def api_parse(self, mediaVideoItem):
items = []
try:
mediaItem = mediaVideoItem['media']
logging.log(logging.INFO,
'api parse pid: %s' % mediaItem['cont_id'])
self.api_media_info(mediaItem)
if 'title' in mediaItem:
videoItems = []
pagenum = 1
while True:
videos_url = self.other_album_api % (mediaItem['cont_id'],
pagenum)
result = Util.get_url_content(videos_url)
page_items = self.other_album_resolve(text=result,
meta={
'url':
videos_url,
'pagenum':
pagenum
})
if not page_items:
break
videoItems = videoItems + page_items
pagenum = pagenum + 1
if videoItems:
if 'url' not in mediaItem:
mediaItem['url'] = videoItems[0]['url']
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())
def api_episode_info(self, mvItem=None, playlistId='', cat_id=''):
# 应该保证mvItem,playlistId不为空，且包含mid或者sid、untrack_id,包含channel_id、site_id
items = []
try:
mvitem = mvItem
ep_item = mvitem["media"]
url = self.album_api % (playlistId, 1)
logging.log(logging.INFO, 'api_episode_info, info url %s' % url)
info = self.httpdownload.get_data(url)
info = info.decode('gbk').encode('utf-8')
info_json = json.loads(info)
actor_list = info_json.get("mainActors")
director_list = info_json.get("directors")
type_list = info_json.get("categories")
if "actor" not in ep_item and actor_list:
ep_item["actor"] = Util.join_list_safely(actor_list)
if "director" not in ep_item and director_list:
ep_item["director"] = Util.join_list_safely(director_list)
if "type" not in ep_item and type_list:
ep_item["type"] = Util.join_list_safely(type_list)
if "title" not in ep_item:
ep_item["title"] = info_json.get("albumName")
if "district" not in ep_item:
ep_item["district"] = info_json.get("area")
if "release_date" not in ep_item and info_json.get("publishYear"):
ep_item["release_date"] = Util.str2date(
str(info_json.get("publishYear")))
if "intro" not in ep_item:
ep_item["intro"] = info_json.get("albumDesc")
if "poster_url" not in ep_item or not str.strip(
str(ep_item["poster_url"])):
ep_item["poster_url"] = info_json.get("pic240_330")
if "cont_id" not in ep_item:
ep_item["cont_id"] = playlistId
ttvitem = []
if ep_item['title']:
mvitem['media'] = ep_item
ttvitem = self.parse_video_item(cat_id, playlistId)
if ttvitem:
mvitem['video'] = ttvitem
if "url" not in mvitem["media"]:
mvitem["media"]["url"] = ttvitem[0]['url']
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)
except Exception as e:
logging.log(logging.ERROR, traceback.format_exc())
return items
def api_media_info(self, mediaVideoItem, vid, prefix_video_url):
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
try:
miu = self.media_info_url % vid
jdata = self.httpdownload.get_data(miu)
if not jdata:
pass
else:
ddata = json.loads(jdata)
assert int(ddata.get('code', 202)) == 200, "接口获取媒体信息失败"
detail = ddata.get('data').get('detail')
assert type(detail) == dict
mediaItem['cont_id'] = str(detail.get('collectionId'))
mediaItem['title'] = detail.get('collectionName')
mediaItem['director'] = Util.join_list_safely(
detail.get('director').split('/'))
mediaItem['actor'] = Util.join_list_safely(
detail.get('player').split('/'))
mediaItem['release_date'] = Util.str2date(
detail.get('publishTime'))
mediaItem['vcount'] = int(detail.get('totalvideocount'))
latest = detail.get('lastseries')
m = re.compile('\D*(\d+)\D*').match(latest)
if m:
mediaItem['latest'] = m.group(1)
if mediaItem['vcount'] == 1:
mediaItem['latest'] = 1
mediaItem['paid'] = detail.get('isvip')
mediaItem['intro'] = detail.get('desc')
mediaItem['poster_url'] = detail.get('image')
mediaItem['site_id'] = self.site_id
mediaItem['channel_id'] = self.channels_name_id[
mediaItem['channel_id']]
info_id = Util.md5hash(Util.summarize(mediaItem))
mediaItem['info_id'] = info_id
vcount = mediaItem['vcount']
if not vcount:
vcount = 1
else:
vcount = int(vcount)
