got = 'md5:' + md5(info_dict.get(info_field))

                     else:
                         got = 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))
+                    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()):
 
                 # 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')
+                    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 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:
         finally:

-            for tc in test_cases:
-                _try_rm(tc['file'])
-                _try_rm(tc['file'] + '.part')
-                _try_rm(tc['file'] + '.info.json')
+            try_rm_tcs_files()

 
     return test_template
 
 ### And add them to TestDownload
 
     return test_template
 
 ### And add them to TestDownload

-for test_case in defs:
+for n, test_case in enumerate(defs):

     test_method = generator(test_case)
     test_method = generator(test_case)

-    test_method.__name__ = "test_{0}".format(test_case["name"])
+    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
 
     setattr(TestDownload, test_method.__name__, test_method)
     del test_method