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 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()
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 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)
