def load_video_urls(self):
items = []
try:
if self.json_data:
cmd = self.json_data['cmd'] if 'cmd' in self.json_data else None
if cmd == 'trig':
stat = self.json_data[
'stat'] if 'stat' in self.json_data else None
res = self.mgr.get_untrack_url(self.site_code, stat)
for item in res:
mediaVideoItem = MediaVideoItem()
mediaVideoItem['sid'] = item['sid']
mediaVideoItem['untrack_id'] = item['untrack_id']
mediaItem = MediaItem()
mediaItem['channel_id'] = item['name']
mediaVideoItem['media'] = mediaItem
url = item['url']
items.append(
Request(url=url,
callback=self.video_parse,
meta={'item': mediaVideoItem}))
elif cmd == 'assign':
tasks = self.json_data[
'task'] if 'task' in self.json_data else None
for task in tasks:
mediaVideoItem = MediaVideoItem()
mediaVideoItem[
'sid'] = task['sid'] if 'sid' in task else None
mediaVideoItem['untrack_id'] = task[
'untrack_id'] if 'untrack_id' in task else None
mediaItem = MediaItem()
mediaItem['channel_id'] = task['name']
mediaVideoItem['media'] = mediaItem
url = task['url']
items.append(
Request(url=url,
callback=self.video_parse,
meta={'item': mediaVideoItem}))
elif cmd == 'test':
channel_id = self.json_data[
'id'] if 'id' in self.json_data else None
url = self.json_data[
'url'] if 'url' in self.json_data else None
if url and channel_id:
list_channel = self.mgr.get_channel_name(channel_id)
if list_channel:
list_channel = list_channel['name']
items.append(
Request(url=url,
callback=self.list_parse,
meta={
'first': False,
'id': list_channel
}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
def video_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'video url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else None
#播放页获取详细信息
sels = response.xpath('//script[@type="text/javascript"]')
hunantv_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//div[@class="play-xxmes clearfix"]')
hunantv_extract.media_info_extract(sels, mediaItem)
#获得媒体页地址
url_express = '(http://www\.hunantv\.com/v/[\d]+/[\d]+)/[a-zA-Z]/[\d]+\.html'
url_regex = re.compile(url_express)
match_results = url_regex.search(request_url)
if match_results:
url_content = match_results.groups()[0]
mediaItem['url'] = url_content
#获取正片地址
url_exist = False
sels = response.xpath(
'//div[@class="play-index-con-til clearfix"]//*[@class="mppl-til"]'
)
for sel in sels:
results = hunantv_extract.album_extract(sel)
if results:
item = results[0]
url = item['url']
url = Util.get_absolute_url(url, prefix_url)
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=url,
callback=self.album_parse,
meta={
'url': request_url,
'item': mediaVideoItem
}))
url_exist = True
break
#不存在正在播放的链接，如“芒果捞星闻”
if 'url' in mediaItem and not url_exist:
year_api = mediaItem['url'] + '/s/json.year.js'
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=year_api,
callback=self.album_json_parse,
meta={
'item': mediaVideoItem,
'url': year_api
}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'video url: %s' % request_url)
def play_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'play url: %s' % request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
route_url_list = response.xpath(
'//div[@class="play-content"]//div[@class="v-panel-route"]/a/@href'
).extract()
media_url = ''
if route_url_list:
media_url = route_url_list[-1]
if media_url:
# 有媒体页url，媒体页抓取媒体信息
items.append(
Request(url=media_url,
callback=self.media_parse,
meta={
'url': request_url,
'item': mediaVideoItem
}))
else:
# 电影没有媒体页，在播放页抓取媒体信息
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
title_class = "v-info v-info-film e-follow"
div_class = "v-meta v-meta-film"
v_title = '//div[@class="%s"]//h1[@class="title"]/text()'
title_list = response.xpath(v_title % title_class).extract()
title = Util.join_list_safely(title_list)
if title:
mediaItem['title'] = title
mediaItem = self.pack_media_info(response, mediaItem,
title_class, div_class)
# 没有媒体页，播放地址作为媒体地址
mediaItem['url'] = Util.normalize_url(request_url,
self.site_code)
mediaVideoItem['media'] = mediaItem
r = re.compile('.*/(\d+).html')
m = r.match(mediaItem['url'])
if m:
vid = m.group(1)
prefix_video_url = re.sub(vid, '%s', mediaItem['url'])
items.append(
self.api_media_info(mediaVideoItem, vid,
prefix_video_url))
else:
items.append(mediaVideoItem)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'play url: %s' % request_url)
def video_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'video url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
sels = response.xpath('//script[@type="text/javascript"]')
letv_extract.media_info_extract(sels, mediaItem)
sels = None
if not sels:
#Detail电视剧、综艺、动漫
sels = response.xpath(
'//div[@data-statectn="play_info"]//ul[@class="intro_box"]'
)
if not sels:
#Info:普通影片,动漫
sels = response.xpath(
'//div[@data-statectn="newplay_info"]//ul[@class="info_list"]'
)
if not sels:
#收费影片
sels = response.xpath(
'//div[@class="Player"]//span[@class="video_info"]')
if sels:
results = letv_extract.media_extract(sels)
if results:
item = results[0]
url = Util.get_absolute_url(item['url'], prefix_url)
mediaItem['url'] = url
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=url,
callback=self.media_parse,
meta={'item': mediaVideoItem}))
if not items:
#视频播放页找不到媒体页地址,尝试直接采用接口爬取
if 'cont_id' in mediaItem:
self.api_parse(mediaVideoItem)
else:
logging.log(logging.INFO,
'该视频播放页找不到媒体页地址,也无法直接采用接口: %s' % request_url)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'video url: %s' % request_url)
def list_html_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'list html url: %s' % request_url)
page = response.request.meta[
'page'] if 'page' in response.request.meta else 1
if page > self.max_update_page:
return items
channel_id = response.request.meta[
'id'] if 'id' in response.request.meta else None
postfix_url = response.request.meta[
'postfix_url'] if 'postfix_url' in response.request.meta else None
if u'电影' == channel_id:
'''
is_hj:是否合集的标志，爬虫目前舍弃合集的链接
is_virtual:本站点是否存在
'''
sels = response.xpath(
'//a[@class="ui-list-ct" and @is_hj="0" and @is_virtual="0"]'
)
else:
sels = response.xpath(
'//a[@class="ui-list-ct" and @is_virtual="0"]')
if sels:
#表明仍有下一页
for sel in sels:
mediaVideoItem = MediaVideoItem()
mediaItem = MediaItem()
mediaItem['channel_id'] = channel_id
urls = sel.xpath('./@href').extract()
mediaItem['url'] = urls[0]
pptv_extract.media_info_extract(sel, mediaItem)
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=mediaItem['url'],
callback=self.video_parse,
meta={'item': mediaVideoItem}))
#下一页
page = page + 1
url = self.list_prefix_url + '?' + postfix_url + '&page=%s' % page
items.append(
Request(url=url,
callback=self.list_html_parse,
meta={
'page': page,
'id': channel_id,
'postfix_url': postfix_url
}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
def compose_mvitem(self, response, title_list, pers, dirs, play_url,
cat_id, poster_url, text):
try:
cat_id = response.request.meta['cat_id']
poster_url = response.request.meta['poster_url']
untrack_id = ""
sid = ""
if "untrack_id" in response.request.meta:
untrack_id = response.request.meta['untrack_id']
