yt-dlc/youtube_dl/extractor/googledrive.py

214 lines
7.7 KiB
Python
Raw Normal View History

2015-12-21 03:05:34 +01:00
from __future__ import unicode_literals
2015-06-29 08:53:21 +02:00
import re
2015-06-24 02:13:23 +02:00
from .common import InfoExtractor
from ..utils import (
ExtractorError,
2015-12-21 03:05:34 +01:00
int_or_none,
lowercase_escape,
update_url_query,
)
2015-06-24 02:13:23 +02:00
2015-12-21 03:05:34 +01:00
class GoogleDriveIE(InfoExtractor):
_VALID_URL = r'https?://(?:(?:docs|drive)\.google\.com/(?:uc\?.*?id=|file/d/)|video\.google\.com/get_player\?.*?docid=)(?P<id>[a-zA-Z0-9_-]{28,})'
_TESTS = [{
2015-12-21 03:05:34 +01:00
'url': 'https://drive.google.com/file/d/0ByeS4oOUV-49Zzh4R1J6R09zazQ/edit?pli=1',
'md5': 'd109872761f7e7ecf353fa108c0dbe1e',
2015-06-29 08:53:21 +02:00
'info_dict': {
2015-12-21 03:05:34 +01:00
'id': '0ByeS4oOUV-49Zzh4R1J6R09zazQ',
2015-06-29 08:53:21 +02:00
'ext': 'mp4',
2015-12-21 03:05:34 +01:00
'title': 'Big Buck Bunny.mp4',
'duration': 45,
2015-06-29 08:53:21 +02:00
}
}, {
# video id is longer than 28 characters
'url': 'https://drive.google.com/file/d/1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ/edit',
'md5': 'c230c67252874fddd8170e3fd1a45886',
'info_dict': {
'id': '1ENcQ_jeCuj7y19s66_Ou9dRP4GKGsodiDQ',
'ext': 'mp4',
'title': 'Andreea Banica feat Smiley - Hooky Song (Official Video).mp4',
'duration': 189,
},
'only_matching': True
}]
2015-12-21 03:05:34 +01:00
_FORMATS_EXT = {
'5': 'flv',
'6': 'flv',
'13': '3gp',
'17': '3gp',
'18': 'mp4',
'22': 'mp4',
'34': 'flv',
'35': 'flv',
'36': '3gp',
'37': 'mp4',
'38': 'mp4',
'43': 'webm',
'44': 'webm',
'45': 'webm',
'46': 'webm',
'59': 'mp4',
}
_BASE_URL_CAPTIONS = 'https://drive.google.com/timedtext'
_CAPTIONS_ENTRY_TAG = {
'subtitles': 'track',
'automatic_captions': 'target',
}
_caption_formats_ext = []
_captions_xml = None
2015-06-29 08:53:21 +02:00
@staticmethod
def _extract_url(webpage):
mobj = re.search(
r'<iframe[^>]+src="https?://(?:video\.google\.com/get_player\?.*?docid=|(?:docs|drive)\.google\.com/file/d/)(?P<id>[a-zA-Z0-9_-]{28,})',
2015-06-29 08:53:21 +02:00
webpage)
if mobj:
return 'https://drive.google.com/file/d/%s' % mobj.group('id')
def _download_subtitles_xml(self, video_id, subtitles_id, hl):
if self._captions_xml:
return
self._captions_xml = self._download_xml(
self._BASE_URL_CAPTIONS, video_id, query={
'id': video_id,
'vid': subtitles_id,
'hl': hl,
'v': video_id,
'type': 'list',
'tlangs': '1',
'fmts': '1',
'vssids': '1',
}, note='Downloading subtitles XML',
errnote='Unable to download subtitles XML', fatal=False)
if self._captions_xml:
for f in self._captions_xml.findall('format'):
if f.attrib.get('fmt_code') and not f.attrib.get('default'):
self._caption_formats_ext.append(f.attrib['fmt_code'])
def _get_captions_by_type(self, video_id, subtitles_id, caption_type,
origin_lang_code=None):
if not subtitles_id or not caption_type:
return
captions = {}
for caption_entry in self._captions_xml.findall(
self._CAPTIONS_ENTRY_TAG[caption_type]):
caption_lang_code = caption_entry.attrib.get('lang_code')
if not caption_lang_code:
continue
caption_format_data = []
for caption_format in self._caption_formats_ext:
query = {
'vid': subtitles_id,
'v': video_id,
'fmt': caption_format,
'lang': (caption_lang_code if origin_lang_code is None
else origin_lang_code),
'type': 'track',
