mirror of
https://github.com/yt-dlp/yt-dlp
synced 2024-11-18 10:05:58 +01:00
ArteTvIE: rewrite the extract process to support the new site (fixes #875)
The video can be downloaded with rtmp or http, but the best quality format seems to always use rtmp. Deleted the old methods.
This commit is contained in:
Jaime Marquínez Ferrándiz
parent
bcd606c0fe
commit
75c9481224
1 changed files with 37 additions and 87 deletions
|
|
@ -1,53 +1,21 @@
|
|||
import re
|
||||
import socket
|
||||
import json
|
||||
|
||||
from .common import InfoExtractor
|
||||
from ..utils import (
|
||||
compat_http_client,
|
||||
compat_str,
|
||||
compat_urllib_error,
|
||||
# This is used by the not implemented extractLiveStream method
|
||||
compat_urllib_parse,
|
||||
compat_urllib_request,
|
||||
|
||||
ExtractorError,
|
||||
unified_strdate,
|
||||
)
|
||||
|
||||
class ArteTvIE(InfoExtractor):
|
||||
"""arte.tv information extractor."""
|
||||
|
||||
_VALID_URL = r'(?:http://)?videos\.arte\.tv/(?:fr|de)/videos/.*'
|
||||
_VALID_URL = r'(?:http://)?www\.arte.tv/guide/(?:fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
|
||||
_LIVE_URL = r'index-[0-9]+\.html$'
|
||||
|
||||
IE_NAME = u'arte.tv'
|
||||
|
||||
def fetch_webpage(self, url):
|
||||
request = compat_urllib_request.Request(url)
|
||||
try:
|
||||
self.report_download_webpage(url)
|
||||
webpage = compat_urllib_request.urlopen(request).read()
|
||||
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
|
||||
raise ExtractorError(u'Unable to retrieve video webpage: %s' % compat_str(err))
|
||||
except ValueError as err:
|
||||
raise ExtractorError(u'Invalid URL: %s' % url)
|
||||
return webpage
|
||||
|
||||
def grep_webpage(self, url, regex, regexFlags, matchTuples):
|
||||
page = self.fetch_webpage(url)
|
||||
mobj = re.search(regex, page, regexFlags)
|
||||
info = {}
|
||||
|
||||
if mobj is None:
|
||||
raise ExtractorError(u'Invalid URL: %s' % url)
|
||||
|
||||
for (i, key, err) in matchTuples:
|
||||
if mobj.group(i) is None:
|
||||
raise ExtractorError(err)
|
||||
else:
|
||||
info[key] = mobj.group(i)
|
||||
|
||||
return info
|
||||
|
||||
# TODO implement Live Stream
|
||||
# def extractLiveStream(self, url):
|
||||
# video_lang = url.split('/')[-4]
|
||||
|
|
@ -75,62 +43,44 @@ class ArteTvIE(InfoExtractor):
|
|||
# )
|
||||
# video_url = u'%s/%s' % (info.get('url'), info.get('path'))
|
||||
|
||||
def extractPlus7Stream(self, url):
|
||||
video_lang = url.split('/')[-3]
|
||||
info = self.grep_webpage(
|
||||
url,
|
||||
r'param name="movie".*?videorefFileUrl=(http[^\'"&]*)',
|
||||
0,
|
||||
[
|
||||
(1, 'url', u'Invalid URL: %s' % url)
|
||||
]
|
||||
)
|
||||
next_url = compat_urllib_parse.unquote(info.get('url'))
|
||||
info = self.grep_webpage(
|
||||
next_url,
|
||||
r'<video lang="%s" ref="(http[^\'"&]*)' % video_lang,
|
||||
0,
|
||||
[
|
||||
(1, 'url', u'Could not find <video> tag: %s' % url)
|
||||
]
|
||||
)
|
||||
next_url = compat_urllib_parse.unquote(info.get('url'))
|
||||
|
||||
info = self.grep_webpage(
|
||||
next_url,
|
||||
r'<video id="(.*?)".*?>.*?' +
|
||||
'<name>(.*?)</name>.*?' +
|
||||
'<dateVideo>(.*?)</dateVideo>.*?' +
|
||||
'<url quality="hd">(.*?)</url>',
|
||||
re.DOTALL,
|
||||
[
|
||||
(1, 'id', u'could not extract video id: %s' % url),
|
||||
(2, 'title', u'could not extract video title: %s' % url),
|
||||
(3, 'date', u'could not extract video date: %s' % url),
|
||||
(4, 'url', u'could not extract video url: %s' % url)
|
||||
]
|
||||
)
|
||||
|
||||
return {
|
||||
'id': info.get('id'),
|
||||
'url': compat_urllib_parse.unquote(info.get('url')),
|
||||
'uploader': u'arte.tv',
|
||||
'upload_date': unified_strdate(info.get('date')),
|
||||
'title': info.get('title').decode('utf-8'),
|
||||
'ext': u'mp4',
|
||||
'format': u'NA',
|
||||
'player_url': None,
|
||||
}
|
||||
|
||||
def _real_extract(self, url):
|
||||
video_id = url.split('/')[-1]
|
||||
self.report_extraction(video_id)
|
||||
mobj = re.match(self._VALID_URL, url)
|
||||
name = mobj.group('name')
|
||||
# This is not a real id, it can be for example AJT for the news
|
||||
# http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
|
||||
video_id = mobj.group('id')
|
||||
|
||||
if re.search(self._LIVE_URL, video_id) is not None:
|
||||
raise ExtractorError(u'Arte live streams are not yet supported, sorry')
|
||||
# self.extractLiveStream(url)
|
||||
# return
|
||||
else:
|
||||
info = self.extractPlus7Stream(url)
|
||||
|
||||
return [info]
|
||||
webpage = self._download_webpage(url, video_id)
|
||||
json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
|
||||
|
||||
json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
|
||||
self.report_extraction(video_id)
|
||||
info = json.loads(json_info)
|
||||
player_info = info['videoJsonPlayer']
|
||||
|
||||
info_dict = {'id': player_info['VID'],
|
||||
'title': player_info['VTI'],
|
||||
'description': player_info['VDE'],
|
||||
'upload_date': unified_strdate(player_info['VDA'].split(' ')[0]),
|
||||
'thumbnail': player_info['programImage'],
|
||||
}
|
||||
|
||||
formats = player_info['VSR'].values()
|
||||
# We order the formats by quality
|
||||
formats = sorted(formats, key=lambda f: int(f['height']))
|
||||
# Pick the best quality
|
||||
format_info = formats[-1]
|
||||
if format_info['mediaType'] == u'rtmp':
|
||||
info_dict['url'] = format_info['streamer']
|
||||
info_dict['play_path'] = 'mp4:' + format_info['url']
|
||||
info_dict['ext'] = 'mp4'
|
||||
else:
|
||||
info_dict['url'] = format_info['url']
|
||||
info_dict['ext'] = 'mp4'
|
||||
|
||||
return info_dict
|
||||
|
|
|
|||
Loading…
Reference in a new issue