Update upstream source from tag 'upstream/2019.09.01'

[youtubedl] / test / test_utils.py

diff --git a/test/test_utils.py b/test/test_utils.py
index e6887be9f29847984be257a73af67e5750f4225c..659c6ece53d480a5d709366ef44e29eedbec3027 100644 (file)
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -18,7 +18,9 @@ import xml.etree.ElementTree
 from youtube_dl.utils import (
     age_restricted,
     args_to_str,
 from youtube_dl.utils import (
     age_restricted,
     args_to_str,

+    encode_base_n,

     clean_html,
     clean_html,

+    date_from_str,

     DateRange,
     detect_exe_version,
     determine_ext,
     DateRange,
     detect_exe_version,
     determine_ext,


@@ -27,39 +29,65 @@ from youtube_dl.utils import (
     encodeFilename,
     escape_rfc3986,
     escape_url,
     encodeFilename,
     escape_rfc3986,
     escape_url,

+    extract_attributes,

     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,

+    float_or_none,
+    get_element_by_class,
+    get_element_by_attribute,
+    get_elements_by_class,
+    get_elements_by_attribute,

     InAdvancePagedList,
     InAdvancePagedList,

+    int_or_none,

     intlist_to_bytes,
     is_html,
     js_to_json,
     limit_length,
     intlist_to_bytes,
     is_html,
     js_to_json,
     limit_length,

+    merge_dicts,
+    mimetype2ext,
+    month_by_name,
+    multipart_encode,

     ohdave_rsa_encrypt,
     OnDemandPagedList,
     orderedSet,
     ohdave_rsa_encrypt,
     OnDemandPagedList,
     orderedSet,

+    parse_age_limit,

     parse_duration,
     parse_filesize,
     parse_duration,
     parse_filesize,

+    parse_count,

     parse_iso8601,
     parse_iso8601,

+    parse_resolution,
+    parse_bitrate,
+    pkcs1pad,

     read_batch_urls,
     sanitize_filename,
     sanitize_path,
     read_batch_urls,
     sanitize_filename,
     sanitize_path,

+    sanitize_url,
+    expand_path,

     prepend_extension,
     replace_extension,
     prepend_extension,
     replace_extension,

+    remove_start,
+    remove_end,

     remove_quotes,
     shell_quote,
     smuggle_url,
     str_to_int,
     strip_jsonp,
     remove_quotes,
     shell_quote,
     smuggle_url,
     str_to_int,
     strip_jsonp,

-    struct_unpack,
+    strip_or_none,

     timeconvert,
     unescapeHTML,
     unified_strdate,
     timeconvert,
     unescapeHTML,
     unified_strdate,

+    unified_timestamp,

     unsmuggle_url,
     uppercase_escape,
     lowercase_escape,
     url_basename,
     unsmuggle_url,
     uppercase_escape,
     lowercase_escape,
     url_basename,

+    url_or_none,
+    base_url,
+    urljoin,

     urlencode_postdata,
     urlencode_postdata,

+    urshift,
+    update_url_query,

     version_tuple,
     xpath_with_ns,
     xpath_element,
     version_tuple,
     xpath_with_ns,
     xpath_element,


@@ -72,9 +100,16 @@ from youtube_dl.utils import (
     cli_option,
     cli_valueless_option,
     cli_bool_option,
     cli_option,
     cli_valueless_option,
     cli_bool_option,

+    parse_codecs,

 )
 from youtube_dl.compat import (
 )
 from youtube_dl.compat import (

+    compat_chr,

     compat_etree_fromstring,
     compat_etree_fromstring,

+    compat_getenv,
+    compat_os_name,
+    compat_setenv,
+    compat_urlparse,
+    compat_parse_qs,

 )
 
 
 )
 
 


@@ -131,8 +166,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual('yes_no', sanitize_filename('yes? no', restricted=True))
         self.assertEqual('this_-_that', sanitize_filename('this: that', restricted=True))
 
         self.assertEqual('yes_no', sanitize_filename('yes? no', restricted=True))
         self.assertEqual('this_-_that', sanitize_filename('this: that', restricted=True))
 

-        tests = 'a\xe4b\u4e2d\u56fd\u7684c'
-        self.assertEqual(sanitize_filename(tests, restricted=True), 'a_b_c')
+        tests = 'aäb\u4e2d\u56fd\u7684c'
+        self.assertEqual(sanitize_filename(tests, restricted=True), 'aab_c')

         self.assertTrue(sanitize_filename('\xf6', restricted=True) != '')  # No empty filename
 
         forbidden = '"\0\\/&!: \'\t\n()[]{}$;`^,#'
         self.assertTrue(sanitize_filename('\xf6', restricted=True) != '')  # No empty filename
 
         forbidden = '"\0\\/&!: \'\t\n()[]{}$;`^,#'


@@ -147,6 +182,10 @@ class TestUtil(unittest.TestCase):
         self.assertTrue(sanitize_filename('-', restricted=True) != '')
         self.assertTrue(sanitize_filename(':', restricted=True) != '')
 
         self.assertTrue(sanitize_filename('-', restricted=True) != '')
         self.assertTrue(sanitize_filename(':', restricted=True) != '')
 

+        self.assertEqual(sanitize_filename(
+            'ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', restricted=True),
+            'AAAAAAAECEEEEIIIIDNOOOOOOOOEUUUUUYTHssaaaaaaaeceeeeiiiionooooooooeuuuuuythy')
+

     def test_sanitize_ids(self):
         self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw')
         self.assertEqual(sanitize_filename('_BD_eEpuzXw', is_id=True), '_BD_eEpuzXw')
     def test_sanitize_ids(self):
         self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw')
         self.assertEqual(sanitize_filename('_BD_eEpuzXw', is_id=True), '_BD_eEpuzXw')


@@ -188,6 +227,24 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(sanitize_path('./abc'), 'abc')
         self.assertEqual(sanitize_path('./../abc'), '..\\abc')
 
         self.assertEqual(sanitize_path('./abc'), 'abc')
         self.assertEqual(sanitize_path('./../abc'), '..\\abc')
 

+    def test_sanitize_url(self):
+        self.assertEqual(sanitize_url('//foo.bar'), 'http://foo.bar')
+        self.assertEqual(sanitize_url('httpss://foo.bar'), 'https://foo.bar')
+        self.assertEqual(sanitize_url('rmtps://foo.bar'), 'rtmps://foo.bar')
+        self.assertEqual(sanitize_url('https://foo.bar'), 'https://foo.bar')
+
+    def test_expand_path(self):
+        def env(var):
+            return '%{0}%'.format(var) if sys.platform == 'win32' else '${0}'.format(var)
+
+        compat_setenv('YOUTUBE_DL_EXPATH_PATH', 'expanded')
+        self.assertEqual(expand_path(env('YOUTUBE_DL_EXPATH_PATH')), 'expanded')
+        self.assertEqual(expand_path(env('HOME')), compat_getenv('HOME'))
+        self.assertEqual(expand_path('~'), compat_getenv('HOME'))
+        self.assertEqual(
+            expand_path('~/%s' % env('YOUTUBE_DL_EXPATH_PATH')),
+            '%s/expanded' % compat_getenv('HOME'))
+

     def test_prepend_extension(self):
         self.assertEqual(prepend_extension('abc.ext', 'temp'), 'abc.temp.ext')
         self.assertEqual(prepend_extension('abc.ext', 'temp', 'ext'), 'abc.temp.ext')
     def test_prepend_extension(self):
         self.assertEqual(prepend_extension('abc.ext', 'temp'), 'abc.temp.ext')
         self.assertEqual(prepend_extension('abc.ext', 'temp', 'ext'), 'abc.temp.ext')