video_list = self.api_video_list(vid, vcount, prefix_video_url,
mediaItem['channel_id'])
if video_list:
Util.set_ext_id(mediaItem, video_list)
mediaVideoItem['video'] = video_list
mediaVideoItem['media'] = mediaItem
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.ERROR, vid)
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 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 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 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 None
mediaItem = mediaVideoItem[
'media'] if 'media' in mediaVideoItem else MediaItem()
sels = response.xpath('//script[@type="text/javascript"]')
letv_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//div[@class="play"]')
letv_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//dl[@class="textInfo"]')
if sels:
#电视剧、综艺、动漫
letv_extract.media_info_extract(sels, mediaItem)
else:
#电影
sels = response.xpath('//div[@class="detail"]')
letv_extract.media_info_extract(sels, mediaItem)
#获取正片的url
videoItems = []
if u'电影' == mediaItem['channel_id']:
pagenum = 1
videos_url = self.other_album_api % (mediaItem['cont_id'],
pagenum)
result = Util.get_url_content(videos_url)
page_items = self.other_album_resolve(text=result,
meta={
'url': videos_url,
'pagenum': pagenum
})
videoItems = page_items
#综艺
elif u'综艺' == mediaItem['channel_id']:
sels = response.xpath(
'//div[@class="listTab"]//div[@data-statectn="n_click"]')
if sels:
year_month_sels = sels.xpath('.//a')
for year_month_sel in year_month_sels:
years = year_month_sel.xpath('./@list-year').extract()
months = year_month_sel.xpath(
'./@list-month').extract()
year = None
month = None
if years:
year = years[0]
if months:
month = months[0]
if year and month:
videos_url = self.zongyi_album_api % (
year, month, mediaItem['cont_id'])
result = Util.get_url_content(videos_url)
videoItems = videoItems + self.zongyi_album_resolve(
text=result,
meta={
'url': videos_url,
'year': year,
'month': month
})
elif mediaItem['channel_id'] in [u'电视剧', u'动漫']:
pagenum = 1
while True:
videos_url = self.other_album_api % (mediaItem['cont_id'],
pagenum)
result = Util.get_url_content(videos_url)
page_items = self.other_album_resolve(text=result,
meta={
'url':
videos_url,
'pagenum':
pagenum
})
if not page_items:
break
videoItems = videoItems + page_items
pagenum = pagenum + 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, '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 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.variety_id):
tvideoitems = self.parse_variety(response)
if tvideoitems:
for titem in tvideoitems:
videoitems.append(titem)
elif '/Play/show' not in url:
#if int(cat_id) != int(self.movie_id):
#ul_list = response.xpath('//div[@class="teleplay_gather tab_box"]/div[@class="list_tabs_cont"]/ul/li')
ul_list = response.xpath(
'//div[@class="teleplay_gather tab_box"]/div/ul/li')
if ul_list:
#http://www.wasu.cn/Tele/index/id/6539647
for li in ul_list:
yugaopian = li.xpath('.//i[@class="yugao"]').extract()
if yugaopian:
continue
url = li.xpath('./a/@href').extract()
ttitle = li.xpath('./a/@title').extract()
snum = li.xpath('./a/text()').extract()
play_num = ""
if snum:
play_num = self.get_play_num(snum[0])
if int(cat_id) == int(self.variety_id):
play_num1 = self.getvnum(self.url_prefix + url[0])
if play_num1:
play_num = play_num1
if not ttitle:
ttitle = [play_num]
vitem = None
if self.site_name == Util.guess_site(url[0]):
vitem = self.compose_vitem([url[0]],
[title[0].strip()],
play_num)
else:
vitem = self.compose_vitem(
[self.url_prefix + url[0]], [title[0].strip()],
play_num)
if 'url' in vitem:
videoitems.append(vitem)
if not ul_list:
#http://www.wasu.cn/Tele/index/id/6786984
ul_list = response.xpath(
'//div[@class="tab_box"]//div[ends-with(@class, "col2")]'
)
for li in ul_list:
yugaopian = li.xpath('.//i[@class="yugao"]').extract()
