def parse_list(self,response):
#Get seller attributes
sel = Selector(response)
self.page += 1
for s in sel.xpath(Seller.base_xpath):
seller_loader = ItemLoader(Seller(),selector=s)
# iterate over fields and add xpaths to the seller_loader
seller_loader.add_value('page',self.page)
seller_loader.add_value('flag','Seller')
for key,value in Seller.item_fields.iteritems():
seller_loader.add_xpath(key,value)
yield seller_loader.load_item()
#Get commodity attributes
for s in sel.xpath(Commodity.base_xpath):
comm_loader = ItemLoader(Commodity(),selector=s)
comm_loader.add_value('page',self.page)
comm_loader.add_value('flag','Commodity')
for key,value in Commodity.item_fields.iteritems():
comm_loader.add_xpath(key,value)
yield comm_loader.load_item()
#Next page
if(sel.xpath(self.next_page_xpath)):
yield Request("http://spu.taobao.com/spu/3c/detail.htm" +
sel.xpath(self.next_page_xpath).extract()[0],
callback=self.parse_list)
def parse(self,response):
l = ItemLoader(item = timeItem(),response = response)
#l.add_xpath('topnews','//*[@id="article-container"]/div/div[1]/section/div/article[*]/div/p/text()')
l.add_xpath('topnews','//*[@id="article-container"]/div/div[1]/section/div/article[*]/div/h2/a/text()')
l.add_xpath('topnews','//*[@id="article-container"]/div/div[1]/section/div/article[1]/div/div/div[2]/div[*]/h3/a/text()')
l.add_xpath('sectionnews','//a[contains(@class,"home-columnists-title")]/text()')
l.add_xpath('sectionnews','//a[contains(@data-event,"hp-news")]/text()')
x = l.load_item()
nytdict = dict()
datelist = []
datalist = datetime.date.today()
topnewslist = []
sectionnewslist = []
nytdict['date'] = str(datalist)
for t in x['topnews']:
topnewslist.append(str(t.encode('ascii','ignore')).strip())
nytdict['topnews']=topnewslist
for t in x['sectionnews']:
sectionnewslist.append(str(t.encode('ascii','ignore')).strip())
nytdict['sectionnews']=sectionnewslist
filename = datetime.date.today()
f=open('{}.json'.format(filename),'w')
json.dump(nytdict,f)
return l.load_item()
def parse_review(self, response):
sel = Selector(response)
if not self._is_right_category(sel):
self.log('Skip URL: %s' % response.url, level=log.INFO)
return
self.log('Parse URL: %s' % response.url, level=log.INFO)
loader = ItemLoader(item=YelpReview(), selector=sel)
loader.add_value('crawl_date', '%s' % datetime.utcnow())
loader.add_value('page_url', response.url)
# Loop over all the fields we need to extract.
for field, selector in self._item_selectors.iteritems():
loader.add_xpath(field, selector)
master_review = loader.load_item()
review_selectors = sel.xpath('//div[contains(@class, "review")][@itemprop="review"]')
for rev_sel in review_selectors:
review_loader = ItemLoader(item=master_review.copy(), selector=rev_sel)
for field, selector in self._review_selectors.iteritems():
review_loader.add_xpath(field, selector)
yield review_loader.load_item()
return
def parse_book_url(self, response):
book_item = BookDetails(book_id="", book_type="pdf")
bil = ItemLoader(item=book_item, response=response)
bil.add_xpath("book_id",
"/*//script/text()",
re=r'bookId\s*:\s*(.*),.*')
bil.add_xpath("book_path",
"/*//script/text()",
re=r'getDownloadUrl\s*:\s*\"(.*)\".*')
#bil.get_xpath()
bil.load_item()
download_url = self.base_url + book_item['book_path'][0]
post_data = "book_id=" + book_item['book_id'][
0] + "&" + "type=" + book_item['book_type']
#post_data = "book_id=" + "2759" + "&" + "type=" + book_item['book_type']
#set header
post_header = {}
post_header[
"Content-Type"] = "application/x-www-form-urlencoded; charset=UTF-8"
post_header["User-Agent"] = "Mozilla/5.0"
#print post_header
#print curl_cmd
yield Request(download_url,
self.get_book_link,
headers=post_header,
method='POST',
body=post_data)
def get_new(self, response):
sel = Selector(response)
il = ItemLoader(item=New())
il.add_value('tema', ['Marketing y Publicidad'])
il.add_value('titulo', sel.xpath('//h1[@itemprop="headline"]/a/text()').extract())
il.add_value('texto', sel.xpath('//div[@itemprop="articleBody"]').extract())
il.add_value('fecha', sel.xpath('//div[@itemprop="datePublished"]/text()').extract())
il.add_value('keywords', sel.xpath('//div[contains(@class,"nota-tags")]//h3/a/text()').extract())
item = il.load_item()
if 'titulo' in item:
pass
else:
iln = ItemLoader(item=New())
iln.add_value('tema', ['Marketing y Publicidad'])
iln.add_value('titulo', sel.xpath('//h1/text()').extract())
iln.add_value('texto', sel.xpath('//div[@id="principal"]/div[@class="nota"]/div[3]').extract())
iln.add_value('fecha', sel.xpath('//div[@class="fecha-nota"]/text()').extract())
iln.add_value('keywords', sel.xpath('//div[contains(@class,"nota-tags")]//h3/a/text()').extract())
item = iln.load_item()
if 'keywords' in item:
pass
else:
item['keywords'] = ['Marketing y Publicidad']
if 'fecha' in item:
item['fecha'] = self.parse_date(item['fecha'])
else:
item['fecha'] = '10/05/2015'
if 'titulo' in item:
if 'texto' in item:
yield item
def parse_content(self, response):
goods_loader = ItemLoader(item=AlibbItem(), response = response)
url = str(response.url)
goods_loader.add_value('url', url)
goods_loader.add_value('url_hash',hashlib.sha1(url).hexdigest())
goods_loader.add_xpath('name', self._x_query['title'].encode('utf-8'))
# detail data
iDetailDataPattern=re.compile("iDetailData.*};",re.DOTALL)
detail_data_list=response.xpath('//script').re(iDetailDataPattern)
detail_data=detail_data_list[0].replace("iDetailData = {","{")
detail_data=detail_data.replace("};","}")
detail_data=detail_data.replace("\t|\n|\\","")
detail_data_json=json.loads(detail_data)
if len(detail_data_json)!=0:
properties=detail_data_json['sku']['skuMap'].keys()
goods_loader.add_value('properties',[property.replace(">",",") for property in properties])