if "sid" in response.request.meta:
sid = response.request.meta['sid']
videoitems = []
ep_item = MediaItem()
if title_list:
ep_item["title"] = title_list[0].strip()
ep_item["actor"] = pers
ep_item["director"] = dirs
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
ep_item["poster_url"] = poster_url
ep_item["url"] = Util.normalize_url(response.request.url, "wasu")
if len(text) > 0:
ep_item["intro"] = text[0].strip()
mvitem = MediaVideoItem()
mvitem["media"] = ep_item
mid = self.getshowid(response.request.url)
mvitem["media"]["cont_id"] = mid
ttvitem = {}
if title_list:
ttvitem = self.parse_video_item(response, cat_id, play_url,
title_list, None)
if ttvitem:
if 'video' in ttvitem and len(ttvitem['video']) > 0:
mvitem['video'] = ttvitem['video']
mvitem["media"]["info_id"] = Util.md5hash(
Util.summarize(mvitem["media"]))
Util.set_ext_id(mvitem["media"], mvitem["video"])
if untrack_id and sid:
mvitem["untrack_id"] = untrack_id
mvitem["sid"] = sid
res = self.check_url(mvitem)
if not res:
return None
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return mvitem
def video_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'video url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
pps_extract.media_extract(response, mediaItem)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'video url: %s' % request_url)
def album_json_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'json url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else None
url = response.request.meta[
'url'] if 'url' in response.request.meta else None
if url != request_url:
#被重定向，表明不存在
return items
year_express = '(\[.*\])'
year_regex = re.compile(year_express)
match_results = year_regex.search(response.body)
if match_results:
videoItems = []
year_content = match_results.groups()[0]
years = json.loads(year_content)
for year in years:
video_url = mediaItem['url'] + '/s/json.%s.js' % year
result = Util.get_url_content(video_url)
videoItems = videoItems + self.album_tag_json_resolve(
text=result, meta={'url': video_url})
if videoItems:
Util.set_ext_id(mediaItem, videoItems)
#进入媒体页，获取相关信息
result = Util.get_url_content(mediaItem['url'])
if result:
mediaItem = self.media_resolve(text=result,
meta={
'item': mediaItem,
'url':
mediaItem['url']
})
self.set_media_info(mediaItem)
mediaVideoItem['media'] = mediaItem
mediaVideoItem['video'] = videoItems
items.append(mediaVideoItem)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'json url: %s' % request_url)
def video_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'video url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
if prefix_url == self.vip_prefix_url:
mediaItem['paid'] = '1'
else:
mediaItem['paid'] = '0'
#http://vod.kankan.com/v/87/87998.shtml
sels = response.xpath('//ul[@class="movieinfo"]')
if sels:
kankan_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//p[@id="movie_info_intro_l"]')
if sels:
kankan_extract.media_info_extract(sels, mediaItem)
#普通电影，电视剧，综艺，动漫
sels = response.xpath('//div[@class="header_title"]')
if sels:
results = kankan_extract.media_extract(sels)
else:
#http://vip.kankan.com/vod/88365.html
sels = response.xpath('//div[@class="movie_info"]')
if sels:
kankan_extract.media_info_extract(sels, mediaItem)
results = kankan_extract.media_extract(sels)
else:
#http://vip.kankan.com/vod/88169.html?fref=kk_search_sort_01
sels = response.xpath(
'//div[@class="aside"]//div[@class="intro"]')
results = kankan_extract.media_extract(sels)
for item in results:
mediaItem['url'] = item['url']
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=item['url'],
callback=self.media_parse,
meta={'item': mediaVideoItem}))
break
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'video url: %s' % request_url)
def parse_video_item(self, response, cat_id, url, title, playlistId):
#logging.log(logging.INFO, 'parse_video_item , info url %s,paly_url: %s,cat id %s,title %s' % (response.request.url,url,cat_id,title))
videoitems = []
ep_item = MediaItem()
item = MediaVideoItem()
item["media"] = ep_item
item["video"] = videoitems
try:
if int(cat_id) != int(self.movie_id):
ul_list = response.xpath(
'//div[@class="episodes clearfix "]/a')
if not ul_list:
ul_list = response.xpath(
'//div[@class="episodes clearfix enc-episodes-detail"]/a'
)
for li in ul_list:
url = li.xpath('./@href').extract()
ttitle = li.xpath('./@title').extract()
snum = li.xpath('./text()').extract()
if snum:
play_num = self.get_play_num(snum[0])
if int(cat_id) == int(self.variety_id):
play_num = self.getvnum(self.url_prefix + url[0])
if not ttitle:
ttitle = [play_num]
vitem = self.compose_vitem([self.url_prefix + url[0]],
title, play_num)
if 'url' in vitem:
videoitems.append(vitem)
elif int(cat_id) == int(self.movie_id):
if url:
vitem = self.compose_vitem([url], title, 1)
if 'url' in vitem:
videoitems.append(vitem)
if videoitems:
item["video"] = videoitems
item["media"]["url"] = response.request.url
Util.set_ext_id(item["media"], item["video"])
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return item
def parse_media(self, response, **kwargs):
try:
channel = kwargs['channel']
logging.log(logging.INFO, response.request.url)
items = []
mv = MediaVideoItem()
mv["video"] = []
media_info = self.parser[channel].parse_media_info(response)
#print media_info
if media_info:
media_info['site_id'] = self.site_id
media_info['channel_id'] = self.channel_map[channel]
media_info['url'] = Util.normalize_url(response.request.url)
media_info['ext_id'] = Util.md5hash(media_info['url'])
media_info['info_id'] = Util.md5hash(
Util.summarize(media_info))
mv["media"] = media_info
ext_video = self.parser[channel].parse_video(response)
#print ext_video
if ext_video:
mv["ext_video"] = {
'site_id': self.site_id,
'channel_id': media_info['channel_id'],
'urls': ext_video,
'media_ext_id': media_info['ext_id']
}
reviews = self.parser[channel].parse_review(response)
if reviews:
mv["review"] = {
'urls': reviews,
'media_ext_id': media_info['ext_id']
}
items.append(mv)
return items
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
def list_html_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'list html url: %s' % request_url)
page = response.request.meta[
'page'] if 'page' in response.request.meta else 1
if page > self.max_update_page:
return items
channel_id = response.request.meta[
'id'] if 'id' in response.request.meta else None
results = pps_extract.media_extract(response)
for item in results:
mediaVideoItem = MediaVideoItem()
item['channel_id'] = channel_id
url = item['url']
url = Util.get_absolute_url(url, request_url)
item['url'] = url
mediaVideoItem['media'] = item
items.append(
Request(url=url,
callback=self.video_parse,
meta={'item': mediaVideoItem}))
#下一页
page = page + 1
results = pps_extract.next_page_extract(response)
for item in results:
url = Util.get_absolute_url(item, request_url)
items.append(
Request(url=url,
callback=self.list_html_parse,
meta={
'page': page,
'id': channel_id
}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