@@ -204,6 +261,16 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(replace_extension('.abc', 'temp'), '.abc.temp')
         self.assertEqual(replace_extension('.abc.ext', 'temp'), '.abc.temp')
 
         self.assertEqual(replace_extension('.abc', 'temp'), '.abc.temp')
         self.assertEqual(replace_extension('.abc.ext', 'temp'), '.abc.temp')
 

+    def test_remove_start(self):
+        self.assertEqual(remove_start(None, 'A - '), None)
+        self.assertEqual(remove_start('A - B', 'A - '), 'B')
+        self.assertEqual(remove_start('B - A', 'A - '), 'B - A')
+
+    def test_remove_end(self):
+        self.assertEqual(remove_end(None, ' - B'), None)
+        self.assertEqual(remove_end('A - B', ' - B'), 'A')
+        self.assertEqual(remove_end('B - A', ' - B'), 'B - A')
+

     def test_remove_quotes(self):
         self.assertEqual(remove_quotes(None), None)
         self.assertEqual(remove_quotes('"'), '"')
     def test_remove_quotes(self):
         self.assertEqual(remove_quotes(None), None)
         self.assertEqual(remove_quotes('"'), '"')


@@ -226,6 +293,16 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unescapeHTML('/'), '/')
         self.assertEqual(unescapeHTML('é'), 'é')
         self.assertEqual(unescapeHTML('�'), '�')
         self.assertEqual(unescapeHTML('/'), '/')
         self.assertEqual(unescapeHTML('é'), 'é')
         self.assertEqual(unescapeHTML('�'), '�')

+        self.assertEqual(unescapeHTML('&a"'), '&a"')
+        # HTML5 entities
+        self.assertEqual(unescapeHTML('.''), '.\'')
+
+    def test_date_from_str(self):
+        self.assertEqual(date_from_str('yesterday'), date_from_str('now-1day'))
+        self.assertEqual(date_from_str('now+7day'), date_from_str('now+1week'))
+        self.assertEqual(date_from_str('now+14day'), date_from_str('now+2week'))
+        self.assertEqual(date_from_str('now+365day'), date_from_str('now+1year'))
+        self.assertEqual(date_from_str('now+30day'), date_from_str('now+1month'))

 
     def test_daterange(self):
         _20century = DateRange("19000101", "20000101")
 
     def test_daterange(self):
         _20century = DateRange("19000101", "20000101")


@@ -249,8 +326,39 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(
             unified_strdate('2/2/2015 6:47:40 PM', day_first=False),
             '20150202')
         self.assertEqual(
             unified_strdate('2/2/2015 6:47:40 PM', day_first=False),
             '20150202')

+        self.assertEqual(unified_strdate('Feb 14th 2016 5:45PM'), '20160214')

         self.assertEqual(unified_strdate('25-09-2014'), '20140925')
         self.assertEqual(unified_strdate('25-09-2014'), '20140925')

+        self.assertEqual(unified_strdate('27.02.2016 17:30'), '20160227')

         self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None)
         self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None)

+        self.assertEqual(unified_strdate('Feb 7, 2016 at 6:35 pm'), '20160207')
+        self.assertEqual(unified_strdate('July 15th, 2013'), '20130715')
+        self.assertEqual(unified_strdate('September 1st, 2013'), '20130901')
+        self.assertEqual(unified_strdate('Sep 2nd, 2013'), '20130902')
+
+    def test_unified_timestamps(self):
+        self.assertEqual(unified_timestamp('December 21, 2010'), 1292889600)
+        self.assertEqual(unified_timestamp('8/7/2009'), 1247011200)
+        self.assertEqual(unified_timestamp('Dec 14, 2012'), 1355443200)
+        self.assertEqual(unified_timestamp('2012/10/11 01:56:38 +0000'), 1349920598)
+        self.assertEqual(unified_timestamp('1968 12 10'), -33436800)
+        self.assertEqual(unified_timestamp('1968-12-10'), -33436800)
+        self.assertEqual(unified_timestamp('28/01/2014 21:00:00 +0100'), 1390939200)
+        self.assertEqual(
+            unified_timestamp('11/26/2014 11:30:00 AM PST', day_first=False),
+            1417001400)
+        self.assertEqual(
+            unified_timestamp('2/2/2015 6:47:40 PM', day_first=False),
+            1422902860)
+        self.assertEqual(unified_timestamp('Feb 14th 2016 5:45PM'), 1455471900)
+        self.assertEqual(unified_timestamp('25-09-2014'), 1411603200)
+        self.assertEqual(unified_timestamp('27.02.2016 17:30'), 1456594200)
+        self.assertEqual(unified_timestamp('UNKNOWN DATE FORMAT'), None)
+        self.assertEqual(unified_timestamp('May 16, 2016 11:15 PM'), 1463440500)
+        self.assertEqual(unified_timestamp('Feb 7, 2016 at 6:35 pm'), 1454870100)
+        self.assertEqual(unified_timestamp('2017-03-30T17:52:41Q'), 1490896361)
+        self.assertEqual(unified_timestamp('Sep 11, 2013 | 5:49 AM'), 1378878540)
+        self.assertEqual(unified_timestamp('December 15, 2017 at 7:49 am'), 1513324140)
+        self.assertEqual(unified_timestamp('2018-03-14T08:32:43.1493874+00:00'), 1521016363)

 
     def test_determine_ext(self):
         self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4')
 
     def test_determine_ext(self):
         self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4')


@@ -258,6 +366,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(determine_ext('http://example.com/foo/bar.nonext/?download', None), None)
         self.assertEqual(determine_ext('http://example.com/foo/bar/mp4?download', None), None)
         self.assertEqual(determine_ext('http://example.com/foo/bar.m3u8//?download'), 'm3u8')
         self.assertEqual(determine_ext('http://example.com/foo/bar.nonext/?download', None), None)
         self.assertEqual(determine_ext('http://example.com/foo/bar/mp4?download', None), None)
         self.assertEqual(determine_ext('http://example.com/foo/bar.m3u8//?download'), 'm3u8')

