def parse_detail(self, response):
type_name = None
for k,v in self.movie_category.items():
if str(v) == re.search('type=(.*?)&', unquote(response.url)).group(1):
type_name = k
datas = json.loads((response.text))
for data in datas:
type_name = type_name
titie = data.get('title')
rank = data.get('rank')
url = data.get('url')
actors = data.get('actors')
cover_url = data.get('cover_url')
regions = data.get('regions')
release_date = data.get('release_date')
score = data.get('score')
types = data.get('types')
vote_count = data.get('vote_count')
douban_item = DoubanMovieItem()
for field in douban_item.fields:
try:
douban_item[field] = eval(field)
except:
print('Field is Not Defined', field)
yield douban_item
def parse(self, response):
#print(response.body.decode())
item=DoubanMovieItem()
#movie_list=json.loads(response.body.decode())
# chrome_options = Options()
# chrome_options.add_argument('--headless')
# chrome_options.add_argument('--disable-gpu')
#chrome_options.add_argument('--no-sandbox')
#chrome_driver=r'C:\Users\igogle\Desktop\douban2\douban_movie\chromedriver.exe'
#browser=webdriver.Chrome(executable_path=chrome_driver)
#browser.get('https://www.amazon.com/ref=nav_logo')
#browser.implicitly_wait(5)
print('s')
print('s')
#movies=browser.find_element_by_xpath('//*[@id="desktop-4"]')
#test=movies[0].get_attribute("li")
movies= response.xpath('//*[@id="desktop-4"]//li/@data-sgproduct')
print('s')
print('s')
print(movies)
print(str(len(movies)))
print('s')
print('s')
#print(response.xpath('//*[@id="content"]/div/div[1]/div/div[4]/div/a[1]/p/text()'))
for movie in movies:
print('s')
print(movie.xpath('//p/text()').extract()[0])
#names=movie.xpath('.//p/text()').extract()[0]
#item['name']=names
#item['score']=movie['score']
#item['url']=movie['url']
#yield item
def parse_item(self, response):
sel = Selector(response)
item = DoubanMovieItem()
title = sel.xpath('//*[@id="content"]/h1/span[1]/text()').extract()
director = sel.xpath(
'//*[@id="info"]/span[1]/span[2]/a/text()').extract()
actor = sel.xpath('//*[@id="info"]/span[3]/span[2]/a/text()').extract()
#release_time = sel.xpath('//*[@id="info"]/span[11]/text()').extract()
#time = sel.xpath('//*[@id="info"]/span[13]/text()').extract()
star = sel.xpath(
'//*[@id="interest_sectl"]/div[1]/div[2]/strong/text()').extract()
item['title'] = title
item['director'] = director
item['actor'] = actor
#item['release_time'] = release_time
#item['time'] = time
item['star'] = star
yield item
print(title)
print(director)
print(actor)
# print(release_time)
#print(time)
print(star)
def parse(self, response):
time.sleep(random.randint(2, 5))
global START
item = DoubanMovieItem()
# https://movie.douban.com/j/new_search_subjects?sort=T&range=0,10&tags=%E7%94%B5%E5%BD%B1&start=20
# while True:
pat_title = r'title":"(.*?)"'
pat_rate = r'rate":"(.*?)"'
pat_url = r'url":"(.*?)"'