def video_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'video url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
#播放页 - 普通电影
sels = response.xpath(
'//div[@class="film-info clearfix"]//span[@class="summary"]/a/@href'
)
if not sels:
#播放页 - vip电影
sels = response.xpath(
'//div[@class="f_song inner_resumeCon intro"]//div[@class="con"]/a/@href'
)
if not sels:
#播放页 - 预告片电影
sels = response.xpath(
'//div[@class="related-film clear"]//a[@class="rel-film-img"]/@href'
)
if sels:
url = sels.extract()[0]
url = Util.get_absolute_url(url, prefix_url)
mediaItem['url'] = url
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=url,
callback=self.media_parse,
meta={'item': mediaVideoItem}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'video url: %s' % request_url)
def media_parse(self, response):
items = []
try:
media_url = response.request.url
logging.log(logging.INFO, 'media url: %s' % media_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
# 媒体页获取媒体信息
title_class = "v-info v-info-album "
div_class = "v-meta v-meta-album"
v_title = '//div[@class="%s"]//h1[@class="title"]/span/text()'
title_list = response.xpath(v_title % title_class).extract()
title = Util.join_list_safely(title_list)
if title:
mediaItem['title'] = title
mediaItem = self.pack_media_info(response, mediaItem, title_class,
div_class)
mediaItem['url'] = Util.normalize_url(media_url, self.site_code)
request_url = response.meta['url']
request_url = Util.normalize_url(request_url, self.site_code)
r = re.compile('.*/(\d+).html')
m = r.match(request_url)
if m:
vid = m.group(1)
prefix_video_url = re.sub(vid, '%s', request_url)
items.append(
self.api_media_info(mediaVideoItem, vid, prefix_video_url))
else:
pass
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'media url: %s' % request_url)
def parse_episode_info(self,response):
try:
logging.log(logging.INFO, 'parse_episode_info: %s' % response.request.url)
cat_id = response.request.meta['cat_id']
poster_url = response.request.meta['poster_url']
page_id = self.get_youku_pageid(response.request.url)
if not page_id:
log.error('miss content id: %s' % response.request.url)
return
untrack_id = ""
sid = ""
mid = ""
if "untrack_id" in response.request.meta:
untrack_id = response.request.meta['untrack_id']
if "sid" in response.request.meta:
sid = response.request.meta['sid']
if "mid" in response.request.meta:
mid = response.request.meta['mid']
items = []
year_list = []
title = self.parse_title(response,cat_id)
performer_list = self.parse_actor(response)
director_list = self.parse_director(response)
district_list = response.xpath('//ul[@class="baseinfo"]/li/span/label[text()="%s"]/../a/text()' % u'地区:').extract()
type_list = response.xpath('//ul[@class="baseinfo"]/li/span/label[text()="%s"]/../a/text()' % u'类型:').extract()
play_date = self.parse_play_date(response)
total_num = self.parse_total_num(response)
year_list = response.xpath('//div[@class="mod plot"]/ul[@class="filter"]/li[@class="v-year"]/a/em/text()').extract()
pers = Util.join_list_safely(performer_list)
dirs = Util.join_list_safely(director_list)
types = Util.join_list_safely(type_list)
#text
text = response.xpath('//div[@class="detail"]/span/text()').extract()
videoitems = []
ep_item = MediaItem()
if title:
ep_item["title"] = title[0].strip()
if pers:
ep_item["actor"] = pers
if dirs > 0:
ep_item["director"] = dirs
if types:
ep_item["type"] = types
if district_list:
ep_item["district"] = district_list[0].strip()
if play_date:
ep_item["release_date"] = Util.str2date(play_date)
if total_num:
ep_item["vcount"] = total_num
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
ep_item["poster_url"] = poster_url
ep_item["url"] = Util.normalize_url(response.request.url,"youku")
if text:
ep_item["intro"] = text[0].strip()
ep_item["cont_id"] = page_id
ep_item["info_id"] = Util.md5hash(Util.summarize(ep_item))
mvitem = MediaVideoItem();
if mid:
mvitem['mid'] = mid
mvitem["media"] = ep_item;
if untrack_id:
mvitem["untrack_id"] = untrack_id
if sid:
mvitem["sid"] = sid
video_list = self.parse_video_item(response, cat_id, ep_item["title"], page_id)
mvitem['video'] = video_list
Util.set_ext_id(mvitem["media"], mvitem["video"])
items.append(mvitem)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return items
def load_video_urls(self):
items = []
try:
if self.json_data:
cmd = self.json_data['cmd'] if 'cmd' in self.json_data else None
if cmd == 'trig':
'''
#由于当前360爬取策略将m1905 -> 1905，故舍弃该方法
site_codes = ['1905', 'm1905']
for site_code in site_codes:
res = self.mgr.get_untrack_url(site_code=site_code)
for item in res:
mediaVideoItem = MediaVideoItem()
mediaVideoItem['sid'] = item['sid']
mediaVideoItem['untrack_id'] = item['untrack_id']
mediaItem = MediaItem()
mediaItem['channel_id'] = item['name']
mediaVideoItem['media'] = mediaItem
url = item['url']
items.append(Request(url=url, callback=self.video_parse, meta={'item':mediaVideoItem}))
'''
#由于电影网特别的site_code有:1905, m1905,舍弃原因的方式
if self.site_code:
stat = self.json_data[
'stat'] if 'stat' in self.json_data else None
res = self.mgr.get_untrack_url(self.site_code, stat)
for item in res:
mediaVideoItem = MediaVideoItem()
mediaVideoItem['sid'] = item['sid']
mediaVideoItem['untrack_id'] = item['untrack_id']
mediaItem = MediaItem()
mediaItem['channel_id'] = item['name']
mediaVideoItem['media'] = mediaItem
url = item['url']
items.append(
Request(url=url,
callback=self.video_parse,
meta={'item': mediaVideoItem}))
elif cmd == 'assign':
tasks = self.json_data[
'task'] if 'task' in self.json_data else None
for task in tasks:
mediaVideoItem = MediaVideoItem()
mediaVideoItem[
'sid'] = task['sid'] if 'sid' in task else None
mediaVideoItem['untrack_id'] = task[
'untrack_id'] if 'untrack_id' in task else None
mediaItem = MediaItem()
mediaItem['channel_id'] = task['name']
mediaVideoItem['media'] = mediaItem
url = task['url']
items.append(
Request(url=url,
callback=self.video_parse,
meta={'item': mediaVideoItem}))
elif cmd == 'test':
channel_id = self.json_data[
'id'] if 'id' in self.json_data else None
url = self.json_data[
'url'] if 'url' in self.json_data else None
if url and channel_id:
list_channel = self.mgr.get_channel_name(channel_id)
if list_channel:
list_channel = list_channel['name']
level = self.max_mark_depth + 1
items.append(
Request(url=url,
callback=self.list_parse,
meta={
'level': level,
'id': list_channel
}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
def media_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'media url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
#过滤掉skip_types类型的影片
sels = response.xpath('//head//script')
if sels:
regex_express = 'movieInfo\.play_type[ ]?=[ ]?\'(.*)\''
match_result = sels.re(regex_express)
if match_result:
play_type = match_result[0]
if play_type in self.skip_types:
return items
#由于某些URL会有跳转，所以应保存真是的URL
#http://movie.kankan.com/movie/88365 -> http://data.movie.kankan.com/movie/88365
mediaItem['url'] = request_url
sels = response.xpath('//head')
kankan_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//div[@class="info_list"]')
if sels:
kankan_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//ul[@class="detail_ul"]')
if sels:
kankan_extract.media_info_extract(sels, mediaItem)
#获取媒体的剧集信息
videoItems = []
if u'综艺' == mediaItem['channel_id']:
#综艺
sels = response.xpath(
'//div[@id[re:test(., "fenji_[\d]+_[\d]+")]]')
for sel in sels:
video_sels = sel.xpath('.//li')
for video_sel in video_sels:
videoItem = VideoItem()
videoItem['intro'] = mediaItem['channel_id']
kankan_extract.video_info_extract(video_sel, videoItem)
if 'url' in videoItem:
url = videoItem['url']
url = Util.get_absolute_url(url, prefix_url)
videoItem['url'] = url
self.set_video_info(videoItem,
mediaItem['channel_id'])
videoItems.append(videoItem)
elif u'电影' == mediaItem['channel_id']:
#电影，从立即观看中获取
videoItem = VideoItem()
Util.copy_media_to_video(mediaItem, videoItem)
sels = response.xpath('//div[@class="section clearfix s2"]')
if sels:
urls = sels.xpath(
'.//a[starts-with(@class, "foc")]/@href').extract()
thumb_urls = sels.xpath(
'.//a[@class="foc"]/img/@src').extract()
if urls:
url = urls[0]
url = Util.get_absolute_url(url, prefix_url)
videoItem['url'] = url
if thumb_urls:
videoItem['thumb_url'] = thumb_urls[0]
self.set_video_info(videoItem, mediaItem['channel_id'])
videoItems.append(videoItem)
else:
#电视剧
sels = response.xpath(
'//div[@id[re:test(., "fenji_[\d]+_asc")]]')
if not sels:
#动漫,电视剧
sels = response.xpath(
'//ul[@id[re:test(., "fenji_[\d]+_asc")]]')
for sel in sels:
video_sels = sel.xpath('.//li')
for video_sel in video_sels:
videoItem = VideoItem()
videoItem['intro'] = mediaItem['channel_id']
kankan_extract.video_info_extract(video_sel, videoItem)
if 'url' in videoItem:
url = videoItem['url']
url = Util.get_absolute_url(url, prefix_url)
videoItem['url'] = url
self.set_video_info(videoItem,
mediaItem['channel_id'])
videoItems.append(videoItem)
if videoItems:
#设置ext_id
Util.set_ext_id(mediaItem, videoItems)
self.set_media_info(mediaItem)
mediaVideoItem['media'] = mediaItem
mediaVideoItem['video'] = videoItems
items.append(mediaVideoItem)
#self.count = self.count + 1
#logging.log(logging.INFO, 'count: %s' % str(self.count))
else:
logging.log(logging.INFO, '%s: no videos' % request_url)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'media url: %s' % request_url)
def parse_single_episode(self, response):
items = []
try:
logging.log(logging.INFO,
'parse_single_episode: %s' % response.request.url)
cat_id = response.request.meta['cat_id']
untrack_id = response.request.meta['untrack_id']
sid = response.request.meta['sid']
mid = response.request.meta[
'mid'] if 'mid' in response.request.meta else ""
playtype_list = response.selector.re(
re.compile(r'var pagetype = .*?(\D+)'))
#发现新的类型页面，http://tv.sohu.com/20100804/n273985736.shtml
#http://my.tv.sohu.com/us/49390690/29200993.shtml 该URL利用现有的逻辑无法爬取到
urls = response.xpath(
'//div[@id="crumbsBar"]/div[@class="area cfix"]/div[@class="left"]/div[@class="crumbs"]/a[last()]'
)
attributes = urls.xpath('./@*').extract()
size = len(attributes)
urls = urls.xpath('./@href').extract()
if size == 1 and urls and not playtype_list:
for iurl in urls:
surl = Util.normalize_url(iurl, "sohu")
if surl and "http" in surl:
items.append(
Request(url=surl,
callback=self.parse_episode_info,
meta={
'cat_id': cat_id,
'poster_url': '',
'page': 1,
"untrack_id": untrack_id,
"sid": sid,
"mid": mid
}))
#付费电影，不能跳转到媒体页
else:
mvitem = self.parse_episode_play(response, untrack_id, sid)
if mid:
mvitem['mid'] = mid
if mvitem and "media" in mvitem and "url" in mvitem[
"media"] and "ext_id" in mvitem["media"]:
if self.check_url(mvitem):
items.append(mvitem)
if not items:
mvitem = MediaVideoItem()
if mid:
mvitem['mid'] = mid
if untrack_id and sid:
mvitem["untrack_id"] = untrack_id
mvitem["sid"] = sid
ep_item = MediaItem()
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
mvitem["media"] = ep_item
playlistId = ""
playlistId_list = response.selector.re(
re.compile(r'var playlistId.*?(\d+)'))
if not playlistId_list:
playlistId_list = response.selector.re(
re.compile(r'var PLAYLIST_ID.*?(\d+)'))
if not playlistId_list:
playlistId_list = response.selector.re(
re.compile(r'= playlistId.*?(\d+)'))
if playlistId_list:
playlistId = playlistId_list[0]
items += self.api_episode_info(mvItem=mvitem,
playlistId=playlistId,
cat_id=cat_id)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return items
def parse_episode_info(self, response):
items = []
try:
logging.log(logging.INFO,
'parse_episode_info: %s' % response.request.url)
cat_id = response.request.meta['cat_id']
poster_url = response.request.meta['poster_url']
untrack_id = ""
sid = ""
if "untrack_id" in response.request.meta:
untrack_id = response.request.meta['untrack_id']
if "sid" in response.request.meta:
sid = response.request.meta['sid']
year_list = []
lyears = []
title_list = response.xpath(
'//div[@class="aboutThis clearfix"]/div[@class="makeup"]/h3/a/@title'
).extract()
director_list = response.xpath(
'//div[@class="info clearfix"]/span[text()="%s"]/a/text()' %
u'导演：').extract()
performer_list = response.xpath(
'//div[@class="info clearfix"]/span[text()="%s"]/a/text()' %
u'主演：').extract()
type_list = response.xpath(
'//div[@class="info clearfix"]/span[text()="%s"]/a/text()' %
u'类型：').extract()
district_list = response.xpath(
'//div[@class="info clearfix"]/span[text()="%s"]/a/text()' %
u'地区：').extract()
year_info = response.xpath(
'//div[@class="info clearfix"]/span[text()="%s"]/text()' %
u'地区：').extract()
year = None
if len(year_info) >= 2:
year = self.get_year(year_info[1])
#year_list = response.xpath('//div[@class="mod plot"]/ul[@class="filter"]/li[@class="v-year"]/a/em/text()').extract()
pers = Util.join_list_safely(performer_list)
dirs = Util.join_list_safely(director_list)
types = Util.join_list_safely(type_list)
districts = Util.join_list_safely(district_list)
#text
text = response.xpath(
'//div[@class="juqing briefTab"]/div/text()').extract()
#score
score = response.xpath(
'//div[@class="aboutThis clearfix"]/div[@class="makeup"]/div[1]/div[@class="score"]/div[class="score-num"]/strong/text()'
).extract()
play_url = ""
tplay_url = response.xpath(
'//div[@class="aboutThis clearfix"]/div[@class="makeup"]/div[@class="sourcePlay"]/a[@id="moviePlayButton"]/@href'
).extract()
if tplay_url:
play_url = self.url_prefix + tplay_url[0].strip()
videoitems = []
ep_item = MediaItem()
if title_list:
ep_item["title"] = title_list[0]
if ep_item["title"].find(u'预:') >= 0:
print "预告片，url", response.request.url
return items
ep_item["actor"] = pers
ep_item["director"] = dirs
if types:
ep_item["type"] = types
if district_list:
ep_item["district"] = districts
if year:
ep_item["release_date"] = Util.str2date(year)
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
ep_item["poster_url"] = poster_url
ep_item["url"] = Util.normalize_url(response.request.url,
"baofeng")
if len(text) > 0:
ep_item["intro"] = text[0].strip()
mvitem = MediaVideoItem()
mvitem["media"] = ep_item
vurl = ""
videoid = self.getshowid(response.request.url)
mvitem["media"]["cont_id"] = videoid
ttvitem = {}
if title_list:
ttvitem = self.parse_video_item(response, cat_id, play_url,