+        self.assertEqual(determine_ext('foobar', None), None)

 
     def test_find_xpath_attr(self):
         testxml = '''<root>
 
     def test_find_xpath_attr(self):
         testxml = '''<root>


@@ -350,9 +459,32 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(res_url, url)
         self.assertEqual(res_data, None)
 
         self.assertEqual(res_url, url)
         self.assertEqual(res_data, None)
 

+        smug_url = smuggle_url(url, {'a': 'b'})
+        smug_smug_url = smuggle_url(smug_url, {'c': 'd'})
+        res_url, res_data = unsmuggle_url(smug_smug_url)
+        self.assertEqual(res_url, url)
+        self.assertEqual(res_data, {'a': 'b', 'c': 'd'})
+

     def test_shell_quote(self):
         args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')]
     def test_shell_quote(self):
         args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')]

-        self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""")
+        self.assertEqual(
+            shell_quote(args),
+            """ffmpeg -i 'ñ€ß'"'"'.mp4'""" if compat_os_name != 'nt' else '''ffmpeg -i "ñ€ß'.mp4"''')
+
+    def test_float_or_none(self):
+        self.assertEqual(float_or_none('42.42'), 42.42)
+        self.assertEqual(float_or_none('42'), 42.0)
+        self.assertEqual(float_or_none(''), None)
+        self.assertEqual(float_or_none(None), None)
+        self.assertEqual(float_or_none([]), None)
+        self.assertEqual(float_or_none(set()), None)
+
+    def test_int_or_none(self):
+        self.assertEqual(int_or_none('42'), 42)
+        self.assertEqual(int_or_none(''), None)
+        self.assertEqual(int_or_none(None), None)
+        self.assertEqual(int_or_none([]), None)
+        self.assertEqual(int_or_none(set()), None)

 
     def test_str_to_int(self):
         self.assertEqual(str_to_int('123,456'), 123456)
 
     def test_str_to_int(self):
         self.assertEqual(str_to_int('123,456'), 123456)


@@ -368,6 +500,61 @@ class TestUtil(unittest.TestCase):
             url_basename('http://media.w3.org/2010/05/sintel/trailer.mp4'),
             'trailer.mp4')
 
             url_basename('http://media.w3.org/2010/05/sintel/trailer.mp4'),
             'trailer.mp4')
 

+    def test_base_url(self):
+        self.assertEqual(base_url('http://foo.de/'), 'http://foo.de/')
+        self.assertEqual(base_url('http://foo.de/bar'), 'http://foo.de/')
+        self.assertEqual(base_url('http://foo.de/bar/'), 'http://foo.de/bar/')
+        self.assertEqual(base_url('http://foo.de/bar/baz'), 'http://foo.de/bar/')
+        self.assertEqual(base_url('http://foo.de/bar/baz?x=z/x/c'), 'http://foo.de/bar/')
+
+    def test_urljoin(self):
+        self.assertEqual(urljoin('http://foo.de/', '/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin(b'http://foo.de/', '/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de/', b'/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin(b'http://foo.de/', b'/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('//foo.de/', '/a/b/c.txt'), '//foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de/', 'a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de', '/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de', 'a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de/', 'http://foo.de/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de/', '//foo.de/a/b/c.txt'), '//foo.de/a/b/c.txt')
+        self.assertEqual(urljoin(None, 'http://foo.de/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin(None, '//foo.de/a/b/c.txt'), '//foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('', 'http://foo.de/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin(['foobar'], 'http://foo.de/a/b/c.txt'), 'http://foo.de/a/b/c.txt')
+        self.assertEqual(urljoin('http://foo.de/', None), None)
+        self.assertEqual(urljoin('http://foo.de/', ''), None)
+        self.assertEqual(urljoin('http://foo.de/', ['foobar']), None)
+        self.assertEqual(urljoin('http://foo.de/a/b/c.txt', '.././../d.txt'), 'http://foo.de/d.txt')
+        self.assertEqual(urljoin('http://foo.de/a/b/c.txt', 'rtmp://foo.de'), 'rtmp://foo.de')
+        self.assertEqual(urljoin(None, 'rtmp://foo.de'), 'rtmp://foo.de')
+
+    def test_url_or_none(self):
+        self.assertEqual(url_or_none(None), None)
+        self.assertEqual(url_or_none(''), None)
+        self.assertEqual(url_or_none('foo'), None)
+        self.assertEqual(url_or_none('http://foo.de'), 'http://foo.de')
+        self.assertEqual(url_or_none('https://foo.de'), 'https://foo.de')
+        self.assertEqual(url_or_none('http$://foo.de'), None)
+        self.assertEqual(url_or_none('http://foo.de'), 'http://foo.de')
+        self.assertEqual(url_or_none('//foo.de'), '//foo.de')
+
+    def test_parse_age_limit(self):
+        self.assertEqual(parse_age_limit(None), None)
+        self.assertEqual(parse_age_limit(False), None)
+        self.assertEqual(parse_age_limit('invalid'), None)
+        self.assertEqual(parse_age_limit(0), 0)
+        self.assertEqual(parse_age_limit(18), 18)
+        self.assertEqual(parse_age_limit(21), 21)
+        self.assertEqual(parse_age_limit(22), None)
+        self.assertEqual(parse_age_limit('18'), 18)
+        self.assertEqual(parse_age_limit('18+'), 18)
+        self.assertEqual(parse_age_limit('PG-13'), 13)
+        self.assertEqual(parse_age_limit('TV-14'), 14)
+        self.assertEqual(parse_age_limit('TV-MA'), 17)
+        self.assertEqual(parse_age_limit('TV14'), 14)
+        self.assertEqual(parse_age_limit('TV_G'), 0)
+

     def test_parse_duration(self):
         self.assertEqual(parse_duration(None), None)
         self.assertEqual(parse_duration(False), None)
     def test_parse_duration(self):
         self.assertEqual(parse_duration(None), None)
         self.assertEqual(parse_duration(False), None)


@@ -397,6 +584,9 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('01:02:03:04'), 93784)
         self.assertEqual(parse_duration('1 hour 3 minutes'), 3780)
         self.assertEqual(parse_duration('87 Min.'), 5220)
         self.assertEqual(parse_duration('01:02:03:04'), 93784)
         self.assertEqual(parse_duration('1 hour 3 minutes'), 3780)
         self.assertEqual(parse_duration('87 Min.'), 5220)

+        self.assertEqual(parse_duration('PT1H0.040S'), 3600.04)
+        self.assertEqual(parse_duration('PT00H03M30SZ'), 210)
+        self.assertEqual(parse_duration('P0Y0M0DT0H4M20.880S'), 260.88)

 
     def test_fix_xml_ampersands(self):
         self.assertEqual(
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(


@@ -436,9 +626,6 @@ class TestUtil(unittest.TestCase):
         testPL(5, 2, (2, 99), [2, 3, 4])
         testPL(5, 2, (20, 99), [])
 
         testPL(5, 2, (2, 99), [2, 3, 4])
         testPL(5, 2, (20, 99), [])
 

