def parse(self, response):
PrintLog.print_start_flag(self.parse.__name__)
sel = Selector(response)
# pdb.set_trace()
# print response.url
# print response.body
# 取出friendsList
'''
下面的正则表达式要查找和取出字符串‘user_list : [...],’中间的...内容
(?<= # 断言要匹配的文本的前缀开始
user_list : \[ # 查找字符串'user_list : ['
) # 前缀结束
[\s\S]* # 匹配任意文本
(?= # 断言要匹配的文本的后缀开始
\], # 查找字符串'[,'
) # 后缀结束
'''
friends = sel.re(r'(?<=user_list : \[)[\s\S]*(?=\],)')
yield self.parse_friends_list(friends_list=friends)
# 尝试取下一页数据
#pdb.set_trace()
PrintLog.print_log("get next page")
page_count_str_list = sel.re(r'pageCount :\s*(.*)')
if page_count_str_list:
m = re.findall(r"\d", page_count_str_list[0])
self.total_page_count = int(m[0])
# print "page_count_num=", self.total_page_count
self.page_num += 1 # 下一页码
if self.page_num < self.total_page_count:
yield self.request_page(page_idx=self.page_num)
def parse(self, response):
PrintLog.print_start_flag(self.parse.__name__)
sel = Selector(response)
# pdb.set_trace()
# print response.url
# print response.body
# 取出friendsList
'''
下面的正则表达式要查找和取出字符串‘user_list : [...],’中间的...内容
(?<= # 断言要匹配的文本的前缀开始
user_list : \[ # 查找字符串'user_list : ['
) # 前缀结束
[\s\S]* # 匹配任意文本
(?= # 断言要匹配的文本的后缀开始
\], # 查找字符串'[,'
) # 后缀结束
'''
friends = sel.re(r'(?<=user_list : \[)[\s\S]*(?=\],)')
yield self.parse_friends_list(friends_list=friends)
# 尝试取下一页数据
#pdb.set_trace()
PrintLog.print_log("get next page")
page_count_str_list = sel.re(r'pageCount :\s*(.*)')
if page_count_str_list:
m = re.findall(r"\d", page_count_str_list[0])
self.total_page_count = int(m[0])
# print "page_count_num=", self.total_page_count
self.page_num += 1 # 下一页码
if self.page_num < self.total_page_count:
yield self.request_page(page_idx=self.page_num)
def parse(self, response):
sel = Selector(response)
item = MarketScrapyItem()
#item['name'] = 'test'
item['name'] = sel.re('id=\"mdname\"(.*?)</span')
item['bus'] = sel.re('id=\"inforBody\".*')
return item
def parse(self, response):
sel = Selector(response)
item = MarketScrapyItem()
#item['name'] = 'test'
item['name'] = sel.re('id=\"mdname\"(.*?)</span')
item['bus'] = sel.re('id=\"inforBody\".*')
return item
def parse_article(self, response):
item = WechartAccount()
sel = Selector(response)
user_name = sel.re('<span class="profile_meta_value">([^<]*)</span>')
nickname = sel.re('var nickname = "([^"]*)";')
image_url = sel.re('hd_head_img : "([^"]*)"')
item['nickname'] = ''.join(nickname)
item['user_name'] = user_name[0]
item['image_url'] = ''.join(image_url)
return item
def parse_item(self, response):
item = TupianItem()
x = Selector(response)
imgs = x.re('src="(http.*?\.jpg)".*?alt')
imgname = x.re('src.*?alt="(.*?)"')
for i in range(len(imgs)):
item['tupianming'] = imgname[i]
item['images_urls'] = [imgs[i]]
item['leibie'] = x.xpath('//span/h1/text()').extract()
yield item
def parse_item(self, response):
item = TianqiItem()
x = Selector(response)
rq_list = x.re('(\d+年\d+月\d+日)')
tqzk_list = x.re('([\u4e00-\u9fa5]+\s+/[\u4e00-\u9fa5]+)')[::2]
qw_list = x.re('\d+℃\s+/\s+-?\d+℃')
for i in range(len(tqzk_list)):
item['rq'] = rq_list[i]
item['tqzk'] = re.sub('\s+', '', tqzk_list[i])
item['qw'] = re.sub('\s+', '', qw_list[i])
yield item
def parse_product(self, response):
self.response = response
selector = Selector(response=response)
ids = selector.re('"skuId":"(\w+)"')
savings = selector.re('"SAVE":"(\d+)%"')
options = selector.re('"Option":"(.+?)"')
prices = selector.re(r'"RP":".(\d+\.\d{2})"')
name = selector.re('productDisplayName="(.+?)"')