title_list, None)
if ttvitem:
if 'video' in ttvitem and len(ttvitem['video']) > 0:
mvitem['video'] = ttvitem['video']
mvitem["media"]["info_id"] = Util.md5hash(
Util.summarize(mvitem["media"]))
Util.set_ext_id(mvitem["media"], mvitem["video"])
if untrack_id and sid:
mvitem["untrack_id"] = untrack_id
mvitem["sid"] = sid
res = self.check_url(mvitem)
#if self.check_url(mvitem):
if res:
items.append(mvitem)
pass
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return items
def parse_episode_info(self, response):
items = []
try:
logging.log(logging.INFO,
'parse_episode_info: %s' % response.request.url)
cat_id = response.request.meta['cat_id']
poster_url = response.request.meta['poster_url']
untrack_id = ""
sid = ""
if "untrack_id" in response.request.meta:
untrack_id = response.request.meta['untrack_id']
if "sid" in response.request.meta:
sid = response.request.meta['sid']
#title
title = response.xpath(
'//div[@class="mod_video_intro mod_video_intro_rich"]/div[@class="video_title"]/strong/a/text()'
).extract()
if not title or not title[0]:
title = response.xpath(
'//div[@class="mod_box mod_video_info"]/div[@class="mod_hd mod_hd_border"]/h1/strong/@title'
).extract()
if not title or not title[0]:
title = response.xpath(
'//div[@class="mod_box mod_video_info"]/div[@class="mod_hd mod_hd_border"]/h2/strong/@title'
).extract()
if not title or not title[0]:
title = response.xpath(
'//div[@class="mod_page_banner"]/div[@class="banner_pic"]/a/@title'
).extract()
#performer
#performer_list = response.xpath('//div[@class="mod_video_intro mod_video_intro_rich"]/div[2]/div[2]/div[1]/a/span/text()').extract()
performer_list = response.xpath(
'//div[@class="mod_video_intro mod_video_intro_rich"]/div[@class="video_info cf"]/div[@class="info_line cf"]/div[@class="info_cast"]/a/span/text()'
).extract()
if not performer_list:
performer_list = response.xpath(
'//div[@class="video_info cf"]/div[@class="info_line cf"]/p/span[text()="%s"]/../span[@class="content"]/a/span/text()'
% u'主演：').extract()
#director
#director_list=response.xpath('//div[@class="mod_video_intro mod_video_intro_rich"]/div[2]/div[3]/div[1]/a/span/text()').extract()
director_list = response.xpath(
'//div[@class="mod_video_intro mod_video_intro_rich"]/div[@class="video_info cf"]/div[@class="info_line cf"]/div[@class="info_director"]/a/span/text()'
).extract()
if not director_list:
director_list = response.xpath(
'//div[@class="video_info cf"]/div[@class="info_line cf"]/p/span[text()="%s"]/../span[@class="content"]/a/span/text()'
% u'导演：').extract()
#text
text = response.xpath(
'//div[@class="movie_info_wrap"]/div[1]/d1[1]/dd[3]/p[@class="detail_all"]/text()'
).extract()
if not text:
response.xpath(
'//div[@class="mod_video_focus"]/div[@class="info_desc"]/span[@class="desc"]/text()'
).extract()
type_list = response.xpath(
'//div[@class="mod_video_intro mod_video_intro_rich"]/div[@class="video_info cf"]/div[@class="info_line info_line_tags cf"]/div[@class="info_tags"]/a/span/text()'
).extract()
if not type_list:
type_list = response.xpath(
'//div[@class="video_info cf"]/div[@class="info_line cf"]/p/span[text()="%s"]/../span[@class="content"]/a/text()'
% u'类型：').extract()
year_info = response.xpath(
'//div[@class="mod_video_intro mod_video_intro_rich"]/div[@class="video_title"]/span[@class="video_current_state"]/span[@class="current_state"]/text()'
).extract()
if not year_info:
year_info = response.xpath(
'//div[@class="video_info cf"]/div[@class="info_line cf"]/p/span[text()="%s"]/../span[@class="content"]/a/text()'
% u'年份：').extract()
play_date = None
if year_info:
play_date = self.get_year(year_info[0])
#
dirs = Util.join_list_safely(director_list)
types = Util.join_list_safely(type_list)
pers = Util.join_list_safely(performer_list)
#sourceid
sourceid = ""
sourceid_list = response.xpath(
'//div[@class="mod_bd sourceCont"]/@sourceid').extract()
if sourceid_list:
sourceid = sourceid_list[0]
videoitems = []
ep_item = MediaItem()
if len(title) > 0:
ep_item["title"] = title[0]
if len(pers) > 0:
ep_item["actor"] = pers
if len(dirs) > 0:
ep_item["director"] = dirs
if types:
ep_item["type"] = types
if play_date:
ep_item["release_date"] = Util.str2date(play_date)
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
ep_item["url"] = Util.normalize_url(response.request.url, "qq")
ep_item["poster_url"] = poster_url
if len(text) > 0:
ep_item["intro"] = text[0]
mvitem = MediaVideoItem()
mvitem["media"] = ep_item
mvitem["video"] = videoitems
vurl = ""
url_pre = "http://s.video.qq.com/loadplaylist?vkey="
url_tail = "&vtype=2&otype=json&video_type=2&callback=jQuery191048201349820010364_1425370006500&low_login=1"
videoid = self.get_qq_showid(response.request.url)
#videoid = self.get_vid(response.body,response.request.url)
mvitem["media"]["cont_id"] = videoid
mvitem["media"]["info_id"] = Util.md5hash(
Util.summarize(mvitem["media"]))
vurl = url_pre + str(sourceid) + url_tail
tflag = "jQuery191048201349820010364_1425370006500"
tpitem = self.parse_play_list(cat_id, vurl, tflag, response)
#没有sourceid，比如专题页面
if not tpitem:
tpitem = self.parse_topic_play_list(response)
videoids = response.xpath(
'//div[@class="mod_episodes_info episodes_info"]/input[@name="cid"]/@value'
).extract()
if videoids:
mvitem["media"]["cont_id"] = videoids[0]
if tpitem:
mvitem["video"] = tpitem
Util.set_ext_id(mvitem["media"], mvitem["video"])
if untrack_id:
mvitem["untrack_id"] = untrack_id
if sid:
mvitem["sid"] = sid
if self.check_url(mvitem):
items.append(mvitem)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return items
def parse_episode_play(self, response):
mvitem = None
try:
logging.log(logging.INFO,
'parse_episode_play: %s' % response.request.url)
cat_id = response.request.meta['cat_id']
poster_url = ""
untrack_id = ""
sid = ""
if "untrack_id" in response.request.meta:
untrack_id = response.request.meta['untrack_id']
if "sid" in response.request.meta:
sid = response.request.meta['sid']
#items = []
#title
title_list = response.xpath(
'//div[@class="movie_info"]/div[@class="title_wrap"]/h3/a/@title'
).extract()
if not title_list:
title_list = response.xpath(
'//div[@class="intro_lt"]/div[@class="intro_title cf"]/p[@class="title_cn"]/text()'
).extract()
#performer
performer_list = response.xpath(
'//div[@class="movie_info"]/div[@class="movie_detail"]/dl[@class="detail_list"]/dd[@class="actor"]/a/text()'
).extract()
#director
director_list = response.xpath(
'//div[@class="movie_info"]/div[@class="movie_detail"]/dl[@class="detail_list"]/dd[@class="type"]/span[text()="%s"]/a/text()'
% u'导演：').extract()
#type_list = response.xpath('//div[@class="movie_info"]/div[@class="movie_detail"]/dl[@class="detail_list"]/dd[@class="type"]/span[text()="%s"]/a/text()' % u'导演：').extract()
pers = Util.join_list_safely(performer_list)
dirs = Util.join_list_safely(director_list)
#text
text = response.xpath(
'//div[@class="movie_info_wrap"]/div[1]/d1[1]/dd[3]/p[@class="detail_all"]/text()'
).extract()
ep_item = MediaItem()
videoitems = []
#not film
if int(cat_id) != int(self.movie_id):