-    def test_struct_unpack(self):
-        self.assertEqual(struct_unpack('!B', b'\x00'), (0,))
-

     def test_read_batch_urls(self):
         f = io.StringIO('''\xef\xbb\xbf foo
             bar\r
     def test_read_batch_urls(self):
         f = io.StringIO('''\xef\xbb\xbf foo
             bar\r


@@ -452,6 +639,50 @@ class TestUtil(unittest.TestCase):
         data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
         self.assertTrue(isinstance(data, bytes))
 
         data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'})
         self.assertTrue(isinstance(data, bytes))
 

+    def test_update_url_query(self):
+        def query_dict(url):
+            return compat_parse_qs(compat_urlparse.urlparse(url).query)
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'quality': ['HD'], 'format': ['mp4']})),
+            query_dict('http://example.com/path?quality=HD&format=mp4'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'system': ['LINUX', 'WINDOWS']})),
+            query_dict('http://example.com/path?system=LINUX&system=WINDOWS'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'fields': 'id,formats,subtitles'})),
+            query_dict('http://example.com/path?fields=id,formats,subtitles'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'fields': ('id,formats,subtitles', 'thumbnails')})),
+            query_dict('http://example.com/path?fields=id,formats,subtitles&fields=thumbnails'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path?manifest=f4m', {'manifest': []})),
+            query_dict('http://example.com/path'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path?system=LINUX&system=WINDOWS', {'system': 'LINUX'})),
+            query_dict('http://example.com/path?system=LINUX'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'fields': b'id,formats,subtitles'})),
+            query_dict('http://example.com/path?fields=id,formats,subtitles'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'width': 1080, 'height': 720})),
+            query_dict('http://example.com/path?width=1080&height=720'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'bitrate': 5020.43})),
+            query_dict('http://example.com/path?bitrate=5020.43'))
+        self.assertEqual(query_dict(update_url_query(
+            'http://example.com/path', {'test': '第二行тест'})),
+            query_dict('http://example.com/path?test=%E7%AC%AC%E4%BA%8C%E8%A1%8C%D1%82%D0%B5%D1%81%D1%82'))
+
+    def test_multipart_encode(self):
+        self.assertEqual(
+            multipart_encode({b'field': b'value'}, boundary='AAAAAA')[0],
+            b'--AAAAAA\r\nContent-Disposition: form-data; name="field"\r\n\r\nvalue\r\n--AAAAAA--\r\n')
+        self.assertEqual(
+            multipart_encode({'欄位'.encode('utf-8'): '值'.encode('utf-8')}, boundary='AAAAAA')[0],
+            b'--AAAAAA\r\nContent-Disposition: form-data; name="\xe6\xac\x84\xe4\xbd\x8d"\r\n\r\n\xe5\x80\xbc\r\n--AAAAAA--\r\n')
+        self.assertRaises(
+            ValueError, multipart_encode, {b'field': b'value'}, boundary='value')
+

     def test_dict_get(self):
         FALSE_VALUES = {
             'none': None,
     def test_dict_get(self):
         FALSE_VALUES = {
             'none': None,


@@ -474,6 +705,17 @@ class TestUtil(unittest.TestCase):
             self.assertEqual(dict_get(d, ('b', 'c', key, )), None)
             self.assertEqual(dict_get(d, ('b', 'c', key, ), skip_false_values=False), false_value)
 
             self.assertEqual(dict_get(d, ('b', 'c', key, )), None)
             self.assertEqual(dict_get(d, ('b', 'c', key, ), skip_false_values=False), false_value)
 

+    def test_merge_dicts(self):
+        self.assertEqual(merge_dicts({'a': 1}, {'b': 2}), {'a': 1, 'b': 2})
+        self.assertEqual(merge_dicts({'a': 1}, {'a': 2}), {'a': 1})
+        self.assertEqual(merge_dicts({'a': 1}, {'a': None}), {'a': 1})
+        self.assertEqual(merge_dicts({'a': 1}, {'a': ''}), {'a': 1})
+        self.assertEqual(merge_dicts({'a': 1}, {}), {'a': 1})
+        self.assertEqual(merge_dicts({'a': None}, {'a': 1}), {'a': 1})
+        self.assertEqual(merge_dicts({'a': ''}, {'a': 1}), {'a': ''})
+        self.assertEqual(merge_dicts({'a': ''}, {'a': 'abc'}), {'a': 'abc'})
+        self.assertEqual(merge_dicts({'a': None}, {'a': ''}, {'a': 'abc'}), {'a': 'abc'})
+

     def test_encode_compat_str(self):
         self.assertEqual(encode_compat_str(b'\xd1\x82\xd0\xb5\xd1\x81\xd1\x82', 'utf-8'), 'тест')
         self.assertEqual(encode_compat_str('тест', 'utf-8'), 'тест')
     def test_encode_compat_str(self):
         self.assertEqual(encode_compat_str(b'\xd1\x82\xd0\xb5\xd1\x81\xd1\x82', 'utf-8'), 'тест')
         self.assertEqual(encode_compat_str('тест', 'utf-8'), 'тест')


@@ -499,6 +741,30 @@ class TestUtil(unittest.TestCase):
         d = json.loads(stripped)
         self.assertEqual(d, {'status': 'success'})
 
         d = json.loads(stripped)
         self.assertEqual(d, {'status': 'success'})
 

+        stripped = strip_jsonp('window.cb && window.cb({"status": "success"});')
+        d = json.loads(stripped)
+        self.assertEqual(d, {'status': 'success'})
+
+        stripped = strip_jsonp('window.cb && cb({"status": "success"});')
+        d = json.loads(stripped)
+        self.assertEqual(d, {'status': 'success'})
+
+        stripped = strip_jsonp('({"status": "success"});')
+        d = json.loads(stripped)
+        self.assertEqual(d, {'status': 'success'})
+
+    def test_strip_or_none(self):
+        self.assertEqual(strip_or_none(' abc'), 'abc')
+        self.assertEqual(strip_or_none('abc '), 'abc')
+        self.assertEqual(strip_or_none(' abc '), 'abc')
+        self.assertEqual(strip_or_none('\tabc\t'), 'abc')
+        self.assertEqual(strip_or_none('\n\tabc\n\t'), 'abc')
+        self.assertEqual(strip_or_none('abc'), 'abc')
+        self.assertEqual(strip_or_none(''), '')
+        self.assertEqual(strip_or_none(None), None)
+        self.assertEqual(strip_or_none(42), None)
+        self.assertEqual(strip_or_none([]), None)
+

     def test_uppercase_escape(self):
         self.assertEqual(uppercase_escape('aä'), 'aä')
         self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐')
     def test_uppercase_escape(self):
         self.assertEqual(uppercase_escape('aä'), 'aä')
         self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐')


