administration/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-13 20:53:06 +00:00
Code Issues Projects Releases Wiki Activity

Merge remote-tracking branch 'origin/master' into c13cl

Browse Source
This commit is contained in:
Nicolas Dato 2023-12-09 11:40:38 -03:00
commit 6314563ce5
10 changed files with 132 additions and 65 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -1268,7 +1268,7 @@ The field names themselves (the part inside the parenthesis) can also have some
1. **Object traversal**: The dictionaries and lists available in metadata can be traversed by using a dot `.` separator; e.g. `%(tags.0)s`, `%(subtitles.en.-1.ext)s`. You can do Python slicing with colon `:`; E.g. `%(id.3:7:-1)s`, `%(formats.:.format_id)s`. Curly braces `{}` can be used to build dictionaries with only specific keys; e.g. `%(formats.:.{format_id,height})#j`. An empty field name `%()s` refers to the entire infodict; e.g. `%(.{id,title})s`. Note that all the fields that become available using this method are not listed below. Use `-j` to see such fields
1. **Addition**: Addition and subtraction of numeric fields can be done using `+` and `-` respectively. E.g. `%(playlist_index+10)03d`, `%(n_entries+1-playlist_index)d`
1. **Arithmetic**: Simple arithmetic can be done on numeric fields using `+`, `-` and `*`. E.g. `%(playlist_index+10)03d`, `%(n_entries+1-playlist_index)d`
1. **Date/time Formatting**: Date/time fields can be formatted according to [strftime formatting](https://docs.python.org/3/library/datetime.html#strftime-and-strptime-format-codes) by specifying it separated from the field name using a `>`. E.g. `%(duration>%H-%M-%S)s`, `%(upload_date>%Y-%m-%d)s`, `%(epoch-3600>%H-%M-%S)s`

1
test/test_YoutubeDL.py
View File

@ -797,6 +797,7 @@ class TestYoutubeDL(unittest.TestCase):
test('%(title|%)s %(title|%%)s', '% %%')
test('%(id+1-height+3)05d', '00158')
test('%(width+100)05d', 'NA')
test('%(filesize*8)d', '8192')
test('%(formats.0) 15s', ('% 15s' % FORMATS[0], None))
test('%(formats.0)r', (repr(FORMATS[0]), None))
test('%(height.0)03d', '001')

17
test/test_utils.py
View File

@ -2317,23 +2317,6 @@ Line 1
self.assertEqual(traverse_obj({}, (0, slice(1)), traverse_string=True), [],
msg='branching should result in list if `traverse_string`')
# Test is_user_input behavior
_IS_USER_INPUT_DATA = {'range8': list(range(8))}
self.assertEqual(traverse_obj(_IS_USER_INPUT_DATA, ('range8', '3'),
is_user_input=True), 3,
msg='allow for string indexing if `is_user_input`')
self.assertCountEqual(traverse_obj(_IS_USER_INPUT_DATA, ('range8', '3:'),
is_user_input=True), tuple(range(8))[3:],
msg='allow for string slice if `is_user_input`')
self.assertCountEqual(traverse_obj(_IS_USER_INPUT_DATA, ('range8', ':4:2'),
is_user_input=True), tuple(range(8))[:4:2],
msg='allow step in string slice if `is_user_input`')
self.assertCountEqual(traverse_obj(_IS_USER_INPUT_DATA, ('range8', ':'),
is_user_input=True), range(8),
msg='`:` should be treated as `...` if `is_user_input`')
with self.assertRaises(TypeError, msg='too many params should result in error'):
traverse_obj(_IS_USER_INPUT_DATA, ('range8', ':::'), is_user_input=True)
# Test re.Match as input obj
mobj = re.fullmatch(r'0(12)(?P<group>3)(4)?', '0123')
self.assertEqual(traverse_obj(mobj, ...), [x for x in mobj.groups() if x is not None],

15
yt_dlp/YoutubeDL.py
View File

@ -1179,6 +1179,7 @@ class YoutubeDL:
MATH_FUNCTIONS = {
'+': float.__add__,
'-': float.__sub__,
'*': float.__mul__,
}
# Field is of the form key1.key2...
# where keys (except first) can be string, int, slice or "{field, ...}"
@ -1200,6 +1201,15 @@ class YoutubeDL:
(?:\|(?P<default>.*?))?
)$''')
def _from_user_input(field):
if field == ':':
return ...
elif ':' in field:
return slice(*map(int_or_none, field.split(':')))
elif int_or_none(field) is not None:
return int(field)
return field
def _traverse_infodict(fields):
fields = [f for x in re.split(r'\.({.+?})\.?', fields)
for f in ([x] if x.startswith('{') else x.split('.'))]
@ -1209,11 +1219,12 @@ class YoutubeDL:
for i, f in enumerate(fields):
if not f.startswith('{'):
fields[i] = _from_user_input(f)
continue
assert f.endswith('}'), f'No closing brace for {f} in {fields}'
fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
fields[i] = {k: list(map(_from_user_input, k.split('.'))) for k in f[1:-1].split(',')}
return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
return traverse_obj(info_dict, fields, traverse_string=True)
def get_value(mdict):
# Object traversal