'name': '',
'kind': '',
}
if origin_lang_code is not None:
query.update({'tlang': caption_lang_code})
caption_format_data.append({
'url': update_url_query(self._BASE_URL_CAPTIONS, query),
'ext': caption_format,
})
captions[caption_lang_code] = caption_format_data
return captions
def _get_subtitles(self, video_id, subtitles_id, hl):
if not subtitles_id or not hl:
return
self._download_subtitles_xml(video_id, subtitles_id, hl)
if not self._captions_xml:
return
return self._get_captions_by_type(video_id, subtitles_id, 'subtitles')
def _get_automatic_captions(self, video_id, subtitles_id, hl):
if not subtitles_id or not hl:
return
self._download_subtitles_xml(video_id, subtitles_id, hl)
if not self._captions_xml:
return
track = self._captions_xml.find('track')
if track is None:
return
origin_lang_code = track.attrib.get('lang_code')
if not origin_lang_code:
return
return self._get_captions_by_type(
video_id, subtitles_id, 'automatic_captions', origin_lang_code)
2015-06-29 08:53:21 +02:00
def _real_extract(self, url):
video_id = self._match_id(url)
2015-12-21 03:05:34 +01:00
webpage = self._download_webpage(
'http://docs.google.com/file/d/%s' % video_id, video_id)
2015-06-29 08:53:21 +02:00
reason = self._search_regex(
r'"reason"\s*,\s*"([^"]+)', webpage, 'reason', default=None)
2015-12-21 03:05:34 +01:00
if reason:
raise ExtractorError(reason)
2015-06-24 02:13:23 +02:00
2015-12-21 03:05:34 +01:00
title = self._search_regex(r'"title"\s*,\s*"([^"]+)', webpage, 'title')
duration = int_or_none(self._search_regex(
r'"length_seconds"\s*,\s*"([^"]+)', webpage, 'length seconds',
default=None))
2015-12-21 03:05:34 +01:00
fmt_stream_map = self._search_regex(
r'"fmt_stream_map"\s*,\s*"([^"]+)', webpage,
'fmt stream map').split(',')
fmt_list = self._search_regex(
r'"fmt_list"\s*,\s*"([^"]+)', webpage, 'fmt_list').split(',')
2015-06-24 02:13:23 +02:00
resolutions = {}
for fmt in fmt_list:
mobj = re.search(
r'^(?P<format_id>\d+)/(?P<width>\d+)[xX](?P<height>\d+)', fmt)
if mobj:
resolutions[mobj.group('format_id')] = (
int(mobj.group('width')), int(mobj.group('height')))
2015-06-24 02:13:23 +02:00
formats = []
for fmt_stream in fmt_stream_map:
fmt_stream_split = fmt_stream.split('|')
if len(fmt_stream_split) < 2:
continue
format_id, format_url = fmt_stream_split[:2]
f = {
'url': lowercase_escape(format_url),
'format_id': format_id,
'ext': self._FORMATS_EXT[format_id],
}
resolution = resolutions.get(format_id)
if resolution:
f.update({
'width': resolution[0],
'height': resolution[1],
})
formats.append(f)
2015-06-24 02:13:23 +02:00
self._sort_formats(formats)
hl = self._search_regex(
r'"hl"\s*,\s*"([^"]+)', webpage, 'hl', default=None)
subtitles_id = None
ttsurl = self._search_regex(
r'"ttsurl"\s*,\s*"([^"]+)', webpage, 'ttsurl', default=None)
if ttsurl:
# the video Id for subtitles will be the last value in the ttsurl
# query string
subtitles_id = ttsurl.encode('utf-8').decode(
'unicode_escape').split('=')[-1]
2015-06-24 02:13:23 +02:00
return {
'id': video_id,
'title': title,
'thumbnail': self._og_search_thumbnail(webpage, default=None),
2015-12-21 03:05:34 +01:00
'duration': duration,
'formats': formats,
'subtitles': self.extract_subtitles(video_id, subtitles_id, hl),
'automatic_captions': self.extract_automatic_captions(
video_id, subtitles_id, hl),
2015-06-24 02:13:23 +02:00
}