@@ -514,6 +780,58 @@ class TestUtil(unittest.TestCase):
             limit_length('foo bar baz asd', 12).startswith('foo bar'))
         self.assertTrue('...' in limit_length('foo bar baz asd', 12))
 
             limit_length('foo bar baz asd', 12).startswith('foo bar'))
         self.assertTrue('...' in limit_length('foo bar baz asd', 12))
 

+    def test_mimetype2ext(self):
+        self.assertEqual(mimetype2ext(None), None)
+        self.assertEqual(mimetype2ext('video/x-flv'), 'flv')
+        self.assertEqual(mimetype2ext('application/x-mpegURL'), 'm3u8')
+        self.assertEqual(mimetype2ext('text/vtt'), 'vtt')
+        self.assertEqual(mimetype2ext('text/vtt;charset=utf-8'), 'vtt')
+        self.assertEqual(mimetype2ext('text/html; charset=utf-8'), 'html')
+
+    def test_month_by_name(self):
+        self.assertEqual(month_by_name(None), None)
+        self.assertEqual(month_by_name('December', 'en'), 12)
+        self.assertEqual(month_by_name('décembre', 'fr'), 12)
+        self.assertEqual(month_by_name('December'), 12)
+        self.assertEqual(month_by_name('décembre'), None)
+        self.assertEqual(month_by_name('Unknown', 'unknown'), None)
+
+    def test_parse_codecs(self):
+        self.assertEqual(parse_codecs(''), {})
+        self.assertEqual(parse_codecs('avc1.77.30, mp4a.40.2'), {
+            'vcodec': 'avc1.77.30',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.2'), {
+            'vcodec': 'none',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.5,avc1.42001e'), {
+            'vcodec': 'avc1.42001e',
+            'acodec': 'mp4a.40.5',
+        })
+        self.assertEqual(parse_codecs('avc3.640028'), {
+            'vcodec': 'avc3.640028',
+            'acodec': 'none',
+        })
+        self.assertEqual(parse_codecs(', h264,,newcodec,aac'), {
+            'vcodec': 'h264',
+            'acodec': 'aac',
+        })
+        self.assertEqual(parse_codecs('av01.0.05M.08'), {
+            'vcodec': 'av01.0.05M.08',
+            'acodec': 'none',
+        })
+        self.assertEqual(parse_codecs('theora, vorbis'), {
+            'vcodec': 'theora',
+            'acodec': 'vorbis',
+        })
+        self.assertEqual(parse_codecs('unknownvcodec, unknownacodec'), {
+            'vcodec': 'unknownvcodec',
+            'acodec': 'unknownacodec',
+        })
+        self.assertEqual(parse_codecs('unknown'), {})
+

     def test_escape_rfc3986(self):
         reserved = "!*'();:@&=+$,/?#[]"
         unreserved = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_.~'
     def test_escape_rfc3986(self):
         reserved = "!*'();:@&=+$,/?#[]"
         unreserved = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_.~'


@@ -535,11 +853,11 @@ class TestUtil(unittest.TestCase):
         )
         self.assertEqual(
             escape_url('http://тест.рф/фрагмент'),
         )
         self.assertEqual(
             escape_url('http://тест.рф/фрагмент'),

-            'http://тест.рф/%D1%84%D1%80%D0%B0%D0%B3%D0%BC%D0%B5%D0%BD%D1%82'
+            'http://xn--e1aybc.xn--p1ai/%D1%84%D1%80%D0%B0%D0%B3%D0%BC%D0%B5%D0%BD%D1%82'

         )
         self.assertEqual(
             escape_url('http://тест.рф/абв?абв=абв#абв'),
         )
         self.assertEqual(
             escape_url('http://тест.рф/абв?абв=абв#абв'),

-            'http://тест.рф/%D0%B0%D0%B1%D0%B2?%D0%B0%D0%B1%D0%B2=%D0%B0%D0%B1%D0%B2#%D0%B0%D0%B1%D0%B2'
+            'http://xn--e1aybc.xn--p1ai/%D0%B0%D0%B1%D0%B2?%D0%B0%D0%B1%D0%B2=%D0%B0%D0%B1%D0%B2#%D0%B0%D0%B1%D0%B2'

         )
         self.assertEqual(escape_url('http://vimeo.com/56015672#at=0'), 'http://vimeo.com/56015672#at=0')
 
         )
         self.assertEqual(escape_url('http://vimeo.com/56015672#at=0'), 'http://vimeo.com/56015672#at=0')
 


@@ -566,6 +884,24 @@ class TestUtil(unittest.TestCase):
         json_code = js_to_json(inp)
         self.assertEqual(json.loads(json_code), json.loads(inp))
 
         json_code = js_to_json(inp)
         self.assertEqual(json.loads(json_code), json.loads(inp))
 

+        inp = '''{
+            0:{src:'skipped', type: 'application/dash+xml'},
+            1:{src:'skipped', type: 'application/vnd.apple.mpegURL'},
+        }'''
+        self.assertEqual(js_to_json(inp), '''{
+            "0":{"src":"skipped", "type": "application/dash+xml"},
+            "1":{"src":"skipped", "type": "application/vnd.apple.mpegURL"}
+        }''')
+
+        inp = '''{"foo":101}'''
+        self.assertEqual(js_to_json(inp), '''{"foo":101}''')
+
+        inp = '''{"duration": "00:01:07"}'''
+        self.assertEqual(js_to_json(inp), '''{"duration": "00:01:07"}''')
+
+        inp = '''{segments: [{"offset":-3.885780586188048e-16,"duration":39.75000000000001}]}'''
+        self.assertEqual(js_to_json(inp), '''{"segments": [{"offset":-3.885780586188048e-16,"duration":39.75000000000001}]}''')
+

     def test_js_to_json_edgecases(self):
         on = js_to_json("{abc_def:'1\\'\\\\2\\\\\\'3\"4'}")
         self.assertEqual(json.loads(on), {"abc_def": "1'\\2\\'3\"4"})
     def test_js_to_json_edgecases(self):
         on = js_to_json("{abc_def:'1\\'\\\\2\\\\\\'3\"4'}")
         self.assertEqual(json.loads(on), {"abc_def": "1'\\2\\'3\"4"})


@@ -586,12 +922,108 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('["abc", "def",]')
         self.assertEqual(json.loads(on), ['abc', 'def'])
 
         on = js_to_json('["abc", "def",]')
         self.assertEqual(json.loads(on), ['abc', 'def'])
 

+        on = js_to_json('[/*comment\n*/"abc"/*comment\n*/,/*comment\n*/"def",/*comment\n*/]')
+        self.assertEqual(json.loads(on), ['abc', 'def'])
+
+        on = js_to_json('[//comment\n"abc" //comment\n,//comment\n"def",//comment\n]')
+        self.assertEqual(json.loads(on), ['abc', 'def'])
+

         on = js_to_json('{"abc": "def",}')
         self.assertEqual(json.loads(on), {'abc': 'def'})
 
         on = js_to_json('{"abc": "def",}')
         self.assertEqual(json.loads(on), {'abc': 'def'})
 

