diff --git a/test/test_utils.py b/test/test_utils.py index 0953db3719..3d14f61fbf 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -1,6 +1,8 @@ #!/usr/bin/env python # coding: utf-8 +from __future__ import unicode_literals + # Allow direct execution import os import sys @@ -13,7 +15,6 @@ import io import json import xml.etree.ElementTree -#from youtube_dl.utils import htmlentity_transform from youtube_dl.utils import ( DateRange, encodeFilename, @@ -41,11 +42,6 @@ from youtube_dl.utils import ( uppercase_escape, ) -if sys.version_info < (3, 0): - _compat_str = lambda b: b.decode('unicode-escape') -else: - _compat_str = lambda s: s - class TestUtil(unittest.TestCase): def test_timeconvert(self): @@ -67,9 +63,9 @@ class TestUtil(unittest.TestCase): self.assertEqual('this - that', sanitize_filename('this: that')) self.assertEqual(sanitize_filename('AT&T'), 'AT&T') - aumlaut = _compat_str('\xe4') + aumlaut = 'ä' self.assertEqual(sanitize_filename(aumlaut), aumlaut) - tests = _compat_str('\u043a\u0438\u0440\u0438\u043b\u043b\u0438\u0446\u0430') + tests = '\u043a\u0438\u0440\u0438\u043b\u043b\u0438\u0446\u0430' self.assertEqual(sanitize_filename(tests), tests) forbidden = '"\0\\/' @@ -91,9 +87,9 @@ class TestUtil(unittest.TestCase): self.assertEqual('yes_no', sanitize_filename('yes? no', restricted=True)) self.assertEqual('this_-_that', sanitize_filename('this: that', restricted=True)) - tests = _compat_str('a\xe4b\u4e2d\u56fd\u7684c') + tests = 'a\xe4b\u4e2d\u56fd\u7684c' self.assertEqual(sanitize_filename(tests, restricted=True), 'a_b_c') - self.assertTrue(sanitize_filename(_compat_str('\xf6'), restricted=True) != '') # No empty filename + self.assertTrue(sanitize_filename('\xf6', restricted=True) != '') # No empty filename forbidden = '"\0\\/&!: \'\t\n()[]{}$;`^,#' for fc in forbidden: @@ -101,8 +97,8 @@ class TestUtil(unittest.TestCase): self.assertTrue(fbc not in sanitize_filename(fc, restricted=True)) # Handle a common case more neatly - self.assertEqual(sanitize_filename(_compat_str('\u5927\u58f0\u5e26 - Song'), restricted=True), 'Song') - self.assertEqual(sanitize_filename(_compat_str('\u603b\u7edf: Speech'), restricted=True), 'Speech') + self.assertEqual(sanitize_filename('\u5927\u58f0\u5e26 - Song', restricted=True), 'Song') + self.assertEqual(sanitize_filename('\u603b\u7edf: Speech', restricted=True), 'Speech') # .. but make sure the file name is never empty self.assertTrue(sanitize_filename('-', restricted=True) != '') self.assertTrue(sanitize_filename(':', restricted=True) != '') @@ -120,7 +116,9 @@ class TestUtil(unittest.TestCase): self.assertEqual(orderedSet([135, 1, 1, 1]), [135, 1]) def test_unescape_html(self): - self.assertEqual(unescapeHTML(_compat_str('%20;')), _compat_str('%20;')) + self.assertEqual(unescapeHTML('%20;'), '%20;') + self.assertEqual( + unescapeHTML('é'), 'é') def test_daterange(self): _20century = DateRange("19000101","20000101") @@ -138,7 +136,7 @@ class TestUtil(unittest.TestCase): self.assertEqual(unified_strdate('1968-12-10'), '19681210') def test_find_xpath_attr(self): - testxml = u''' + testxml = ''' @@ -151,18 +149,18 @@ class TestUtil(unittest.TestCase): self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'c'), doc[2]) def test_meta_parser(self): - testhtml = u''' + testhtml = ''' ''' get_meta = lambda name: get_meta_content(name, testhtml) - self.assertEqual(get_meta('description'), u'foo & bar') + self.assertEqual(get_meta('description'), 'foo & bar') self.assertEqual(get_meta('author'), 'Plato') def test_xpath_with_ns(self): - testxml = u''' + testxml = ''' The Author http://server.com/download.mp3 @@ -171,8 +169,8 @@ class TestUtil(unittest.TestCase): doc = xml.etree.ElementTree.fromstring(testxml) find = lambda p: doc.find(xpath_with_ns(p, {'media': 'http://example.com/'})) self.assertTrue(find('media:song') is not None) - self.assertEqual(find('media:song/media:author').text, u'The Author') - self.assertEqual(find('media:song/url').text, u'http://server.com/download.mp3') + self.assertEqual(find('media:song/media:author').text, 'The Author') + self.assertEqual(find('media:song/url').text, 'http://server.com/download.mp3') def test_smuggle_url(self): data = {u"ö": u"ö", u"abc": [3]} @@ -187,22 +185,22 @@ class TestUtil(unittest.TestCase): self.assertEqual(res_data, None) def