for attribute in detail_data_json['sku']['skuProps']:
attributes={}
options=[value['name'] for value in attribute['value']]
attributes['name']=attribute['prop']
attributes['options']=options
goods_loader.add_value('attributes',attributes)
else:
goods_loader.add_value('attributes',"")
price=response.xpath('//span[re:test(@class,"value price-length-\d$")]/text()').extract()
goods_loader.add_value('price',price[0] if len(price)>0 else detail_data_json['sku']['price'])
# detail information
detail_info_list=response.xpath(self._x_query['detail_info']).extract()
goods_loader.add_value('parameters', [list(info_list) for info_list in zip(detail_info_list[::2],detail_info_list[1::2])])
print goods_loader.load_item()['url']
# profile img
profile_img_urls=response.xpath('//li/@data-imgs').re("original.*jpg")
for urls in profile_img_urls:
profile_img_url=urls.replace("original\":\"http","http")
goods_loader.add_value("boothes",profile_img_url)
# big img
for link in response.xpath('//*[@id="desc-lazyload-container"]/@data-tfs-url').extract():
yield Request(url = link, meta={'item': goods_loader},callback=self.parse_content_down)
def parse_product(self, response):
product_url = response.url
# sel = self.selenium
#sel.open(response.url)
#time.sleep(2.5)
selector = Selector(response)
# //*[@id="product_detail_view_1"]/div/div[6]/div[2]/span[2]
price = selector.xpath('//*[@id="product_detail_view_1"]/div/div[7]/div[2]/span[2]/text()').extract()
if not price:
price = selector.xpath('//*[@id="product_detail_view_1"]/div/div[6]/div[2]/span[2]/text()').extract()
if not price:
price = selector.xpath(
'//*[@id="product_detail_view_1"]/div/div[5]/div[2]/span[2]/text()').extract()
if not price:
price = selector.xpath(
'//*[@id="product_detail_view_1"]/div/div[4]/div[2]/span[2]/text()').extract()
l = ItemLoader(item=BillionPricesIndiaItem(), response=response)
l.add_xpath('product_name', '//*[@id="inner"]/div[1]/div[1]/div/div/text()')
l.add_xpath('quantity', '//*[@id="product_detail_view_1"]/div/div[1]/div/text()')
l.add_xpath('category', '//*[@id="inner"]/div[1]/div[1]/div/a[1]/text()')
l.add_xpath('product', '//*[@id="inner"]/div[1]/div[1]/div/a[2]/text()')
item = l.load_item()
item['product_url'] = product_url
item['price'] = price
item['vendor'] ='Local Banya'
item['city'] = 'Mumbai'
item['state'] = 'Maharashtra'
item['country'] = 'India'
item['date']=str(time.strftime("%d/%m/%Y"))
return item
def parse_item(self, response):
sel = Selector(response)
# collect xpaths of each player (row in table)
rows = sel.xpath('/html//div[@class="table-container"]/table/tbody/tr')
# loop through players
for row in rows:
loader = ItemLoader(SkatTOIItem(), selector=row)
loader.default_input_processor = MapCompose()
loader.default_output_processor = Join()
# get unique NHL ID number from player's page URL
num = row.xpath("td[2]/a/@href").extract()
sNum = num[0][-7:]
loader.add_value("nhl_num", sNum)
# add season data
loader.add_value("season", str(self.year))
# collect TOI stats after converting from m,mmm:ss to seconds
i = 5
CATEG = ["es_toi", "sh_toi", "pp_toi", "toi"]
while i < 12:
i += 1
if i % 2 == 0:
temp = row.xpath("td[" + str(i) + "]/text()").extract()[0]
sTemp = temp.split(":")
sTemp[0] = sTemp[0].replace(",", "")
loader.add_value(CATEG[(i - 6) / 2], str(60 * int(sTemp[0]) + int(sTemp[1])))
else:
pass
# feed item to pipeline
yield loader.load_item()
def parse_item(self, response):
sel = Selector(response)
# collect xpaths of each player (row in table)
rows = sel.xpath('/html//div[@class="table-container"]/table/tbody/tr')
# loop through players
for row in rows:
loader = ItemLoader(SkatSOItem(), selector=row)
loader.default_input_processor = MapCompose()
loader.default_output_processor = Join()
# get unique NHL ID number from player's page URL
num = row.xpath("td[2]/a/@href").extract()
sNum = num[0][-7:]
loader.add_value("nhl_num", sNum)
# add season data
loader.add_value("season", str(self.year))
# collect stats
loader.add_xpath("so_shots", ".//td[13]/text()")
loader.add_xpath("so_goals", ".//td[14]/text()")
loader.add_xpath("so_pct", ".//td[15]/text()")
loader.add_xpath("game_deciding_goals", ".//td[16]/text()")
# feed item to pipeline
yield loader.load_item()
def process_row(self, row, task):
stats = self.crawler.stats
l = ItemLoader(WV_DrillingPermit())
l.add_value(None, row)
item = l.load_item()
if item['API'] and item['permit_activity_type'] and item[
'permit_activity_date']:
existing_item = self.db.loadItem(
item, {
'API': item['API'],
'permit_activity_type': item['permit_activity_type'],
'permit_activity_date': item['permit_activity_date']
})
if existing_item:
stats.inc_value('_existing_count', spider=self)
else:
stats.inc_value('_new_count', spider=self)
yield item
dt = datetime.strptime(item['permit_activity_date'],
'%Y-%m-%d %H:%M:%S')
# if item['permit_activity_type'] in ('Permit Issued', 'Permit Commenced', 'Permit Completed'):
if item['permit_activity_type'] in (
'Permit Issued', 'Permits Issued'
) and datetime.now() - dt < timedelta(days=365):
for item in self.create_feed_entry(item, task):
yield item
def parse(self, response):
l = ItemLoader(item=Problem(), response=response)
d = pyq(response.body)
l.add_value('id', response.url[-4:])
l.add_value('title', d('#content_body > center:nth-child(1) > span').text())
l.add_value('body', d('#content_body').text())
return l.load_item()
def parse(self, response):
contents = response.xpath(
'//ul[@class="note-list"]/li/div[@class="content"]')
for content in contents:
l = ItemLoader(item=JianshuSummaryItem(),
selector=content,
response=response)
l.add_xpath(
'title', 'a[@class="title"]/text()',
MapCompose(lambda i: i.replace('|', '.').replace('丨', '.')))