+        on = js_to_json('{/*comment\n*/"abc"/*comment\n*/:/*comment\n*/"def"/*comment\n*/,/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'abc': 'def'})
+
+        on = js_to_json('{ 0: /* " \n */ ",]" , }')
+        self.assertEqual(json.loads(on), {'0': ',]'})
+
+        on = js_to_json('{ /*comment\n*/0/*comment\n*/: /* " \n */ ",]" , }')
+        self.assertEqual(json.loads(on), {'0': ',]'})
+
+        on = js_to_json('{ 0: // comment\n1 }')
+        self.assertEqual(json.loads(on), {'0': 1})
+
+        on = js_to_json(r'["<p>x<\/p>"]')
+        self.assertEqual(json.loads(on), ['<p>x</p>'])
+
+        on = js_to_json(r'["\xaa"]')
+        self.assertEqual(json.loads(on), ['\u00aa'])
+
+        on = js_to_json("['a\\\nb']")
+        self.assertEqual(json.loads(on), ['ab'])
+
+        on = js_to_json("/*comment\n*/[/*comment\n*/'a\\\nb'/*comment\n*/]/*comment\n*/")
+        self.assertEqual(json.loads(on), ['ab'])
+
+        on = js_to_json('{0xff:0xff}')
+        self.assertEqual(json.loads(on), {'255': 255})
+
+        on = js_to_json('{/*comment\n*/0xff/*comment\n*/:/*comment\n*/0xff/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'255': 255})
+
+        on = js_to_json('{077:077}')
+        self.assertEqual(json.loads(on), {'63': 63})
+
+        on = js_to_json('{/*comment\n*/077/*comment\n*/:/*comment\n*/077/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'63': 63})
+
+        on = js_to_json('{42:42}')
+        self.assertEqual(json.loads(on), {'42': 42})
+
+        on = js_to_json('{/*comment\n*/42/*comment\n*/:/*comment\n*/42/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'42': 42})
+
+        on = js_to_json('{42:4.2e1}')
+        self.assertEqual(json.loads(on), {'42': 42.0})
+
+    def test_js_to_json_malformed(self):
+        self.assertEqual(js_to_json('42a1'), '42"a1"')
+        self.assertEqual(js_to_json('42a-1'), '42"a"-1')
+
+    def test_extract_attributes(self):
+        self.assertEqual(extract_attributes('<e x="y">'), {'x': 'y'})
+        self.assertEqual(extract_attributes("<e x='y'>"), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x=y>'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x="a \'b\' c">'), {'x': "a 'b' c"})
+        self.assertEqual(extract_attributes('<e x=\'a "b" c\'>'), {'x': 'a "b" c'})
+        self.assertEqual(extract_attributes('<e x="&#121;">'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x="&#x79;">'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x="&amp;">'), {'x': '&'})  # XML
+        self.assertEqual(extract_attributes('<e x="&quot;">'), {'x': '"'})
+        self.assertEqual(extract_attributes('<e x="&pound;">'), {'x': '£'})  # HTML 3.2
+        self.assertEqual(extract_attributes('<e x="&lambda;">'), {'x': 'λ'})  # HTML 4.0
+        self.assertEqual(extract_attributes('<e x="&foo">'), {'x': '&foo'})
+        self.assertEqual(extract_attributes('<e x="\'">'), {'x': "'"})
+        self.assertEqual(extract_attributes('<e x=\'"\'>'), {'x': '"'})
+        self.assertEqual(extract_attributes('<e x >'), {'x': None})
+        self.assertEqual(extract_attributes('<e x=y a>'), {'x': 'y', 'a': None})
+        self.assertEqual(extract_attributes('<e x= y>'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e x=1 y=2 x=3>'), {'y': '2', 'x': '3'})
+        self.assertEqual(extract_attributes('<e \nx=\ny\n>'), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e \nx=\n"y"\n>'), {'x': 'y'})
+        self.assertEqual(extract_attributes("<e \nx=\n'y'\n>"), {'x': 'y'})
+        self.assertEqual(extract_attributes('<e \nx="\ny\n">'), {'x': '\ny\n'})
+        self.assertEqual(extract_attributes('<e CAPS=x>'), {'caps': 'x'})  # Names lowercased
+        self.assertEqual(extract_attributes('<e x=1 X=2>'), {'x': '2'})
+        self.assertEqual(extract_attributes('<e X=1 x=2>'), {'x': '2'})
+        self.assertEqual(extract_attributes('<e _:funny-name1=1>'), {'_:funny-name1': '1'})
+        self.assertEqual(extract_attributes('<e x="Fáilte 世界 \U0001f600">'), {'x': 'Fáilte 世界 \U0001f600'})
+        self.assertEqual(extract_attributes('<e x="décompose&#769;">'), {'x': 'décompose\u0301'})
+        # "Narrow" Python builds don't support unicode code points outside BMP.
+        try:
+            compat_chr(0x10000)
+            supports_outside_bmp = True
+        except ValueError:
+            supports_outside_bmp = False
+        if supports_outside_bmp:
+            self.assertEqual(extract_attributes('<e x="Smile &#128512;!">'), {'x': 'Smile \U0001f600!'})
+        # Malformed HTML should not break attributes extraction on older Python
+        self.assertEqual(extract_attributes('<mal"formed/>'), {})
+

     def test_clean_html(self):
         self.assertEqual(clean_html('a:\nb'), 'a: b')
         self.assertEqual(clean_html('a:\n   "b"'), 'a:    "b"')
     def test_clean_html(self):
         self.assertEqual(clean_html('a:\nb'), 'a: b')
         self.assertEqual(clean_html('a:\n   "b"'), 'a:    "b"')

+        self.assertEqual(clean_html('a<br>\xa0b'), 'a\nb')

 
     def test_intlist_to_bytes(self):
         self.assertEqual(
 
     def test_intlist_to_bytes(self):
         self.assertEqual(


@@ -601,7 +1033,7 @@ class TestUtil(unittest.TestCase):
     def test_args_to_str(self):
         self.assertEqual(
             args_to_str(['foo', 'ba/r', '-baz', '2 be', '']),
     def test_args_to_str(self):
         self.assertEqual(
             args_to_str(['foo', 'ba/r', '-baz', '2 be', '']),

-            'foo ba/r -baz \'2 be\' \'\''
+            'foo ba/r -baz \'2 be\' \'\'' if compat_os_name != 'nt' else 'foo ba/r -baz "2 be" ""'

         )
 
     def test_parse_filesize(self):
         )
 
     def test_parse_filesize(self):


@@ -612,7 +1044,38 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_filesize('2 MiB'), 2097152)
         self.assertEqual(parse_filesize('5 GB'), 5000000000)
         self.assertEqual(parse_filesize('1.2Tb'), 1200000000000)
         self.assertEqual(parse_filesize('2 MiB'), 2097152)
         self.assertEqual(parse_filesize('5 GB'), 5000000000)
         self.assertEqual(parse_filesize('1.2Tb'), 1200000000000)