items = self.load(prices, savings, ids, name[0], options)
return items
def parse_news(self, response):
sel = Selector(response)
pattern = re.match(self.url_pattern, str(response.url))
item = SinaItem()
item['source'] = 'sina' # pattern.group(1)
item['date'] = ListCombiner(str(pattern.group(2)).split('-'))
item['newsId'] = sel.re(r'comment_id:(\d-\d-\d+)')[0]
item['cmtId'] = item['newsId']
item['channelId'] = sel.re(r'comment_channel:(\w+);')[0]
item['comments'] = {'link':str('http://comment5.news.sina.com.cn/comment/skin/default.html?channel='+item['channelId']+'&newsid='+item['cmtId'])}
item['contents'] = {'link':str(response.url), 'title':u'', 'passage':u''}
item['contents']['title'] = sel.xpath("//h1[@id='artibodyTitle']/text()").extract()[0]
item['contents']['passage'] = ListCombiner(sel.xpath('//p/text()').extract())
return item
def parse_news(self, response):
sel = Selector(response)
pattern = re.match(self.url_pattern, str(response.url))
datenow = GetDate()
item = SinaItem()
item['source'] = 'sina' # pattern.group(1)
item['date'] = datenow
item['newsId'] = sel.re(r'comment_id:(\d+-\d-\d+)')[0]
item['cmtId'] = item['newsId']
item['channelId'] = sel.re(r'comment_channel:(\w+);')[0]
item['comments'] = {'link':str('http://comment5.news.sina.com.cn/page/info?format=json&channel='+item['channelId']+'&newsid='+item['cmtId']+'&group=0&compress=1&ie=gbk&oe=gbk&page=1&page_size=100&jsvar=requestId_24')}
item['contents'] = {'link':str(response.url), 'title':u'', 'passage':u''}
item['contents']['title'] = sel.xpath("//h1[@id='artibodyTitle']/text()").extract()[0]
item['contents']['passage'] = ListCombiner(sel.xpath('//p/text()').extract())
yield item
def parse_item(response):
sel = Selector(response)
url = response.request.url
if re.match(r'.*?people.com.cn.*?/\d+/\d+/.*?',
url) and 'BIG' not in url:
content = response.xpath(
'//*[@id="rwb_zw"]/p/text() | //*[@id="rwb_zw"]/p/strong/text()'
).extract()
if content:
item = NewsItem(
domainname='http://people.com.cn',
chinesename='人民网',
url=sel.root.base,
title=sel.css('div.text_title > h1::text').extract_first(),
subtitle=sel.css('.sub::text').extract_first(),
language='中文',
encodingtype='utf-8',
corpustype='网络',
timeofpublish=sel.re(r'\d{4}年\d{2}月\d{2}日\d{2}:\d{2}')[0],
content=''.join(content),
source=sel.css(
'div.box01 > div.fl > a::text').extract_first(),
author=sel.css('p.author::text').extract_first())
item = judge_time_news(item)
if item:
yield item
def parse(self, response):
selector = Selector(response)
PageUrl = None
#抓到下一页URL
try:
PageUrl = selector.re(
"href=\"(http:.*\d)\" class=\"ui_page_item ui_page_next\"")[0]
except IndexError:
self.log('One done!' + response.url)
#抓取本页的15个行程概览
divitems = selector.xpath("//div[@class='items']").extract()
for item in divitems:
result = QYURLItem()
journey_name_pattern = re.compile(u'<dd>(.*)</dd>')
url_pattern = re.compile('<a href=\"(//.*)\" class.*>')
day_pattern = re.compile('<strong>(\d*)</strong>')
line_pattern = re.compile('<p>(.*)</p>')
date_pattern = re.compile(r'<dt>(20.*) 出发</dt>')
lable_pattern = re.compile('<strong>(\W*)</strong>')
result['journey_name'] = journey_name_pattern.findall(item)
result['url'] = url_pattern.findall(item)
result['day'] = day_pattern.findall(item)
result['line'] = line_pattern.findall(item)
result['date'] = date_pattern.findall(item)
result['lable'] = lable_pattern.findall(item)
yield result
# self.log(PageUrl, level=log.DEBUG)\
'''
def parse_item(self, response):
""" Main parse function
"""
sel = Selector(response)
item = ProductItem()
item['source'] = 'tmall'
item['name'] = self.get_product_name( sel )
item['img'] = sel.xpath("//ul[@id='J_UlThumb']/li")[0].xpath(".//a/img/@src").extract()[0]
item['category'] = self.get_category(response)
try:
# 获取TShop字符串，并对TShop字符串进行JSON标准化处理
TShop_str = sel.re('TShop\.Setup\(((.|\n)+?)\);')[0]
# 移除注释，目前只有天猫超市有注释，以逗号开头
regex = re.compile(',\s*\/\/[^\n]*')
TShop_str = re.sub(regex, ',', TShop_str)
TShop = eval( TShop_str, type('Dummy', (dict,), dict(__getitem__=lambda s,n:n))() )
except SyntaxError:
return
item['itemId'] = TShop.get('itemDO').get('itemId', '')
item['url'] = 'http://detail.tmall.com/item.htm?id=' + item['itemId']
item['date'] = date.today().strftime('%Y-%m-%d')
item['attr'], item['brand'] = self.get_attr_and_brand( sel )
skuMap = self.get_sku_chinese_map( sel, TShop )
initApi_url = TShop.get('initApi')
yield Request( initApi_url,
headers={'Referer': 'http://www.google.com.hk/'},
