github/youtube-dl
1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2025-08-23 06:35:51 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

_real_extract() now iterates through possible manifest URLS to find the m3u8.

Browse source
This commit is contained in:
Shiyang Zhu 2022-12-10 15:16:03 -05:00
commit 52524e85f4
1 changed files with 92 additions and 73 deletions
Download patch file Download diff file Expand all files Collapse all files

101
youtube_dl/extractor/senateisvp.py
View file

@ -5,15 +5,16 @@ import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
url_or_none,
parse_duration,
unsmuggle_url, unsmuggle_url,
) )
from ..compat import ( from ..compat import (
compat_parse_qs, compat_parse_qs,
compat_urlparse,
) )
class SenateISVPIE(InfoExtractor): class SenateISVPIE(InfoExtractor):
# committee --> [stream_number, stream_domain, stream_id, msl3]
_COMM_MAP = { _COMM_MAP = {
'ag': [76440, 'https://ag-f.akamaihd.net', 2036803, 'agriculture'], 'ag': [76440, 'https://ag-f.akamaihd.net', 2036803, 'agriculture'],
'aging': [76442, 'https://aging-f.akamaihd.net', 2036801, 'aging'], 'aging': [76442, 'https://aging-f.akamaihd.net', 2036801, 'aging'],
@ -37,7 +38,7 @@ class SenateISVPIE(InfoExtractor):
'intlnarc': [76457, 'https://intlnarc-f.akamaihd.net', None, 'internationalnarcoticscaucus'], 'intlnarc': [76457, 'https://intlnarc-f.akamaihd.net', None, 'internationalnarcoticscaucus'],
'jccic': [85180, 'https://jccic-f.akamaihd.net', 2036778, 'jccic'], 'jccic': [85180, 'https://jccic-f.akamaihd.net', 2036778, 'jccic'],
'jec': [76458, 'https://jec-f.akamaihd.net', 2036789, 'jointeconomic'], 'jec': [76458, 'https://jec-f.akamaihd.net', 2036789, 'jointeconomic'],
'judiciary': [76459, 'https://judiciary-f.akamaihd.net', 2036788,'judiciary'], 'judiciary': [76459, 'https://judiciary-f.akamaihd.net', 2036788, 'judiciary'],
'rpc': [76591, 'https://rpc-f.akamaihd.net', None, 'rpc'], 'rpc': [76591, 'https://rpc-f.akamaihd.net', None, 'rpc'],
'rules': [76460, 'https://rules-f.akamaihd.net', 2036787, 'rules'], 'rules': [76460, 'https://rules-f.akamaihd.net', 2036787, 'rules'],
'saa': [76489, 'https://srs-f.akamaihd.net', 2036780, 'srs_saa'], 'saa': [76489, 'https://srs-f.akamaihd.net', 2036780, 'srs_saa'],
@ -46,9 +47,10 @@ class SenateISVPIE(InfoExtractor):
'uscc': [76487, 'https://srs-f.akamaihd.net', 2036781, 'srs_uscc'], 'uscc': [76487, 'https://srs-f.akamaihd.net', 2036781, 'srs_uscc'],
'vetaff': [76462, 'https://vetaff-f.akamaihd.net', 2036785, 'veteransaffairs'], 'vetaff': [76462, 'https://vetaff-f.akamaihd.net', 2036785, 'veteransaffairs'],
'arch': [None, 'https://ussenate-f.akamaihd.net/', None, None], 'arch': [None, 'https://ussenate-f.akamaihd.net/', None, None],
'uscp': [None, '', 2043685, None, 'uscp'], 'uscp': [None, '', 2043685, 'uscp'],
'cio': [None, '', 2043686, None, 'cio'], 'cio': [None, '', 2043686, 'cio'],
} }
_IE_NAME = 'senate.gov' _IE_NAME = 'senate.gov'
_VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)' _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
_TESTS = [{ _TESTS = [{
@ -56,7 +58,7 @@ class SenateISVPIE(InfoExtractor):
'info_dict': { 'info_dict': {
'id': 'judiciary031715', 'id': 'judiciary031715',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Integrated Senate Video Player', 'title': 'judiciary031715',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$', 'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
}, },
'params': { 'params': {
@ -69,7 +71,8 @@ class SenateISVPIE(InfoExtractor):
'info_dict': { 'info_dict': {
'id': 'intel090613', 'id': 'intel090613',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Integrated Senate Video Player' 'title': 'intel090613',
'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
} }
}, { }, {
# From http://www.c-span.org/video/?96791-1 # From http://www.c-span.org/video/?96791-1
@ -78,67 +81,83 @@ class SenateISVPIE(InfoExtractor):
}] }]
@staticmethod @staticmethod
# returns url from an iframe
def _search_iframe_url(webpage): def _search_iframe_url(webpage):
mobj = re.search( mobj = re.search(
r"<iframe[^>]+src=['\"](?P<url>https?://www\.senate\.gov/isvp/?\?[^'\"]+)['\"]", r'''<iframe\b[^>]+\bsrc\s*=\s*(['"])(?P<url>https?://www\.senate\.gov/isvp/?\?(?:(?!\1)\S)+)''',