l.add_xpath('link', 'a[@class="title"]/@href',
MapCompose(lambda i: urljoin(response.url, i)))
l.add_xpath(
'author',
'div[@class="author"]/div[@class="info"]/a[@class="nickname"]/text()'
)
l.add_xpath(
'author_url',
'div[@class="author"]/div[@class="info"]/a[@class="nickname"]/@href',
MapCompose(lambda i: urljoin(response.url, i)))
l.add_xpath(
'timestamp',
'div[@class="author"]/div[@class="info"]/span[@class="time"]/@data-shared-at'
)
l.add_xpath('read', 'div[@class="meta"]/a[1]/text()[2]',
MapCompose(str.strip, int))
l.add_xpath('reply', 'div[@class="meta"]/a[2]/text()[2]',
MapCompose(str.strip, int))
l.add_xpath('like', 'div[@class="meta"]/span[1]/text()',
MapCompose(str.strip, int))
l.add_xpath('money', 'div[@class="meta"]/span[2]/text()',
MapCompose(str.strip, int))
yield l.load_item()
pass
def parse_content(self, response):
'''Parse content pages.'''
loader = ItemLoader(item=Rede(), response=response)
# Usually, we are only interested in the first item, e.g. for title, place, etc.
loader.default_output_processor = TakeFirst()
# Add fields
loader.add_value('link', response.url)
loader.add_css('title', '.text h1', extract_text)
# Test if text has an abstract
abstract = response.css('.abstract')
if abstract:
loader.add_css('abstract', '.abstract', extract_text)
loader.add_css('text', '.abstract ~ p:not(.picture)',
extract_text, Join('\n'))
else:
loader.add_css('text', '.text p:not(.picture)',
extract_text, Join('\n'))
# Metadata are in dt/dd pairs.
keys = response.css('dl dt::text').extract()
values = response.css('dl dd::text').extract()
for key, value in zip(keys, values):
if key == 'Datum:':
match = re.search(r'(\d{1,2}\.\d{1,2}\.\d{2,4})', value)
if match:
# '22.03.2011' format
value = match.group(1)
dt = datetime.strptime(value.encode(ENC), '%d.%m.%Y')
else:
# '22. März 2011' format
dt = datetime.strptime(value.encode(ENC), '%d. %B %Y')
loader.add_value('date', dt.date())
elif key == 'Ort:':
loader.add_value('place', value)
return loader.load_item()
def parse_items(self, response):
item = ItemLoader(Articulos(), response)
item.add_xpath('title', '//*[@id="MainContainer"]/article/section[1]/div[1]/div/h2/text()')
item.add_xpath('description', '//*[@id="MainContainer"]/article/section[1]/div[2]/ul/li[3]/text()')
yield item.load_item()
# scrapy runspider multiplepages.py -o ../../resources/computrabajo.csv -t csv
def create_tag(self, feed_entry_id, tag, comment=''):
# TODO: create tags
l = ItemLoader(FeedEntryTag())
l.add_value('feed_entry_id', feed_entry_id)
l.add_value('tag', tag)
l.add_value('comment', comment)
return l.load_item()
def parse_item(self, response):
# 解析http://www.meizitu.com/a/5336.html获取图片URL
l = ItemLoader(item=MeizituItem(), response=response)
l.add_xpath('image_urls', "//div[@id='picture']/p/img/@src",
Identity())
l.add_value('url', response.url)
return l.load_item()
def parse_detail(self, response):
url = response.url
item = ItemLoader(item=MeizituItem(), response=response)
item.add_xpath("title", "//h2/a/text()")
item.add_xpath("image_urls", '//div[@id="picture"]//img/@src')
item.add_value('url', url)
return item.load_item()
def parse(self, response):
data = json.loads(response.text)
for key in data.keys():
dataLenth = len(data[key])
logging.info("total size: " + str(dataLenth))
for i in range(1, dataLenth):
logging.info("curIndex: " + str(i))
content = {}
logging.info("title: " + data[key][i]["title"])
loader = ItemLoader(item=NetNews(), response=response)
content["title"] = data[key][i]["title"]
loader.add_value("news_title", content["title"])
content["digest"] = data[key][i]["digest"]
loader.add_value("news_digest", content["digest"])
# print "recSource: "+data[key][i]["recSource"]
content["label"] = data[key][i]["recSource"]
loader.add_value("news_label", content["label"])
content["imageUrl"] = data[key][i]["imgsrc"]
loader.add_value("news_img_url", content["imageUrl"])
content["id"] = data[key][i]["id"]
loader.add_value("news_id", content["id"])
detail = self.getNewsDetail(content["id"])
# print "body: "+detail
if (len(detail) != 0):
loader.add_value("news_detail", detail)
# 添加时间戳
loader.add_value("timestamp", self.get_timestamp())
yield loader.load_item()
def get_new(self, response):
sel = Selector(response)
il = ItemLoader(item=New())
il.add_value('tema', ['Marketing y Publicidad'])
il.add_value('titulo', sel.xpath('//h1[@class="glr-post-title glr-mb-10"]/text()').extract())
il.add_value('texto', sel.xpath('//div[@class="glr-post-entry"]').extract())
il.add_value('fecha', sel.xpath('//span[@class="glr-left glr-post-date"]/text()').extract())
il.add_value('keywords', sel.xpath('//div[@class="post-tags"]//a/text()').extract())
item = il.load_item()
if 'keywords' in item:
pass
else:
item['keywords'] = ['Marketing y Publicidad']
if 'fecha' in item:
item['fecha'] = self.parse_date(item['fecha'])
else:
item['fecha'] = '10/05/2015'
if 'titulo' in item:
if 'texto' in item:
yield item
'''
item = New()
item['tema'] = 'Marketing y Publicidad'
