Mirroring
/
yt-dlp
mirror of https://github.com/yt-dlp/yt-dlp
1
0
Fork 0
Code Issues Releases Wiki Activity

changed based on D Trombett repo

Browse Source 
c747e15cdf
pull/12742/head
edmundman 6 months ago committed by GitHub
parent 5711fa1dc8
commit 56124b0ac4
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 208 additions and 224 deletions
Show Stats Download Patch File Download Diff File

400
yt_dlp/extractor/europa.py
Unescape Escape View File

@ -1,272 +1,256 @@
# -*- coding: utf-8 -*-
from .common import InfoExtractor
from ..utils import (
ExtractorError,
int_or_none,
orderedSet,
parse_duration,
parse_iso8601,
parse_qs,
qualities,
str_or_none,
traverse_obj,
unified_strdate,
url_or_none,
xpath_text,
js_to_json,
urljoin,
filter_dict,
HEADRequest, # Import HEADRequest
)
import re
import json
import urllib.error # Import urllib.error for HEAD check exception
# --- EuropaIE (Unchanged) ---
class EuropaIE(InfoExtractor):
_WORKING = False
_VALID_URL = r'https?://ec\.europa\.eu/avservices/(?:video/player|audio/audioDetails)\.cfm\?.*?\bref=(?P<id>[A-Za-z0-9-]+)'
_TESTS = [
# Existing tests...
]
_TESTS = [{
'url': 'http://ec.europa.eu/avservices/video/player.cfm?ref=I107758',
'md5': '574f080699ddd1e19a675b0ddf010371',
'info_dict': {
'id': 'I107758',
'ext': 'mp4',
'title': 'TRADE - Wikileaks on TTIP',
'description': 'NEW LIVE EC Midday press briefing of 11/08/2015',
'thumbnail': r're:^https?://.*\.jpg$',
'upload_date': '20150811',
'duration': 34,
'view_count': int,
'formats': 'mincount:3',
},
}, {
'url': 'http://ec.europa.eu/avservices/video/player.cfm?sitelang=en&ref=I107786',
'only_matching': True,
}, {
'url': 'http://ec.europa.eu/avservices/audio/audioDetails.cfm?ref=I-109295&sitelang=en',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
playlist = self._download_xml(
f'http://ec.europa.eu/avservices/video/player/playlist.cfm?ID={video_id}', video_id)
def get_item(type_, preference):
items = {}
for item in playlist.findall(f'./info/{type_}/item'):
lang, label = (xpath_text(item, 'lg', default=None), xpath_text(item, 'label', default=None))
if lang and label: items[lang] = label.strip()
lang, label = xpath_text(item, 'lg', default=None), xpath_text(item, 'label', default=None)
if lang and label:
items[lang] = label.strip()
for p in preference:
if items.get(p): return items[p]
if items.get(p):
return items[p]
query = parse_qs(url)
preferred_lang = query.get('sitelang', ('en', ))[0]
preferred_langs = orderedSet((preferred_lang, 'en', 'int'))
title = get_item('title', preferred_langs) or video_id
description = get_item('description', preferred_langs)
thumbnail = xpath_text(playlist, './info/thumburl', 'thumbnail')
upload_date = unified_strdate(xpath_text(playlist, './info/date', 'upload date'))
duration = parse_duration(xpath_text(playlist, './info/duration', 'duration'))
view_count = int_or_none(xpath_text(playlist, './info/views', 'views'))
language_preference = qualities(preferred_langs[::-1])
formats = []
for file_ in playlist.findall('./files/file'):
video_url = xpath_text(file_, './url')
if not video_url: continue
if not video_url:
continue
lang = xpath_text(file_, './lg')
formats.append({'url': video_url, 'format_id': lang, 'format_note': xpath_text(file_, './lglabel'), 'language_preference': language_preference(lang)})