webpage) webpage)
if mobj: if mobj:
return mobj.group('url') return mobj.group('url')
# returns stream_number, stream_domain, stream_id, and msl3
def _get_info_for_comm(self, committee): def _get_info_for_comm(self, committee):
for entry in self._COMM_MAP: return self._COMM_MAP[committee][0:]
if entry[0] == committee:
return entry[1:]
def _real_extract(self, url): def _real_extract(self, url):
# smuggled_data may contain a forced title that should be used
url, smuggled_data = unsmuggle_url(url, {}) url, smuggled_data = unsmuggle_url(url, {})
qs = compat_parse_qs(re.match(self._VALID_URL, url).group('qs')) qs = compat_parse_qs(re.match(self._VALID_URL, url).group('qs'))
if not qs.get('filename') or not qs.get('type') or not qs.get('comm'):
raise ExtractorError('Invalid URL', expected=True)
video_id = re.sub(r'.mp4$', '', qs['filename'][0]) # error handling for invalid URL - specify what
if not qs.get('comm'):
raise ExtractorError('''Invalid URL. Missing committee in query parameters.''', expected=True)
if not qs.get('filename'):
raise ExtractorError('''Invalid URL. Missing filename in query parameters.''', expected=True)
webpage = self._download_webpage(url, video_id) committee = qs.get('comm')[0]
filename = qs.get('filename')[0]
video_id = re.sub(r'\.mp4$', '', filename)
if smuggled_data.get('force_title'): # there is no point in pulling the title from the webpage since it always defaults to 'Integrated Senate Player'
title = filename
if smuggled_data:
title = smuggled_data['force_title'] title = smuggled_data['force_title']
else: stream_number, stream_domain, stream_id, msl3 = self._get_info_for_comm(committee)
title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, video_id)
poster = qs.get('poster')
thumbnail = poster[0] if poster else None
video_type = qs['type'][0] # the possible locations that the video could be contained at.
committee = video_type if video_type == 'arch' else qs['comm'][0] possible_manifest_urls = [
stream_num, domain = self._get_info_for_comm(committee) f'https://www-senate-gov-media-srs.akamaized.net/hls/live/{stream_id}/{committee}/{filename}/master.m3u8',
f'https://www-senate-gov-msl3archive.akamaized.net/{msl3}/{filename}_1/master.m3u8',
f'{stream_domain}/i/{filename}_1@{stream_number}/master.m3u8',
f'https://ussenate-f.akamaihd.net/i/{video_id}',
]
# we iterate through the possible locations until we find a hit, that is when formats is filled.
formats = [] formats = []
if video_type == 'arch': for url in possible_manifest_urls:
filename = video_id if '.' in video_id else video_id + '.mp4' print('!!!')
formats = [{ entries = self._extract_m3u8_formats(
# All parameters in the query string are necessary to prevent a 403 error url,
'url': compat_urlparse.urljoin(domain, filename) + '?v=3.1.0&fp=&r=&g=', video_id,
}] ext='mp4',
else: m3u8_id='hls',
hdcore_sign = 'hdcore=3.1.0' entry_protocol='m3u8_native',
url_params = (domain, video_id, stream_num) fatal=False,
f4m_url = '%s/z/%s_1@%s/manifest.f4m?' % url_params + hdcore_sign )
m3u8_url = '%s/i/%s_1@%s/master.m3u8' % url_params print('|-|-|-|')
for entry in self._extract_f4m_formats(f4m_url, video_id, f4m_id='f4m'):
# URLs without the extra param induce an 404 error for entry in entries:
entry.update({'extra_param_to_segment_url': hdcore_sign}) mobj = re.search(r'(?P<tag>-[pb]).m3u8', entry['url'])
formats.append(entry)
for entry in self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4', m3u8_id='m3u8'):
mobj = re.search(r'(?P<tag>(?:-p|-b)).m3u8', entry['url'])
if mobj: if mobj:
entry['format_id'] += mobj.group('tag') entry['format_id'] += mobj.group('tag')
formats.append(entry) formats.append(entry)
if formats:
break
self._sort_formats(formats) self._sort_formats(formats)
thumbnail = url_or_none(qs.get('poster', [None])[-1])
start_time = parse_duration(qs.get('stt', [None])[-1])
stop_time = parse_duration(qs.get('dur', [None])[-1])
if stop_time is not None:
stop_time += start_time or 0
return { return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'formats': formats, 'formats': formats,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'start_time': start_time,
'stop_time': stop_time,
} }