def parse_content(self, response):
try:
item = NewsspiderItem()
item_type = 'society'
item_url = response.url
item_crawl_time = GetCrawlTime.CrawlTime
item_title = ' '.join(response.xpath('//h2[@id="article_title"]/text()').extract()).strip().replace(u'\xa0', u' ')
item_time = ''.join(response.xpath('//span[@id="pubtime_baidu"]/text()').extract()).strip().replace(u'\xa0', u' ')
item_content = '\n'.join(response.xpath('//div[@id="content"]//p//text()').extract()).strip().replace(u'\xa0', u' ')
item_img = response.xpath('//div[@id="content"]//img/@src').extract()
item_author = ''.join(response.xpath('//div[@class="m-editor"]/text()').extract()).strip().replace(u'\xa0', u' ')
item_tags = ''
item_source = u'南方网'
item_summary = item_content[:100]
item_img = str(CommonParse.array_uft8_parse(item_img))
item_time = item_time[:16]
item_time = TimeParse.time_parse(item_time)
item = item.getItem(url=item_url, title=item_title, tags=item_tags, source=item_source,
publish_date=item_time, content=item_content, author=item_author,
type=item_type, img=item_img, summary=item_summary,
crawl_time=item_crawl_time)
item.parseStr(item)
yield item
except Exception as e:
print '=' * 100
print e, response.url
with open('%s.txt' % self.name, 'a') as f:
f.write('%s - %s\n' % (e, response.url))
def parse_class(self, response):
soup = BeautifulSoup(response.body, "lxml")
meta = response.request.meta
category = meta['category']
current_url = response.url
content = soup.find("div", class_="container cc-body-content")
if content is not None:
item = NewsspiderItem()
item['category'] = category
item['url'] = current_url
item['page'] = content.decode()
yield item
else:
logging.info("failed to get content: " + current_url)
all_reviews = soup.find("div", class_="course-all-reviews")
if all_reviews:
for review_title in all_reviews.find_all(
"div", class_="review-title title-with-image"):
a = review_title.find("a")
if a is not None:
link = self.index_url + a['href']
if link not in self.user_dict.keys():
logging.info("add new user: " + link)
self.user_dict[link] = 'user'
def second_parse(self, response):
# 获取meta参数里面键为'meta_1'的值
meta_1 = response.meta['meta_1']
items = []
# 遍历小类里面的子链接
for each in response.xpath('//a/@href'):
# 获取的子链接，以大类链接开头，以.shtml结尾
if each.extract().encode('utf-8').startswith(
meta_1['parentUrl'].encode('utf-8')) and each.extract(
).encode('utf-8').endswith('.shtml'.encode('utf-8')):
item = NewsspiderItem()
item['parentUrl'] = meta_1['parentUrl']
item['parentTitle'] = meta_1['parentTitle']
item['subUrl'] = meta_1['subUrl']
item['subTitle'] = meta_1['subTitle']
item['subpath'] = meta_1['subpath']
item['sonUrl'] = each.extract()
items.append(item)
# 发送子链接请求
for each in items:
yield scrapy.Request(each['sonUrl'],
meta={'meta_2': each},
callback=self.detail_parse)
def parse(self, response):
# filename = response.url.split("/")[-2]
filename = "huxiu"
with open(filename, 'wf') as f:
for newsCell in response.css("div.mod-b"):
item = NewsspiderItem()
item["title"] = (newsCell.css("h3 a").extract())[0].encode("utf-8")
f.write(item["title"])
yield item
def parse_item(self, response):
i = NewsspiderItem()
i['title'] = response.xpath(
"/html/head/meta[@property='og:title']/@content").extract()
i['keywords'] = response.xpath(
"/html/head/meta[@name='keywords']/@content").extract()
content = response.xpath("//div[@id='artibody']").extract()
i['content'] = str(content[0])
i['link'] = response.url
i['source'] = 'sina.com.cn'
i['author'] = response.xpath(
"//p[@class='article-editor']/text()").extract()
i['publish_time'] = response.xpath(
"//meta[@property='article:published_time']/@content").extract()
return i
def parse_user(self, response):
soup = BeautifulSoup(response.body, "lxml")
meta = response.request.meta
category = 'user' #meta['category']
current_url = response.url
content = soup.find("div", class_="container cc-body-content")
if content is not None:
item = NewsspiderItem()
item['category'] = category
item['url'] = current_url
item['page'] = content.decode()
yield item
else:
logging.info("failed to get content: " + current_url)
def parse_detail(self, response):
news_content = ''
news_img = ''
for p in response.xpath('//*[@id="newscontent"]/div/div[1]/div/div/p'):
content = p.xpath('.//text()').extract()
img = p.xpath('./img/@src').extract()
# 要先判断是否为空，不然可能会出现indexError
if content:
# print("正文："+temp[0])
news_content += content[0] + "\r\n"
if img:
# print("图片："+img[0])
news_img += "http://news.xmu.edu.cn" + img[0] + ";"
item = NewsspiderItem()
item["news_title"] = response.xpath(
'//*[@id="mainContainer"]/div[3]/table/tr/td/span/span/span/text()'
).extract()[0]
item["news_content"] = news_content
item["news_source"] = '厦门大学新闻网'
item["news_link"] = response.url
item["news_release_time"] = response.xpath(
'//*[@id="mainContainer"]/div[4]/table/tr[2]/td/span[1]/span/span/text()'
).extract()[0]
item["news_read_status"] = '1'
item["news_get_time"] = time.strftime('%Y-%m-%d',
time.localtime(time.time()))
item["news_imgs"] = news_img
release_time = item['news_release_time']
print("the latest_release_time from web_xmu_news is {0}".format(
self.latest_release_time))
st1 = time.mktime(time.strptime(release_time, "%Y-%m-%d"))
st2 = time.mktime(time.strptime(self.latest_release_time, "%Y-%m-%d"))
result = int(st1) - int(st2)
print("发布时间：{0} 是否继续：{1}".format(release_time, result))
# 发布时间小于数据库最新时间则停止爬取，且数据若已存在则不保存到数据库
if result <= 0:
self.crawler.engine.close_spider(self, '厦门大学新闻网消息爬取完成!')
