-from __future__ import unicode_literals
-
import errno
-import io
import hashlib
import json
import os.path
import re
-import types
import ssl
import sys
+import types
import yt_dlp.extractor
from yt_dlp import YoutubeDL
-from yt_dlp.compat import (
- compat_os_name,
- compat_str,
-)
-from yt_dlp.utils import (
- preferredencoding,
- write_string,
-)
-
+from yt_dlp.compat import compat_os_name
+from yt_dlp.utils import preferredencoding, try_call, write_string
if 'pytest' in sys.modules:
import pytest
'parameters.json')
LOCAL_PARAMETERS_FILE = os.path.join(os.path.dirname(os.path.abspath(__file__)),
'local_parameters.json')
- with io.open(PARAMETERS_FILE, encoding='utf-8') as pf:
+ with open(PARAMETERS_FILE, encoding='utf-8') as pf:
parameters = json.load(pf)
if os.path.exists(LOCAL_PARAMETERS_FILE):
- with io.open(LOCAL_PARAMETERS_FILE, encoding='utf-8') as pf:
+ with open(LOCAL_PARAMETERS_FILE, encoding='utf-8') as pf:
parameters.update(json.load(pf))
if override:
parameters.update(override)
raise
-def report_warning(message):
+def report_warning(message, *args, **kwargs):
'''
Print the message to stderr, it will be prefixed with 'WARNING:'
If stderr is a tty file the 'WARNING:' will be colored
_msg_header = '\033[0;33mWARNING:\033[0m'
else:
_msg_header = 'WARNING:'
- output = '%s %s\n' % (_msg_header, message)
- if 'b' in getattr(sys.stderr, 'mode', '') or sys.version_info[0] < 3:
+ output = f'{_msg_header} {message}\n'
+ if 'b' in getattr(sys.stderr, 'mode', ''):
output = output.encode(preferredencoding())
sys.stderr.write(output)
# Different instances of the downloader can't share the same dictionary
# some test set the "sublang" parameter, which would break the md5 checks.
params = get_params(override=override)
- super(FakeYDL, self).__init__(params, auto_init=False)
+ super().__init__(params, auto_init=False)
self.result = []
- def to_screen(self, s, skip_eol=None):
+ def to_screen(self, s, *args, **kwargs):
print(s)
- def trouble(self, s, tb=None):
+ def trouble(self, s, *args, **kwargs):
raise Exception(s)
def download(self, x):
# Silence an expected warning matching a regex
old_report_warning = self.report_warning
- def report_warning(self, message):
+ def report_warning(self, message, *args, **kwargs):
if re.match(regex, message):
return
- old_report_warning(message)
+ old_report_warning(message, *args, **kwargs)
self.report_warning = types.MethodType(report_warning, self)
def gettestcases(include_onlymatching=False):
for ie in yt_dlp.extractor.gen_extractors():
- for tc in ie.get_testcases(include_onlymatching):
+ yield from ie.get_testcases(include_onlymatching)
+
+
+def getwebpagetestcases():
+ for ie in yt_dlp.extractor.gen_extractors():
+ for tc in ie.get_webpage_testcases():
+ tc.setdefault('add_ie', []).append('Generic')
yield tc
-md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
+md5 = lambda s: hashlib.md5(s.encode()).hexdigest()
def expect_value(self, got, expected, field):
- if isinstance(expected, compat_str) and expected.startswith('re:'):
+ if isinstance(expected, str) and expected.startswith('re:'):
match_str = expected[len('re:'):]
match_rex = re.compile(match_str)
self.assertTrue(
- isinstance(got, compat_str),
- 'Expected a %s object, but got %s for field %s' % (
- compat_str.__name__, type(got).__name__, field))
+ isinstance(got, str),
+ f'Expected a {str.__name__} object, but got {type(got).__name__} for field {field}')
self.assertTrue(
match_rex.match(got),
- 'field %s (value: %r) should match %r' % (field, got, match_str))
- elif isinstance(expected, compat_str) and expected.startswith('startswith:'):
+ f'field {field} (value: {got!r}) should match {match_str!r}')
+ elif isinstance(expected, str) and expected.startswith('startswith:'):
start_str = expected[len('startswith:'):]
self.assertTrue(