return {'id': video_id, 'title': title, 'description': description, 'thumbnail': thumbnail, 'upload_date': upload_date, 'duration': duration, 'view_count': view_count, 'formats': formats}
formats.append({
'url': video_url,
'format_id': lang,
'format_note': xpath_text(file_, './lglabel'),
'language_preference': language_preference(lang),
})
return {
'id': video_id,
'title': title,
'description': description,
'thumbnail': thumbnail,
'upload_date': upload_date,
'duration': duration,
'view_count': view_count,
'formats': formats,
}
# --- EuroParlWebstreamIE (Using JSON from iframe) ---
class EuroParlWebstreamIE(InfoExtractor):
_VALID_URL = r'''(?x)
https?://(?:
multimedia\.europarl\.europa\.eu/(?:\w+/)?webstreaming/(?:[\w-]+_)?(?P<id>[\w-]+)| # Webstreaming page URL
live\.media\.eup\.glcloud\.eu/hls/live/(?P<live_id>\d+)/(?P<channel>channel-\d+-\w+|[\w-]+)/(?P<stream_type>index-archive|index|master|playlist|norsk-archive)(?:\.m3u8)? # Direct HLS URL base
)
https?://multimedia\.europarl\.europa\.eu/
(?P<lang>[^/]*/)?webstreaming/(?:[^_]*_)?(?P<id>[\w-]+)
'''
_TESTS = [
{
'url': 'https://multimedia.europarl.europa.eu/en/webstreaming/committee-on-agriculture-and-rural-development_20250327-0900-COMMITTEE-AGRI',
_TESTS = [{
'url': 'https://multimedia.europarl.europa.eu/pl/webstreaming/plenary-session_20220914-0900-PLENARY',
'md5': '16420ad9c602663759538ac1ca16a8db',
'info_dict': {
'id': '20250327-0900-COMMITTEE-AGRI',
'title': r're:^Committee on Agriculture and Rural Development \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
'is_live': False,
'id': '20220914-0900-PLENARY',
'ext': 'mp4',
'title': 'Plenary session',
'description': '',
'duration': 45147,
'thumbnail': 'https://storage.eup.glcloud.eu/thumbnail/default_thumbnail.png',
'release_timestamp': 1663139069,
'release_date': '20220914',
'modified_timestamp': 1663650921,
'modified_date': '20220920',
'live_status': 'was_live',
},
'params': {'skip_download': True},
# Uses the iframe JSON parsing which should yield 2113752 / channel-06-bxl
}, {
'url': 'https://multimedia.europarl.europa.eu/en/webstreaming/euroscola_20221115-1000-SPECIAL-EUROSCOLA',
'md5': '8b4304f9e15a6e133100248fb55a5dce',
'info_dict': {
'ext': 'mp4',
'id': '20221115-1000-SPECIAL-EUROSCOLA',
'release_timestamp': 1668502798,
'title': 'Euroscola',
'release_date': '20221115',
'live_status': 'was_live',
'description': '',
'duration': 9587,
'thumbnail': 'https://storage.eup.glcloud.eu/thumbnail/default_thumbnail.png',
'modified_timestamp': 1668945274,
'modified_date': '20221120',
},
{
'url': 'https://multimedia.europarl.europa.eu/en/webstreaming/pre-session-briefing_20250328-1100-SPECIAL-PRESSEr',
}, {
'url': 'https://multimedia.europarl.europa.eu/en/webstreaming/committee-on-culture-and-education_20230301-1130-COMMITTEE-CULT',
'md5': '0ca01cf33009d866e6f5e1cd3088c10c',
'info_dict': {
'id': '20250328-1100-SPECIAL-PRESSEr',
'title': r're:^Pre-session briefing \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
'is_live': False,
'id': '20230301-1130-COMMITTEE-CULT',
'ext': 'mp4',
'release_date': '20230301',
'title': 'Committee on Culture and Education',
'release_timestamp': 1677666641,
'description': 'Committee on Culture and Education',
'duration': 1003,
'thumbnail': 'https://storage.eup.glcloud.eu/thumbnail/default_thumbnail.png',
'modified_timestamp': 1732475771,
'modified_date': '20241124',
'live_status': 'was_live',
},