item['title'] = re.compile(pat_title).findall(
str(response.body.decode('utf-8')))
item['rate'] = re.compile(pat_rate).findall(
str(response.body.decode('utf-8')))
item['url'] = re.compile(pat_url).findall(
str(response.body.decode('utf-8')))
print(item['title'])
if item['title'] == []:
exit()
print('return item success')
yield item
print('yield item success')
next_url = 'https://movie.douban.com/j/new_search_subjects?sort=T&range=0,10&tags=%E7%94%B5%E5%BD%B1&start='
START = START + 20
print('next url :' + next_url + str(START))
yield Request(
next_url + str(START),
callback=self.parse,
headers={
'User-Agent':
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/22.0.1207.1 Safari/537.1"
},
encoding='utf-8')
print('yield request success')
def next_parse(self, response):
item2 = DoubanMovieItem()
print(item2)
item2['_name'] = response.xpath(
'//div[@id="content"]/h1/span[@property="v:itemreviewed"]/text()'
).getall()
item2['_year'] = response.xpath(
'//div[@id="content"]/h1/span[@class="year"]/text()').getall()
item2['_pic'] = response.xpath(
'//div[@id="mainpic"]/a/img/@src').getall()
print(item2)
selector_next = response.xpath('//div[@id="info"]')
item2['_director'] = selector_next.xpath(
'//span[@class="attrs"]/a/text()').getall()[:2]
item2['_writer'] = selector_next.xpath(
'//span[@class="attrs"]/a/text()').getall()[2:]
item2['_cast'] = selector_next.xpath(
'//span[@class="actor"]/span[@class="attrs"]/span/a/text()'
).getall()
item2['_type'] = selector_next.xpath(
'span[@property="v:genre"]/text()').getall()
item2['_country'] = selector_next.xpath('text()').getall()[7]
item2['_language'] = selector_next.xpath('text()').getall()[9]
item2['_premiere'] = selector_next.xpath(
'span[@property="v:initialReleaseDate"]/text()').getall()
item2['_episode'] = selector_next.xpath('text()').getall()[13]
item2['_runningtime'] = selector_next.xpath('text()').getall()[15]
item2['_plot'] = response.xpath(
'//span[@property="v:summary"]/text()').getall()
return item2
def parse(self, response):
selector = Selector(response)
ol_li = selector.xpath('//div[@class="item"]')
for li in ol_li:
movie = DoubanMovieItem()
movie['_id'] = str(ObjectId())
movie['rank'] = li.xpath(
'div[@class="pic"]/em/text()').extract_first()
movie['link'] = li.xpath(
'div[@class="pic"]/a/@href').extract_first()
movie['img'] = li.xpath(
'div[@class="pic"]/a/img/@src').extract_first()
movie['title'] = li.xpath(
'div[@class="pic"]/a/img/@alt').extract_first()
movie['star'] = li.xpath(
'div[@class="info"]/div[@class="bd"]/div[@class="star"]/span[@class="rating_num"]/text()'
).extract_first()
movie['quote'] = li.xpath(
'div[@class="info"]/div[@class="bd"]/p[@class="quote"]/span[@class="inq"]/text()'
).extract_first()
yield movie
next_page = response.xpath('//span[@class="next"]/a/@href')
if next_page:
url = 'https://movie.douban.com/top250' + next_page[0].extract()
yield Request(url=url, callback=self.parse)
def parse_movie(self, response):
global cookies
global headers
# 从返回的正文中取出[{...}]之间的内容
dict_str = re.search('\[\{.*}]', response.body).group()
# 取出所有{。。。}之间的内容，是dict格式的字符串
temp = re.findall('\{.*?}', dict_str)
# 替换布尔量，否则报错
temp1 = [x.replace('false', 'False') for x in temp]
temp2 = [x.replace('true', 'True') for x in temp1]
# 替换转义符号
temp3 = [x.replace('\\', '') for x in temp2]
# 使用eval把正确的字符串转为对应的类型
dict_list = [eval(x) for x in temp3]
for x in dict_list:
item = DoubanMovieItem()
item['title'] = x['title']
item['post_urls'] = [
x['cover'],
]
# pdb.set_trace()
yield scrapy.Request(
url=x['url'],
meta={'item': item}, # 通过meta把item传送到另外一个页面抓取中
cookies=cookies,
headers=headers,
callback=self.parse_intro,
# dont_filter=True,
)