41
yt_dlp/extractor/francetv.py
View File

@ -1,12 +1,14 @@
from .common import InfoExtractor
from .dailymotion import DailymotionIE
from ..utils import (
determine_ext,
ExtractorError,
determine_ext,
format_field,
int_or_none,
join_nonempty,
parse_iso8601,
parse_qs,
)
from .dailymotion import DailymotionIE
class FranceTVBaseInfoExtractor(InfoExtractor):
@ -82,6 +84,8 @@ class FranceTVIE(InfoExtractor):
videos = []
title = None
subtitle = None
episode_number = None
season_number = None
image = None
duration = None
timestamp = None
@ -112,7 +116,9 @@ class FranceTVIE(InfoExtractor):
if meta:
if title is None:
title = meta.get('title')
# XXX: what is meta['pre_title']?
# meta['pre_title'] contains season and episode number for series in format "S<ID> E<ID>"
season_number, episode_number = self._search_regex(
r'S(\d+)\s*E(\d+)', meta.get('pre_title'), 'episode info', group=(1, 2), default=(None, None))
if subtitle is None:
subtitle = meta.get('additional_title')
if image is None:
@ -191,19 +197,19 @@ class FranceTVIE(InfoExtractor):
} for sheet in spritesheets]
})
if subtitle:
title += ' - %s' % subtitle
title = title.strip()
return {
'id': video_id,
'title': title,
'title': join_nonempty(title, subtitle, delim=' - ').strip(),
'thumbnail': image,
'duration': duration,
'timestamp': timestamp,
'is_live': is_live,
'formats': formats,
'subtitles': subtitles,
'episode': subtitle if episode_number else None,
'series': title if episode_number else None,
'episode_number': int_or_none(episode_number),
'season_number': int_or_none(season_number),
}
def _real_extract(self, url):
@ -230,14 +236,31 @@ class FranceTVSiteIE(FranceTVBaseInfoExtractor):
'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
'ext': 'mp4',
'title': '13h15, le dimanche... - Les mystères de Jésus',
'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
'timestamp': 1502623500,
'duration': 2580,
'thumbnail': r're:^https?://.*\.jpg$',
'upload_date': '20170813',
},
'params': {
'skip_download': True,
},
'add_ie': [FranceTVIE.ie_key()],
}, {
'url': 'https://www.france.tv/enfants/six-huit-ans/foot2rue/saison-1/3066387-duel-au-vieux-port.html',
'info_dict': {
'id': 'a9050959-eedd-4b4a-9b0d-de6eeaa73e44',
'ext': 'mp4',
'title': 'Foot2Rue - Duel au vieux port',
'episode': 'Duel au vieux port',
'series': 'Foot2Rue',
'episode_number': 1,
'season_number': 1,
'timestamp': 1642761360,
'upload_date': '20220121',
'season': 'Season 1',
'thumbnail': r're:^https?://.*\.jpg$',
'duration': 1441,
},
}, {
# france3
'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',

18
yt_dlp/extractor/mediastream.py
View File

@ -3,8 +3,11 @@ import re
from .common import InfoExtractor
from ..utils import (
clean_html,
filter_dict,
parse_qs,
remove_end,
traverse_obj,
update_url_query,
urljoin,
)
@ -108,7 +111,9 @@ class MediaStreamIE(MediaStreamBaseIE):
for message in [
'Debido a tu ubicación no puedes ver el contenido',
'You are not allowed to watch this video: Geo Fencing Restriction'
'You are not allowed to watch this video: Geo Fencing Restriction',
'Este contenido no está disponible en tu zona geográfica.',
'El contenido sólo está disponible dentro de',
]:
if message in webpage:
self.raise_geo_restricted()
@ -118,7 +123,16 @@ class MediaStreamIE(MediaStreamBaseIE):
formats, subtitles = [], {}
for video_format in player_config['src']:
if video_format == 'hls':
fmts, subs = self._extract_m3u8_formats_and_subtitles(player_config['src'][video_format], video_id)
params = {
'at': 'web-app',
'access_token': traverse_obj(parse_qs(url), ('access_token', 0)),
}
for name, key in (('MDSTRMUID', 'uid'), ('MDSTRMSID', 'sid'), ('MDSTRMPID', 'pid'), ('VERSION', 'av')):
params[key] = self._search_regex(
rf'window\.{name}\s*=\s*["\']([^"\']+)["\'];', webpage, key, default=None)
fmts, subs = self._extract_m3u8_formats_and_subtitles(
update_url_query(player_config['src'][video_format], filter_dict(params)), video_id)
formats.extend(fmts)
self._merge_subtitles(subs, target=subtitles)
elif video_format == 'mpd':