'params': {'skip_download': True},
# Uses the iframe JSON parsing which should yield 2113747 / channel-01-bxl
}, {
'url': 'https://multimedia.europarl.europa.eu/en/webstreaming/committee-on-environment-public-health-and-food-safety_20230524-0900-COMMITTEE-ENVI',
'md5': 'f2e8c30935f956a7165c2f4f4b4ee090',
'info_dict': {
'id': '20230524-0900-COMMITTEE-ENVI',
'ext': 'mp4',
'release_date': '20230524',
'title': 'Committee on Environment, Public Health and Food Safety',
'release_timestamp': 1684912288,
'live_status': 'was_live',
'description': 'Committee on Environment, Public Health and Food Safety',
'duration': 4831,
'thumbnail': 'https://storage.eup.glcloud.eu/thumbnail/default_thumbnail.png',
'modified_timestamp': 1732475771,
'modified_date': '20241124',
},
{ # Test direct HLS URL with archive times
'url': 'https://live.media.eup.glcloud.eu/hls/live/2113752/channel-06-bxl/index-archive.m3u8?startTime=1743068400&endTime=1743079800',
}, {
'url': 'https://multimedia.europarl.europa.eu/en/webstreaming/20240320-1345-SPECIAL-PRESSER',
'md5': '518758eb706471c4c4ef3a134034a5bd',
'info_dict': {
'id': 'index-archive',
'title': 'European Parliament Stream 2113752/channel-06-bxl',
'is_live': False, # Should be detected as not live from lack of live tags/duration
'id': '20240320-1345-SPECIAL-PRESSER',
'ext': 'mp4',
'release_date': '20240320',
'title': 'md5:7c6c814cac55dea5e2d87bf8d3db2234',
'release_timestamp': 1710939767,
'description': 'md5:7c6c814cac55dea5e2d87bf8d3db2234',
'duration': 927,
'thumbnail': 'https://storage.eup.glcloud.eu/thumbnail/default_thumbnail.png',
'modified_timestamp': 1732475771,
'modified_date': '20241124',
'live_status': 'was_live',
},
'params': {'skip_download': True},
}, {
'url': 'https://multimedia.europarl.europa.eu/en/webstreaming/20250328-1600-SPECIAL-PRESSER',
'md5': 'dd1c5e67eb55e609998583d7c2966105',
'info_dict': {
'id': '20250328-1600-SPECIAL-PRESSER',
'ext': 'mp4',
'title': 'md5:04a2ab70c183dabe891a7cd190c3121d',
'description': '',
'duration': 1023,
'thumbnail': 'https://storage.eup.glcloud.eu/thumbnail/default_thumbnail.png',
'release_timestamp': 1743177199,
'release_date': '20250328',
'modified_timestamp': 1743180924,
'modified_date': '20250328',
'live_status': 'was_live',
},
# Potentially add a known live stream test if one is available
]
def _log_debug(self, msg):
self.to_screen(f"[EuroParlWebstream] {msg}")
def _extract_title_from_webpage(self, webpage, display_id):
"""Extracts title from the main webstreaming page."""
title_element = self._search_regex(r'<h1[^>]*>(.*?)</h1>', webpage, 'title element', default=None)
if title_element:
# Clean up potential extra whitespace and HTML entities
title = re.sub(r'\s+', ' ', title_element).strip()
title = self._html_search_meta(['og:title', 'twitter:title'], webpage, default=title)
else:
# Fallback using meta tags or just the ID
title = self._html_search_meta(
['og:title', 'twitter:title'], webpage, default=display_id)
return title.replace('_', ' ') # Replace underscores often used in IDs
def _perform_head_check(self, url, display_id, note=''):
"""Performs a HEAD request to check if the HLS URL likely exists."""
self._log_debug(f'[{display_id}] Performing HEAD check {note}on: {url}')
try:
self._request_webpage(HEADRequest(url), display_id, note=f'HEAD check {note}')
self._log_debug(f'[{display_id}] HEAD check {note}successful.')