def parse_movie(self, response):
hxs = HtmlXPathSelector(response)
movie = DoubanMovieItem()
movie['url_in_douban'] = unicode(response.url)
movie['name_in_douban'] = "".join(
hxs.select("//span[@property='v:itemreviewed']/text()").extract()
).strip()
movie['year'] = ("".join(
hxs.select("//span[@class='year']/text()").extract()).strip()
)[1:-1]
movie['length'] = "".join(
hxs.select(
"//span[@property='v:runtime']/@content").extract()).strip()
movie['url_in_imdb'] = "".join(
hxs.select("//div[@id='info']/a[last()]/@href").extract()).strip()
movie['score'] = "".join(
hxs.select(
"//strong[@class='ll rating_num'][@property='v:average']/text()"
).extract()).strip()
movie['scored_num'] = "".join(
hxs.select(
"//span[@property='v:votes']/text()").extract()).strip()
tag_names = hxs.select("//div[@class='tags-body']/a/text()").extract()
tag_times = hxs.select(
"//div[@class='tags-body']/a/span/text()").extract()
tag_num = len(tag_names)
tags = dict()
for i in range(tag_num):
tags[tag_names[i]] = tag_times[i][1:-1]
movie['tags'] = tags
return [movie]
def parse_movie(self, response):
print(response.status)
print(response.xpath('//li/span[@class="rec"]/@id'))
print(response.xpath('//span[@class="rating_per"]/text()'))
_setDNSCache()
movie_item = DoubanMovieItem()
# movie id
movie_item['movie_id'] = response.xpath(
'//li/span[@class="rec"]/@id').extract()
# movie title
movie_item['movie_title'] = response.xpath(
'//*[@id="content"]/h1/span[1]').extract()
# release_date
movie_item['release_date'] = response.xpath(
'.//h1/span[@class="year"]/text()').extract()
# 导演
movie_item['directedBy'] = response.xpath(
'.//a[@rel="v:directedBy"]/text()').extract()
# 电影主演
movie_item['starring'] = response.xpath(
'.//a[@rel="v:starring"]/text()').extract()
# 电影类别
movie_item['genre'] = response.xpath(
'.//span[@property="v:genre"]/text()').extract()
# 电影时长
movie_item['runtime'] = response.xpath(
'.//span[@property="v:runtime"]/text()').extract()
# # 电影的国别和语言
# temp = response.xpath('.//div[@id="info"]/text()').extract()
# movie_item['country'] = [p for p in temp if (p.strip() != '') & (p.strip() != '/')][0].strip()
# movie_item['language'] = [p for p in temp if (p.strip() != '') & (p.strip() != '/')][1].strip()
# 电影的评分
movie_item['rating_num'] = response.xpath(
'.//strong[@class="ll rating_num"]/text()').extract()
# 评分的人数
movie_item['vote_num'] = response.xpath(
'.//span[@property="v:votes"]/text()').extract()
# 电影1-5星的百分比
# movie_item['rating_per_stars5'] = response.xpath('.//span[@class="rating_per"]/text()').extract()[0].strip()
# movie_item['rating_per_stars4'] = response.xpath('.//span[@class="rating_per"]/text()').extract()[1].strip()
# movie_item['rating_per_stars3'] = response.xpath('.//span[@class="rating_per"]/text()').extract()[2].strip()
# movie_item['rating_per_stars2'] = response.xpath('.//span[@class="rating_per"]/text()').extract()[3].strip()
# movie_item['rating_per_stars1'] = response.xpath('.//span[@class="rating_per"]/text()').extract()[4].strip()
# 电影的剧情简介
intro = response.xpath('.//span[@class="all hidden"]/text()').extract()
if len(intro):
movie_item['intro'] = intro
else:
movie_item['intro'] = response.xpath(
'.//span[@property="v:summary"]/text()').extract()
# 电影的短评数
# movie_item['comment_num'] = response.xpath('.//div[@class="mod-hd"]/h2/span/a/text()').extract()[0].strip()
# # 电影的提问数
# movie_item['question_num'] = response.xpath('.//div[@class="mod-hd"]/h2/span/a/text()').extract()[1].strip()
# 最后输出
yield movie_item
def parse(self, response):
item = DoubanMovieItem()
json_text = response.text
movie_dict = json.loads(json_text)
for one_movie in movie_dict["subjects"]:
item["title"] = one_movie["title"]