17
yt_dlp/extractor/ondemandkorea.py
View File

@ -3,7 +3,6 @@ import re
import uuid
from .common import InfoExtractor
from ..networking import HEADRequest
from ..utils import (
ExtractorError,
OnDemandPagedList,
@ -84,15 +83,17 @@ class OnDemandKoreaIE(InfoExtractor):
def try_geo_bypass(url):
return traverse_obj(url, ({parse_qs}, 'stream_url', 0, {url_or_none})) or url
def try_upgrade_quality(url):
mod_url = re.sub(r'_720(p?)\.m3u8', r'_1080\1.m3u8', url)
return mod_url if mod_url != url and self._request_webpage(
HEADRequest(mod_url), video_id, note='Checking for higher quality format',
errnote='No higher quality format found', fatal=False) else url
formats = []
for m3u8_url in traverse_obj(data, (('sources', 'manifest'), ..., 'url', {url_or_none}, {try_geo_bypass})):
formats.extend(self._extract_m3u8_formats(try_upgrade_quality(m3u8_url), video_id, fatal=False))
mod_url = re.sub(r'_720(p?)\.m3u8', r'_1080\1.m3u8', m3u8_url)
if mod_url != m3u8_url:
mod_format = self._extract_m3u8_formats(
mod_url, video_id, note='Checking for higher quality format',
errnote='No higher quality format found', fatal=False)
if mod_format:
formats.extend(mod_format)
continue
formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, fatal=False))
subtitles = {}
for track in traverse_obj(data, ('text_tracks', lambda _, v: url_or_none(v['url']))):

56
yt_dlp/extractor/pr0gramm.py
View File

@ -4,7 +4,14 @@ from urllib.parse import unquote
from .common import InfoExtractor
from ..compat import functools
from ..utils import ExtractorError, make_archive_id, urljoin
from ..utils import (
ExtractorError,
float_or_none,
int_or_none,
make_archive_id,
mimetype2ext,
urljoin,
)
from ..utils.traversal import traverse_obj
@ -26,6 +33,7 @@ class Pr0grammIE(InfoExtractor):
'dislike_count': int,
'age_limit': 0,
'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
'_old_archive_ids': ['pr0grammstatic 5466437'],
},
}, {
# Tags require account
@ -43,6 +51,7 @@ class Pr0grammIE(InfoExtractor):
'dislike_count': int,
'age_limit': 0,
'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
'_old_archive_ids': ['pr0grammstatic 3052805'],
},
}, {
# Requires verified account
@ -60,6 +69,7 @@ class Pr0grammIE(InfoExtractor):
'dislike_count': int,
'age_limit': 18,
'thumbnail': r're:^https://thumb\.pr0gramm\.com/.*\.jpg',
'_old_archive_ids': ['pr0grammstatic 5848332'],
},
}, {
'url': 'https://pr0gramm.com/static/5466437',
@ -110,37 +120,61 @@ class Pr0grammIE(InfoExtractor):
return data
@staticmethod
def _create_source_url(path):
return urljoin('https://img.pr0gramm.com', path)
def _real_extract(self, url):
video_id = self._match_id(url)
video_info = traverse_obj(
self._call_api('get', video_id, {'id': video_id, 'flags': self._maximum_flags}),
('items', 0, {dict}))
source = urljoin('https://img.pr0gramm.com', video_info.get('image'))
source = video_info.get('image')
if not source or not source.endswith('mp4'):
self.raise_no_formats('Could not extract a video', expected=bool(source), video_id=video_id)
tags = None
if self._is_logged_in:
metadata = self._call_api('info', video_id, {'itemId': video_id})
metadata = self._call_api('info', video_id, {'itemId': video_id}, note='Downloading tags')
tags = traverse_obj(metadata, ('tags', ..., 'tag', {str}))
# Sorted by "confidence", higher confidence = earlier in list
confidences = traverse_obj(metadata, ('tags', ..., 'confidence', ({int}, {float})))
if confidences:
tags = [tag for _, tag in sorted(zip(confidences, tags), reverse=True)]
formats = traverse_obj(video_info, ('variants', ..., {
'format_id': ('name', {str}),
'url': ('path', {self._create_source_url}),
'ext': ('mimeType', {mimetype2ext}),
'vcodec': ('codec', {str}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
'bitrate': ('bitRate', {float_or_none}),
'filesize': ('fileSize', {int_or_none}),
})) if video_info.get('variants') else [{
'ext': 'mp4',
'format_id': 'source',
**traverse_obj(video_info, {
'url': ('image', {self._create_source_url}),
'width': ('width', {int_or_none}),
'height': ('height', {int_or_none}),
}),
}]
subtitles = {}
for subtitle in traverse_obj(video_info, ('subtitles', lambda _, v: v['language'])):
subtitles.setdefault(subtitle['language'], []).append(traverse_obj(subtitle, {
'url': ('path', {self._create_source_url}),
'note': ('label', {str}),
}))
return {
'id': video_id,
'title': f'pr0gramm-{video_id} by {video_info.get("user")}',
'formats': [{
'url': source,
'ext': 'mp4',
**traverse_obj(video_info, {
'width': ('width', {int}),
'height': ('height', {int}),
}),
}],
'tags': tags,
'formats': formats,
'subtitles': subtitles,
'age_limit': 18 if traverse_obj(video_info, ('flags', {0b110.__and__})) else 0,
'_old_archive_ids': [make_archive_id('Pr0grammStatic', video_id)],
**traverse_obj(video_info, {