return True
except ExtractorError as e:
# Specifically catch HTTP errors, especially 404
if isinstance(e.cause, urllib.error.HTTPError):
self._log_debug(f'[{display_id}] HEAD check {note}failed: {e.cause.code} {e.cause.reason}')
else:
self._log_debug(f'[{display_id}] HEAD check {note}failed: {e}')
return False
}, {
'url': 'https://multimedia.europarl.europa.eu/webstreaming/briefing-for-media-on-2024-european-elections_20240429-1000-SPECIAL-OTHER',
'only_matching': True,
}]
def _real_extract(self, url):
mobj = self._match_valid_url(url)
display_id = mobj.group('id')
live_id_direct = mobj.group('live_id')
# --- Handle Direct HLS URL Input ---
if live_id_direct:
self._log_debug(f"Processing Direct HLS URL: {url}")
channel_direct = mobj.group('channel')
stream_type_direct = mobj.group('stream_type') or 'stream' # Default name if not specified
base_url = f'https://live.media.eup.glcloud.eu/hls/live/{live_id_direct}/{channel_direct}/{stream_type_direct}'
query_params_str = mobj.group(0).split('?', 1)[1] if '?' in mobj.group(0) else None
query_params = parse_qs(query_params_str) if query_params_str else {}
start_time_direct = traverse_obj(query_params, ('startTime', 0, {int_or_none}))
end_time_direct = traverse_obj(query_params, ('endTime', 0, {int_or_none}))
# Construct the final URL ensuring .m3u8 is present
final_url = base_url + ('' if base_url.endswith('.m3u8') else '.m3u8')
if start_time_direct and end_time_direct:
final_url += f"?startTime={start_time_direct}&endTime={end_time_direct}"
elif query_params_str: # Append original query if not start/end time based
final_url += f"?{query_params_str}"
# Basic title for direct URL
title = f'European Parliament Stream {live_id_direct}/{channel_direct}'
# HEAD check is good even for direct URLs
if not self._perform_head_check(final_url, f"{live_id_direct}-{channel_direct}", '(direct)'):
raise ExtractorError(f'Direct HLS URL HEAD check failed: {final_url}', expected=True)
formats, subtitles = self._extract_m3u8_formats_and_subtitles(
final_url, display_id or stream_type_direct, 'mp4', m3u8_id='hls', fatal=True)
if not formats: raise ExtractorError(f'Could not extract formats from direct HLS URL: {final_url}', expected=True)
return {
'id': display_id or stream_type_direct,
'title': title,
'formats': formats,
'subtitles': subtitles,
'is_live': not (start_time_direct and end_time_direct) and '.m3u8' not in stream_type_direct # Guess based on URL structure
lang, video_id = self._match_valid_url(url).group('lang', 'id')
query = {
'lang': lang,
'audio': lang,
'autoplay': 'true',
'logo': 'false',
'muted': 'false',
'fullscreen': 'true',
'disclaimer': 'false',
'multicast': 'true',
'analytics': 'false',
}
# --- Handle Webstreaming Page URL ---
if not display_id: raise ExtractorError('Could not parse display ID from URL', expected=True)
self._log_debug(f"Processing Webstreaming Page: {display_id}")
webpage = self._download_webpage(url, display_id)
title = self._extract_title_from_webpage(webpage, display_id) # Get title early
self._log_debug(f'[{display_id}] Extracting metadata and iframe URL...')
nextjs_data = self._search_nextjs_data(webpage, display_id, default={})
media_info = traverse_obj(nextjs_data, ('props', 'pageProps', 'mediaItem')) or {}
# Get initial start time, but prioritize iframe JSON later
initial_start_timestamp = traverse_obj(media_info, ('mediaDate', {parse_iso8601}, {int_or_none}))
iframe_url = traverse_obj(media_info, 'iframeUrls') # Usually just one URL string
self._log_debug(f'[{display_id}] Initial Start Time={initial_start_timestamp}, Iframe URL={iframe_url}')
if not iframe_url:
raise ExtractorError(f'[{display_id}] Could not find iframe URL in page metadata.', expected=True)
# --- Attempt Extraction from Iframe JSON ---
self._log_debug(f'[{display_id}] Attempting extraction from iframe: {iframe_url}')
try:
iframe_content = self._download_webpage(iframe_url, display_id, note='Downloading iframe content')
json_data_str = self._search_regex(
r'<script id="ng-state" type="application/json"[^>]*>\s*({.+?})\s*</script>',
iframe_content, 'iframe JSON data', default=None)
if not json_data_str:
raise ExtractorError('Could not find ng-state JSON in iframe content.')
iframe_json = self._parse_json(json_data_str, display_id, fatal=True)
# Extract required info from the JSON structure
player_url_base = traverse_obj(iframe_json, ('contentEventKey', 'playerUrl'))
start_time = traverse_obj(iframe_json, ('contentEventKey', 'startTime', {int_or_none}))
end_time = traverse_obj(iframe_json, ('contentEventKey', 'endTime', {int_or_none}))
is_live = traverse_obj(iframe_json, ('contentEventKey', 'live')) # boolean
# Use title from JSON if available and seems better
json_title = traverse_obj(iframe_json, ('contentEventKey', 'title'))
if json_title: title = json_title
self._log_debug(f'[{display_id}] Found in iframe JSON: playerUrl={player_url_base}, startTime={start_time}, endTime={end_time}, is_live={is_live}')
if not player_url_base:
raise ExtractorError('Could not extract playerUrl from iframe JSON.')
