Compare commits

...

4 Commits

Author SHA1 Message Date
Simon Sawicki
716492e82d
Improve readme even more than more 2023-10-02 22:46:48 +02:00
Simon Sawicki
5e867eb794
Improve readme even more 2023-10-02 21:40:30 +02:00
Simon Sawicki
636751f1fd
Improve readme 2023-10-02 21:33:30 +02:00
Simon Sawicki
c3e9646ecc
Rework to use flat loop 2023-10-02 21:00:33 +02:00
2 changed files with 57 additions and 49 deletions

View File

@ -1809,7 +1809,7 @@ The following extractors use this feature:
* `formats`: Change the types of formats to return. `dashy` (convert HTTP to DASH), `duplicate` (identical content but different URLs or protocol; includes `dashy`), `incomplete` (cannot be downloaded completely - live dash and post-live m3u8)
* `innertube_host`: Innertube API host to use for all API requests; e.g. `studio.youtube.com`, `youtubei.googleapis.com`. Note that cookies exported from one subdomain will not work on others
* `innertube_key`: Innertube API key to use for all API requests
* `raise_incomplete_data`: `Incomplete Data` error should raise an error instead of a warning.
* `raise_incomplete_data`: `Incomplete Data Received` raises an error instead of reporting a warning
#### youtubetab (YouTube playlists, channels, feeds, etc.)
* `skip`: One or more of `webpage` (skip initial webpage download), `authcheck` (allow the download of playlists requiring authentication when no initial webpage is downloaded. This may cause unwanted behavior, see [#1122](https://github.com/yt-dlp/yt-dlp/pull/1122) for more details)

View File

@ -941,57 +941,65 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None,
ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None,
default_client='web'):
for icd_retry in self.RetryManager(
fatal=self._configuration_arg('raise_incomplete_data', [False], ie_key=YoutubeIE)[0] is not False
):
for retry in self.RetryManager():
try:
response = self._call_api(
ep=ep, fatal=True, headers=headers,
video_id=item_id, query=query, note=note,
context=self._extract_context(ytcfg, default_client),
api_key=self._extract_api_key(ytcfg, default_client),
api_hostname=api_hostname, default_client=default_client)
except ExtractorError as e:
if not isinstance(e.cause, network_exceptions):
return self._error_or_warning(e, fatal=fatal)
elif not isinstance(e.cause, HTTPError):
retry.error = e
continue
first_bytes = e.cause.response.read(512)
if not is_html(first_bytes):
yt_error = try_get(
self._parse_json(
self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
lambda x: x['error']['message'], str)
if yt_error:
self._report_alerts([('ERROR', yt_error)], fatal=False)
# Downloading page may result in intermittent 5xx HTTP error
# Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
# We also want to catch all other network exceptions since errors in later pages can be troublesome
# See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
if e.cause.status not in (403, 429):
retry.error = e
continue
raise_for_incomplete = bool(self._configuration_arg('raise_incomplete_data', ie_key=YoutubeIE))
icd_retries = iter(self.RetryManager(fatal=raise_for_incomplete))
icd_rm = next(icd_retries)
main_retries = iter(self.RetryManager())
main_rm = next(main_retries)
for _ in range(main_rm.retries + icd_rm.retries + 1):
try:
response = self._call_api(
ep=ep, fatal=True, headers=headers,
video_id=item_id, query=query, note=note,
context=self._extract_context(ytcfg, default_client),
api_key=self._extract_api_key(ytcfg, default_client),
api_hostname=api_hostname, default_client=default_client)
except ExtractorError as e:
if not isinstance(e.cause, network_exceptions):
return self._error_or_warning(e, fatal=fatal)
try:
self._extract_and_report_alerts(response, only_once=True)
except ExtractorError as e:
# YouTube servers may return errors we want to retry on in a 200 OK response
# See: https://github.com/yt-dlp/yt-dlp/issues/839
if 'unknown error' in e.msg.lower():
retry.error = e
continue
return self._error_or_warning(e, fatal=fatal)
# Youtube sometimes sends incomplete data
# See: https://github.com/ytdl-org/youtube-dl/issues/28194
if not traverse_obj(response, *variadic(check_get_keys)):
icd_retry.error = ExtractorError('Incomplete data received', expected=True)
elif not isinstance(e.cause, HTTPError):
main_rm.error = e
next(main_retries)
continue
return response
first_bytes = e.cause.response.read(512)
if not is_html(first_bytes):
yt_error = try_get(
self._parse_json(
self._webpage_read_content(e.cause.response, None, item_id, prefix=first_bytes) or '{}', item_id, fatal=False),
lambda x: x['error']['message'], str)
if yt_error:
self._report_alerts([('ERROR', yt_error)], fatal=False)
# Downloading page may result in intermittent 5xx HTTP error
# Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
# We also want to catch all other network exceptions since errors in later pages can be troublesome
# See https://github.com/yt-dlp/yt-dlp/issues/507#issuecomment-880188210
if e.cause.status not in (403, 429):
main_rm.error = e
next(main_retries)
continue
return self._error_or_warning(e, fatal=fatal)
try:
self._extract_and_report_alerts(response, only_once=True)
except ExtractorError as e:
# YouTube servers may return errors we want to retry on in a 200 OK response
# See: https://github.com/yt-dlp/yt-dlp/issues/839
if 'unknown error' in e.msg.lower():
main_rm.error = e
next(main_retries)
continue
return self._error_or_warning(e, fatal=fatal)
# Youtube sometimes sends incomplete data
# See: https://github.com/ytdl-org/youtube-dl/issues/28194
if not traverse_obj(response, *variadic(check_get_keys)):
icd_rm.error = ExtractorError('Incomplete data received', expected=True)
should_retry = next(icd_retries, None)
if not should_retry:
return None
continue
return response
@staticmethod
def is_music_url(url):