paradizelost/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Updated _real_extract() to iterate through possible manifest URLs to find the m3u8. Added start_time and stop_time metadata.

Browse Source
This commit is contained in:
jsmnhou 2023-04-09 16:10:06 -04:00
parent 213d1d91bf
commit d59b83f04b
1 changed files with 51 additions and 36 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

85
youtube_dl/extractor/senateisvp.py
View File

@ -5,7 +5,9 @@ import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
ExtractorError, ExtractorError,
parse_duration,
unsmuggle_url, unsmuggle_url,
url_or_none,
) )
from ..compat import ( from ..compat import (
compat_parse_qs, compat_parse_qs,
@ -87,67 +89,80 @@ 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, 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$', '', qs['filename'][0]) # error handling for invalid URL - specify which error
if not qs.get('filename'):
raise ExtractorError('Invalid URL. Missing filename in query parameters', expected=True)
if not qs.get('comm'):
raise ExtractorError('Invalid URL. Missing committee in query parameters', expected=True)
webpage = self._download_webpage(url, video_id) committee = qs.get('comm')[0]
filename = qs.get('filename')[0]
video_id = re.sub(r'\.mp4$', '', 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 = smuggled_data['force_title'] title = smuggled_data.get('force_title') or filename
else:
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] # extract more info about committee (for matching to possible locations)
committee = video_type if video_type == 'arch' else qs['comm'][0] stream_number, stream_domain, stream_id, msl3 = self._get_info_for_comm(committee)
stream_num, domain = self._get_info_for_comm(committee)
# possible locations that m3u8 could be located at
possible_manifest_urls = [
'https://www-senate-gov-media-srs.akamaized.net/hls/live/%d/%s/%s/master.m3u8' % (stream_id, committee, filename),
'https://www-senate-gov-msl3archive.akamaized.net/%s/%s_1/master.m3u8' % (msl3, filename),
'{stream_domain}/i/%s_1@%d/master.m3u8' % (filename, stream_number),
'https://ussenate-f.akamaihd.net/i/%s' % video_id,
]
# iterate through possible locations until we find a match (match found 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' entries = self._extract_m3u8_formats(
formats = [{ url,
# All parameters in the query string are necessary to prevent a 403 error video_id,
'url': compat_urlparse.urljoin(domain, filename) + '?v=3.1.0&fp=&r=&g=', ext='mp4',
}] m3u8_id='hls',
else: entry_protocal='mu38_native',
hdcore_sign = 'hdcore=3.1.0' fatal=False
url_params = (domain, video_id, stream_num) )
f4m_url = '%s/z/%s_1@%s/manifest.f4m?' % url_params + hdcore_sign
m3u8_url = '%s/i/%s_1@%s/master.m3u8' % url_params for entry in entries:
for entry in self._extract_f4m_formats(f4m_url, video_id, f4m_id='f4m'): mobj = re.search(r'(?P<tag>-[pb]).m3u8', entry['url'])
# URLs without the extra param induce an 404 error
entry.update({'extra_param_to_segment_url': hdcore_sign})
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,
} }