mirror of
https://github.com/yt-dlp/yt-dlp
synced 2024-12-26 21:59:08 +01:00
[ie/twitter] Work around API rate-limit (#8825)
Closes #8762 Authored by: bashonly
This commit is contained in:
parent
e7d22348e7
commit
116c268438
1 changed files with 42 additions and 31 deletions
|
@ -10,6 +10,7 @@ from ..compat import (
|
||||||
compat_urllib_parse_unquote,
|
compat_urllib_parse_unquote,
|
||||||
compat_urllib_parse_urlparse,
|
compat_urllib_parse_urlparse,
|
||||||
)
|
)
|
||||||
|
from ..networking.exceptions import HTTPError
|
||||||
from ..utils import (
|
from ..utils import (
|
||||||
ExtractorError,
|
ExtractorError,
|
||||||
dict_get,
|
dict_get,
|
||||||
|
@ -1317,41 +1318,51 @@ class TwitterIE(TwitterBaseIE):
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
def _extract_status(self, twid):
|
def _call_syndication_api(self, twid):
|
||||||
if self.is_logged_in or self._selected_api == 'graphql':
|
self.report_warning(
|
||||||
status = self._graphql_to_legacy(self._call_graphql_api(self._GRAPHQL_ENDPOINT, twid), twid)
|
'Not all metadata or media is available via syndication endpoint', twid, only_once=True)
|
||||||
|
status = self._download_json(
|
||||||
elif self._selected_api == 'legacy':
|
'https://cdn.syndication.twimg.com/tweet-result', twid, 'Downloading syndication JSON',
|
||||||
status = self._call_api(f'statuses/show/{twid}.json', twid, {
|
headers={'User-Agent': 'Googlebot'}, query={
|
||||||
'cards_platform': 'Web-12',
|
'id': twid,
|
||||||
'include_cards': 1,
|
# TODO: token = ((Number(twid) / 1e15) * Math.PI).toString(36).replace(/(0+|\.)/g, '')
|
||||||
'include_reply_count': 1,
|
'token': ''.join(random.choices('123456789abcdefghijklmnopqrstuvwxyz', k=10)),
|
||||||
'include_user_entities': 0,
|
|
||||||
'tweet_mode': 'extended',
|
|
||||||
})
|
})
|
||||||
|
if not status:
|
||||||
|
raise ExtractorError('Syndication endpoint returned empty JSON response')
|
||||||
|
# Transform the result so its structure matches that of legacy/graphql
|
||||||
|
media = []
|
||||||
|
for detail in traverse_obj(status, ((None, 'quoted_tweet'), 'mediaDetails', ..., {dict})):
|
||||||
|
detail['id_str'] = traverse_obj(detail, (
|
||||||
|
'video_info', 'variants', ..., 'url', {self._MEDIA_ID_RE.search}, 1), get_all=False) or twid
|
||||||
|
media.append(detail)
|
||||||
|
status['extended_entities'] = {'media': media}
|
||||||
|
|
||||||
elif self._selected_api == 'syndication':
|
return status
|
||||||
self.report_warning(
|
|
||||||
'Not all metadata or media is available via syndication endpoint', twid, only_once=True)
|
def _extract_status(self, twid):
|
||||||
status = self._download_json(
|
if self._selected_api not in ('graphql', 'legacy', 'syndication'):
|
||||||
'https://cdn.syndication.twimg.com/tweet-result', twid, 'Downloading syndication JSON',
|
raise ExtractorError(f'{self._selected_api!r} is not a valid API selection', expected=True)
|
||||||
headers={'User-Agent': 'Googlebot'}, query={
|
|
||||||
'id': twid,
|
try:
|
||||||
# TODO: token = ((Number(twid) / 1e15) * Math.PI).toString(36).replace(/(0+|\.)/g, '')
|
if self.is_logged_in or self._selected_api == 'graphql':
|
||||||
'token': ''.join(random.choices('123456789abcdefghijklmnopqrstuvwxyz', k=10)),
|
status = self._graphql_to_legacy(self._call_graphql_api(self._GRAPHQL_ENDPOINT, twid), twid)
|
||||||
|
elif self._selected_api == 'legacy':
|
||||||
|
status = self._call_api(f'statuses/show/{twid}.json', twid, {
|
||||||
|
'cards_platform': 'Web-12',
|
||||||
|
'include_cards': 1,
|
||||||
|
'include_reply_count': 1,
|
||||||
|
'include_user_entities': 0,
|
||||||
|
'tweet_mode': 'extended',
|
||||||
})
|
})
|
||||||
if not status:
|
except ExtractorError as e:
|
||||||
raise ExtractorError('Syndication endpoint returned empty JSON response')
|
if not isinstance(e.cause, HTTPError) or not e.cause.status == 429:
|
||||||
# Transform the result so its structure matches that of legacy/graphql
|
raise
|
||||||
media = []
|
self.report_warning('Rate-limit exceeded; falling back to syndication endpoint')
|
||||||
for detail in traverse_obj(status, ((None, 'quoted_tweet'), 'mediaDetails', ..., {dict})):
|
status = self._call_syndication_api(twid)
|
||||||
detail['id_str'] = traverse_obj(detail, (
|
|
||||||
'video_info', 'variants', ..., 'url', {self._MEDIA_ID_RE.search}, 1), get_all=False) or twid
|
|
||||||
media.append(detail)
|
|
||||||
status['extended_entities'] = {'media': media}
|
|
||||||
|
|
||||||
else:
|
if self._selected_api == 'syndication':
|
||||||
raise ExtractorError(f'"{self._selected_api}" is not a valid API selection', expected=True)
|
status = self._call_syndication_api(twid)
|
||||||
|
|
||||||
return traverse_obj(status, 'retweeted_status', None, expected_type=dict) or {}
|
return traverse_obj(status, 'retweeted_status', None, expected_type=dict) or {}
|
||||||
|
|
||||||
|
|
Loading…
Reference in a new issue