meta={'item': item, 'skuMap': skuMap},
callback=self.parse_initapi )
def search(q):
'''根据书名检索馆藏信息
'''
fdata = {
'tag': 'search',
'subtag': 'simsearch',
'gcbook': 'yes',
'viewtype': '',
'flword': '',
'viewtype': '',
'q': q
}
resp = requests.post(search_url, data=fdata)
#得到记录条数
s_res = Selector(text=resp.content.decode('utf-8')).xpath('//p[@id="page"]/span/text()')
#如没有检索到记录,result_list为空
result_list = s_res.extract()
if len(result_list) == 0:
return "没有检索到记录"
result_str = result_list[0]
num = int(s_res.re('[\d]+')[0])
if num > 3:
note = ""
if num > 10:
note = "\n注:只显示前10条结果,得到所有检索结果:" + search_url + "\n======"
return result_str + "\n======" + note + getManyLinks(resp, num)
else:
return result_str + "\n======" + getdetail(resp, num)
def parse_product(self, response):
sel = Selector(response)
price = sel.re(re.compile('jsProductPrice = \'(.*)\';'))
categories = sel.xpath('//div[@id="navBreadCrumb"]/a/text()')[1:].extract()
brand = sel.xpath('//span[@class="product_manufacturer"]/text()').re('Manufactured by: (.*)')
brand = brand[0].strip() if brand else ''
sku = sel.xpath('//span[@class="product_model"]/text()').re('Ref: (.*)')
sku = sku[0].strip() if sku else ''
identifier = re.search('p-(.*)\.html', response.url).group(1)
image_url = response.xpath('//div[@id="replace_image_zoom"]//img[@class="zoom_pic"]/@src').extract()
if image_url:
image_url = response.urljoin(image_url[0])
name = sel.xpath('//h1[@class="productGeneral"]/text()').extract()
loader = ProductLoader(item=Product(), response=response)
loader.add_value('identifier', identifier)
loader.add_value('sku', sku)
loader.add_value('name', name)
loader.add_value('price', price)
price = loader.get_output_value('price')
if price and Decimal(price) < Decimal('400.0'):
loader.add_value('shipping_cost', Decimal('35.00'))
loader.add_value('url', response.url)
if image_url:
loader.add_value('image_url', image_url)
for category in categories:
loader.add_value('category', category)
loader.add_value('brand', brand)
yield loader.load_item()
def myre(self, response, key_list):
date = Selector(response)
for key in key_list:
#llprint key
if date.re(key):
return 1
return 0
class ChainReactionReviews(ChainReaction):
name = 'chain-reaction-reviews'
response = None
selector = None
item = None
loader = None
def _register(self, response):
self.response = response
self.selector = Selector(response=response)
self.item = response.meta['item'] if 'item' in response.meta.keys() else Review()
self.loader = ChainReactionReviewLoader(self.item, response=self.response)
def parse_product(self, response):
self._register(response)
self.loader.add_value('slug', self.selector.re('productDisplayName="(.+?)"'))
self.loader.add_value('name', self.selector.re('productDisplayName="(.+?)"'))
self.loader.add_value('retailer', RETAILER)
self.loader.add_value('manufacturer', MANUFACTURER)
request = Request(response.url + '/reviews.djs?format=embeddedhtml', callback=self.parse_reviews)
request.meta['item'] = self.loader.load_item()
return request
def parse_reviews(self, response):
self._register(response)
self.loader.add_value('review', 'review')
self.loader.add_value('author', 'author')
self.loader.add_value('date', 'date')
return self.loader.load_item()
def parse_item(response):
sel = Selector(response)
url = response.request.url
if re.match(r'.*?sina.com.*?/\d{4}-\d{2}-\d{2}/.*?', url):
content = response.xpath(
'//*[@id="artibody"]//p//text()').extract()
# 移除编辑
editor = response.xpath(
'//*[@class="article-editor"]/text()').extract_first()
if editor:
content.remove(editor)
publish_time = sel.re(r'\d{4}年\d{2}月\d{2}日.{0,1}\d{2}:\d{2}')[0]
if ' ' in publish_time:
publish_time = publish_time.replace(' ', '')
if content:
item = NewsItem(
domainname='http://sina.com.cn',
chinesename='新浪网',
url=sel.root.base,
title=sel.css('#artibodyTitle::text, #main_title::text'
).extract_first(),
subtitle=sel.css('.sub::text').extract_first(),
language='中文',
encodingtype='utf-8',
corpustype='网络',
timeofpublish=publish_time,
content=''.join(content),
source=sel.xpath(
'//*[@data-sudaclick="media_name"]/text() | //*[@data-sudaclick="media_name"]/a/text()'
).extract_first(),
author=None)
item = judge_time_news(item)
if item:
yield item
def myre(self,response,key_list): date=Selector(response) for key in key_list: #llprint key if date.re(key): return 1 return 0
def parse_item(response):
sel = Selector(response)
print(sel)
url = response.request.url
