mirror of
https://github.com/ytdl-org/youtube-dl.git
synced 2024-06-10 05:49:37 +00:00
Model match-filter string comparison after format selection
This commit is contained in:
parent
3b74d490e0
commit
9ec3eb2a3d
30
README.md
30
README.md
|
@ -237,24 +237,20 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
|
||||||
compare against a number, key = 'LITERAL'
|
compare against a number, key = 'LITERAL'
|
||||||
(like "uploader = 'Mike Smith'", also works
|
(like "uploader = 'Mike Smith'", also works
|
||||||
with !=) to match against a string literal
|
with !=) to match against a string literal
|
||||||
and & to require multiple matches. Values
|
and & to require multiple matches. The
|
||||||
which are not known are excluded unless you
|
string comparisons ^= (starts with), *=
|
||||||
put a question mark (?) after the operator.
|
(contains) and $= (ends with). Every
|
||||||
For example, to only match videos that have
|
comparison operator can be prefixed by ! to
|
||||||
been liked more than 100 times and disliked
|
negate it. Values which are not known are
|
||||||
less than 50 times (or the dislike
|
excluded unless you put a question mark (?)
|
||||||
functionality is not available at the given
|
after the operator. For example, to only
|
||||||
service), but who also have a description,
|
match videos that have been liked more than
|
||||||
use --match-filter "like_count > 100 &
|
100 times and disliked less than 50 times
|
||||||
|
(or the dislike functionality is not
|
||||||
|
available at the given service), but who
|
||||||
|
also have a description, use
|
||||||
|
--match-filter "like_count > 100 &
|
||||||
dislike_count <? 50 & description" .
|
dislike_count <? 50 & description" .
|
||||||
For matching strings, the oparators ~= and
|
|
||||||
!~= check for string containment and
|
|
||||||
exclusion. The operators *= and !*= search
|
|
||||||
for a regular expression.
|
|
||||||
For example, to only match videos which
|
|
||||||
have neither 'sponsored' nor 'Sponsored' in
|
|
||||||
the title, use --match-filter "title !*=
|
|
||||||
'[Ss]ponsored'"
|
|
||||||
--no-playlist Download only the video, if the URL refers
|
--no-playlist Download only the video, if the URL refers
|
||||||
to a video and a playlist.
|
to a video and a playlist.
|
||||||
--yes-playlist Download the playlist, if the URL refers to
|
--yes-playlist Download the playlist, if the URL refers to
|
||||||
|
|
|
@ -1178,7 +1178,6 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')
|
||||||
'9999 51')
|
'9999 51')
|
||||||
|
|
||||||
def test_match_str(self):
|
def test_match_str(self):
|
||||||
self.assertRaises(ValueError, match_str, 'xy>foobar', {})
|
|
||||||
self.assertFalse(match_str('xy', {'x': 1200}))
|
self.assertFalse(match_str('xy', {'x': 1200}))
|
||||||
self.assertTrue(match_str('!xy', {'x': 1200}))
|
self.assertTrue(match_str('!xy', {'x': 1200}))
|
||||||
self.assertTrue(match_str('x', {'x': 1200}))
|
self.assertTrue(match_str('x', {'x': 1200}))
|
||||||
|
@ -1195,6 +1194,17 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')
|
||||||
self.assertTrue(match_str('y=foobar42', {'y': 'foobar42'}))
|
self.assertTrue(match_str('y=foobar42', {'y': 'foobar42'}))
|
||||||
self.assertFalse(match_str('y!=foobar42', {'y': 'foobar42'}))
|
self.assertFalse(match_str('y!=foobar42', {'y': 'foobar42'}))
|
||||||
self.assertTrue(match_str('y!=foobar2', {'y': 'foobar42'}))
|
self.assertTrue(match_str('y!=foobar2', {'y': 'foobar42'}))
|
||||||
|
self.assertTrue(match_str('y^=foo', {'y': 'foobar42'}))
|
||||||
|