if yugaopian:
continue
url = li.xpath(
'./div[@class="ws_des"]/p[1]/a/@href').extract()
ttitle = li.xpath(
'./div[@class="ws_des"]/p[2]/span/text()').extract(
)
snum = li.xpath(
'./div[@class="ws_des"]/p[1]/a/text()').extract()
play_num = ""
if snum:
play_num = self.get_play_num(snum[0])
if int(cat_id) == int(self.variety_id):
play_num1 = self.getvnum(self.url_prefix + url[0])
if play_num1:
play_num = play_num1
if not ttitle:
ttitle = [play_num]
vitem = None
if self.site_name == Util.guess_site(url[0]):
vitem = self.compose_vitem([url[0]],
[title[0].strip()],
play_num)
else:
vitem = self.compose_vitem(
[self.url_prefix + url[0]], [title[0].strip()],
play_num)
if 'url' in vitem:
videoitems.append(vitem)
else:
#elif int(cat_id) == int(self.movie_id):
#无媒体页的播放页
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_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 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 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_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 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()
videoItems = []
#普通媒体页
channel_id_fun = mediaItem['channel_id']
sels = response.xpath('//div[@id="qitancommonarea"]')
iqiyi_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//script[@type="text/javascript"]')
iqiyi_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//div[@class="mod_search_topic mb20"]')
if not sels:
sels = response.xpath('.//div[@id="block-B"]')
iqiyi_extract.media_info_extract(sels, mediaItem)
#特辑媒体页
iqiyi_extract.media_info_extract(response, mediaItem)
cont_id = mediaItem['cont_id'] if 'cont_id' in mediaItem else None
title = mediaItem['title'] if 'title' in mediaItem else None
if cont_id and title:
cont_ids = cont_id.split('|')
cont_id = cont_ids[0]
cont_type = cont_ids[1]
'''
vip_url = self.vip_api % cont_id
try:
result = Util.get_url_content(vip_url)
if result:
json_data = json.loads(result)
if json_data['code'] == self.api_success_code:
mediaItem['paid'] = '1'
else:
mediaItem['paid'] = '0'
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'vip url: %s' % vip_url)
logging.log(logging.INFO, '-------json data----------')
logging.log(logging.INFO, result)
'''
mediaItem['channel_id'] = channel_id_fun
channel_id_site = iqiyi_extract.list_channels_id[
channel_id_fun]
if cont_type == 'source_id':
#年份，都采用统一的api来获取
#years = response.xpath('//div[@data-widget="album-sourcelist"]//div[@data-widget-year="album-yearlist"]//a/@data-year').extract()
#快乐大本营，天天向上的等,提供的是接口
url = self.source_year_api % (channel_id_site, cont_id)
result = Util.get_url_content(url)
years = self.source_year_json_resolve(result, url)
for year in years:
url = self.source_media_api % (
channel_id_site, cont_id, year, channel_id_site,
cont_id, year)
result = Util.get_url_content(url)
videoItems = videoItems + self.source_media_json_resolve(
result, mediaItem, url)
elif cont_type == 'album_id':
#其他，其他的接口
page = 1
url = self.album_media_api % (cont_id, page, cont_id, page)
result = Util.get_url_content(url)
videoItems = videoItems + self.album_media_json_resolve(
result, mediaItem, url)
if not videoItems:
#特殊节目暂时不爬取,http://www.iqiyi.com/yule/cjkgbj.html
#不作任何处理
videoItems = videoItems
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))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'media 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 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):
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']
title = response.request.meta['title']
actor = response.request.meta['actor']
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 = []
if not poster_url:
poster_url_list = response.xpath('//div[@class="cover_img"]/div[@class="pack pack_album"]/div[@class="pic"]/img/@src').extract()