if re.match(r'.*?tibet.people.com.cn/.*?', url):
print('---------------------')
print(url)
content = response.xpath('//html/body/div[2]/div[4]/div[1]/div[2]/div[2]/text()').extract()
# '//*[@id="rwb_zw"]/p/text() | //*[@id="rwb_zw"]/p/strong/text()'| //*[@id="content"]/p[2]/span/span/text()
print(content)
if content:
item = NewsItem(
domainname='http://tibet.people.com.cn',
chinesename='people',
url=sel.root.base,
title=sel.css('.gq_content > h1:nth-child(2)::text').extract_first(),
subtitle=sel.css('.sub::text').extract_first(),
language='tibet',
encodingtype='utf-8',
corpustype='网络',
timeofpublish=sel.re(r'\d{4}.*?\d{2}.*?\d{2}.*?\d{2}:\d{2}')[0].replace('ལོའི་ཟླ་ ', '年').replace('ཚེས་', '月').replace('ཉིན། ', '日'),
# timeofpublish = re.search(r'\d{4}.*？\d{2}.*?\d{2}',sel.css('.title_hd > p:nth-child(2)::text').extract_first()).group(0),
content=''.join(content),
# source=sel.css('.title_hd > p:nth-child(2)::text').extract_first(),
# author=sel.css('.title_hd > p:nth-child(2)::text').extract_first()
)
print(item.get("title", None))
print(item.get("timeofpublish", None))
print(item.get("source", None))
print(item.get("author", None))
item = judge_time_news_people(item)
if item:
yield item
def parse_item(self, response):
items = DoubanItem()
x = Selector(response)
lb = x.xpath('//div[@class="hd"]/h1/text()').extract()[0]
zz_yz = x.re('作者</span.*?</a></span></span></p>')
x_k = x.xpath('//div[@class="info"]')
sm = x_k.xpath('./div[@class="title"]/a/text()').extract()
jj = x_k.xpath('./div[@class="article-desc-brief"]/text()').extract()
for i in range(len(sm)):
items['lb'] = lb
items['sm'] = sm[i]
zz_k = re.findall('作者</span.*?</a></span></span>', zz_yz[i])
items['zz'] = re.findall('([〕〔\u4e00-\u9fa5·\s]{2,})', zz_k[0])[1:]
yz_k = re.findall('译者</span.*?</a></span></span>', zz_yz[i])
if not yz_k:
items['yz'] = None
else:
items['yz'] = re.findall('([〕〔\u4e00-\u9fa5·]+)', yz_k[0])[1:]
pf = x_k[i].xpath(
'./div/span[@class="rating-average"]/text()').extract()
if not pf:
items['pf'] = None
else:
items['pf'] = pf[0]
items['jj'] = jj[i]
yield items
#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()
return i
def parse_post(self, response):
sel = Selector(response)
post_has_been_removed = sel.xpath(self.post_parse_rules['invalid'])
if len(post_has_been_removed) > 0:
return None
title = sel.xpath(self.post_parse_rules['title'])
# postinfo = sel.xpath("//div[@class='postinginfos']")
date_posted = sel.xpath(self.post_parse_rules['date_posted'])
date_updated = sel.xpath(self.post_parse_rules['date_updated'])
userbody = sel.xpath(self.post_parse_rules['body'])
text = sel.xpath(self.post_parse_rules['text'])
pictures = sel.re(self.post_parse_rules['picture'])
match = re.search("(\d+)\.\w+$", response.url)
filename, pid = (match.group(0), match.group(1))
item = PostItem()
item['pid'] = pid
item['filename'] = filename
item['region'] = self.region
item['domain'] = self.domain
item['url'] = response.url
item['body'] = response.body
item['pictures'] = list(set(pictures)) # Eliminate duplicates
item['title'] = title.extract()[0]
item['date_posted'] = date_posted.extract()[0]
if len(date_updated) > 0:
item['date_updated'] = date_updated.extract()[0]
item['userbody'] = userbody.extract()[0]
item['text'] = text.extract()[0]
return item
def down_file(self,response):
'''
下载软件
该请求返回一段js，拿出中的跳转路径，用curl进行下载
:param response:
:return:
'''
try:
item = response.meta['item']
sel = Selector(response);
#匹配出软件下载的真正链接
url = sel.re(r"window.location.href\s+=\s+\'([^\']+)")
if url:
#软件下载url是编过码的，替换里面的乱码，分割出软件名称
otherurl = url[0].replace('%2','/')
#软件文件的父路径
parent_file = os.path.sep+settings.PARENT_FILE_NAME
if not os.path.exists(parent_file):
os.mkdir(parent_file)
#软件路径
item['file_name'] = otherurl.split('/')[-1]
filename = parent_file + os.path.sep + otherurl.split('/')[-1]
#生成curl命令，-i代表断点续传，-o代表存储文件
commond = 'curl -i -o ' + filename + ' ' + url[0]
recode = subprocess.call(commond,shell=True)
print 'successful!'
self.col.update({'vendor': self.vendor}, {'$set': {'state': 'crawling'}})
yield item
except Exception as e:
print e.message
self.col.update({'vendor': self.vendor}, {'$set': {'state': 'error'}})
def parseTmall(self, response):
""" Tmall parser
"""
sel = Selector(response)
item = ProductItem()
item['surl'] = response.url
item['source'] = 'tmall'
item['name'] = self.get_product_name(sel)
try:
# 获取TShop字符串，并对TShop字符串进行JSON标准化处理
TShop_str = sel.re('TShop\.Setup\(((.|\n)+?)\);')[0]
