X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Ffacebook.py;h=f5bbd39d2d0e90996c118e3fae325034fc2bbb6d;hb=2e7e561c1d9dedf1a8e5a206e1ef86cfa4599956;hp=f9fd8ed4ec7950a4e35f70125eaa071632e0a3d7;hpb=2e6e742c3cda5ed4846bff4ef894aac21434e3d4;p=youtube-dl
diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py
index f9fd8ed4e..f5bbd39d2 100644
--- a/youtube_dl/extractor/facebook.py
+++ b/youtube_dl/extractor/facebook.py
@@ -6,9 +6,11 @@ import socket
from .common import InfoExtractor
from ..compat import (
+ compat_etree_fromstring,
compat_http_client,
compat_urllib_error,
compat_urllib_parse_unquote,
+ compat_urllib_parse_unquote_plus,
)
from ..utils import (
error_to_compat_str,
@@ -32,9 +34,12 @@ class FacebookIE(InfoExtractor):
video/video\.php|
photo\.php|
video\.php|
- video/embed
- )\?(?:.*?)(?:v|video_id)=|
- [^/]+/videos/(?:[^/]+/)?
+ video/embed|
+ story\.php
+ )\?(?:.*?)(?:v|video_id|story_fbid)=|
+ [^/]+/videos/(?:[^/]+/)?|
+ [^/]+/posts/|
+ groups/[^/]+/permalink/
)|
facebook:
)
@@ -44,6 +49,11 @@ class FacebookIE(InfoExtractor):
_CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
_NETRC_MACHINE = 'facebook'
IE_NAME = 'facebook'
+
+ _CHROME_USER_AGENT = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.97 Safari/537.36'
+
+ _VIDEO_PAGE_TEMPLATE = 'https://www.facebook.com/video/video.php?v=%s'
+
_TESTS = [{
'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
'md5': '6a40d33c0eccbb1af76cf0485a052659',
@@ -65,6 +75,43 @@ class FacebookIE(InfoExtractor):
'expected_warnings': [
'title'
]
+ }, {
+ 'note': 'Video with DASH manifest',
+ 'url': 'https://www.facebook.com/video.php?v=957955867617029',
+ 'md5': '54706e4db4f5ad58fbad82dde1f1213f',
+ 'info_dict': {
+ 'id': '957955867617029',
+ 'ext': 'mp4',
+ 'title': 'When you post epic content on instagram.com/433 8 million followers, this is ...',
+ 'uploader': 'Demy de Zeeuw',
+ },
+ }, {
+ 'url': 'https://www.facebook.com/maxlayn/posts/10153807558977570',
+ 'md5': '037b1fa7f3c2d02b7a0d7bc16031ecc6',
+ 'info_dict': {
+ 'id': '544765982287235',
+ 'ext': 'mp4',
+ 'title': '"What are you doing running in the snow?"',
+ 'uploader': 'FailArmy',
+ }
+ }, {
+ 'url': 'https://m.facebook.com/story.php?story_fbid=1035862816472149&id=116132035111903',
+ 'md5': '1deb90b6ac27f7efcf6d747c8a27f5e3',
+ 'info_dict': {
+ 'id': '1035862816472149',
+ 'ext': 'mp4',
+ 'title': 'What the Flock Is Going On In New Zealand Credit: ViralHog',
+ 'uploader': 'S. Saint',
+ },
+ }, {
+ 'note': 'swf params escaped',
+ 'url': 'https://www.facebook.com/barackobama/posts/10153664894881749',
+ 'md5': '97ba073838964d12c70566e0085c2b91',
+ 'info_dict': {
+ 'id': '10153664894881749',
+ 'ext': 'mp4',
+ 'title': 'Facebook video #10153664894881749',
+ },
}, {
'url': 'https://www.facebook.com/video.php?v=10204634152394104',
'only_matching': True,
@@ -77,6 +124,9 @@ class FacebookIE(InfoExtractor):
}, {
'url': 'facebook:544765982287235',
'only_matching': True,
+ }, {
+ 'url': 'https://www.facebook.com/groups/164828000315060/permalink/764967300301124/',
+ 'only_matching': True,
}]
def _login(self):
@@ -145,15 +195,40 @@ class FacebookIE(InfoExtractor):
def _real_initialize(self):
self._login()
- def _real_extract(self, url):
- video_id = self._match_id(url)
- url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
- webpage = self._download_webpage(url, video_id)
+ def _extract_from_url(self, url, video_id, fatal_if_no_video=True):
+ req = sanitized_Request(url)
+ req.add_header('User-Agent', self._CHROME_USER_AGENT)