self.assertFalse(match_str('y!^=foo', {'y': 'foobar42'}))
|
||||||
|
self.assertFalse(match_str('y^=bar', {'y': 'foobar42'}))
|
||||||
|
self.assertTrue(match_str('y!^=bar', {'y': 'foobar42'}))
|
||||||
|
self.assertRaises(ValueError, match_str, 'x^=42', {'x': 42})
|
||||||
|
self.assertTrue(match_str('y*=bar', {'y': 'foobar42'}))
|
||||||
|
self.assertFalse(match_str('y!*=bar', {'y': 'foobar42'}))
|
||||||
|
self.assertFalse(match_str('y*=baz', {'y': 'foobar42'}))
|
||||||
|
self.assertTrue(match_str('y!*=baz', {'y': 'foobar42'}))
|
||||||
|
self.assertTrue(match_str('y$=42', {'y': 'foobar42'}))
|
||||||
|
self.assertFalse(match_str('y$=43', {'y': 'foobar42'}))
|
||||||
self.assertFalse(match_str(
|
self.assertFalse(match_str(
|
||||||
'like_count > 100 & dislike_count <? 50 & description',
|
'like_count > 100 & dislike_count <? 50 & description',
|
||||||
{'like_count': 90, 'description': 'foo'}))
|
{'like_count': 90, 'description': 'foo'}))
|
||||||
|
|
|
@ -4368,15 +4368,13 @@ def _match_one(filter_part, dct):
|
||||||
'>': operator.gt,
|
'>': operator.gt,
|
||||||
'>=': operator.ge,
|
'>=': operator.ge,
|
||||||
'=': operator.eq,
|
'=': operator.eq,
|
||||||
'!=': operator.ne,
|
'*=': operator.contains,
|
||||||
'~=': operator.contains,
|
'^=': lambda attr, value: attr.startswith(value),
|
||||||
'!~=': lambda left, right: not operator.contains(left, right),
|
'$=': lambda attr, value: attr.endswith(value),
|
||||||
'*=': lambda left, right: bool(re.search(right, left)),
|
|
||||||
'!*=': lambda left, right: not bool(re.search(right, left)),
|
|
||||||
}
|
}
|
||||||
operator_rex = re.compile(r'''(?x)\s*
|
operator_rex = re.compile(r'''(?x)\s*
|
||||||
(?P<key>[a-z_]+)
|
(?P<key>[a-z_]+)
|
||||||
\s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
|
\s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
|
||||||
(?:
|
(?:
|
||||||
(?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
|
(?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)|
|
||||||
(?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)|
|
(?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)|
|
||||||
|
@ -4386,7 +4384,11 @@ def _match_one(filter_part, dct):
|
||||||
''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
|
''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys())))
|
||||||
m = operator_rex.search(filter_part)
|
m = operator_rex.search(filter_part)
|
||||||
if m:
|
if m:
|
||||||
op = COMPARISON_OPERATORS[m.group('op')]
|
unnegated_op = COMPARISON_OPERATORS[m.group('op')]
|
||||||
|
if m.group('negation'):
|
||||||
|
op = lambda attr, value: not unnegated_op(attr, value)
|
||||||
|
else:
|
||||||
|
op = unnegated_op
|
||||||
actual_value = dct.get(m.group('key'))
|
actual_value = dct.get(m.group('key'))
|
||||||
if (m.group('quotedstrval') is not None
|
if (m.group('quotedstrval') is not None
|
||||||
or m.group('strval') is not None
|
or m.group('strval') is not None
|
||||||
|
@ -4401,7 +4403,7 @@ def _match_one(filter_part, dct):
|
||||||
if quote is not None:
|
if quote is not None:
|
||||||
comparison_value = comparison_value.replace(r'\%s' % quote, quote)
|
comparison_value = comparison_value.replace(r'\%s' % quote, quote)
|
||||||
else:
|
else:
|
||||||
if m.group('op') in ('~=', '!~=', '*=', '!*='):
|
if m.group('op') in ('*=', '^=', '$='):
|
||||||
raise ValueError(
|
raise ValueError(
|
||||||
'Operator %s only supports string values!' % m.group('op'))
|
'Operator %s only supports string values!' % m.group('op'))
|
||||||
try:
|
try:
|
||||||
|
|
Loading…
Reference in a new issue