#video list
#video_list = response.xpath('//div[@class="mod_player_side_inner"]/div[2]/div[1]/div[1]/div[1]/div[1]/ul[1]/li')
video_list = response.xpath(
'//div[@class="tabcont_warp tabcont_warp_yespadding"]/div[@class="tabcont_album"]/ul[@class="album_list cf"]/li'
)
i = 0
for tvideo in video_list:
lurl = tvideo.xpath('./a/@href').extract()
surl = ""
#lnum = tvideo.xpath('./a/@title').extract()
lnum = tvideo.xpath('./a/span/text()').extract()
vitem = VideoItem()
if lnum and lurl:
vitem["vnum"] = lnum[0]
surl = "http://film.qq.com" + lurl[0]
vitem["os_id"] = self.os_id
vitem["site_id"] = self.site_id
#vitem["cont_id"] = self.get_vid(response.body,surl)
turl = ""
if cat_id == self.tv_id:
turl = Util.normalize_url(surl, "qq", "tv")
if cat_id == self.cartoon_id:
turl = Util.normalize_url(surl, "qq", "cartoon")
else:
turl = Util.normalize_url(surl, "qq")
if turl:
vitem["ext_id"] = Util.md5hash(turl)
vitem["url"] = turl
vitem["cont_id"] = self.get_qq_showid(vitem["url"])
else:
continue
videoitems.append(vitem)
else:
vitem = VideoItem()
if title_list:
vitem["title"] = title_list[0]
vitem["vnum"] = "1"
vitem["os_id"] = self.os_id
vitem["site_id"] = self.site_id
#vitem["cont_id"] = self.get_vid(response.body,response.request.url)
turl = Util.normalize_url(response.request.url, "qq")
vitem["url"] = turl
vitem["ext_id"] = Util.md5hash(turl)
vitem["cont_id"] = self.get_qq_showid(vitem["url"])
videoitems.append(vitem)
if len(title_list) > 0:
ep_item["title"] = title_list[0]
if len(pers) > 0:
ep_item["actor"] = pers
if len(dirs) > 0:
ep_item["director"] = dirs
if len(text) > 0:
ep_item["intro"] = text[0]
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
ep_item["poster_url"] = poster_url
videoid = self.get_qq_showid(response.request.url)
#videoid = self.get_vid(response.body,response.request.url)
ep_item["cont_id"] = videoid
mvitem = MediaVideoItem()
mvitem["media"] = ep_item
mvitem["video"] = videoitems
#mvitem["media"]["url"] = response.request.url
mvitem["media"]["url"] = Util.normalize_url(
response.request.url, "qq")
#mvitem["ext_id"] = Util.md5hash(mvitem["media"]["url"])
if untrack_id:
mvitem["untrack_id"] = untrack_id
if sid:
mvitem["sid"] = sid
mvitem["media"]["info_id"] = Util.md5hash(
Util.summarize(mvitem["media"]))
Util.md5hash(Util.summarize(mvitem["media"]))
Util.set_ext_id(mvitem["media"], mvitem["video"])
#items.append(mvitem)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return mvitem
def api_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'api prase url: %s' % request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
mediaItem['url'] = request_url
sel = response.xpath('.//script[@type="text/javascript"]')
pidl = response.xpath('.//script[@type="text/javascript"]').re(
'\"pid\"\D?(\d+)')
vidl = response.xpath('.//script[@type="text/javascript"]').re(
'\"id\"\D?(\d+)')
if pidl and vidl:
pid = pidl[0]
vid = vidl[0]
app_api = self.app_api % (self.get_auth(), pid)
ismovie = False
isvariaty = False
if u'电影' == mediaItem['channel_id']:
ismovie = True
app_api = self.app_api % (self.get_auth(), vid)
mediaItem['cont_id'] = str(vid)
elif u'综艺' == mediaItem['channel_id']:
isvariaty = True
app_api = self.app_api % (self.get_auth(), pid)
mediaItem['cont_id'] = str(pid)
else:
app_api = self.app_api % (self.get_auth(), pid)
mediaItem['cont_id'] = str(pid)
xpara = self.get_xdata(url=app_api)
mediaItem = self.resolve_media_info(xpara,
mediaItem,
ismovie=ismovie)
mediaItem['url'] = Util.normalize_url(request_url,
self.site_code)
mediaItem['site_id'] = self.site_id
mediaItem['channel_id'] = self.channels_name_id[
mediaItem['channel_id']]
mediaItem['info_id'] = Util.md5hash(Util.summarize(mediaItem))
max_page = self.get_max_page(xpara)
video_list = []
if ismovie:
videoItem = VideoItem()
videoItem['title'] = mediaItem[
'title'] if 'title' in mediaItem else None
videoItem['thumb_url'] = mediaItem[
'poster_url'] if 'poster_url' in mediaItem else None
videoItem['url'] = mediaItem[
'url'] if 'url' in mediaItem else None
videoItem['os_id'] = self.os_id
videoItem['site_id'] = self.site_id
videoItem['ext_id'] = Util.md5hash(
mediaItem['url']) if 'url' in mediaItem else None
videoItem['vnum'] = mediaItem[
'vcount'] if 'vcount' in mediaItem else 1
videoItem['cont_id'] = mediaItem[
'cont_id'] if 'cont_id' in mediaItem else None
video_list.append(videoItem)
else:
for i in range(1, max_page):
web_api = self.web_api % (pid, i)
dpara = self.get_ddata(url=web_api)
video_list += self.resolve_video_item(
dpara, page_num=i, isvariaty=isvariaty)
if isvariaty:
video_list = self.revise_video_item(video_list, xpara)
if video_list:
Util.set_ext_id(mediaItem, video_list)
mediaVideoItem['media'] = mediaItem
mediaVideoItem['video'] = video_list
items.append(mediaVideoItem)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
def parse_episode_info(self, response):
items = []
try:
logging.log(logging.INFO,
'parse_episode_info: %s' % response.request.url)
cat_id = response.request.meta['cat_id']
poster_url = response.request.meta['poster_url']
untrack_id = ""
sid = ""
mid = ""
if "untrack_id" in response.request.meta:
untrack_id = response.request.meta['untrack_id']
if "sid" in response.request.meta:
sid = response.request.meta['sid']
if "mid" in response.request.meta:
mid = response.request.meta['mid']
year_list = []
lyears = []
playlistId = ""
playlistId_list = response.selector.re(
re.compile(r'var playlistId.*?(\d+)'))
if not playlistId_list:
playlistId_list = response.selector.re(
re.compile(r'var PLAYLIST_ID.*?(\d+)'))
if not playlistId_list:
playlistId_list = response.selector.re(
re.compile(r'= playlistId.*?(\d+)'))
if playlistId_list:
playlistId = playlistId_list[0]
if not playlistId:
logging.log(
logging.INFO,
"parse_episode_info error,not find playlistid,url:%s " %
response.request.url)
return items
title_list = self.parse_title(response, cat_id)
performer_list = self.parse_actor(response)
director_list = self.parse_director(response)
district_list = self.parse_district(response)
type_list = self.parse_type_list(response)
#year_list = response.xpath('//div[@class="mod plot"]/ul[@class="filter"]/li[@class="v-year"]/a/em/text()').extract()
year_list = self.parse_year(response)
year = None
if year_list:
year = year_list[0]
#pers = "|".join([t.strip() for t in performer_list])
#dirs = "|".join([t.strip() for t in director_list])
pers = Util.join_list_safely(performer_list)
dirs = Util.join_list_safely(director_list)
types = Util.join_list_safely(type_list)
district = Util.join_list_safely(district_list)
#text
text = response.xpath(
'//div[@class="movieCont mod"]/p[1]/span[@class="full_intro"]/text()'
).extract()
play_url = ""
play_url = response.xpath(
'//div[@class="cfix movie-info"]/div[2]/div[@class="cfix bot"]/a[@class="btn-playFea"]/@href'
).extract()
videoitems = []
ep_item = MediaItem()
if title_list:
ep_item["title"] = title_list[0]
ep_item["actor"] = pers
ep_item["director"] = dirs
if types:
ep_item["type"] = types
if district:
ep_item["district"] = district
if year:
ep_item["release_date"] = Util.str2date(year)
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
ep_item["poster_url"] = poster_url
ep_item["url"] = Util.normalize_url(response.request.url, "sohu")
playlistId = str(playlistId)
ep_item["cont_id"] = playlistId
if len(text) > 0:
ep_item["intro"] = text[0].strip()
mvitem = MediaVideoItem()
if mid:
mvitem['mid'] = mid
if untrack_id and sid:
mvitem["untrack_id"] = untrack_id
mvitem["sid"] = sid
mvitem["media"] = ep_item
vurl = ""
ttvitem = []
if title_list:
ttvitem = self.parse_video_item(cat_id, playlistId)
if ttvitem:
mvitem['video'] = ttvitem
mvitem["media"]["info_id"] = Util.md5hash(
Util.summarize(mvitem["media"]))
Util.set_ext_id(mvitem["media"], mvitem["video"])
if self.check_url(mvitem):
items.append(mvitem)
if not items and playlistId:
items += self.api_episode_info(mvitem,
playlistId,
cat_id=cat_id)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return items
def album_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'album url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
video_url = response.request.meta[
'url'] if 'url' in response.request.meta else None
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
videoItems = []
sels = response.xpath(
'//div[@class="page-videolist-tag-main"]//p[@class="pa1-nav"]')
if sels:
#存在tag页
#http://list.hunantv.com/album/56.html
results = hunantv_extract.album_tag_extract(sels)
for item in results:
url = Util.get_absolute_url(item['url'], prefix_url)
result = Util.get_url_content(url)
videoItems = videoItems + self.album_tag_resolve(
text=result, meta={'url': url})
else:
#不存在tag页
#http://list.hunantv.com/album/2905.html
video_sels = response.xpath(
'//div[@class="page-videolist clearfix"]')
if video_sels:
result = video_sels.extract()[0]
videoItems = videoItems + self.album_tag_resolve(
text=result, meta={'url': request_url})
else:
#无正片页地址
#http://www.hunantv.com/v/7/102831/f/1043648.html，有正片集的URL，但该URL是无效的
if video_url:
videoItem = VideoItem()
Util.copy_media_to_video(mediaItem, videoItem)
videoItem['url'] = video_url
Util.copy_media_to_video(mediaItem, videoItem)
video_url_express = 'http://www\.hunantv\.com/v/[\d]+/[\d]+/[a-zA-Z]/([\d]+)\.html'
video_url_regex = re.compile(video_url_express)
#获取视频id
match_results = video_url_regex.search(video_url)
if match_results:
id = match_results.groups()[0]
videoItem['cont_id'] = id
self.set_video_info(videoItem)
videoItems.append(videoItem)
if videoItems:
#设置ext_id
Util.set_ext_id(mediaItem, videoItems)
#进入媒体页，获取相关信息
result = Util.get_url_content(mediaItem['url'])
if result:
mediaItem = self.media_resolve(text=result,
meta={
'item': mediaItem,
'url': mediaItem['url']
})
self.set_media_info(mediaItem)
mediaVideoItem['media'] = mediaItem
mediaVideoItem['video'] = videoItems
items.append(mediaVideoItem)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'album url: %s' % request_url)
def parse_episode_play(self, response, untrack_id, sid):
mvitem = None
try:
logging.log(logging.INFO,
'parse_episode_play: %s' % response.request.url)
cat_id = response.request.meta['cat_id']
#vip
title_list = response.xpath(
'//div[@id="crumbsBar"]/div[@class="area cfix"]/div[@class="left"]/h2/@title'
).extract()
director_list = response.xpath(
'//div[@class="info info-con"]/ul/li[text()="%s"]/a/text()' %
u'导演：').extract()
performer_list = response.xpath(
'//div[@class="info info-con"]/ul/li[text()="%s"]/a/text()' %
u'主演：').extract()
text = response.xpath(
'//div[@class="info info-con"]/p[@class="intro"]/text()'
).extract()
pers = "|".join([t.strip() for t in performer_list])
dirs = "|".join([t.strip() for t in director_list])
playlistId = ""
playlistId_list = response.selector.re(
re.compile(r'var playlistId.*?(\d+)'))
if not playlistId_list:
playlistId_list = response.selector.re(
re.compile(r'var PLAYLIST_ID.*?(\d+)'))
if not playlistId_list:
playlistId_list = response.selector.re(
re.compile(r'= playlistId.*?(\d+)'))
if playlistId_list:
playlistId = playlistId_list[0]
vid = ""
vid_list = response.selector.re(re.compile(r'var vid.*?(\d+)'))
if vid_list:
vid = vid_list[0]
if not playlistId or not vid:
return mvitem
ep_item = MediaItem()
ep_item["cont_id"] = playlistId
if title_list:
ep_item["title"] = title_list[0]
ep_item["actor"] = pers
ep_item["director"] = dirs
ep_item["site_id"] = self.site_id
ep_item["channel_id"] = cat_id
ep_item["url"] = Util.normalize_url(response.request.url, "sohu")
if text:
ep_item["intro"] = text[0].strip()
mvitem = MediaVideoItem()
mvitem["media"] = ep_item
if untrack_id:
mvitem["untrack_id"] = untrack_id
if sid:
mvitem["sid"] = sid
vitem = VideoItem()
vitem["title"] = ep_item["title"] if 'title' in ep_item else None
vitem["url"] = ep_item["url"]
vitem["vnum"] = "1"
vitem["os_id"] = self.os_id
vitem["ext_id"] = Util.md5hash(ep_item["url"])
vitem["site_id"] = self.site_id
vitem["cont_id"] = vid
videoitems = []
videoitems.append(vitem)
mvitem["video"] = videoitems
mvitem["media"]["info_id"] = Util.md5hash(
Util.summarize(mvitem["media"]))
Util.set_ext_id(mvitem["media"], mvitem["video"])
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return mvitem
def list_json_parse(self, response):
items = []
try:
origin_url = response.request.meta['url']
request_url = response.request.url
logging.log(logging.INFO, 'json api url: %s' % request_url)
page = response.request.meta[
'page'] if 'page' in response.request.meta else 1
if page > self.max_update_page:
return items
channel_id = response.request.meta[
'id'] if 'id' in response.request.meta else None
list_json_postfix_url = response.request.meta[
'postfix_url'] if 'postfix_url' in response.request.meta else None
json_datas = json.loads(response.body)
videos = []
if json_datas:
videos = json_datas[
'data_list'] if 'data_list' in json_datas else []
if videos:
#表明仍有下一页
video_url = 'http://www.letv.com/ptv/vplay/%s.html'
for item in videos:
mediaVideoItem = MediaVideoItem()
mediaItem = MediaItem()
mediaItem['channel_id'] = channel_id
if 'rating' in item and item['rating']:
mediaItem['score'] = item['rating']
subCategoryName = item['subCategoryName']
mediaItem['type'] = subCategoryName.replace(',', ';')
mediaVideoItem['media'] = mediaItem
release_date = item['releaseDate']
if release_date:
release_date = float(release_date)
if release_date > 0:
release_date = release_date / 1000
release_date = time.localtime(release_date)
release_date = '%s-%s-%s' % (release_date.tm_year,
release_date.tm_mon,
release_date.tm_mday)
mediaItem['release_date'] = Util.str2date(
release_date)
vid = ''
if 'vids' in item:
vids = item['vids']
vids = vids.split(',')
vid = vids[0]
elif 'vid' in item:
vid = item['vid']
if vid:
url = video_url % vid
items.append(
Request(url=url,
callback=self.video_parse,
meta={'item': mediaVideoItem}))
#下一页
page = page + 1
url = self.list_json_prefix_url + list_json_postfix_url + 'p=%s' % page
items.append(
Request(url=url,
callback=self.list_json_parse,
meta={
'page': page,
'id': channel_id,
'postfix_url': list_json_postfix_url,
'url': url
}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'json api url: %s' % request_url)