+ webpage = self._download_webpage(req, video_id)
- BEFORE = '{swf.addParam(param[0], param[1]);});\n'
+ video_data = None
+
+ BEFORE = '{swf.addParam(param[0], param[1]);});'
AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
- m = re.search(re.escape(BEFORE) + '(.*?)' + re.escape(AFTER), webpage)
- if not m:
+ m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage)
+ if m:
+ swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"')
+ data = dict(json.loads(swf_params))
+ params_raw = compat_urllib_parse_unquote(data['params'])
+ video_data = json.loads(params_raw)['video_data']
+
+ def video_data_list2dict(video_data):
+ ret = {}
+ for item in video_data:
+ format_id = item['stream_type']
+ ret.setdefault(format_id, []).append(item)
+ return ret
+
+ if not video_data:
+ server_js_data = self._parse_json(self._search_regex(
+ r'handleServerJS\(({.+})\);', webpage, 'server js data', default='{}'), video_id)
+ for item in server_js_data.get('instances', []):
+ if item[1][0] == 'VideoConfig':
+ video_data = video_data_list2dict(item[2][0]['videoData'])
+ break
+
+ if not video_data:
+ if not fatal_if_no_video:
+ return webpage, False
m_msg = re.search(r'class="[^"]*uiInterstitialContent[^"]*">(.*?)
', webpage)
if m_msg is not None:
raise ExtractorError(
@@ -161,26 +236,32 @@ class FacebookIE(InfoExtractor):
expected=True)
else:
raise ExtractorError('Cannot parse data')
- data = dict(json.loads(m.group(1)))
- params_raw = compat_urllib_parse_unquote(data['params'])
- params = json.loads(params_raw)
formats = []
- for format_id, f in params['video_data'].items():
+ for format_id, f in video_data.items():
if not f or not isinstance(f, list):
continue
for quality in ('sd', 'hd'):
for src_type in ('src', 'src_no_ratelimit'):
src = f[0].get('%s_%s' % (quality, src_type))
if src:
+ preference = -10 if format_id == 'progressive' else 0
+ if quality == 'hd':
+ preference += 5
formats.append({
'format_id': '%s_%s_%s' % (format_id, quality, src_type),
'url': src,
- 'preference': -10 if format_id == 'progressive' else 0,
+ 'preference': preference,
})
+ dash_manifest = f[0].get('dash_manifest')
+ if dash_manifest:
+ formats.extend(self._parse_mpd_formats(
+ compat_etree_fromstring(compat_urllib_parse_unquote_plus(dash_manifest))))
if not formats:
raise ExtractorError('Cannot find video formats')
+ self._sort_formats(formats)
+
video_title = self._html_search_regex(
r']*class="uiHeaderTitle"[^>]*>([^<]*)', webpage, 'title',
default=None)
@@ -193,9 +274,36 @@ class FacebookIE(InfoExtractor):
video_title = 'Facebook video #%s' % video_id
uploader = clean_html(get_element_by_id('fbPhotoPageAuthorName', webpage))
- return {
+ info_dict = {
'id': video_id,
'title': video_title,
'formats': formats,
'uploader': uploader,
}
+
+ return webpage, info_dict
+
+ def _real_extract(self, url):
+ video_id = self._match_id(url)
+
+ real_url = self._VIDEO_PAGE_TEMPLATE % video_id if url.startswith('facebook:') else url
+ webpage, info_dict = self._extract_from_url(real_url, video_id, fatal_if_no_video=False)
+
+ if info_dict:
+ return info_dict
+
+ if '/posts/' in url:
+ entries = [
+ self.url_result('facebook:%s' % vid, FacebookIE.ie_key())
+ for vid in self._parse_json(
+ self._search_regex(
+ r'(["\'])video_ids\1\s*:\s*(?P\[.+?\])',
+ webpage, 'video ids', group='ids'),
+ video_id)]
+
+ return self.playlist_result(entries, video_id)
+ else:
+ _, info_dict = self._extract_from_url(
+ self._VIDEO_PAGE_TEMPLATE % video_id,
+ video_id, fatal_if_no_video=True)
+ return info_dict