item["rate"] = one_movie["rate"]
yield item
def parse_item(self, response):
for item in response.xpath(
'//div[@class="body-bg"]/div[@class="w1000"]/div/div/ul/li'):
l = DoubanMovieItemLoader(DoubanMovieItem(), item)
l.add_xpath('ebtang_id', './a/@href')
l.add_xpath('title', './a/text()')
l.add_xpath('date', './span/text()')
l.add_value('crawl_time', datetime.now())
yield l.load_item()
def parse(self, response):
json_string = response.body.decode('utf-8')
content = json.loads(json_string)
for movie in content['subjects']:
item = DoubanMovieItem()
item['movie_info'] = movie['url']
item['movie_pic'] = movie['cover']
item['movie_title'] = movie['title']
item['movie_score'] = movie['rate']
yield item
def parse_movie_item(self, response):
item = DoubanMovieItem()
item['url'] = response.url
item['name'] = response.xpath(
'//span[@property="v:itemreviewed"]/text()').extract_first()
item['summary'] = response.xpath(
'//span[@property="v:summary"]/text()').extract_first()
item['score'] = response.xpath(
'//strong[@property="v:average"]/text()').extract_first()
return item
def parse(self, response):
movies = Selector(response=response).xpath('//div[@class="hd"]')
for movie in movies:
item = DoubanMovieItem()
title = movie.xpath('./a/span[1]/text()').extract()
link = str(
movie.xpath('./a/@href').extract()[0]) + 'comments?status=P'
item['title'] = title
item['link'] = link
yield scrapy.Request(url=link,
meta={'item': item},
callback=self.parse2)
def parse_item(self, response):
i = {}
#i['domain_id'] = response.xpath('//input[@id="sid"]/@value').extract()
#i['name'] = response.xpath('//div[@id="name"]').extract()
#i['description'] = response.xpath('//div[@id="description"]').extract()
i['url'] = response.url
i['name'] = response.xpath(
'//span[@property="v:itemreviewed"]/text()').extract_first()
i['summary'] = response.xpath(
'//span[@property="v:summary"]/text()').extract_first().strip()
i['score'] = float(response.xpath('//strong/text()').extract_first())
return DoubanMovieItem(i)
def parse_item(self, response):
for item in response.xpath('//div[@id="content"]/div/div[1]/ol/li'):
l = DoubanMovieItemLoader(DoubanMovieItem(), item)
l.add_xpath('rank', './div/div/em/text()')
l.add_xpath('picture', './div/div/a/img/@src')
l.add_xpath('title', './div/div/div/a/span/text()')
l.add_xpath('info', './div/div/div/p/text()')
l.add_css('star', 'div.star span.rating_num::text')
l.add_xpath('people', './div/div/div/div/span[4]/text()')
l.add_css('quote', 'p.quote > span.inq::text')
l.add_value('crawl_time', datetime.now())
yield l.load_item()
def parse_page(self, response):
item = DoubanMovieItem()
try:
soup = BeautifulSoup(response.body,
'html.parser',
from_encoding='utf-8')
movie_name_tag = soup.find('div', id='content').findChild('h1')
no = soup.find('span', 'top250-no').get_text()
# no = response.xpath('//span[@class=top250-no]/text()').extract()
movie_name = movie_name_tag.findChildren()[0].get_text(
) + movie_name_tag.findChildren()[1].get_text()
# movie_name = response.xpath('//h1/span/text()').extract()
# print(no,movie_name)
director = soup.find('a', rel='v:directedBy').get_text()
writer = soup.find('span',
text='编剧').next_sibling.next_sibling.text
actor = '/'.join(star.text
for star in soup.findAll('a', rel='v:starring'))
type = '/'.join(
genre.text
for genre in soup.findAll('span', property='v:genre'))
region = soup.find('span', text='制片国家/地区:').next_sibling
language = soup.find('span', text='语言:').next_sibling
date = soup.find('span', property='v:initialReleaseDate').text
length_tag = soup.find('span', property='v:runtime')
if str(length_tag.next_sibling) != '<br/>':
length = length_tag.text + str(length_tag.next_sibling)