# 移除注释，目前只有天猫超市有注释，以逗号开头
regex = re.compile(',\s*\/\/[^\n]*')
TShop_str = re.sub(regex, ',', TShop_str)
TShop = eval(
TShop_str,
type('Dummy', (dict, ), dict(__getitem__=lambda s, n: n))())
except SyntaxError:
return
item['itemId'] = TShop.get('itemDO').get('itemId', '')
item['url'] = 'http://detail.tmall.com/item.htm?id=' + item['itemId']
initApi_url = TShop.get('initApi')
yield Request(initApi_url,
headers={'Referer': 'http://www.google.com.hk/'},
meta={'item': item},
callback=self.parse_initapi)
def parse(self,response):
"""
第一次解析时，由于是论坛主页,XPath解析规则不同,之后根据从主页上获取子
论坛URL后，对每个子论坛的URL调用Parse
"""
sel=Selector(response)
items=[]
if self.IndexFlag:
self.IndexFlag=False
##由于使用BBS论坛页面原因，使用Xpath方式获取不是很好的选择
urls=sel.re(r"(http://bbs2.99nets.me/forum.*html)")[1:-2]
for url in urls:
msg("------URL: %s -----" % url,level="DEBUG")
yield Request(url,callback=self.parse)
msg("-----Crawl END----",level="INFO")
else:
sites=sel.xpath("//form/table/tbody/tr/th/a[@class='s xst']")
for site in sites:
item=BbssecondItem()
item['title']=site.xpath("text()").extract()
item['link'] =site.xpath("@href").extract()
yield item
url,text=response.url,sel.xpath("//title/text()").extract()
msg("---CURRENT URL:%s \n ---TEXT:%s" % (url,str(text).encode("UTF8")),
level='INFO')
##子论坛翻页
rule="//div[@class='pg']/a[@class='nxt']/"
if sel.xpath(rule+"text()").extract():
yield Request(sel.xpath(rule+"@href").extract()[0]
,callback=self.parse)
def parse_item(response):
sel = Selector(response)
url = response.request.url
if re.match(r'.*?sohu.com.*?/\d{4}\d{2}\d{2}/.*?', url):
content = response.xpath(
'//*[@itemprop="articleBody"]//p//text()').extract()
# 有的段落并不是在p标签下，所以
if len(content) < 3:
content = response.xpath(
'//*[@itemprop="articleBody"]//p//text() | //*[@id="contentText"]//div/text()'
).extract()
publish_time = sel.re(
r'\d{4}-\d{2}-\d{2} {0,1}\d{2}:\d{2}:\d{2}')[0]
if content:
item = NewsItem(
domainname='http://sohu.com',
chinesename='搜狐网',
url=sel.root.base,
title=sel.xpath(
'//*[@itemprop="headline"]/text()').extract_first(),
subtitle=sel.css('.sub::text').extract_first(),
language='中文',
encodingtype='utf-8',
corpustype='网络',
timeofpublish=publish_time,
content=''.join(content),
source=sel.xpath(
'//*[@id="media_span"]/span/text()').extract_first(),
author=sel.xpath(
'//*[@id="author_baidu"]/text()').extract_first())
item = judge_time_news(item)
if item:
yield item
def parse_page(self, response):
print response.url
sel = Selector(response)
email = sel.re('(\w+@[a-zA-Z_]+?\.[a-zA-Z]{2,6})')[0]
print email
if email not in email_in_file and email not in added_email:
file.write(email+'\n')
added_email.append(email)
def parse_page(self, response):
print response.url
sel = Selector(response)
email = sel.re('(\w+@[a-zA-Z_]+?\.[a-zA-Z]{2,6})')[0]
print email
if email not in email_in_file and email not in added_email:
file.write(email+'\n')
added_email.append(email)
def parse_item(self,response):
sel= Selector(response)
item = CommitsspiderItem()
item['cve'] = sel.re(self.patt)
if item['cve'] != []:
item['url'] = response.url
yield item
return
def parse_page(self, response):
sel = Selector(response)
emails = sel.re('(\w+@[a-zA-Z_]+?\.[a-zA-Z]{2,6})')
emails = list(filter(lambda x: x != '*****@*****.**', emails))
if bool(emails):
for email in emails:
if email not in email_in_file and email not in current_session_emails:
file.write(email+'\n')
current_session_emails.append(email)
def parse(self, response):
sel = Selector(response)
link = sel.re('((http).*?(\.mp4))')[0]
name = sel.xpath('//title/text()').extract()[0]
items = []
item = MaizieduCourseItem()
item['link'] = link
item['name'] = name
items.append(item)
return items
def parse_item(self, response):
response = Selector(response)
itme = S80Item()
itme['电影'] = response.xpath('//h1/text()').extract()
itme['类型'] = response.xpath(
'//span[@class="span_block"][1]/a/text()').extract()
itme['演员'] = response.re('a href="/actor/.*?>([\u4e00-\u9fa5·]+)<')
itme['地区'] = response.xpath(
'//div[@class="clearfix"]/span[2]/a/text()')[0].extract()
itme['语言'] = response.xpath(
'//div[@class="clearfix"]/span[3]/a/text()').extract()
itme['导演'] = response.xpath(
'//div[@class="clearfix"]/span[4]/a/text()').extract()
itme['片长'] = response.xpath(
'//div[@class="clearfix"]/span[6]/text()').extract()
itme['上映时间'] = response.xpath(
'//div[@class="clearfix"]/span[5]/text()').extract()
itme['下载链接'] = list(set(response.re('"(thunder://.*?)"')))
