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

[weibo] Misc improvements

This commit is contained in:
Yen Chi Hsuan 2018-01-09 18:12:55 +08:00
parent 1dd38dc0f4
commit 5eca00a2e3
1 changed files with 66 additions and 59 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

125
youtube_dl/extractor/weibo.py
View File

@ -8,7 +8,8 @@ import random
import re import re
from ..compat import ( from ..compat import (
compat_urlparse, compat_parse_qs,
compat_str,
) )
from ..utils import ( from ..utils import (
js_to_json, js_to_json,
@ -31,70 +32,71 @@ class WeiboIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
# to get Referer url for genvisitor # to get Referer url for genvisitor
webpage, urlh = self._download_webpage_handle(url, video_id, note="first visit the page") webpage, urlh = self._download_webpage_handle(url, video_id)
visitor_url = urlh.geturl() visitor_url = urlh.geturl()
headers = {
'Referer': visitor_url
}
fp = { if 'passport.weibo.com' in visitor_url:
"os": "2", # first visit
"browser": "Gecko57,0,0,0", visitor_data = self._download_json(
"fonts": "undefined", 'https://passport.weibo.com/visitor/genvisitor', video_id,
"screenInfo": "1440*900*24", note='Generating first-visit data',
"plugins": "" transform_source=strip_jsonp,
} headers={'Referer': visitor_url},
data = urlencode_postdata({ data=urlencode_postdata({
"cb": "gen_callback", 'cb': 'gen_callback',
"fp": json.dumps(fp), 'fp': json.dumps({
}) 'os': '2',
'browser': 'Gecko57,0,0,0',
'fonts': 'undefined',
'screenInfo': '1440*900*24',
'plugins': '',
}),
}))
genvisitor_url = 'https://passport.weibo.com/visitor/genvisitor' tid = visitor_data['data']['tid']
webpage = self._download_webpage(genvisitor_url, video_id, data=data, headers=headers, note="gen visitor") cnfd = '%03d' % visitor_data['data']['confidence']
p = strip_jsonp(webpage) self._download_webpage(
i1 = p.find('{') 'https://passport.weibo.com/visitor/visitor', video_id,
i2 = p.rfind('}') note='Running first-visit callback',
j = p[i1:i2 + 1] # get JSON object query={
d = json.loads(j) 'a': 'incarnate',
tid = d["data"]["tid"] 't': tid,
cnfd = "%03d" % d["data"]["confidence"] 'w': 2,
'c': cnfd,
'cb': 'cross_domain',
'from': 'weibo',
'_rand': random.random(),
})
query = { webpage = self._download_webpage(
'a': 'incarnate', url, video_id, note='Revisiting webpage')
't': tid,
'w': 2,
'c': cnfd,
'cb': 'cross_domain',
'from': 'weibo',
'_rand': random.random()
}
gencallback_url = "https://passport.weibo.com/visitor/visitor"
self._download_webpage(gencallback_url, video_id, note="gen callback", query=query)
webpage = self._download_webpage(url, video_id, note="retry to visit the page") title = self._html_search_regex(
r'<title>(.+?)</title>', webpage, 'title')
title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title') video_formats = compat_parse_qs(self._search_regex(
r'video-sources=\\\"(.+?)\"', webpage, 'video_sources'))
video_sources_text = self._search_regex(r'video-sources=\\\"(.+?)\"', webpage, 'video_sources')
video_formats = compat_urlparse.parse_qs(video_sources_text)
formats = [] formats = []
supported_resolutions = ('720', '480') supported_resolutions = (480, 720)
for res in supported_resolutions: for res in supported_resolutions:
f = video_formats.get(res) vid_urls = video_formats.get(compat_str(res))
if isinstance(f, list): if not vid_urls or not isinstance(vid_urls, list):
if len(f) > 0: continue
vid_url = f[0]
formats.append({ vid_url = vid_urls[0]
'url': vid_url, formats.append({
'format': 'mp4', 'url': vid_url,
'height': int(res), 'height': res,
}) })
self._sort_formats(formats) self._sort_formats(formats)
uploader = self._og_search_property('nick-name', webpage, 'uploader', default=None)
uploader = self._og_search_property(
'nick-name', webpage, 'uploader', default=None)
return { return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
@ -118,12 +120,17 @@ class WeiboMobileIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
# to get Referer url for genvisitor # to get Referer url for genvisitor
webpage = self._download_webpage(url, video_id, note="visit the page") webpage = self._download_webpage(url, video_id, note='visit the page')
js_code = self._search_regex(r'var\s+\$render_data\s*=\s*\[({.*})\]\[0\] \|\| {};', webpage, 'js_code', flags=re.DOTALL)
weibo_info = self._parse_json(js_code, video_id, transform_source=js_to_json) weibo_info = self._parse_json(self._search_regex(
page_info = weibo_info.get('status').get('page_info') r'var\s+\$render_data\s*=\s*\[({.*})\]\[0\]\s*\|\|\s*{};',
title = weibo_info.get('status').get('status_title') webpage, 'js_code', flags=re.DOTALL),
uploader = weibo_info.get('status').get('user').get('screen_name') video_id, transform_source=js_to_json)
status_data = weibo_info.get('status', {})
page_info = status_data.get('page_info')
title = status_data['status_title']
uploader = status_data.get('user', {}).get('screen_name')
return { return {
'id': video_id, 'id': video_id,