if poster_url_list:
poster_url = poster_url_list[0]
if not title:
title_list = response.xpath('//div[@class="cover_info"]/h2/strong/@title').extract()
if title_list:
title = title_list[0]
if not actor:
#actor_list = response.xpath('//div[@class="cover_keys"]/span/a/text()').extract()
actor_list = response.xpath('//div[@class="cover_keys"]/span/span[text()="%s"]/../a/text()' % u' 主演：').extract()
if actor_list:
actor = Util.join_list_safely(actor_list)
#actor = "|".join([t.strip() for t in actor_list])
#performer
pers = actor
type_list = response.xpath('//div[@class="cover_keys"]/span/span[text()="%s"]/../a/text()' % u'类型：\n').extract()
district_list = response.xpath('//div[@class="cover_keys"]/span/span[text()="%s"]/../a/text()' % u'地区：').extract()
release_date_list = response.xpath('//div[@class="cover_keys"]/span/span[text()="%s"]/../a/text()' % u'年代：').extract()
types = None
if type_list:
types = Util.join_list_safely(type_list)
#director
director_list = response.xpath('//div[@class="cover_keys"]/span/span[text()="%s"]/../a/text()' % u'编导：').extract()
if not director_list:
director_list = response.xpath('//div[@class="cover_keys"]/span/span[text()="%s"]/../a/text()' % u'导演：').extract()
dirs = Util.join_list_safely(director_list)
#dirs = "|".join([t.strip() for t in director_list])
#text
text = response.xpath('//div[@class="cover_info"]/div[@class="desc"]/p/text()').extract()
#sourceid
sourceid = self.get_tudou_showid(response.request.url)
videoitems = []
ep_item = MediaItem()
if len(title) > 0:
ep_item["title"] = title
if len(pers) > 0:
ep_item["actor"] = pers
if len(dirs) > 0:
ep_item["director"] = dirs
if types:
ep_item["type"] = types
if district_list:
ep_item["district"] = district_list[0].strip()
if release_date_list:
ep_item["release_date"] = Util.str2date(release_date_list[0])
#ep_item["info_id"] = Util.md5hash(tinfo)
ep_item["cont_id"] = sourceid
ep_item["site_id"] = self.site_id
ep_item["url"] = response.request.url
ep_item["channel_id"] = cat_id
ep_item["poster_url"] = poster_url
if len(text) > 0:
ep_item["intro"] = text[0]
mvitem = MediaVideoItem();
mvitem["media"] = ep_item;
mvitem["video"] = videoitems
lurl = "http://www.tudou.com/crp/getAlbumvoInfo.action?charset=utf-8&areaCode=110000&acode=" + str(sourceid)
info = self.httpdownload.get_data(lurl)
jinfo = json.loads(info)
if "items" in jinfo:
for sitem in jinfo["items"]:
vitem = VideoItem()
vitem["title"] = sitem["itemTitle"]
vitem["vnum"] = sitem["episode"]
vitem["os_id"] = self.os_id
trailer = sitem['trailer']
if not sitem["itemPlayUrl"]:
continue
#预告片
if trailer:
continue
turl = Util.normalize_url(sitem["itemPlayUrl"],"tudou")
vitem["url"] = turl
vitem["os_id"] = self.os_id
vitem["site_id"] = self.site_id
vitem["ext_id"] = Util.md5hash(turl)
vitem["cont_id"] = self.get_tudou_showid(turl)
#if "ext_id" not in mvitem["media"]:
# mvitem["media"]["ext_id"] = vitem["ext_id"]
#vitem["media_ext_id"] = vitem["ext_id"]
mvitem["video"].append(vitem)
if len(mvitem["video"]) > 0:
Util.set_ext_id(mvitem["media"],mvitem["video"])
mvitem["media"]["info_id"] = Util.md5hash(Util.summarize(mvitem["media"]))
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, 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 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()
#获取播放地址
videoItems = []
videoItem = VideoItem()
Util.copy_media_to_video(mediaItem, videoItem)
sels = response.xpath('//div[@class="laMovPIC fl pr22"]')
dy1905_extract.video_info_extract(sels, videoItem)
if 'url' not in videoItem:
#如果videoItem['url']为空，则表示只有影片资料，无播放地址，直接扔掉
logging.log(logging.INFO, '该影片找不到播放地址: %s' % request_url)
return items
url = videoItem['url']
url = Util.get_absolute_url(url, prefix_url)
videoItem['url'] = url
self.set_video_info(videoItem)
videoItems.append(videoItem)
#媒体属性