- isinstance(got, compat_str),
- 'Expected a %s object, but got %s for field %s' % (
- compat_str.__name__, type(got).__name__, field))
+ isinstance(got, str),
+ f'Expected a {str.__name__} object, but got {type(got).__name__} for field {field}')
self.assertTrue(
got.startswith(start_str),
- 'field %s (value: %r) should start with %r' % (field, got, start_str))
- elif isinstance(expected, compat_str) and expected.startswith('contains:'):
+ f'field {field} (value: {got!r}) should start with {start_str!r}')
+ elif isinstance(expected, str) and expected.startswith('contains:'):
contains_str = expected[len('contains:'):]
self.assertTrue(
- isinstance(got, compat_str),
- 'Expected a %s object, but got %s for field %s' % (
- compat_str.__name__, type(got).__name__, field))
+ isinstance(got, str),
+ f'Expected a {str.__name__} object, but got {type(got).__name__} for field {field}')
self.assertTrue(
contains_str in got,
- 'field %s (value: %r) should contain %r' % (field, got, contains_str))
+ f'field {field} (value: {got!r}) should contain {contains_str!r}')
elif isinstance(expected, type):
self.assertTrue(
isinstance(got, expected),
- 'Expected type %r for field %s, but got value %r of type %r' % (expected, field, got, type(got)))
+ f'Expected type {expected!r} for field {field}, but got value {got!r} of type {type(got)!r}')
elif isinstance(expected, dict) and isinstance(got, dict):
expect_dict(self, got, expected)
elif isinstance(expected, list) and isinstance(got, list):
index, field, type_expected, type_got))
expect_value(self, item_got, item_expected, field)
else:
- if isinstance(expected, compat_str) and expected.startswith('md5:'):
+ if isinstance(expected, str) and expected.startswith('md5:'):
self.assertTrue(
- isinstance(got, compat_str),
- 'Expected field %s to be a unicode object, but got value %r of type %r' % (field, got, type(got)))
+ isinstance(got, str),
+ f'Expected field {field} to be a unicode object, but got value {got!r} of type {type(got)!r}')
got = 'md5:' + md5(got)
- elif isinstance(expected, compat_str) and re.match(r'^(?:min|max)?count:\d+', expected):
+ elif isinstance(expected, str) and re.match(r'^(?:min|max)?count:\d+', expected):
self.assertTrue(
isinstance(got, (list, dict)),
- 'Expected field %s to be a list or a dict, but it is of type %s' % (
- field, type(got).__name__))
+ f'Expected field {field} to be a list or a dict, but it is of type {type(got).__name__}')
op, _, expected_num = expected.partition(':')
expected_num = int(expected_num)
if op == 'mincount':
return
self.assertEqual(
expected, got,
- 'Invalid value for field %s, expected %r, got %r' % (field, expected, got))
+ f'Invalid value for field {field}, expected {expected!r}, got {got!r}')
def expect_dict(self, got_dict, expected_dict):
expect_value(self, got, expected, info_field)
+def sanitize_got_info_dict(got_dict):
+ IGNORED_FIELDS = (
+ *YoutubeDL._format_fields,
+
+ # Lists
+ 'formats', 'thumbnails', 'subtitles', 'automatic_captions', 'comments', 'entries',
+
+ # Auto-generated
+ 'autonumber', 'playlist', 'format_index', 'video_ext', 'audio_ext', 'duration_string', 'epoch', 'n_entries',
+ 'fulltitle', 'extractor', 'extractor_key', 'filename', 'filepath', 'infojson_filename', 'original_url',
+
+ # Only live_status needs to be checked
+ 'is_live', 'was_live',
+ )
+
+ IGNORED_PREFIXES = ('', 'playlist', 'requested', 'webpage')
+
+ def sanitize(key, value):
+ if isinstance(value, str) and len(value) > 100 and key != 'thumbnail':
+ return f'md5:{md5(value)}'
+ elif isinstance(value, list) and len(value) > 10:
+ return f'count:{len(value)}'
+ elif key.endswith('_count') and isinstance(value, int):
+ return int
+ return value
+
+ test_info_dict = {
+ key: sanitize(key, value) for key, value in got_dict.items()
+ if value is not None and key not in IGNORED_FIELDS and (
