[utils] Add extract_attributes for extracting html tag attributes

[youtube-dl] / test / test_utils.py

diff --git a/test/test_utils.py b/test/test_utils.py
index d0736f435093403442c303b64e7024f4389a70ae..cb85e18f08574c84ecd2c8616d69e36aff123717 100644 (file)
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -18,6 +18,7 @@ import xml.etree.ElementTree
 from youtube_dl.utils import (
     age_restricted,
     args_to_str,
+    encode_base_n,
     clean_html,
     DateRange,
     detect_exe_version,
@@ -27,6 +28,7 @@ from youtube_dl.utils import (
     encodeFilename,
     escape_rfc3986,
     escape_url,
+    extract_attributes,
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
@@ -74,6 +76,7 @@ from youtube_dl.utils import (
     cli_bool_option,
 )
 from youtube_dl.compat import (
+    compat_chr,
     compat_etree_fromstring,
 )
 
@@ -590,6 +593,44 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('{"abc": "def",}')
         self.assertEqual(json.loads(on), {'abc': 'def'})
 
+    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!'})
+
     def test_clean_html(self):
         self.assertEqual(clean_html('a:\nb'), 'a: b')
         self.assertEqual(clean_html('a:\n   "b"'), 'a:    "b"')
@@ -802,5 +843,16 @@ The first line
             ohdave_rsa_encrypt(b'aa111222', e, N),
             '726664bd9a23fd0c70f9f1b84aab5e3905ce1e45a584e9cbcf9bcc7510338fc1986d6c599ff990d923aa43c51c0d9013cd572e13bc58f4ae48f2ed8c0b0ba881')
 
+    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)
+
 if __name__ == '__main__':
     unittest.main()