item['titulo'] = self.parse_html(sel.xpath('//h1[@class="glr-post-title glr-mb-10"]/text()').extract()[0].strip())
item['texto'] = self.parse_html(sel.xpath('//div[@class="glr-post-entry"]').extract()[0].strip())
item['fecha'] = self.parse_date(sel.xpath('//span[@class="glr-left glr-post-date"]/text()').extract()[0].strip())
'''
#yield item
'''res = []
def parse(self, response):
for sel in response.css("ul#channels-browse-content-grid > li"):
loader = ItemLoader(YoutubeVideo(), selector=sel)
loader.add_xpath('link', './/h3/a/@href')
yield loader.load_item()
def parse(self, response):
content = response.body
page = response.url.split("/")[-1]
"""
content = Selector(response=response).xpath("//div[@class='body textStyle']").extract()
if (len(content)):
content = content[0]
#踢除标签
strip = StripTags()
content = strip.filterTags(content)
#写文件
filename = 'quotes-%s' % page
with open(filename, 'w') as f:
f.write(str(content))
self.log('Saved file %s' % filename)
"""
loader = ItemLoader(item=TutorialItem(), response=response)
loader.add_xpath('title', "//title/text()")
loader.add_xpath('content', "//div[@class='body textStyle']")
data = loader.load_item()
downFile = DownFile(data['content'][0], 'http://www.admin10000.com')
downFile.downImgFile()
mongo = Mongo("articles")
mongo.setTable("admin10000")
content = data['content'][0]
# 踢除标签
strip = StripTags()
content = strip.filterTags(content)
article = {'title': data['title'][0], 'content': content}
mongo.add(article)
def parse_by_product(self, response):
"""
For the 'Bundles' category, grab the product details for the first
product listed.
"""
self.selector = Selector(response)
self.results = self.selector.xpath('//*[@id="ctl00_tdMainPanel"]')
loader = ItemLoader(item = VisionsProduct(),
selector = self.results[0])
self.field_xpaths = {
'product': ('div[contains(@class, "catalogueTitle")]'
'/h3/text()'),
'price': ('div[@id="ctl00_ContentPlaceHolder1_pnl'
'Bundle"]/div[@id="divProductDetails"]/div'
'[contains(@class, "priceAddToCart")]/div[1]/span'
'[contains(@id, "SalePrice")]/text()')
}
# Extract and load product details
loader.add_xpath('product', self.field_xpaths['product'])
loader.add_xpath('price', self.field_xpaths['price'],
re = '\$[\d]*[,]*[\d]*\.[\d]*')
loader.add_value('availability', 'Not Limited/Clearance Item')
# Because it's an individual product page, manually set the category
self.category = '/'.join(['Home', response.url.split('/')[4]])
loader.add_value('category', self.category)
yield loader.load_item()
def parse(self, response):
sel = response.xpath('.//*[@class="post_info"]')
if not sel:
self.log('posts are not find')
return
self.group_id = response.xpath('.//div[@id="group_followers"]/a/@href').re('group.=(\d+?)$')[0]
for s in sel:
wall_text = s.xpath('div[@class="wall_text"]')
text = wall_text.xpath('div/div[@class="wall_post_text"]').extract()
spam_words = get_spam_words_from_msg(text, self.spam_words_from_file)
if spam_words:
l = ItemLoader(item=VkItem(), selector=s, response=response)
date = s.xpath('div[@class="replies"]/div/small/a[1]/span/text()').extract()
date = l.get_value(date, MapCompose(normalize_date), TakeFirst())
if is_date_less_last_date(date, self.days_count_to_parse):
return
l.add_value('id', wall_text.xpath('div/a/@data-from-id').extract())
l.add_value('name', wall_text.xpath('div/a/text()').extract())
l.add_value('text', text)
l.add_value('date', date)
l.add_value('words', spam_words)
yield l.load_item()
#ban => Request()
replies_hidden = s.xpath('.//a[@class="wr_header"]/@onclick')
if replies_hidden:
url = get_url_hided_replies(replies_hidden[0].extract(), self.main_page)
yield Request(url=url, callback=self.get_hided_items)
else:
replies = s.xpath('.//div[@class="reply_table"]').extract()
for reply in replies:
raw_html = ''.join(reply.splitlines()).encode('utf-8')
html_response = HtmlResponse(url=response.url, body=raw_html)
for i in self.get_replies_items(html_response):
yield i.load_item()
yield Request(url=self.get_next_msgs_url(), method='POST', callback=self.parse, body=self.get_post_body_for_next_msgs())
def get_new(self, response):
sel = Selector(response)
il = ItemLoader(item=New())
il.add_value('tema', ['Marketing y Publicidad'])
il.add_value('titulo', sel.xpath('//h1/text()').extract())
il.add_value('texto', sel.xpath('//div[contains(@class,"post-detalle")]').extract())
il.add_value('fecha', sel.xpath('//p[@itemprop="datePublished"]/text()').extract())
il.add_value('keywords', sel.xpath('//div[contains(@class,"tags")]/a/text()').extract())
item = il.load_item()
if 'titulo' in item:
pass
else:
print item['titulo']
print item['texto']
if 'keywords' in item:
pass
else:
item['keywords'] = ['Marketing y Publicidad']
if 'fecha' in item:
item['fecha'] = self.parse_date(item['fecha'])
else:
item['fecha'] = '10/05/2015'
if 'titulo' in item:
if 'texto' in item:
yield item
def create_tag_items (self, task_id, item_id):
tag_items = []
l = ItemLoader (FeedEntryTag())