return itme
def parse_page(self, response):
sel = Selector(response)
email = sel.re('(\w+@[a-zA-Z_]+?\.[a-zA-Z]{2,6})')
if bool(email):
email = email[0]
if email + "\n" not in email_in_file and email not in added_email:
file.write(email+'\n')
added_email.append(email)
print "Spider: Mandy. Email {0} added to file".format(email)
else:
print "Spider: Mandy. Email {0} already in the file".format(email)
def get_asn(self, response):
parent = response.meta['parent']
base = parent.get('base')
if base:
sel = Selector(response)
asn = ''.join(sel.re('([\d]+)\s+\<')) + ''.join(
sel.xpath("//span[1]/text()").extract()).strip()
print '______________________________'
print asn
base['asn'] = asn
yield parent
def parse_item(self, response):
selector = Selector(response)
# Use regex to find valid emails
emails = selector.re(self.EMAIL_REGEX)
if (emails):
item = CompanyItem({'link': response.url})
item['emails'] = emails
return item
return
def parse_news(self, response):
sel = Selector(response)
pattern = re.match(self.url_pattern, str(response.url))
item = NewsItem()
item['source'] = 'news.sina.com.cn' # pattern.group(1)
item['date'] = ListCombiner(str(pattern.group(2)).split('-'))
item['newsId'] = sel.re(r'comment_id:(\d-\d-\d+)')[0]
item['contents'] = {'link':str(response.url), 'title':u'', 'passage':u''}
item['contents']['title'] = sel.xpath("//h1[@id='artibodyTitle']/text()").extract()[0]
item['contents']['passage'] = ListCombiner(sel.xpath('//p/text()').extract())
return item
def parse_item(self, response):
selector = Selector(response)
# Use regex to find valid emails
emails = selector.re(self.EMAIL_REGEX)
if (emails):
item = CompanyItem({'link': response.url})
item['emails'] = emails
return item
return
def parse_get_user(self,response):
log.msg("parse_get_user: "******"id=(.*?)\\"')
for user in users:
url='http://www.weibo.com/'+user
yield Request(url=url,cookies=self.login_cookie,callback=self.parse_user,meta={'url':url})
except Exception, e:
log.msg("Error for parse_get_user: " + response.url, level=log.ERROR)
log.msg(str(e), level=log.ERROR)
def parse_page(self, response):
sel = Selector(response)
emails = sel.re('(\w+@[a-zA-Z_]+?\.[a-zA-Z]{2,6})')
emails = list(filter(lambda x: x != '*****@*****.**', emails))
if bool(emails):
for email in emails:
if email + "\n" not in email_in_file and email not in current_session_emails:
file.write(email+'\n')
current_session_emails.append(email)
print 'Spider: ProductionHub. Email {0} added to file'.format(email)
else:
print 'Spider: ProductionHub. Email {0} already in the file'.format(email)
def parseTmall(self, response):
""" Tmall parser
"""
def _referer():
referer = response.request.headers.get('Referer')
if referer and referer.find('list.tmall.com') > -1:
rto = 'http://list.tmall.com/search_product.htm?'
resultC = re.compile('[\?&]cat=(\d+)').search(referer)
if resultC: rto += 'cat=%s' % resultC.group(1)
resultQ = re.compile('[\?&]q=([^&]+)').search(referer)
if resultQ:
if resultC: rto += '&q=%s' % resultQ.group(1)
else: rto += 'q=%s' % resultQ.group(1)
if not 'http://list.tmall.com/search_product.htm?' == rto:
return rto
elif not referer and response.url.find('detail.tmall.com') > -1:
return response.url
return ''
sel = Selector(response)
item = ProductItem()
item['source'] = 'tmall'
item['name'] = self.get_product_name(sel)
item['start_url'] = _referer()
store = ''.join(
sel.xpath('//input[@name="seller_nickname"]/@value').extract())
item['tm_store'] = '[%s] %s' % (store[-3:],
store) if len(store) > 3 else store
try:
# 获取TShop字符串，并对TShop字符串进行JSON标准化处理
TShop_str = sel.re('TShop\.Setup\(((.|\n)+?)\);')[0]
# 移除注释，目前只有天猫超市有注释，以逗号开头
regex = re.compile(',\s*\/\/[^\n]*')
TShop_str = re.sub(regex, ',', TShop_str)
TShop = eval(
TShop_str,
type('Dummy', (dict, ), dict(__getitem__=lambda s, n: n))())
except SyntaxError:
return
item['itemId'] = TShop.get('itemDO').get('itemId', '')
item['url'] = response.url
initApi_url = TShop.get('initApi')
yield Request(initApi_url,
headers={'Referer': 'http://www.google.com.hk/'},
meta={'item': item},
dont_filter=True,
callback=self.parse_initapi)
def parse(self, response):
sel = Selector(response)
products = json.loads("{"+sel.re("\"products\":\[.*\]")[0]+"}")
items = []
for product in products["products"]:
#print(product)
item = SephoraItem()
item['display_name'] = product["display_name"]
item['list_price'] = product["derived_sku"]["list_price"]
item['brand_name'] = product["brand_name"]
item['rating'] = product["rating"]
items.append(item)
return items
def parse_news(self, response):
sel = Selector(response)
pattern = re.match(self.url_pattern, str(response.url))
item = TencentItem()
item['source'] = 'tencent' # pattern.group(1)
item['date'] = pattern.group(2)
item['newsId'] = pattern.group(3)
item['cmtId'] = (sel.re(r"cmt_id = (.*);"))[0] # unicode string
item['comments'] = {'link':str('http://coral.qq.com/')+item['cmtId']}
item['contents'] = {'link':str(response.url), 'title':u'', 'passage':u''}
item['contents']['title'] = sel.xpath('//h1/text()').extract()[0]
item['contents']['passage'] = ListCombiner(sel.xpath('//p/text()').extract())
return item
def parse(self, response):
self.driver.get(response.url)
self.driver.execute_script("window.scrollTo(0, 1600)")
self.driver.find_element_by_xpath(
"//div[@class = 'index-showMoreText']").click()
mount_root = Selector(text=self.driver.page_source)
print(mount_root.re(r'url\("([^\")]+)'))
row_keys = mount_root.xpath(
"//*[@class = 'index-rowKey']/text()").getall()
row_values = mount_root.xpath(
"//*[@class = 'index-rowValue']/text()").getall()