# For recorded streams (archives), startTime and endTime are essential
if not is_live and (start_time is None or end_time is None):
raise ExtractorError('Missing startTime or endTime in iframe JSON for recorded stream.')
# Construct the final URL
# Ensure base URL doesn't already have query params before adding ours
player_url_base = player_url_base.split('?')[0]
if not player_url_base.endswith('.m3u8'):
player_url_base += '.m3u8' # Ensure correct extension
if is_live:
final_player_url = player_url_base # Live streams don't use start/end times
webpage = self._download_webpage(f'https://control.eup.glcloud.eu/content-manager/content-page/{video_id}',
video_id, 'Downloading iframe', query=query)
stream_info = self._search_json(r'<script [^>]*id="ng-state"[^>]*>', webpage, 'stream info', video_id)['contentEventKey']
player_url = stream_info.get('playerUrl')
# status = traverse_obj(stream_info, ('media_item', 'mediaSubType'))
# base = 'https://control.eup.glcloud.eu/content-manager/api/v1/socket.io/?EIO=4&transport=polling'
# headers = {'referer': f'https://control.eup.glcloud.eu/content-manager/content-page/{video_id}'}
# sid = self._download_socket_json(base, video_id, note='Opening socket', headers=headers)['sid']
# base += '&sid=' + sid
# self._download_webpage(base, video_id, 'Polling socket with payload', data=b'40/content,', headers=headers)
# self._download_webpage(base, video_id, 'Polling socket', headers=headers)
# self._download_socket_json(base, video_id, 'Getting broadcast metadata from socket', headers=headers)
if player_url:
live_status = 'was_live'
query = None if stream_info.get('finalVod') else traverse_obj(stream_info, {
'startTime': ('startTime', {str_or_none}),
'endTime': ('endTime', {str_or_none}),
})
formats, subtitles = self._extract_m3u8_formats_and_subtitles(player_url, video_id, query=query, ext='mp4')
else:
final_player_url = f"{player_url_base}?startTime={start_time}&endTime={end_time}"
# Perform HEAD check on the constructed URL
if not self._perform_head_check(final_player_url, display_id, '(dynamic)'):
raise ExtractorError(f'Dynamic HLS URL from iframe failed HEAD check: {final_player_url}')
# Extract formats
self._log_debug(f'[{display_id}] Extracting formats from {final_player_url}')
formats, subtitles = self._extract_m3u8_formats_and_subtitles(
final_player_url, display_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=True) # Use fatal=True, if extraction fails, it's an error
if not formats:
raise ExtractorError(f'Could not extract M3U8 formats from {final_player_url}', expected=True)
formats = None
subtitles = None
live_status = 'is_upcoming'
self.raise_no_formats('Stream didn\'t start yet', True, video_id)
if stream_info.get('live'):
live_status = 'is_live'
return {
'id': display_id,
'title': title,
'formats': formats,
'subtitles': subtitles,
'is_live': is_live,
'timestamp': start_time if not is_live else None, # Use JSON start time for VOD
'duration': (end_time - start_time) if not is_live and start_time and end_time else None,
'live_status': live_status,
**traverse_obj(stream_info, {
'id': ('commonId', {str_or_none}),
'title': ('title', {str_or_none}),
'description': ('description', {str_or_none}),
'release_timestamp': ('startTime', {int_or_none}),
'duration': ('endTime', {lambda e: e and (s := stream_info.get('startTime')) and (e - s)}),
'thumbnail': ('posterFrame', {url_or_none}),
'modified_timestamp': ('meta', 'updatedAt', {parse_iso8601}),
}),
}
except ExtractorError as e:
# Re-raise specific extractor errors
raise e
except Exception as e:
# Wrap unexpected errors
raise ExtractorError(f'[{display_id}] Error processing iframe content: {e}', cause=e)
# This part should ideally not be reached if iframe extraction is mandatory
raise ExtractorError(f'[{display_id}] Failed to extract stream information from iframe.', expected=True)

Write Preview
Loading…
Cancel
Save