l.add_value ('feed_entry_id', item_id)
l.add_value ('tag', 'NRC')
tag_items.append(l.load_item())
nrc_tags = self.db.loadNrcTags(task_id)
for t in nrc_tags:
l = ItemLoader (FeedEntryTag())
l.add_value ('feed_entry_id', item_id)
l.add_value ('tag', t['tag'])
l.add_value ('comment', t['comment'])
tag_items.append(l.load_item())
return tag_items
def parse_movie(self, response):
loader = ItemLoader(item=MovieItem(), response=response)
loader.add_xpath(
'name',
'//div[@id="title-overview-widget"]/div[2]/div[2]/div/div[2]/div[2]/h1/text()'
)
loader.add_xpath('year', "//h1/span[@id='titleYear']/a/text()")
loader.add_xpath(
'rate',
"//div[@id='title-overview-widget']/div[2]/div[2]/div/div[1]/div[1]/div[1]/strong/span/text()"
)
loader.add_xpath('director',
"//div[2]/div[1]/div[2]/span/a/span/text()")
loader.add_xpath('director',
"//div[3]/div[1]/div[2]/span/a/span/text()")
loader.add_xpath('storyline',
"//div[@id='titleStoryLine']/div[1]/p/text()")
user_review_url = response.xpath(
"//div[@id='titleUserReviewsTeaser']/div/div[3]/a[2]/@href"
).extract()
item = loader.load_item()
user_review_another_url = response.xpath(
"//div[@id='titleUserReviewsTeaser']/div/div[2]/a[2]/@href"
).extract()
if user_review_url or user_review_another_url:
full_url = 0
if not user_review_another_url:
full_url = urljoin(response.url, user_review_url.pop())
elif not user_review_url:
full_url = urljoin(response.url, user_review_another_url.pop())
request = Request(urljoin(response.url, full_url),
callback=self.parse_audience_review)
request.meta['item'] = item
return request
return item
def parse(self, response):
items = ItemLoader(item=XsContentItem(), response=response)
#章节标题
items.add_xpath('title', '//*[@class="bookname"]/h1/text()')
#正文
items.add_xpath('text', '//*[@id="content"]/text()')
yield items.load_item()
def parse_item(self, response):
sel = Selector(response)
# collect xpaths of each player (row in table)
rows = sel.xpath('/html//div[@class="table-container"]/table/tbody/tr')
# prepare to adjust for shootout stats if necessary
shootout = 0
if self.year > 2005:
shootout = 1
# loop through players
for row in rows:
loader = ItemLoader(SkatEngItem(), selector=row)
loader.default_input_processor = MapCompose()
loader.default_output_processor = Join()
# get unique NHL ID number from player's page URL
num = row.xpath("td[2]/a/@href").extract()
sNum = num[0][-7:]
loader.add_value("nhl_num", sNum)
# add season data
loader.add_value("season", str(self.year))
# collect stats
if shootout:
loader.add_xpath("en_goals", ".//td[20]/text()")
loader.add_xpath("ps_goals", ".//td[21]/text()")
else:
loader.add_xpath("en_goals", ".//td[21]/text()")
loader.add_xpath("ps_goals", ".//td[22]/text()")
# feed item to pipeline
yield loader.load_item()
def parse_item(self, response):
sel = Selector(response)
# collect xpaths of each player (row in table)
rows = sel.xpath('/html//div[@class="table-container"]/table/tbody/tr')
# loop through players
for row in rows:
loader = ItemLoader(SkatRTSItem(), selector=row)
loader.default_input_processor = MapCompose()
loader.default_output_processor = Join()
# get unique NHL ID number from player's page URL
num = row.xpath("td[2]/a/@href").extract()
sNum = num[0][-7:]
loader.add_value("nhl_num", sNum)
# add season data
loader.add_value("season", str(self.year))
# collect stats
loader.add_xpath("hits", ".//td[6]/text()")
loader.add_xpath("blocked_shots", ".//td[7]/text()")
loader.add_xpath("missed_shots", ".//td[8]/text()")
loader.add_xpath("giveaways", ".//td[9]/text()")
loader.add_xpath("takeaways", ".//td[10]/text()")
loader.add_xpath("faceoff_wins", ".//td[11]/text()")
loader.add_xpath("faceoff_losses", ".//td[12]/text()")
# feed item to pipeline
yield loader.load_item()
def parse_item(self, response):
"""This method will not populate such fields:
locality, mobile_number, country, email
"""
il = ItemLoader(item=UKBusinessItem(), response=response)
il.add_value('url', unicode(response.url))
il.add_xpath('name', '//h3[@class="biz"]/text()')
il.add_xpath('category', '//div[@id="breadcrumbs"]/a[2]/text()')
bcon_list = response.xpath('//ul[@class="bcon"]/li')
for li in bcon_list:
li_text = cond_set_value(li.xpath('.//b/text()').extract())
if li_text == 'Tel:':
phone_number = cond_set_value(li.xpath('text()').extract())
il.add_value('phone_number', phone_number)
if li_text == 'Web:':
website = cond_set_value(li.xpath('.//a/text()').extract())
il.add_value('website', website)
if li_text == 'Fax:':
fax_number = cond_set_value(li.xpath('text()').extract())
il.add_value('fax_number', fax_number)
address_list = response.xpath('//ul[@class="bad"]/li/text()').extract()
if address_list:
address_without_postal_code = u', '.join(address_list[:-1])
postal_code = address_list[-1]
il.add_value('address', address_without_postal_code)
il.add_value('postal_code', postal_code)
il.add_xpath('latitude', '//div[@id="lat"]/text()')
il.add_xpath('longitude', '//div[@id="lng"]/text()')
return il.load_item()