else:
length = length_tag.text
another_name = soup.find('span', text='又名:').next_sibling
introduction = soup.find('span', property='v:summary').text
grade = soup.find('strong', property='v:average').text
comment_times = soup.find('span', property='v:votes').text
item['no'] = no
item['movie_name'] = movie_name
item['director'] = director
item['writer'] = writer
item['actor'] = actor
item['type'] = type
item['region'] = region
item['language'] = language
item['date'] = date
item['length'] = length
item['another_name'] = another_name
item['introduction'] = introduction
item['grade'] = grade
item['comment_times'] = comment_times
except exception as e:
print('Parse error:', e)
return item
def parse_item(self, response):
yield DoubanMovieItem({
'url':
response.url,
'name':
response.xpath(
'//span[@property="v:itemreviewed"]/text()').extract(),
'summary':
response.xpath(
'//span[@property="v:summary"]/text()').extract_first(),
'score':
response.xpath(
'//strong[@property="v:average"]/text()').extract_first()
})
def parse(self, response):
datas = json.loads(response.text).get('data')
logger.info("request successful")
for data in datas:
item = DoubanMovieItem()
for field in item.fields:
if field in data:
item[field] = data[field]
id = item['movie_id'] = data['id']
if redis_db.add_movie_id(id):
# logger.info('add %s in redis' %data['title'])
yield item
else:
self.repeat_count += 1
def parse(self, response):
item = DoubanMovieItem()
divlist = response.xpath(r"//div[@class='pic']")
for div in divlist:
item['Url'] = div.xpath("a/@href").extract_first()
item['Img_url'] = div.xpath('a/img/@src').extract_first()
yield item
yield scrapy.Request(url=item['Url'],
meta={'item': item},
callback=self.parse_movie)
next_href = response.xpath('//link[@rel="next"]/@href').extract()
if next_href:
nexturl = 'http://movie.douban.com/top250' + next_href[0]
yield scrapy.Request(url=nexturl, headers=self.headers)
def parse(self, response):
sel = Selector(response)
movie_name = sel.xpath("//div[@class='pl2']/a/text()").extract()
movie_url = sel.xpath("//div[@class='pl2']/a/@href").extract()
movie_star = sel.xpath(
"//div[@class='pl2']/div/span[@class='rating_nums']/text()"
).extract()
item = DoubanMovieItem()
item['movie_name'] = [n for n in movie_name]
item['movie_url'] = [n for n in movie_url]
item['movie_star'] = [n for n in movie_star]
yield item
def parse(self, response):
datas = json.loads(response.body)
item = DoubanMovieItem()
if datas:
for data in datas:
item['ranking'] = data['rank']
item['movie_name'] = data['title']
item['score_num'] = data['vote_count']
yield item
# 如果datas存在数据则对下一页进行采集
page_num = re.search(r'start=(\d+)', response.url).group(1)
page_num = 'start=' + str(int(page_num) + 20)
next_url = re.sub(r'start=\d+', page_num, response.url)
yield Request(next_url, headers=self.headers)
def parse(self, response):
item = DoubanMovieItem()
movies = response.xpath('//ol[@class="grid_view"]/li')
for movie in movies:
item['rank'] = movie.xpath('.//div[@class="pic"]/em/text()').extract()[0]
item['score'] = movie.xpath('.//div[@class="star"]/span[@class="rating_num"]/text()').extract()[0]
item['name'] = movie.xpath('.//div[@class="hd"]/a/span[1]/text()').extract()[0]
yield item
next_url = response.xpath('//span[@class="next"]/a/@href').extract()
if next_url:
next_url = 'https://movie.douban.com/top250' + next_url[0]
yield Request(next_url, headers=self.headers)
def parse_item(self, response):
# hxs=HtmlXPathSelector(response)
# sel=Selector(response)
item = DoubanMovieItem()
# items.py定义的内容
# url=Field()
# ID=Field()
# name=Field()
# director=Field()
# writer=Field()
# role=Field()
# types=Field()
# summary=Field()
item['url'] = re.match(