#设置媒体付费属性
video_prefix_url = Util.prefix_url_parse(url)
if video_prefix_url in self.vip_prefix_urls:
mediaItem['paid'] = '1'
else:
mediaItem['paid'] = '0'
sels = response.xpath('//div[@class="laMovPIC fl pr22"]')
dy1905_extract.media_info_extract(sels, mediaItem)
sels = response.xpath('//div[@class="laMovMAIN fl"]')
dy1905_extract.media_info_extract(sels, mediaItem)
#剧情与演职人员
nav_sels = response.xpath(
'//ul[@class="navSMb"]//li[@class="mdbpLeft2"]//div[@class="nowDefLine DefBOttom"]//a'
)
if nav_sels:
for sel in nav_sels:
labels = sel.xpath('./text()').extract()
urls = sel.xpath('./@href').extract()
if labels and urls:
label = labels[0].strip()
if label.startswith(u'剧情') or label.startswith('演职人员'):
url = urls[0]
url = Util.get_absolute_url(url, prefix_url)
result = Util.get_url_content(url)
dy1905_extract.media_more_info_resolve(
result, mediaItem)
#设置绝对路径
url = mediaItem['url']
url = Util.get_absolute_url(url, prefix_url)
mediaItem['url'] = url
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, 'media url: %s' % request_url)
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 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()
channel_id_fun = mediaItem['channel_id']
sels = response.xpath('//script[@type="text/javascript"]')
iqiyi_extract.media_info_extract(sels, mediaItem)
mediaItem['channel_id'] = channel_id_fun
sels = response.xpath(
'//div[@itemtype="http://schema.org/ShowEpisode"]')
iqiyi_extract.media_info_extract(sels, mediaItem)
#播放页 - 用于直接从播放页进入
sels = response.xpath(
'//div[@class="crumb_bar" or @class="mod-crumb_bar"]')
iqiyi_extract.media_info_extract(sels, mediaItem)
url = mediaItem['url'] if 'url' in mediaItem else ''
url_type = iqiyi_extract.url_type_resolve(url)
if url_type == URL_TYPE_MEDIA:
mediaVideoItem['media'] = mediaItem
url = mediaItem['url']
items.append(
Request(url=url,
callback=self.media_parse,
meta={'item': mediaVideoItem}))
else:
cont_id = mediaItem[
'cont_id'] if 'cont_id' in mediaItem else None
title = mediaItem['title'] if 'title' in mediaItem else None
if cont_id and title:
cont_ids = cont_id.split('|')
cont_id = cont_ids[0]
cont_type = cont_ids[1]
'''
vip_url = self.vip_api % cont_id
try:
result = Util.get_url_content(vip_url)
if result:
json_data = json.loads(result)
if json_data['code'] == self.api_success_code:
mediaItem['paid'] = '1'
else:
mediaItem['paid'] = '0'
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'vip url: %s' % vip_url)
'''
videoItems = []
if cont_type == 'source_id':
#年份，都采用统一的api来获取
#years = response.xpath('//div[@data-widget="album-sourcelist"]//div[@data-widget-year="album-yearlist"]//a/@data-year').extract()
#快乐大本营，天天向上的等,提供的是接口
url = self.source_year_api % (channel_id_site, cont_id)
result = Util.get_url_content(url)
years = self.source_year_json_resolve(result, url)
for year in years:
url = self.source_media_api % (
channel_id_site, cont_id, year,
channel_id_site, cont_id, year)
result = Util.get_url_content(url)
videoItems = videoItems + self.source_media_json_resolve(
result, mediaItem, url)
elif cont_type == 'album_id':
#其他，其他的接口
page = 1
url = self.album_media_api % (cont_id, page, cont_id,
page)
result = Util.get_url_content(url)
videoItems = videoItems + self.album_media_json_resolve(
result, mediaItem, url)
if not videoItems:
#特殊节目暂时不爬取,http://www.iqiyi.com/yule/cjkgbj.html
#不作任何处理
videoItems = videoItems
if videoItems:
#设置ext_id
Util.set_ext_id(mediaItem, videoItems)
self.set_media_info(mediaItem)
mediaVideoItem['media'] = mediaItem
mediaVideoItem['video'] = videoItems
print mediaVideoItem
items.append(mediaVideoItem)
#self.count = self.count + 1
#logging.log(logging.INFO, 'count: %s' % str(self.count))
except Exception, e:
logging.log(logging.ERROR, traceback.format_exc())
logging.log(logging.INFO, 'video url: %s' % request_url)