11
yt_dlp/extractor/youtube.py
View File

@ -6469,6 +6469,9 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
def _has_tab(self, tabs, tab_id):
return any(self._extract_tab_id_and_name(tab)[0] == tab_id for tab in tabs)
def _empty_playlist(self, item_id, data):
return self.playlist_result([], item_id, **self._extract_metadata_from_tabs(item_id, data))
@YoutubeTabBaseInfoExtractor.passthrough_smuggled_data
def _real_extract(self, url, smuggled_data):
item_id = self._match_id(url)
@ -6534,6 +6537,10 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
selected_tab_id, selected_tab_name = self._extract_tab_id_and_name(selected_tab, url) # NB: Name may be translated
self.write_debug(f'Selected tab: {selected_tab_id!r} ({selected_tab_name}), Requested tab: {original_tab_id!r}')
# /about is no longer a tab
if original_tab_id == 'about':
return self._empty_playlist(item_id, data)
if not original_tab_id and selected_tab_name:
self.to_screen('Downloading all uploads of the channel. '
'To download only the videos in a specific tab, pass the tab\'s URL')
@ -6546,7 +6553,7 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
if not extra_tabs and selected_tab_id != 'videos':
# Channel does not have streams, shorts or videos tabs
if item_id[:2] != 'UC':
raise ExtractorError('This channel has no uploads', expected=True)
return self._empty_playlist(item_id, data)
# Topic channels don't have /videos. Use the equivalent playlist instead
pl_id = f'UU{item_id[2:]}'
@ -6554,7 +6561,7 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
try:
data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True)
except ExtractorError:
raise ExtractorError('This channel has no uploads', expected=True)
return self._empty_playlist(item_id, data)
else:
item_id, url = pl_id, pl_url
self.to_screen(

19
yt_dlp/utils/traversal.py
View File

@ -8,7 +8,7 @@ from ._utils import (
IDENTITY,
NO_DEFAULT,
LazyList,
int_or_none,
deprecation_warning,
is_iterable_like,
try_call,
variadic,
@ -17,7 +17,7 @@ from ._utils import (
def traverse_obj(
obj, *paths, default=NO_DEFAULT, expected_type=None, get_all=True,
casesense=True, is_user_input=False, traverse_string=False):
casesense=True, is_user_input=NO_DEFAULT, traverse_string=False):
"""
Safely traverse nested `dict`s and `Iterable`s
@ -63,10 +63,8 @@ def traverse_obj(
@param get_all If `False`, return the first matching result, otherwise all matching ones.
@param casesense If `False`, consider string dictionary keys as case insensitive.
The following are only meant to be used by YoutubeDL.prepare_outtmpl and are not part of the API
`traverse_string` is only meant to be used by YoutubeDL.prepare_outtmpl and is not part of the API
@param is_user_input Whether the keys are generated from user input.
If `True` strings get converted to `int`/`slice` if needed.
@param traverse_string Whether to traverse into objects as strings.
If `True`, any non-compatible object will first be
converted into a string and then traversed into.
@ -80,6 +78,9 @@ def traverse_obj(
If no `default` is given and the last path branches, a `list` of results
is always returned. If a path ends on a `dict` that result will always be a `dict`.
"""
if is_user_input is not NO_DEFAULT:
deprecation_warning('The is_user_input parameter is deprecated and no longer works')
casefold = lambda k: k.casefold() if isinstance(k, str) else k
if isinstance(expected_type, type):
@ -195,14 +196,6 @@ def traverse_obj(
key = None
for last, key in lazy_last(variadic(path, (str, bytes, dict, set))):
if is_user_input and isinstance(key, str):
if key == ':':
key = ...
elif ':' in key:
key = slice(*map(int_or_none, key.split(':')))
elif int_or_none(key) is not None:
key = int(key)
if not casesense and isinstance(key, str):
key = key.casefold()