elif myMysql.myMysql().columnExist(tableName="web_news",
columnValue=item['news_link']):
return
else:
yield item
def second_parse(self, response):
meta_1 = response.meta['meta_1']
items = []
# 循环遍历获取文章url
for each in response.xpath('//a/@href'):
if each.extract().encode('utf-8').startswith(
meta_1['first_url'].encode('utf-8')) and each.extract(
).encode('utf-8').endswith('.shtml'.encode('utf-8')):
item = NewsspiderItem()
item['first_url'] = meta_1['first_url']
item['second_url'] = meta_1['second_url']
item['article_url'] = each.extract()
items.append(item)
# 获取文章请求
for each in items:
yield scrapy.Request(each['article_url'],
meta={'meta_2': each},
callback=self.detail_parse)
def parse(self, response):
# 以一级目录的“地方站”作为根来循环遍历所有的一级目录url和title
for each in response.xpath(
"//div[@id='tab01']/div[@data-sudaclick!='citynav']"):
# 获取一级目录的url
first_url = each.xpath('./h3/a/@href').extract()[0]
# 循环遍历二级目录url
for other in each.xpath("./ul/li/a"):
# 获取二级目录的url
if other.xpath('./@href').extract()[0].startswith(first_url):
item = NewsspiderItem()
second_url = other.xpath('./@href').extract()[0]
item['first_url'] = first_url
item['second_url'] = second_url
# 获取二级目录请求
yield scrapy.Request(url=item['second_url'],
meta={'meta_1': item},
callback=self.second_parse)
def parse_detail(self, response):
print(response.status)
print(response.xpath('//h1[@class="main-title"]/text()').extract()
[0]) #标题
news = NewsspiderItem()
news["url"] = response.url
news["title"] = response.xpath(
'//h1[@class="main-title"]/text()').extract()[0]
news["time"] = response.xpath(
'//*[@id="top_bar"]/div/div[2]/span/text()').extract()[0]
news["origin"] = response.xpath(
'//*[@id="top_bar"]/div/div[2]/a/text()').extract()
news["origin_url"] = response.xpath(
'//*[@id="top_bar"]/div/div[2]/a/@href').extract()[0]
news["detail"] = "\n".join(response.xpath('//div[@class="article"]/div/p/text()').extract())+\
"\n".join(response.xpath('//div[@class="article"]/p/text()').extract())+\
"\n".join(response.xpath('//div[@class="article"]/div/div/text()'))
yield news
def parse(self, response):
# 通过某节点作为根节点进行大类链接遍历
for each in response.xpath(
"//div[@id='tab01']/div[@data-sudaclick!='citynav']"):
# 获取大类链接和大类标题
#encode('utf-8') string编码为bytes
parentUrl = each.xpath('./h3/a/@href').extract()[0]
#parentTitle = each.xpath('./h3/a/text()').extract()[0].encode('utf-8')
parentTitle = each.xpath('./h3/a/text()').extract()[0]
# 设置大类存储路径
parentpath = './data/' + parentTitle
#parentpath = parentTitle
if not os.path.exists(parentpath):
os.makedirs(parentpath)
# 遍历小类链接
for other in each.xpath("./ul/li/a"):
# 获取以大类链接开头的小类链接
if other.xpath('./@href').extract()[0].startswith(parentUrl):
# 注意item的位置，不同的位置会导致不同的结果。尽量不要把item的数据在外循环和内循环里面分别获取，如必须这样做，则创建空列表添加item来解决。
item = NewsspiderItem()
subUrl = other.xpath('./@href').extract()[0]
subTitle = other.xpath('./text()').extract()[0]
subpath = parentpath + '/' + subTitle
item['parentUrl'] = parentUrl
item['parentTitle'] = parentTitle
item['subUrl'] = subUrl
item['subTitle'] = subTitle
item['subpath'] = subpath
if not os.path.exists(subpath):
os.makedirs(subpath)
# 发送小类链接请求，使用meta参数把item数据传递到回调函数里面，通过response.meta['']得到数据
yield scrapy.Request(url=item['subUrl'],
meta={'meta_1': item},
callback=self.second_parse)
def archive(self, response):
title_tag = response.css('.noBorder::text').extract_first().strip()
art_author = response.css('.art_author::text').extract_first().strip()
art_authors = art_author.encode('utf-8').split("：")
if len(art_authors) == 2:
art_author = art_authors[1]
else:
art_author = u'佚名'
art_publish = response.css(
'.art_publish::text').extract_first().strip()
art_publishs = art_publish.encode('utf-8').split("：")
if len(art_publishs) == 2:
art_publish = art_publishs[1]
else:
art_publish = '2001-01-01'
art_con = response.css('.atr_con').extract_first()
item = NewsspiderItem()
item['title'] = title_tag
item['author'] = art_author
item['date'] = art_publish
item['content'] = art_con
item['url'] = response.url
# print item['title']
yield item