string=''.join(response.url),
pattern='(https://movie.douban.com/subject/\d+)/.*').group(1)
item['movieid'] = item['url'].split('/')[-1]
item['ID'] = '/'.join(
response.xpath('//*/a[contains(@href,"subject")]/@href').re(
'movie.douban.com/subject/(\d+)/(?:\?from|$)'))
# item['ID']=''.join(response.xpath('//*[@id="content"]/div/div[1]/div[1]/div[3]/ul/li[5]/span/@id').extract())
item['name'] = ''.join(
response.xpath('//*[@id="content"]/h1/span[1]/text()').extract())
item['director'] = '/'.join(
response.xpath(
'//*[@id="info"]/span[1]/span[2]/a/text()').extract())
item['writer'] = '/'.join(
response.xpath(
'//*[@id="info"]/span[2]/span[2]/a/text()').extract())
item['role'] = '/'.join(
response.xpath(
'//*[@id="info"]/span[3]/span[2]/a/text()').extract())
item['types'] = '/'.join(
response.xpath('//span[@property="v:genre"]/text()').extract())
item['summary'] = ''.join(
response.xpath('//span[@property="v:summary"]/text()').extract())
item['summary'] = item['summary'].strip().\
replace('<br />', '').replace('\t', ' ').\
replace('\n', ' ').replace('&', '').replace('"','').replace(u'\u3000', '')
item['summary'] = re.sub(r' {1,}', ' ', item['summary'])
if self.count == self.MAX_MOVIE:
while True:
print 'You have got {0} movies, please quit!'.format(
self.MAX_MOVIE)
time.sleep(2)
self.count += 1
yield item
def parse(self, response):
print(response.text)
item = DoubanMovieItem()
selector = response.xpath('//div[@class="list-wp"]/a')
item['name'] = selector.xpath('p/span[@class="title"]/text()').getall()
item['score'] = selector.xpath('p/span[@class="rate"]/text()').getall()
item['pic'] = selector.xpath('div/span/img/@src').getall()
item['link'] = selector.xpath('@href').getall()
yield item
print(type(item['link']), len(item['link']))
for link in item['link']:
yield response.follow(link,
callback=self.next_parse,
meta={'depth': 1})
'''
def parse(self, response):
movie_list = response.xpath('//ol[@class="grid_view"]//li')
for movie in movie_list:
item = DoubanMovieItem()
item['movie_pic'] = movie.xpath(
'.//div[@class="item"]//div[@class="pic"]//img/@src'
).extract_first()
item['movie_title'] = movie.xpath(
'.//div[@class="item"]//div[@class="hd"]//a//span[@class="title"]/text()'
).extract_first()
span2 = movie.xpath(
'.//div[@class="item"]//div[@class="hd"]//a//span[2]/text()'
).extract_first()
span3 = movie.xpath(
'.//div[@class="item"]//div[@class="hd"]//a//span[3]/text()'
).extract_first()
item['movie_other'] = ((span2 if span2 else '') +
(span3 if span3 else '')).replace(
u'\xa0', u' ')
item['movie_introduce'] = movie.xpath(
'normalize-space(.//div[@class="item"]//div[@class="bd"]//p[1]/text())'
).extract_first().replace(u'\xa0', u' ')
item['movie_star'] = movie.xpath(
'.//div[@class="bd"]//div[@class="star"]/span[2]/text()'
).extract_first()
item['evaluate_num'] = movie.xpath(
'.//div[@class="bd"]//div[@class="star"]/span[4]/text()'
).extract_first()
item['movie_description'] = movie.xpath(
'.//div[@class="bd"]//p[@class="quote"]/span/text()'
).extract_first()
yield item
next_url = response.xpath(
'//span[@class="next"]/a/@href').extract_first()
if next_url:
print(self.start_urls[0] + next_url)
yield scrapy.Request(self.start_urls[0] + next_url,
callback=self.parse)
def parse(self, response):
item = DoubanMovieItem()
film_list = json.loads(response.body.decode())
if film_list == [] or self.offset > 1000:
return
for film in film_list['data']:
item['film_name'] = film['title']
item['film_directors'] = film['directors']
item['film_rate'] = film['rate']
item['film_actors'] = film['casts']
item['film_image_url'] = film['cover']
urllib.request.urlretrieve(