specifications = dict(zip(row_keys, row_values))
def parse4(self, response):
print("PARSE4 GO！！！")
selector = Selector(response=response)
li = selector.re(r'<ul class="listContent">(.*?)</ul>')
'''遍历所有页数，判断li标签是否存在'''
if not li:
# 通过XPath获取详情页url信息
url_pattern = r'//div[@class="content"]/div[1]/ul[@class="listContent"]/li/a/@href'
url_list = selector.xpath(url_pattern).extract()
print("======Redis连接信息：Host:{} Port:{}======".format(
self.settings.get('REDIS_HOST'),
self.settings.get('REDIS_PORT')))
# 将URL信息写入Redis数据库
print('PARSE1 开始写入')
for u in url_list:
print("准备写入{}".format(u))
self.r_link.rpush("Lianjia:detail_url", u)
print("{}写入成功!".format(u))
print('=' * 30, '\n', "共计写入url：{}个".format(len(url_list)), '\n',
'=' * 30)
elif li[0] != '':
# 通过XPath获取详情页url信息
url_pattern = r'//div[@class="content"]/div[1]/ul[@class="listContent"]/li/a/@href'
url_list = selector.xpath(url_pattern).extract()
# 启用Redis服务
r_link = redis.Redis(port=self.settings.get('REDIS_PORT'),
host=self.settings.get('REDIS_HOST'),
decode_responses=True,
db=1)
print("======Redis连接信息：Host:{} Port:{}======".format(
self.settings.get('REDIS_HOST'),
self.settings.get('REDIS_PORT')))
# 将URL信息写入Redis数据库
print('PARSE4 开始写入')
for u in url_list:
print("准备写入{}".format(u))
r_link.rpush("Lianjia:detail_url", u)
print("{}写入成功!".format(u))
print('=' * 30, '\n', "共计写入url：{}个".format(len(url_list)), '\n',
'=' * 30)
elif li[0] == '':
pass
def parse_itme(self, response):
print('1')
x = Selector(response)
names = x.xpath('//ul/li/b/text()').extract()
leibie = x.re('小说分类：([\u4e00-\u9fa5]+)')
links = LinkExtractor(allow=('/down/\d+.html'))
link_list = links.extract_links(response)
for link in link_list:
yield Request(url=link.url,
callback=self.parse_itme1,
meta={
'name': names,
'leibie': leibie
})
def parseTmall(self, response):
""" Tmall parser
"""
def _referer():
referer = response.request.headers.get('Referer')
if referer and referer.find('list.tmall.com') > -1:
rto = 'http://list.tmall.com/search_product.htm?'
resultC = re.compile('[\?&]cat=(\d+)').search( referer )
if resultC: rto += 'cat=%s' % resultC.group(1)
resultQ = re.compile('[\?&]q=([^&]+)').search( referer )
if resultQ:
if resultC: rto += '&q=%s' % resultQ.group(1)
else: rto += 'q=%s' % resultQ.group(1)
if not 'http://list.tmall.com/search_product.htm?' == rto:
return rto
elif not referer and response.url.find('detail.tmall.com') > -1:
return response.url
return ''
sel = Selector(response)
item = ProductItem()
item['source'] = 'tmall'
item['name'] = self.get_product_name( sel )
item['start_url'] = _referer()
store = ''.join( sel.xpath('//input[@name="seller_nickname"]/@value').extract() )
item['tm_store'] = '[%s] %s' % (store[-3:], store) if len(store) > 3 else store
try:
# 获取TShop字符串，并对TShop字符串进行JSON标准化处理
TShop_str = sel.re('TShop\.Setup\(((.|\n)+?)\);')[0]
# 移除注释，目前只有天猫超市有注释，以逗号开头
regex = re.compile(',\s*\/\/[^\n]*')
TShop_str = re.sub(regex, ',', TShop_str)
TShop = eval( TShop_str, type('Dummy', (dict,), dict(__getitem__=lambda s,n:n))() )
except SyntaxError:
return
item['itemId'] = TShop.get('itemDO').get('itemId', '')
item['url'] = response.url
initApi_url = TShop.get('initApi')
yield Request( initApi_url,
headers={'Referer': 'http://www.google.com.hk/'},
meta={'item': item},
dont_filter=True,
callback=self.parse_initapi )
def parse_url(self, response):
hxs = Selector(response)
download_url = hxs.re('jjvod_url = \'(.+)\';')
meta = response.request.meta
imgs = meta.get('imgs')
title = meta.get('title')
url = meta.get('url')
item = CCUrlItem()
item['url'] = url
item['download_url'] = download_url
item['title'] = title
item['image_urls'] = imgs
yield item
def parse_news(self, response):
sel = Selector(response)
pattern = re.match(self.url_pattern, str(response.url))
item = NeteaseItem()
item['source'] = 'netease' # pattern.group(1)
item['date'] = '20' + pattern.group(2) + pattern.group(3)
item['newsId'] = pattern.group(4)
item['cmtId'] = item['newsId']
item['boardId'] = sel.re(r"boardId = \"(.*)\"")[0]
item['comments'] = {'link':str('http://comment.news.163.com/'+item['boardId']+'/'+item['cmtId']+'.html')}
item['contents'] = {'link':str(response.url), 'title':u'', 'passage':u''}
item['contents']['title'] = sel.xpath("//h1[@id='h1title']/text()").extract()[0]
item['contents']['passage'] = ListCombiner(sel.xpath('//p/text()').extract())
return item
def parse(self, response):
selector = Selector(response)
p1 = re.compile('<ul class="clearfix" style="display:none;">(.*?)</ul>', re.S) # 这个正则表达式会找的7个匹配<ul></ul>
rp1 = selector.re(p1)
p2 = re.compile(u'<a data-id="\d*" data-key="\D*" href="(http://.*)">(.*)</a>')
r = Redis()
for index, i in enumerate(rp1):
rp2 = p2.findall(i)
for j in rp2:
con_item = ContryItem()
con_item['name'] = j[1]
con_item['url'] = j[0]
con_item['con_id'] = index+1 #continent ID fk
Redis_utils.server.lpush('myspider:qyconpage_urls', j[0])
yield con_item
def parse(self, response):
selector = Selector(response=response)
results = selector.re(self.WIF_REGEX)
#self.root_logger.debug("REGEX: " + str(results))
for wif in results:
if self.balance_checker.is_valid(wif):
info = self.balance_checker.get_key_info(wif)
if info.balance > 0:
self.root_logger.critical("CASH: " + str(info))
self.root_logger.critical("URL: " + str(response.url))
self.root_logger.critical("BODY: " + str(response.body))
links = LinkExtractor(canonicalize=True,
unique=True).extract_links(response)
for link in links:
yield scrapy.Request(link.url, callback=self.parse)
def parse_stop_data(self, response):
"""Parse stop data from a response where the "from" and "to"
stops were selected in the request."""