logging.log(logging.INFO, 'origin url: %s' % origin_url)
def parse_play_list(self, cat_id, url, flag, response):
item = None
videoitems = []
try:
ep_item = MediaItem()
item = MediaVideoItem()
item["media"] = ep_item
item['video'] = videoitems
info = None
try:
info = self.httpdownload.get_data(url)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return videoitems
if not info or len(info) < 2:
return videoitems
msg = info
bodylen = len(msg) - 1
index = msg.find(flag) + len(flag) + 1
info = msg[index:bodylen]
jinfo = json.loads(info)
if "video_play_list" not in jinfo:
return videoitems
itemlist = jinfo["video_play_list"]["playlist"]
for titem in itemlist:
if "episode_number" not in titem:
continue
info = titem["episode_number"]
if info and titem["title"].find(u"预告") < 0 and url.find(
"qq.com") >= 0:
vitem = VideoItem()
vitem["title"] = titem["title"]
tvnum = string.replace(info, "-", "")
#集数不是数字，是字符串，http://v.qq.com/detail/x/xk98t8hntls72f4.html
tvnum_list = re.findall(r'[\D]+', tvnum)
if not tvnum_list:
vitem["vnum"] = string.replace(info, "-", "")
else:
continue
vitem["os_id"] = self.os_id
vitem["site_id"] = self.site_id
turl = ""
if int(cat_id) == int(self.tv_id) or int(cat_id) == int(
self.cartoon_id):
turl = Util.normalize_url(titem["url"], "qq", "tv")
else:
turl = Util.normalize_url(titem["url"], "qq")
if turl:
vitem["ext_id"] = Util.md5hash(turl)
#vitem["cont_id"] = self.get_vid(response.body,turl)
vitem["url"] = turl
vitem["cont_id"] = self.get_qq_showid(vitem["url"])
else:
continue
videoitems.append(vitem)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return videoitems
def list_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
first = response.request.meta[
'first'] if 'first' in response.request.meta else False
channel_id = response.request.meta[
'id'] if 'id' in response.request.meta else None
if first:
sels = response.xpath('//div[@class="tab_box"]//a')
for sel in sels:
texts = sel.xpath('.//span/text()').extract()
if texts:
text = texts[0].replace(' ', '')
if text == u'最新':
urls = sel.xpath('./@href').extract()
url = urls[0]
items.append(
Request(url=url,
callback=self.list_parse,
meta={'id': channel_id}))
break
else:
page = response.request.meta[
'page'] if 'page' in response.request.meta else 1
if page > self.max_update_page:
return items
#list列表
sels = response.xpath('//ul[@class="movielist"]/li')
for sel in sels:
results = kankan_extract.video_extract(sel)
for item in results:
mediaVideoItem = MediaVideoItem()
mediaItem = MediaItem()
mediaItem['channel_id'] = channel_id
kankan_extract.media_info_extract(sel, mediaItem)
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=item['url'],
callback=self.video_parse,
meta={'item': mediaVideoItem}))
break
#下一页
sels = response.xpath('//p[@class="list-pager-v2"]')
results = kankan_extract.next_page_extract(sels)
page = page + 1
for item in results:
url = Util.get_absolute_url(item, prefix_url)
items.append(
Request(url=url,
callback=self.list_parse,
meta={
'page': page,
'id': channel_id
}))
break
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'url: %s' % request_url)
def video_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'video url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
mediaVideoItem = response.request.meta[
'item'] if 'item' in response.request.meta else MediaVideoItem(
)
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
if prefix_url == self.vip_prefix_url:
mediaItem['paid'] = '1'
else:
mediaItem['paid'] = '0'
mediaItem['url'] = request_url
pptv_extract.media_info_extract(response, mediaItem)
videoItems = []
if u'电影' == mediaItem['channel_id']:
if 'cont_id' not in mediaItem or not mediaItem['cont_id']:
return items
videoItem = VideoItem()
videoItem['url'] = mediaItem['url']
videoItem['cont_id'] = mediaItem['cont_id']
Util.copy_media_to_video(mediaItem, videoItem)
self.set_video_info(videoItem)
videoItems.append(videoItem)
else:
sel = response.xpath('//script[@type="text/javascript"]')
#获取pid&cid用于获取电视剧，综艺，动漫的剧集信息
if sel:
pids = sel.re('\"pid\"[ ]?:[ ]?(\d+)')
cids = sel.re('\"cat_id\"[ ]?:[ ]?(\d+)')
vids = sel.re('\"id\"[ ]?:[ ]?(\d+)')
if pids and cids and vids:
pid = pids[0]
cid = cids[0]
vid = vids[0]
page = 1
#给media的cont_id赋值
mediaItem['cont_id'] = pid
while True:
meta = {
'pid': pid,
'cid': cid,
'vid': vid,
'page': page
}
url = self.album_api % (pid, cid, vid, page)
result = Util.get_url_content(url)
page_result = self.album_json_resolve(
result, mediaItem, meta)
if not page_result['items']:
#该接口暂时由于获取不到video url，暂不提供
#for auth in self.auths:
# url = self.auth_album_api % (pid, auth)
# result = Util.get_url_content(url)
# page_items = self.auth_album_xml_resolve(result, mediaItem, meta)
# if page_items:
# videoItems = page_items
# break
break
else:
videoItems = videoItems + page_result['items']
page = page + 1
if videoItems:
#设置ext_id
Util.set_ext_id(mediaItem, videoItems)
self.set_media_info(mediaItem)
mediaVideoItem['media'] = mediaItem
mediaVideoItem['video'] = videoItems
items.append(mediaVideoItem)
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'video url: %s' % request_url)
def list_parse(self, response):
items = []
try:
request_url = response.request.url
logging.log(logging.INFO, 'url: %s' % request_url)
prefix_url = Util.prefix_url_parse(request_url)
level = response.request.meta[
'level'] if 'level' in response.request.meta else -1
if level == 0:
#第一次进入list页面
sels = response.xpath(
'//div[@id="hony-searchtag-condition"]/p')
for list_channel in hunantv_extract.list_channels:
list_postfix_urls = sels.xpath(
'.//a[normalize-space(text())="%s"]/@href' %
list_channel).extract()
if list_postfix_urls:
list_postfix_url = list_postfix_urls[0]
url = Util.get_absolute_url(list_postfix_url,
prefix_url)
items.append(
Request(url=url,
callback=self.list_parse,
meta={'id': list_channel}))
else:
page = response.request.meta[
'page'] if 'page' in response.request.meta else 1
channel_id = response.request.meta[
'id'] if 'id' in response.request.meta else None
if page > self.max_update_page:
return items
#获取播放地址
sels = response.xpath('//div[@class="play-index-con-box"]')
results = hunantv_extract.video_extract(sels)
for item in results:
mediaVideoItem = MediaVideoItem()
mediaItem = MediaItem()
mediaItem['channel_id'] = channel_id
video_sels = sels.xpath('.//a[@href="%s"]/..' %
item['url'])
hunantv_extract.media_info_extract(video_sels, mediaItem)
mediaItem['poster_url'] = Util.get_absolute_url(
mediaItem['poster_url'], prefix_url)
url = Util.get_absolute_url(item['url'], prefix_url)
mediaVideoItem['media'] = mediaItem
items.append(
Request(url=url,
callback=self.video_parse,
meta={'item': mediaVideoItem}))
#下一页
results = hunantv_extract.next_page_extract(response)
if results:
result = results[0]
result = Util.get_absolute_url(result, prefix_url)
page = page + 1
items.append(
Request(url=result,
callback=self.list_parse,
meta={
'page': page,
'id': channel_id
}))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'url: %s' % request_url)