item['film_image_url'],
self.file_path + "/" + item['film_name'] + "." +
item['film_image_url'].split(".")[-1])
yield item
self.offset = self.offset + 20
new_url = 'https://movie.douban.com/j/new_search_subjects?sort=U&range=0,10&tags=1&start=' + str(
self.offset)
yield scrapy.Request(url=new_url, callback=self.parse)
def parse(self, response):
movie_blocks = response.xpath('//ol[@class="grid_view"]/li')
for block in movie_blocks:
name = block.css('span.title::text').extract_first()
# name = block.xpath()
star = block.xpath(
".//span[@class='rating_num']/text()").extract_first()
e = block.xpath(
".//div[@class='star']/span[4]/text()").extract_first()
evaluation = self.eval_re.search(e).group()
#group（）和groups（），前一个输出字符串，后一个输出元组
introduction = block.css('span.inq::text').extract_first()
item = DoubanMovieItem()
item['name'] = name
item['star'] = star
item['evaluation'] = evaluation
item['introduction'] = introduction
yield item #下一次迭代从yield之后开始
next_url = response.css('span.next > a::attr(href)').extract_first()
# 注意地址取法，之前都是去text，而这里取的是<>里的，指定attr和对应名称。
# <a href="?start=25&filter=" style="background: rgb(204, 136, 136); border: 2px solid red;">后页></a>
if next_url:
next_url = response.urljoin(next_url) # ruljoin创建完整的url
yield scrapy.Request(next_url, callback=self.parse)
def parse_item(self, response):
global failed_count
global real_parse_count
item = DoubanMovieItem()
try:
real_parse_count += 1
print("real parse count = %d" % (real_parse_count))
# get movie id
url = response.url
id = url.split('/')[-2].strip()
item["movie_id"] = id
# get movie name
name = response.xpath(
'//div[@id="content"]/h1/span[1]/text()').extract_first()
item["movie_name"] = name.strip() if name else ""
#get movie year
year = response.xpath(
'//div[@id="content"]/h1/span[2]/text()').extract_first()
item["movie_year"] = year.strip("（）() ") if year else ""
# get movie rate
rate = response.xpath(
"//div[@class='rating_self clearfix']/strong/text()"
).extract_first()
item["movie_rate"] = float(rate.strip() if rate else "-1")
# get movie rate people
rate_num = response.xpath(
"//span[@property='v:votes']/text()").extract_first()
item["movie_rate_people"] = int(
rate_num.strip() if rate_num else "-1")
# get hot short comments
comments = response.xpath(
"//div[@id='hot-comments']//div[@class='comment-item']//div[@class='comment']/p/text()"
).extract()
votes = response.xpath(
"//div[@id='hot-comments']//div[@class='comment-item']//div[@class='comment']//span[@class='votes pr5']/text()"
).extract()
rates = response.xpath(
"//div[@id='hot-comments']//div[@class='comment-item']//span[@class='comment-info']/span[1]/@title"
).extract()
if len(comments) == len(votes) and len(votes) == len(rates):
commentsarray = []
for i in range(len(votes)):
short_comments = {}
short_comments['comment'] = comments[i]
short_comments['votes'] = int(votes[i])
short_comments['rates'] = rates[i]
commentsarray.append(short_comments)
item["movie_hot_short_comments"] = commentsarray
seenwish = response.xpath(
"//div[@class='subject-others-interests-ft']//a//text()"
).extract()
if seenwish and len(seenwish) == 2:
item['movie_seen'] = int(seenwish[0][:-3])
item['movie_wishes'] = int(seenwish[1][:-3])
# get movie info
info = response.xpath("//div[@id='info']")
infoarray = info.extract()
infostr = ''.join(infoarray).strip()
director = info.xpath("span[1]/span[2]/a/text()").extract()
self.add_array("movie_director", director, item)
writor = info.xpath("span[2]/span[2]/a/text()").extract()
self.add_array("movie_writor", writor, item)
actors = info.xpath("span[3]/span[2]/a/text()").extract()
self.add_array("movie_actors", actors, item)
time = info.xpath(
