def parse_episode(self, response):
try:
log.msg('%s' % response.request.url)
thumb_url = response.request.meta['thumb_url']
upload_time = response.request.meta['upload_time']
category = response.request.meta['category']
kw_id = response.request.meta[
'kw_id'] if 'kw_id' in response.request.meta else 1
items = []
#owner
owner = response.xpath(
'//div[@class="yt-user-info"]/a/@data-ytid').extract()
owner_url = response.xpath(
'//div[@class="yt-user-info"]/a/@href').extract()
owner_show_id = None
if owner:
owner_show_id = owner[0]
items.append(
Request(url=self.url_prefix + owner_url[0] + "/about",
callback=self.parse_about))
#video info
title = response.xpath('//span[@id="eow-title"]/text()').extract()
#category = response.xpath('//p[@id="eow-category"]/a/text()').extract()
tag = response.xpath(
'./head/meta[@name="keywords"]/@content').extract()
#upload = response.xpath('//p[@id="watch-uploader-info"]/strong/text()').extract()
description = response.xpath(
'//p[@id="eow-description"]/descendant-or-self::*/text()'
).extract()
played = response.xpath(
'//div[@class="watch-view-count"]/text()').extract()
#other info
sts = re.search(r'\"sts\": ?(\d+)', response.body)
ep_item = EpisodeItem()
ep_item['show_id'] = Util.get_youtube_showid(response.request.url)
if owner_show_id:
ep_item['owner_show_id'] = owner_show_id
if title:
ep_item['title'] = title[0].strip()
if tag:
ep_item['tag'] = tag[0].replace(', ', '|')
if category:
#ep_item['category'] = category[0].replace('&', '|')
ep_item['category'] = category
'''
if upload:
ptime = Util.get_youtube_publish(upload[0])
if ptime:
ep_item['upload_time'] = ptime
'''
if upload_time:
t = Util.get_youtube_upload_time(upload_time[0].strip())
if t:
ep_item['upload_time'] = Util.get_datetime_delta(
datetime.now(), t)
if description:
ep_item['description'] = "\n".join(description)
if thumb_url:
ep_item['thumb_url'] = thumb_url[0]
if played:
pld = Util.normalize_played(played[0])
if pld:
ep_item['played'] = Util.normalize_played(played[0])
else:
ep_item['played'] = '0'
ep_item['spider_id'] = self.spider_id
ep_item['site_id'] = self.site_id
ep_item['url'] = Util.normalize_youtube_url(response.request.url)
ep_item['kw_id'] = kw_id
query = Util.encode({'video_id': ep_item['show_id'], \
'eurl': 'https://youtube.googleapis.com/v/' + ep_item['show_id'], \
'sts': sts.groups()[0] if sts else ''})
items.append(
Request(url='http://www.youtube.com/get_video_info?' + query,
callback=self.parse_other_info,
meta={'item': ep_item}))
return items
except Exception as e:
log.msg(traceback.format_exc(), level=log.ERROR)
def video_parse(self, response):
items = []
try:
kw_id = response.request.meta[
'kw_id'] if 'kw_id' in response.request.meta else None
pg_id = response.request.meta[
'pg_id'] if 'pg_id' in response.request.meta else None
cat_id = response.request.meta[
'cat_id'] if 'cat_id' in response.request.meta else None
subject_id = response.request.meta[
'subject_id'] if 'subject_id' in response.request.meta else None
show_id = Util.get_youtube_showid(response.request.url)
if not show_id:
return items
#owner
owner = response.xpath(
'//div[@class="yt-user-info"]/a/@data-ytid').extract()
owner_url = response.xpath(
'//div[@class="yt-user-info"]/a/@href').extract()
owner_show_id = None
if owner:
owner_show_id = owner[0]
items.append(
Request(url=self.youtube_url_prefix + owner_url[0] +
"/about",
callback=self.video_about_parse))
#video info
title = response.xpath('//span[@id="eow-title"]/text()').extract()
tag = response.xpath(
'./head/meta[@name="keywords"]/@content').extract()
description = response.xpath(
'//p[@id="eow-description"]/descendant-or-self::*/text()'
).extract()
played = response.xpath(
'//div[@class="watch-view-count"]/text()').extract()
category = response.xpath(
'//div[@id="watch-description"]//ul[@class="content watch-info-tag-list"]/li/a/text()'
).extract()
upload = response.xpath(
'//meta[@itemprop="datePublished"]/@content').extract()
#该方法获取的缩略图
thumb_url = response.xpath(
'//link[@itemprop="thumbnailUrl"]/@href').extract()
#other info
sts = re.search(r'\"sts\": ?(\d+)', response.body)
ep_item = EpisodeItem()
ep_item['show_id'] = show_id
#这里缩略图采用合成的方式得到['default', 'mqdefault', 'hqdefault', 'sddefault', 'maxresdefault']
#ep_item['thumb_url'] = self.thumb_url_prefix + '/' + show_id + '/default.jpg'
if owner_show_id:
ep_item['owner_show_id'] = owner_show_id
if title:
ep_item['title'] = title[0].strip()
if tag:
ep_item['tag'] = tag[0].replace(', ', '|')
if description:
ep_item['description'] = "\n".join(description)
if played:
pld = Util.normalize_played(played[0])
if pld:
ep_item['played'] = Util.normalize_played(played[0])
else:
ep_item['played'] = '0'
if kw_id:
ep_item['kw_id'] = kw_id
if pg_id:
ep_item['pg_id'] = pg_id
if cat_id:
ep_item['cat_id'] = cat_id
if subject_id:
ep_item['subject_id'] = subject_id
if thumb_url:
ep_item['thumb_url'] = thumb_url[0]
if category:
category = category[0].strip()
#https://www.youtube.com/watch?v=lwy4qwaByVQ
ep_item['category'] = category.replace('&', '|')
if upload:
upload = upload[0].strip()
struct_time = None
struct_time = time.strptime(upload, '%b %d, %Y')
if not struct_time:
struct_time = time.strptime(upload, '%Y年%m月%d日')
if struct_time:
time_str = time.strftime('%Y-%m-%d %H:%M:%S', struct_time)
#time_str = "%s-%s-%s %s" % (struct_time.tm_year, struct_time.tm_mon, struct_time.tm_mday, time_str)
ep_item['upload_time'] = time_str
ep_item['spider_id'] = self.spider_id
ep_item['site_id'] = self.site_id
ep_item['url'] = Util.normalize_youtube_url(response.request.url)
query = Util.encode({'video_id': ep_item['show_id'], \
'eurl': 'https://youtube.googleapis.com/v/' + ep_item['show_id'], \
'sts': sts.groups()[0] if sts else ''})
items.append(
Request(url='http://www.youtube.com/get_video_info?' + query,
callback=self.video_other_info_parse,
meta={'item': ep_item}))
except Exception, e:
log.msg(traceback.format_exc(), level=log.ERROR)