def detail(self, response):
log.msg(response.url)
hxs = HtmlXPathSelector(response)
product_name = hxs.xpath('//*[@id="vip_content_section"]/div[2]/h1/text()').extract()
# //*[@id="vip_content_section"]/div[2]/h1
if (len(product_name) != 0):
product_name = hxs.xpath('//*[@id="vip_content_section"]/div[2]/h1/text()').extract()[0]
product_price = hxs.xpath('//*[@id="price-val"]/text()').extract()
if (len(product_price) != 0):
product_price = hxs.xpath('//*[@id="price-val"]/text()').extract()[0]
if (len(product_price) != 0 or product_price != None) and (len(product_name) or product_name != None):
l = ItemLoader(item=BillionPricesIndiaItem(), response=response)
l.add_xpath('product_name', '//*[@id="vip_content_section"]/div[2]/h1/text()')
# l.add_xpath('quantity', '//*[@id="product_detail_view_1"]/div/div[1]/div/text()')
l.add_xpath('category', '//*[@id="cat_crum"]/@value')
l.add_xpath('product', '//*[@id="overview_tab"]/div/div/p/text()')
item = l.load_item()
item['product_url'] = response.url
item['price'] = product_price
item['vendor'] = 'PepperFry'
item['city'] = 'Mumbai'
item['state'] = 'Maharashtra'
item['country'] = 'India'
item['date'] = str(time.strftime("%d/%m/%Y"))
return item
def parse_item(self, response):
xpath = './/div[@class="content_left"]'
sel = response.xpath(xpath)
if not sel:
return
l = ItemLoader(item=HabrahabrItem(), selector=sel, response=response)
l.add_xpath('title', '//h1/span/text()')
l.add_xpath('image_urls', '//div[@class="content html_format"]/img/@src')
comments_items = []
comments = sel.xpath('//div[starts-with(@class, "message html_format")]').extract()
for comment in comments:
comment_item = ItemLoader(item=HabrahabrComment(), selector=sel, response=response)
comment_item.add_value('comment', comment)
comments_items.append(comment_item.load_item())
l.add_value('comments', comments_items)
yield l.load_item()
def detail(self, response):
log.msg(response.url)
hxs = HtmlXPathSelector(response)
product_name = hxs.xpath(
'//*[@id="vip_content_section"]/div[2]/h1/text()').extract()
# //*[@id="vip_content_section"]/div[2]/h1
if (len(product_name) != 0):
product_name = hxs.xpath(
'//*[@id="vip_content_section"]/div[2]/h1/text()').extract()[0]
product_price = hxs.xpath('//*[@id="price-val"]/text()').extract()
if (len(product_price) != 0):
product_price = hxs.xpath(
'//*[@id="price-val"]/text()').extract()[0]
if (len(product_price) != 0
or product_price != None) and (len(product_name)
or product_name != None):
l = ItemLoader(item=BillionPricesIndiaItem(), response=response)
l.add_xpath('product_name',
'//*[@id="vip_content_section"]/div[2]/h1/text()')
# l.add_xpath('quantity', '//*[@id="product_detail_view_1"]/div/div[1]/div/text()')
l.add_xpath('category', '//*[@id="cat_crum"]/@value')
l.add_xpath('product', '//*[@id="overview_tab"]/div/div/p/text()')
item = l.load_item()
item['product_url'] = response.url
item['price'] = product_price
item['vendor'] = 'PepperFry'
item['city'] = 'Mumbai'
item['state'] = 'Maharashtra'
item['country'] = 'India'
item['date'] = str(time.strftime("%d/%m/%Y"))
return item
def get_product_details(self, response):
crumbs = self.get_breadcrumbs(response)
loader = ItemLoader(item=VisionsProduct())
loader.add_value('breadcrumbs', crumbs)
loader.add_value('url', response.url)
if isinstance(crumbs, basestring):
loader.add_value('category', crumbs)
# Ensure we aren't wasting time extracting from an empty page
if extract_helper(response, self.EMPTY_PAGE_CHECK):
for d in self.PRODUCT_DETAILS:
if '_' not in d.name: # Don't load price
loader.add_value(d.name, 'N/A')
else:
productDetails = detailsRunner(self.PRODUCT_DETAILS,
response=response)
if not productDetails['price']:
productDetails['price'] = productDetails['price_gif']
productDetails.pop('price_gif')
# Fix truncated image urls
if productDetails['image']:
productDetails['image'] = add_schema(response.url,
productDetails['image'])
for d in productDetails:
loader.add_value(d, productDetails[d])
yield loader.load_item()
def get_app(self, response):
il = ItemLoader(item=PlayStoreItems(), response=response)
il.add_css('app_id', '.details-wrapper::attr(data-docid)')
il.add_css('name', '.document-title div::text')
il.add_css('category', '.category span::text')
il.add_css(
'category_url', '.category::attr(href)',
Compose(lambda urls: [urljoin(response.url, url) for url in urls]))
il.add_css('price', '.details-actions .price span::text')
il.add_css('offers_in_app_purchases', '.inapp-msg::text')
il.add_css('stars_count', '.stars-count::text')
il.add_css('video', '.details-trailer > span::attr(data-video-url)')
il.add_css('screenshots', '.screenshot::attr(src)')
il.add_xpath(
'description',
'//div[contains(@class, "show-more-content")]/div//text()')
il.add_css('update_date', '[itemprop="datePublished"]::text')
il.add_css('file_size', '[itemprop="fileSize"]::text')
il.add_css('installs', '[itemprop="numDownloads"]::text')
il.add_css('current_version', '[itemprop="softwareVersion"]::text')