"span[@property='v:runtime']/@content").extract_first()
item["movie_time"] = float(time.strip() if time else "-1")
types = info.xpath("span[@property='v:genre']/text()").extract()
self.add_array("movie_type", types, item)
try:
lang = re.search(language_pattern, infostr)
if lang:
language = lang.group(1).strip()
item["movie_language"] = language.strip()
except:
pass
try:
regionmatch = re.search(region_pattern, infostr)
if regionmatch:
region = regionmatch.group(1).strip()
item["movie_region"] = region.strip()
except:
pass
try:
dialectmatch = re.search(dialect_pattern, infostr)
if dialectmatch:
dialect = dialectmatch.group(1).strip()
item["movie_dialect"] = dialect.strip()
except:
pass
desc = response.xpath("//span[@property='v:summary']/node()"
).extract_first().strip()
item["movie_desc"] = desc.strip() if desc else ""
tags = response.xpath(
"//div[@class='tags-body']/a/text()").extract()
self.add_array("movie_tags", tags, item)
pic = response.xpath(
"//div[@id='mainpic']/a/img/@src").extract_first()
item["movie_pic_url"] = pic
yield item
except Exception, e:
# do nothing
logging.info("Parse error:%s" % (str(e)))
print("failed_count = %d" % (failed_count + 1))
failed_count += 1
pass
def parse_content(self, response):
movieid = self.movie[0]
tag = self.movie[1]
title = self.movie[2]
director = self.movie[3]
actor = self.movie[4]
rate = self.movie[5]
star = self.movie[6]
cover = self.movie[7]
html = BeautifulSoup(response.body, 'lxml')
info = html.select('#info')
if len(info) == 0:
print(response.text)
return [-2]
info = html.select('#info')[0].get_text().split('\n')
print(info)
# print(len(info))
category = ''
district = ''
showtime = ''
length = ''
for item in info:
item = item.split(':')
if item[0] == '类型':
category = item[-1].strip()
elif item[0] == '制片国家/地区':
district = item[-1].strip()
elif item[0] == '上映日期':
showtime = item[-1].strip().split('-')[0]
elif item[0] == '片长':
length = item[-1].strip()
length = re.findall('\d+', length)[0]
category = category.replace(r'/', ',')
if len(district) > 0:
district = district[:50]
if len(category) > 0:
category = category[:30]
rate_count = html.select(
'#interest_sectl > div.rating_wrap.clearbox > div.rating_self.clearfix > div > div.rating_sum > a > span'
)[0].get_text()
# interest_sectl > div.rating_wrap.clearbox > div.ratings-on-weight > div:nth-child(1) > span.rating_per
rate5 = html.select(
'#interest_sectl > div.rating_wrap.clearbox > div.ratings-on-weight > div:nth-of-type(1) > span.rating_per'
)[0].get_text().split('%')[0]
rate4 = html.select(
'#interest_sectl > div.rating_wrap.clearbox > div.ratings-on-weight > div:nth-of-type(2) > span.rating_per'
)[0].get_text().split('%')[0]
rate3 = html.select(
'#interest_sectl > div.rating_wrap.clearbox > div.ratings-on-weight > div:nth-of-type(3) > span.rating_per'
)[0].get_text().split('%')[0]
rate2 = html.select(
'#interest_sectl > div.rating_wrap.clearbox > div.ratings-on-weight > div:nth-of-type(4) > span.rating_per'
)[0].get_text().split('%')[0]
rate1 = html.select(
'#interest_sectl > div.rating_wrap.clearbox > div.ratings-on-weight > div:nth-of-type(5) > span.rating_per'
)[0].get_text().split('%')[0]
item = DoubanMovieItem()
item['movieid'] = movieid
item['title'] = title
item['tag'] = tag
item['directors'] = director
item['actors'] = actor
item['showtime'] = showtime
item['length'] = length
item['district'] = district
item['category'] = category
item['star'] = star
item['rate'] = rate
item['rate_count'] = rate_count
item['rate5'] = rate5
item['rate4'] = rate4
item['rate3'] = rate3
item['rate2'] = rate2
item['rate1'] = rate1
item['cover'] = cover
print('###### ')
print(item)
print('######')