+ not any(key.startswith(f'{prefix}_') for prefix in IGNORED_PREFIXES)
+ or key == '_old_archive_ids')
+ }
+
+ # display_id may be generated from id
+ if test_info_dict.get('display_id') == test_info_dict.get('id'):
+ test_info_dict.pop('display_id')
+
+ # release_year may be generated from release_date
+ if try_call(lambda: test_info_dict['release_year'] == int(test_info_dict['release_date'][:4])):
+ test_info_dict.pop('release_year')
+
+ # Check url for flat entries
+ if got_dict.get('_type', 'video') != 'video' and got_dict.get('url'):
+ test_info_dict['url'] = got_dict['url']
+
+ return test_info_dict
+
+
def expect_info_dict(self, got_dict, expected_dict):
expect_dict(self, got_dict, expected_dict)
# Check for the presence of mandatory fields
for key in mandatory_fields:
self.assertTrue(got_dict.get(key), 'Missing mandatory field %s' % key)
# Check for mandatory fields that are automatically set by YoutubeDL
- for key in ['webpage_url', 'extractor', 'extractor_key']:
- self.assertTrue(got_dict.get(key), 'Missing field: %s' % key)
+ if got_dict.get('_type', 'video') == 'video':
+ for key in ['webpage_url', 'extractor', 'extractor_key']:
+ self.assertTrue(got_dict.get(key), 'Missing field: %s' % key)
+
+ test_info_dict = sanitize_got_info_dict(got_dict)
- # Are checkable fields missing from the test case definition?
- test_info_dict = dict((key, value if not isinstance(value, compat_str) or len(value) < 250 else 'md5:' + md5(value))
- for key, value in got_dict.items()
- if value and key in ('id', 'title', 'description', 'uploader', 'upload_date', 'timestamp', 'uploader_id', 'location', 'age_limit'))
missing_keys = set(test_info_dict.keys()) - set(expected_dict.keys())
if missing_keys:
def _repr(v):
- if isinstance(v, compat_str):
+ if isinstance(v, str):
return "'%s'" % v.replace('\\', '\\\\').replace("'", "\\'").replace('\n', '\\n')
+ elif isinstance(v, type):
+ return v.__name__
else:
return repr(v)
- info_dict_str = ''
- if len(missing_keys) != len(expected_dict):
- info_dict_str += ''.join(
- ' %s: %s,\n' % (_repr(k), _repr(v))
- for k, v in test_info_dict.items() if k not in missing_keys)
-
- if info_dict_str:
- info_dict_str += '\n'
+ info_dict_str = ''.join(
+ f' {_repr(k)}: {_repr(v)},\n'
+ for k, v in test_info_dict.items() if k not in missing_keys)
+ if info_dict_str:
+ info_dict_str += '\n'
info_dict_str += ''.join(
- ' %s: %s,\n' % (_repr(k), _repr(test_info_dict[k]))
+ f' {_repr(k)}: {_repr(test_info_dict[k])},\n'
for k in missing_keys)
- write_string(
- '\n\'info_dict\': {\n' + info_dict_str + '},\n', out=sys.stderr)
+ info_dict_str = '\n\'info_dict\': {\n' + info_dict_str + '},\n'
+ write_string(info_dict_str.replace('\n', '\n '), out=sys.stderr)
self.assertFalse(
missing_keys,
'Missing keys in test definition: %s' % (
def assertGreaterEqual(self, got, expected, msg=None):
if not (got >= expected):
if msg is None:
- msg = '%r not greater than or equal to %r' % (got, expected)
+ msg = f'{got!r} not greater than or equal to {expected!r}'
self.assertTrue(got >= expected, msg)
def assertLessEqual(self, got, expected, msg=None):
if not (got <= expected):
if msg is None:
- msg = '%r not less than or equal to %r' % (got, expected)
+ msg = f'{got!r} not less than or equal to {expected!r}'
self.assertTrue(got <= expected, msg)
def assertEqual(self, got, expected, msg=None):
if not (got == expected):
if msg is None:
- msg = '%r not equal to %r' % (got, expected)
+ msg = f'{got!r} not equal to {expected!r}'
self.assertTrue(got == expected, msg)
def expect_warnings(ydl, warnings_re):
real_warning = ydl.report_warning
- def _report_warning(w):
+ def _report_warning(w, *args, **kwargs):
if not any(re.search(w_re, w) for w_re in warnings_re):
- real_warning(w)
+ real_warning(w, *args, **kwargs)
ydl.report_warning = _report_warning