+        self.assertEqual(parse_filesize('1.2tb'), 1200000000000)

         self.assertEqual(parse_filesize('1,24 KB'), 1240)
         self.assertEqual(parse_filesize('1,24 KB'), 1240)

+        self.assertEqual(parse_filesize('1,24 kb'), 1240)
+        self.assertEqual(parse_filesize('8.5 megabytes'), 8500000)
+
+    def test_parse_count(self):
+        self.assertEqual(parse_count(None), None)
+        self.assertEqual(parse_count(''), None)
+        self.assertEqual(parse_count('0'), 0)
+        self.assertEqual(parse_count('1000'), 1000)
+        self.assertEqual(parse_count('1.000'), 1000)
+        self.assertEqual(parse_count('1.1k'), 1100)
+        self.assertEqual(parse_count('1.1kk'), 1100000)
+        self.assertEqual(parse_count('1.1kk '), 1100000)
+        self.assertEqual(parse_count('1.1kk views'), 1100000)
+
+    def test_parse_resolution(self):
+        self.assertEqual(parse_resolution(None), {})
+        self.assertEqual(parse_resolution(''), {})
+        self.assertEqual(parse_resolution('1920x1080'), {'width': 1920, 'height': 1080})
+        self.assertEqual(parse_resolution('1920×1080'), {'width': 1920, 'height': 1080})
+        self.assertEqual(parse_resolution('1920 x 1080'), {'width': 1920, 'height': 1080})
+        self.assertEqual(parse_resolution('720p'), {'height': 720})
+        self.assertEqual(parse_resolution('4k'), {'height': 2160})
+        self.assertEqual(parse_resolution('8K'), {'height': 4320})
+
+    def test_parse_bitrate(self):
+        self.assertEqual(parse_bitrate(None), None)
+        self.assertEqual(parse_bitrate(''), None)
+        self.assertEqual(parse_bitrate('300kbps'), 300)
+        self.assertEqual(parse_bitrate('1500kbps'), 1500)
+        self.assertEqual(parse_bitrate('300 kbps'), 300)

 
     def test_version_tuple(self):
         self.assertEqual(version_tuple('1'), (1,))
 
     def test_version_tuple(self):
         self.assertEqual(version_tuple('1'), (1,))


@@ -692,6 +1155,18 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')
         self.assertFalse(match_str(
             'like_count > 100 & dislike_count <? 50 & description',
             {'like_count': 190, 'dislike_count': 10}))
         self.assertFalse(match_str(
             'like_count > 100 & dislike_count <? 50 & description',
             {'like_count': 190, 'dislike_count': 10}))

+        self.assertTrue(match_str('is_live', {'is_live': True}))
+        self.assertFalse(match_str('is_live', {'is_live': False}))
+        self.assertFalse(match_str('is_live', {'is_live': None}))
+        self.assertFalse(match_str('is_live', {}))
+        self.assertFalse(match_str('!is_live', {'is_live': True}))
+        self.assertTrue(match_str('!is_live', {'is_live': False}))
+        self.assertTrue(match_str('!is_live', {'is_live': None}))
+        self.assertTrue(match_str('!is_live', {}))
+        self.assertTrue(match_str('title', {'title': 'abc'}))
+        self.assertTrue(match_str('title', {'title': ''}))
+        self.assertFalse(match_str('!title', {'title': 'abc'}))
+        self.assertFalse(match_str('!title', {'title': ''}))

 
     def test_parse_dfxp_time_expr(self):
         self.assertEqual(parse_dfxp_time_expr(None), None)
 
     def test_parse_dfxp_time_expr(self):
         self.assertEqual(parse_dfxp_time_expr(None), None)


@@ -715,7 +1190,7 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')
                     <p begin="3" dur="-1">Ignored, three</p>
                 </div>
             </body>
                     <p begin="3" dur="-1">Ignored, three</p>
                 </div>
             </body>

-            </tt>'''
+            </tt>'''.encode('utf-8')

         srt_data = '''1
 00:00:00,000 --> 00:00:01,000
 The following line contains Chinese characters and special symbols
         srt_data = '''1
 00:00:00,000 --> 00:00:01,000
 The following line contains Chinese characters and special symbols


@@ -740,7 +1215,7 @@ Line
                     <p begin="0" end="1">The first line</p>
                 </div>
             </body>
                     <p begin="0" end="1">The first line</p>
                 </div>
             </body>

-            </tt>'''
+            </tt>'''.encode('utf-8')

         srt_data = '''1
 00:00:00,000 --> 00:00:01,000
 The first line
         srt_data = '''1
 00:00:00,000 --> 00:00:01,000
 The first line


@@ -748,10 +1223,72 @@ The first line
 '''
         self.assertEqual(dfxp2srt(dfxp_data_no_default_namespace), srt_data)
 
 '''
         self.assertEqual(dfxp2srt(dfxp_data_no_default_namespace), srt_data)
 

+        dfxp_data_with_style = '''<?xml version="1.0" encoding="utf-8"?>
+<tt xmlns="http://www.w3.org/2006/10/ttaf1" xmlns:ttp="http://www.w3.org/2006/10/ttaf1#parameter" ttp:timeBase="media" xmlns:tts="http://www.w3.org/2006/10/ttaf1#style" xml:lang="en" xmlns:ttm="http://www.w3.org/2006/10/ttaf1#metadata">
+  <head>
+    <styling>
+      <style id="s2" style="s0" tts:color="cyan" tts:fontWeight="bold" />
+      <style id="s1" style="s0" tts:color="yellow" tts:fontStyle="italic" />
+      <style id="s3" style="s0" tts:color="lime" tts:textDecoration="underline" />
+      <style id="s0" tts:backgroundColor="black" tts:fontStyle="normal" tts:fontSize="16" tts:fontFamily="sansSerif" tts:color="white" />
+    </styling>
+  </head>
+  <body tts:textAlign="center" style="s0">
+    <div>
+      <p begin="00:00:02.08" id="p0" end="00:00:05.84">default style<span tts:color="red">custom style</span></p>
+      <p style="s2" begin="00:00:02.08" id="p0" end="00:00:05.84"><span tts:color="lime">part 1<br /></span><span tts:color="cyan">part 2</span></p>
+      <p style="s3" begin="00:00:05.84" id="p1" end="00:00:09.56">line 3<br />part 3</p>
+      <p style="s1" tts:textDecoration="underline" begin="00:00:09.56" id="p2" end="00:00:12.36"><span style="s2" tts:color="lime">inner<br /> </span>style</p>
+    </div>
+  </body>
+</tt>'''.encode('utf-8')
+        srt_data = '''1
+00:00:02,080 --> 00:00:05,839
+<font color="white" face="sansSerif" size="16">default style<font color="red">custom style</font></font>
+
+2
+00:00:02,080 --> 00:00:05,839
+<b><font color="cyan" face="sansSerif" size="16"><font color="lime">part 1
+</font>part 2</font></b>
+
+3
+00:00:05,839 --> 00:00:09,560
+<u><font color="lime">line 3
+part 3</font></u>
+
+4
+00:00:09,560 --> 00:00:12,359
+<i><u><font color="yellow"><font color="lime">inner
+ </font>style</font></u></i>
+
+'''
+        self.assertEqual(dfxp2srt(dfxp_data_with_style), srt_data)
+
+        dfxp_data_non_utf8 = '''<?xml version="1.0" encoding="UTF-16"?>
+            <tt xmlns="http://www.w3.org/ns/ttml" xml:lang="en" xmlns:tts="http://www.w3.org/ns/ttml#parameter">
+            <body>
+                <div xml:lang="en">
+                    <p begin="0" end="1">Line 1</p>
+                    <p begin="1" end="2">第二行</p>
+                </div>
+            </body>
+            </tt>'''.encode('utf-16')
+        srt_data = '''1
+00:00:00,000 --> 00:00:01,000
+Line 1
+
+2
+00:00:01,000 --> 00:00:02,000
+第二行
+
+'''
+        self.assertEqual(dfxp2srt(dfxp_data_non_utf8), srt_data)
+