sel = Selector(response)
stop_name = sel.css('#confirm1_hlFrom').xpath('text()').extract()[0]
stop_location = sel.css('#divFrom').xpath('p[1]/text()').extract()[0]
latlong = sel.re(r'Location\(([\d\.\-,]+)\)')[0]
(stop_lat, stop_long) = latlong.split(',')
return StopItem(
stop_name=stop_name,
stop_location=self.sanitize_stop_location(stop_location),
lat=stop_lat,
long=stop_long
)
def parse(self, response):
sel = Selector(response)
data = sel.re(r'\(([^)]+)\)')[0]
try:
data = json.loads(data)
except Exception as e:
print e
self.log(e)
print data
self.log(data)
print response.body
url = urlparse.unquote(response.url)
nums = re.search(r'"startPoint":{"latitude":(\d+.\d+),"longitude":(\d+.\d+)},"endPoint":{"latitude":(\d+.\d+),"longitude":(\d+.\d+)}',url).groups()
slat, slong, elat, elong = [float(n) for n in nums]
l = abs(slat - elat)
w = abs(slong - elong)
print 'EXPLORING:', (slat,slong), (elat,elong)
self.log('EXPLORING: (%f, %f) (%f,%f)' %(slat,slong,elat,elong))
quarters = [{'start':(slat, slong), 'wid':w/2, 'len':l/2},
{'start':(slat, slong + w/2), 'wid':w/2, 'len':l/2},
{'start':(slat - l/2, slong), 'wid':w/2, 'len':l/2},
{'start':(slat - l/2, slong + w/2), 'wid':w/2, 'len':l/2}]
if not data:
self.log('Failed: No data')
return
for point in data['pointDataList']:
coord = point['coordinate']
if point['count'] > 1:
for i,a in enumerate(quarters):
if self.intersect(a, coord) and a['len'] >= self.minLen:
slat2, slong2 = a['start']
w2, l2 = a['wid'], a['len']
newUrl = self.getUrl((slat2,slong2), (slat2-l2, slong2+w2))
quarters.pop(i)
yield Request(newUrl)
break
else:
celldata = point['cellData']
yield Cell(ID=celldata['cellId'], networkID=celldata['mnc'], latitude=coord['latitude'], longitude=coord['longitude'])
def parse(self, response):
#log.msg( response.body, level=log.INFO )
sel = Selector(response)
#1. all items in current page
urls = sel.re('<a href="(http://finance.sina.com.cn/stock/jsy/\\d+/\\d+.shtml)" target="_blank">')
for url in urls:
log.msg( url, level=log.INFO )
yield Request( url, callback=self.parse_item)
#2. next page detect
#pageBar = sel.css('#Main > div.listBlk > table:nth-child(1) > tbody > tr > td > div > span.pagebox_next')
#pageBar = sel.xpath( '//div[@id="Main"]/div[3]/table[1]/tbody/tr/td/div')
#pageBar = response.xpath( '//div[@id="Main"]/div[3]/table[1]/tbody/tr/td/div')
pageBar = response.xpath('//span[@class="pagebox_next"]')
if pageBar != None and len(pageBar) > 0 :
pageTxt = pageBar.extract()[0]
log.msg( 'matched txt:'+pageTxt, level=log.INFO )
tail_url = self.reg_next_page.search( pageTxt )
log.msg('NEXT PAGE: '+tail_url.group(1), level=log.INFO )
yield Request( 'http://roll.finance.sina.com.cn/finance/zq1/gsjsy/'+tail_url.group(1), callback=self.parse )
def parse_article(self, response):
article = response.meta["article"]
# We'll be using a regex for pub_time so we can't use selector
# response shortcuts.
sel = Selector(response=response)
# TW adds a newline at the beginning of the title, so strip the string.
article["title"] = sel.xpath(
"//span[@class='titlea']/text()").extract_first().strip()
# Publication date and time is grouped in a span together with the
# number of times the article was accessed. Therefore, use a regex to
# extract just the date and time. Matches eg.:
# "04. July 2015. [17:16:00 UTC]"
article["pub_time"] = sel.re(r"\d{2}\. \w+ \d{4}\. \[\d{2}:\d{2}:\d{2} UTC\]")[0]
article["body"] = sel.xpath("//div[@class='articlebody']").extract_first()
yield article
def parse_item(self, response):
sel = Selector(response)
items = sel.xpath(TABLE_TR_XPATH)
for item in items:
problem_id = item.xpath(
PROBLEM_ID_XPATH).extract()[0].strip()
submit_time = item.xpath(
SUBMIT_TIME_XPATH).extract()[0].split(' ')[0]
self.solved[problem_id] = submit_time
if not sel.re(RULE_REGEX):
yield AccountItem(**dict(
origin_oj=self.origin_oj,
username=self.username,
solved=self.solved
))
raise CloseSpider('Crawl finished')
return