test_shell_quote(self): - args = ['ffmpeg', '-i', encodeFilename(u'ñ€ß\'.mp4')] - self.assertEqual(shell_quote(args), u"""ffmpeg -i 'ñ€ß'"'"'.mp4'""") + args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')] + self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""") def test_str_to_int(self): self.assertEqual(str_to_int('123,456'), 123456) self.assertEqual(str_to_int('123.456'), 123456) def test_url_basename(self): - self.assertEqual(url_basename(u'http://foo.de/'), u'') - self.assertEqual(url_basename(u'http://foo.de/bar/baz'), u'baz') - self.assertEqual(url_basename(u'http://foo.de/bar/baz?x=y'), u'baz') - self.assertEqual(url_basename(u'http://foo.de/bar/baz#x=y'), u'baz') - self.assertEqual(url_basename(u'http://foo.de/bar/baz/'), u'baz') + self.assertEqual(url_basename('http://foo.de/'), '') + self.assertEqual(url_basename('http://foo.de/bar/baz'), 'baz') + self.assertEqual(url_basename('http://foo.de/bar/baz?x=y'), 'baz') + self.assertEqual(url_basename('http://foo.de/bar/baz#x=y'), 'baz') + self.assertEqual(url_basename('http://foo.de/bar/baz/'), 'baz') self.assertEqual( - url_basename(u'http://media.w3.org/2010/05/sintel/trailer.mp4'), - u'trailer.mp4') + url_basename('http://media.w3.org/2010/05/sintel/trailer.mp4'), + 'trailer.mp4') def test_parse_duration(self): self.assertEqual(parse_duration(None), None) @@ -256,16 +254,16 @@ class TestUtil(unittest.TestCase): testPL(5, 2, (20, 99), []) def test_struct_unpack(self): - self.assertEqual(struct_unpack(u'!B', b'\x00'), (0,)) + self.assertEqual(struct_unpack('!B', b'\x00'), (0,)) def test_read_batch_urls(self): - f = io.StringIO(u'''\xef\xbb\xbf foo + f = io.StringIO('''\xef\xbb\xbf foo bar\r baz # More after this line\r ; or after this bam''') - self.assertEqual(read_batch_urls(f), [u'foo', u'bar', u'baz', u'bam']) + self.assertEqual(read_batch_urls(f), ['foo', 'bar', 'baz', 'bam']) def test_urlencode_postdata(self): data = urlencode_postdata({'username': 'foo@bar.com', 'password': '1234'}) @@ -282,8 +280,8 @@ class TestUtil(unittest.TestCase): self.assertEqual(d, [{"id": "532cb", "x": 3}]) def test_uppercase_escape(self): - self.assertEqual(uppercase_escape(u'aä'), u'aä') - self.assertEqual(uppercase_escape(u'\\U0001d550'), u'𝕐') + self.assertEqual(uppercase_escape('aä'), 'aä') + self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐') if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 4ebdf6a784..4f02108720 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -298,30 +298,6 @@ def xpath_with_ns(path, ns_map): replaced.append('{%s}%s' % (ns_map[ns], tag)) return '/'.join(replaced) -def htmlentity_transform(matchobj): - """Transforms an HTML entity to a character. - - This function receives a match object and is intended to be used with - the re.sub() function. - """ - entity = matchobj.group(1) - - # Known non-numeric HTML entity - if entity in compat_html_entities.name2codepoint: - return compat_chr(compat_html_entities.name2codepoint[entity]) - - mobj = re.match(u'(?u)#(x?\\d+)', entity) - if mobj is not None: - numstr = mobj.group(1) - if numstr.startswith(u'x'): - base = 16 - numstr = u'0%s' % numstr - else: - base = 10 - return compat_chr(int(numstr, base)) - - # Unknown entity in name, return its literal representation - return (u'&%s;' % entity) compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix class BaseHTMLParser(compat_html_parser.HTMLParser): @@ -543,13 +519,33 @@ def orderedSet(iterable): return res +def _htmlentity_transform(entity): + """Transforms an HTML entity to a character.""" + # Known non-numeric HTML entity + if entity in compat_html_entities.name2codepoint: + return compat_chr(compat_html_entities.name2codepoint[entity]) + + mobj = re.match(r'#(x?[0-9]+)', entity) + if mobj is not None: + numstr = mobj.group(1) + if numstr.startswith(u'x'): + base = 16 + numstr = u'0%s' % numstr + else: + base = 10 + return compat_chr(int(numstr, base)) + + # Unknown entity in name, return its literal representation + return (u'&%s;' % entity) + + def unescapeHTML(s): if s is None: return None assert type(s) == compat_str - result = re.sub(r'(?u)&(.+?);', htmlentity_transform, s) - return result + return re.sub( + r'&([^;]+);', lambda m: _htmlentity_transform(m.group(1)), s) def encodeFilename(s, for_subprocess=False):