il.add_css('requires_android', '[itemprop="operatingSystems"]::text')
il.add_css('offered_by', '[itemprop="author"] > a span::text')
il.add_css(
'offered_by_url', '[itemprop="author"] > a::attr(href)',
Compose(lambda urls: [urljoin(response.url, url) for url in urls]))
yield il.load_item()
def parse(self, response):
l = ItemLoader(item=JianshuArticleItem(), response=response)
l.add_xpath(
'content',
'//div[@class="article"]/div[@class="show-content"]/p/text()')
l.add_value('url', response.url)
return l.load_item()
def parse_item(self, response):
#l=用ItemLoader载入MeizituItem()
re = []
l = ItemLoader(item=MeizituItem(), response=response)
#名字
l.add_xpath('name', '//h2/a/text()')
#标签
l.add_xpath('tags', "//div[@id='maincontent']/div[@class='postmeta clearfix']/div[@class='metaRight']/p")
#图片连接
l.add_xpath('image_urls', "//div[@id='picture']/p/img/@src", Identity())
#url
l.add_value('url', response.url)
re.append(l.load_item())
print re
#return re
return l.load_item()
def create_tag (self, feed_entry_id, tag, comment = ''):
# TODO: create tags
l = ItemLoader (FeedEntryTag())
l.add_value ('feed_entry_id', feed_entry_id)
l.add_value ('tag', tag)
l.add_value ('comment', comment)
return l.load_item()
def parse_item(self, response):
sel = Selector(response)
# collect xpaths of each player (row in table)
rows = sel.xpath('/html//div[@class="table-container"]/table/tbody/tr')
for row in rows:
loader = ItemLoader(GoalSTItem(), selector=row)
loader.default_input_processor = MapCompose()
loader.default_output_processor = Join()
# get unique NHL ID number from player's page URL
num = row.xpath('td[2]/a/@href').extract()
sNum = num[0][-7:]
loader.add_value('nhl_num', sNum)
# add season data
loader.add_value('season', str(self.year))
# collect additional stats
loader.add_xpath('es_shots_against', './/td[6]/text()')
loader.add_xpath('es_goals_against', './/td[7]/text()')
loader.add_xpath('es_saves', './/td[8]/text()')
loader.add_xpath('es_save_pct', './/td[9]/text()')
loader.add_xpath('pp_shots_against', './/td[10]/text()')
loader.add_xpath('pp_goals_against', './/td[11]/text()')
loader.add_xpath('pp_saves', './/td[12]/text()')
loader.add_xpath('pp_save_pct', './/td[13]/text()')
loader.add_xpath('sh_shots_against', './/td[14]/text()')
loader.add_xpath('sh_goals_against', './/td[15]/text()')
loader.add_xpath('sh_saves', './/td[16]/text()')
loader.add_xpath('sh_save_pct', './/td[17]/text()')
# feed item to pipeline
yield loader.load_item()
def parse_news(self, response):
item = ItemLoader(item=NewsItem(), response=response)
item.add_value('url', response.url)
item.add_value('title', response.xpath("//h1[@class='single-post__title']/text()").extract()[0])
item.add_value('content', response.xpath("//section[@class='article']/p/text()").extract())
return item.load_item()
def parse(self, response):
l = ItemLoader(item=MyItem(), response=response)
l.add_xpath(
"title",
"""//div[@class="carousel"]/div[@class="songlist-slides slide-page"]/ul[@class="list-songlist slide-item"]/li[@class="songlist-item"]/a[@class="lnk-songlist"]/@title""",
)
return l.load_item()
def parse_product(self, response):
p = ItemLoader(item=Product(), response=response)
p.add_css('nome', 'h1.livedata::text')
p.add_value('url', response.url)
p.add_css('descricaoLonga', '.desc-info')
p.add_css('image',
'div.container-product-image a.image-link > img',
re='src=[\"|\'](?P<src>[^\"\']+)[\"|\']')
p.add_css('categorias', 'span[itemprop=title]::text')
yield p.load_item()
#executar no mongo
#db.produto.remove({'categorias.0': {$exists: false}})
#db.produto.remove({'categorias.0': {$nin: [' Games', ' Livros', ' DVDs e Blu-ray']}})
#deleta produtos duplicados
#var duplicates = [];
#db.produto_novo.aggregate([
#{"$group" : { "_id": "$nome", "count": { "$sum": 1 }, "dups": { "$addToSet": "$_id" }, }},
#{"$match": {"_id" :{ "$ne" : null } , "count" : {"$gt": 1} } }]
#,{allowDiskUse: true},{cursor:{}}
#).result.forEach(function(doc) {
#doc.dups.shift();
#doc.dups.forEach( function(dupId){
#duplicates.push(dupId);
#}
#)
#})
#printjson(duplicates);
#db.produto_novo.remove({_id:{$in:duplicates}})
def parse_parts2(self, response):
log.msg("\tparse_parts time: %s" % int(time.time()), level=log.DEBUG)
ua = response.request.headers['User-Agent']
log.msg("\tua: %s" % ua, level=log.DEBUG)
for part in response.css('table.parts > tbody > tr'):
il = ItemLoader(item=CarPart(), selector=part)
il.add_xpath('shop_city', "td[@class='shop']/a/text()")
il.add_xpath('shop_name', "td[@class='shop']/a/strong/text()")
shop_url = il.get_xpath("td[@class='shop']/a/@href", TakeFirst())
photo_url = il.get_xpath("td[@class='photo']/a/@href", TakeFirst())
il.add_value('shop_url', urljoin(self.main_url, shop_url))
il.add_value('ext_link', urljoin(self.main_url, photo_url))
il.add_xpath('info', "td[@class='info']//text()")
il.add_xpath('price', "td[@class='price']//text()")
il.add_value('brand', response.meta.get('brand'))
il.add_value('model', response.meta.get('model'))