     def test_cli_option(self):
         self.assertEqual(cli_option({'proxy': '127.0.0.1:3128'}, '--proxy', 'proxy'), ['--proxy', '127.0.0.1:3128'])
         self.assertEqual(cli_option({'proxy': None}, '--proxy', 'proxy'), [])
         self.assertEqual(cli_option({}, '--proxy', 'proxy'), [])
     def test_cli_option(self):
         self.assertEqual(cli_option({'proxy': '127.0.0.1:3128'}, '--proxy', 'proxy'), ['--proxy', '127.0.0.1:3128'])
         self.assertEqual(cli_option({'proxy': None}, '--proxy', 'proxy'), [])
         self.assertEqual(cli_option({}, '--proxy', 'proxy'), [])

+        self.assertEqual(cli_option({'retries': 10}, '--retries', 'retries'), ['--retries', '10'])

 
     def test_cli_valueless_option(self):
         self.assertEqual(cli_valueless_option(
 
     def test_cli_valueless_option(self):
         self.assertEqual(cli_valueless_option(


@@ -792,6 +1329,10 @@ The first line
             cli_bool_option(
                 {'nocheckcertificate': False}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
             ['--check-certificate=true'])
             cli_bool_option(
                 {'nocheckcertificate': False}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
             ['--check-certificate=true'])

+        self.assertEqual(
+            cli_bool_option(
+                {}, '--check-certificate', 'nocheckcertificate', 'false', 'true', '='),
+            [])

 
     def test_ohdave_rsa_encrypt(self):
         N = 0xab86b6371b5318aaa1d3c9e612a9f1264f372323c8c0f19875b5fc3b3fd3afcc1e5bec527aa94bfa85bffc157e4245aebda05389a5357b75115ac94f074aefcd
 
     def test_ohdave_rsa_encrypt(self):
         N = 0xab86b6371b5318aaa1d3c9e612a9f1264f372323c8c0f19875b5fc3b3fd3afcc1e5bec527aa94bfa85bffc157e4245aebda05389a5357b75115ac94f074aefcd


@@ -801,5 +1342,69 @@ The first line
             ohdave_rsa_encrypt(b'aa111222', e, N),
             '726664bd9a23fd0c70f9f1b84aab5e3905ce1e45a584e9cbcf9bcc7510338fc1986d6c599ff990d923aa43c51c0d9013cd572e13bc58f4ae48f2ed8c0b0ba881')
 
             ohdave_rsa_encrypt(b'aa111222', e, N),
             '726664bd9a23fd0c70f9f1b84aab5e3905ce1e45a584e9cbcf9bcc7510338fc1986d6c599ff990d923aa43c51c0d9013cd572e13bc58f4ae48f2ed8c0b0ba881')
 

+    def test_pkcs1pad(self):
+        data = [1, 2, 3]
+        padded_data = pkcs1pad(data, 32)
+        self.assertEqual(padded_data[:2], [0, 2])
+        self.assertEqual(padded_data[28:], [0, 1, 2, 3])
+
+        self.assertRaises(ValueError, pkcs1pad, data, 8)
+
+    def test_encode_base_n(self):
+        self.assertEqual(encode_base_n(0, 30), '0')
+        self.assertEqual(encode_base_n(80, 30), '2k')
+
+        custom_table = '9876543210ZYXWVUTSRQPONMLKJIHGFEDCBA'
+        self.assertEqual(encode_base_n(0, 30, custom_table), '9')
+        self.assertEqual(encode_base_n(80, 30, custom_table), '7P')
+
+        self.assertRaises(ValueError, encode_base_n, 0, 70)
+        self.assertRaises(ValueError, encode_base_n, 0, 60, custom_table)
+
+    def test_urshift(self):
+        self.assertEqual(urshift(3, 1), 1)
+        self.assertEqual(urshift(-3, 1), 2147483646)
+
+    def test_get_element_by_class(self):
+        html = '''
+            <span class="foo bar">nice</span>
+        '''
+
+        self.assertEqual(get_element_by_class('foo', html), 'nice')
+        self.assertEqual(get_element_by_class('no-such-class', html), None)
+
+    def test_get_element_by_attribute(self):
+        html = '''
+            <span class="foo bar">nice</span>
+        '''
+
+        self.assertEqual(get_element_by_attribute('class', 'foo bar', html), 'nice')
+        self.assertEqual(get_element_by_attribute('class', 'foo', html), None)
+        self.assertEqual(get_element_by_attribute('class', 'no-such-foo', html), None)
+
+        html = '''
+            <div itemprop="author" itemscope>foo</div>
+        '''
+
+        self.assertEqual(get_element_by_attribute('itemprop', 'author', html), 'foo')
+
+    def test_get_elements_by_class(self):
+        html = '''
+            <span class="foo bar">nice</span><span class="foo bar">also nice</span>
+        '''
+
+        self.assertEqual(get_elements_by_class('foo', html), ['nice', 'also nice'])
+        self.assertEqual(get_elements_by_class('no-such-class', html), [])
+
+    def test_get_elements_by_attribute(self):
+        html = '''
+            <span class="foo bar">nice</span><span class="foo bar">also nice</span>
+        '''
+
+        self.assertEqual(get_elements_by_attribute('class', 'foo bar', html), ['nice', 'also nice'])
+        self.assertEqual(get_elements_by_attribute('class', 'foo', html), [])
+        self.assertEqual(get_elements_by_attribute('class', 'no-such-foo', html), [])
+
+

 if __name__ == '__main__':
     unittest.main()
 if __name__ == '__main__':
     unittest.main()