il.add_value('car_part', response.meta.get('car_part'))
il.add_value('category', response.meta.get('category'))
item = il.load_item()
if item.is_valid():
yield item
def parse_item(self, response):
request_again = self.error_handler(response)
if request_again:
return request_again
il = ItemLoader(item=UKBusinessItem(), response=response)
# From the OG section at the top
il.add_xpath('name', '//meta[@property="og:title"]/@content')
il.add_xpath('url', '//meta[@property="og:url"]/@content')
il.add_xpath('latitude', '//meta[@property="og:latitude"]/@content')
il.add_xpath('longitude', '//meta[@property="og:longitude"]/@content')
il.add_xpath('address', '//meta[@property="og:street-address"]/@content')
il.add_xpath('locality', '//meta[@property="og:locality"]/@content')
il.add_xpath('postal_code', '//meta[@property="og:postal-code"]/@content')
il.add_xpath('country', '//meta[@property="og:country-name"]/@content')
# XPaths below are from the display
il.add_xpath('name', '//span[@class="busname"]/text()')
# No OG for this
il.add_xpath('phone_number', '//span[@class="bustel"]/text()')
il.add_xpath('website', '//a[@id="linkWebsite"]/@href')
il.add_xpath('address', '//span[@data-yext="address.address"]/text()')
il.add_xpath('locality', '//span[@itemprop="addressLocality"]/text()')
il.add_xpath('postal_code', '//span[@itemprop="postalCode"]/text()')
# Unicoded so it can share an input processor with the rest
il.add_value('url', unicode(response.url))
return il.load_item()
def parse_course_item(self, response):
url_obj = urlparse(response.url)
l = ItemLoader(item=CourseItem(), response=response)
l.default_input_processor = MapCompose(unicode.strip)
l.default_output_processor = TakeFirst()
l.add_xpath('code', "/html/head/meta[@name='DC.Subject.ProgramCode']/@content")
l.add_xpath('name', "/html/head/meta[@name='DC.Subject.Description.Short']/@content")
l.add_xpath('career', "/html/head/meta[@name='DC.Subject.Level']/@content")
l.year_in = Identity()
l.add_value('year', ppath.basename(ppath.dirname(url_obj.path)))
l.add_value('src_url', unicode(response.url))
l.add_xpath('uoc', "/html/head/meta[@name='DC.Subject.UOC']/@content")
l.gened_in = MapCompose(unicode.strip, lambda s: s == 'Y')
l.add_xpath('gened', "/html/head/meta[@name='DC.Subject.GenED']/@content")
l.add_xpath('faculty', "/html/head/meta[@name='DC.Subject.Faculty']/@content")
l.add_xpath('school', ( "//div[@class='column content-col']/div[@class='internalContentWrapper']"
"/div[@class='summary']/p[strong[text()[contains(.,'School')]]]/a/text()"))
l.add_xpath('campus', ( "//div[@class='column content-col']/div[@class='internalContentWrapper']"
"/div[@class='summary']/p[strong[text()[contains(.,'Campus')]]]/text()"))
l.add_xpath('prereqs_str', ( "//div[@class='column content-col']/div[@class='internalContentWrapper']"
"/div[@class='summary']/p[text()[contains(.,'Prerequisite:')]]/text()"),
re=r'Prerequisite:\s(.+)')
l.add_xpath('eftsl', ( "//div[@class='column content-col']/div[@class='internalContentWrapper']"
"/div[@class='summary']/p[strong[text()[contains(.,'EFTSL')]]]/text()"))
l.add_xpath('description_markup', ( "//div[@class='column content-col']/div[@class='internalContentWrapper']"
"/h2[text()='Description']/following-sibling::div"))
course_item = l.load_item()
yield course_item
yield Request(url=response.xpath(("//div[@class='column content-col']/div[@class='internalContentWrapper']"
"/div[@class='summary']//a[text()[contains(.,'Timetable')]]/@href")).extract()[0],
callback=self.parse_class_item,
meta=dict(course_identifier={k: course_item.get(k, None) for k in ('code', 'career', 'year', )}))
def parse_item(self, response):
"""Fields not populated by this method:
email, mobile_number, latitude, longitude.
"""
il = ItemLoader(item=UKBusinessItem(), response=response)
il.add_value("url", unicode(response.url))
il.add_xpath("name", './/h1[@class="title"]/a/text() | .//h1[@class="title"]/text()')
address_text = response.xpath(".//address/text()[normalize-space()]").extract()
address_text = [part.strip().rstrip(",") for part in address_text]
address = ", ".join(address_text)
il.add_value("address", address)
il.add_xpath("postal_code", './/h3[@class="postcode"]/text()')
il.add_xpath(
"website",
'.//div[@class="contact-info"]//strong/a/@href |' './/div[@class="contact-info"]/ul/strong/span/text()',
)
il.add_xpath("category", './/ul[contains(@class, "breadcrumb")]/li[last()]/a/text()')
il.add_xpath("linkedin", './/ul[contains(@class, "social")]/li[@class="linkedIn"]/a/@href')
il.add_xpath("description", './/div[@class="about-text"]/p/text()')
phones_sp = response.xpath('.//div[@class="contact-info"]//li/span')
for span in phones_sp:
text = cond_set_value(span.xpath("text()[normalize-space()]").extract(), "")
if "T:" in text:
phone_number = cond_set_value(span.xpath(".//div/text()").extract())
il.add_value("phone_number", phone_number)
if "F:" in text:
fax_number = cond_set_value(span.xpath(".//div/text()").extract())
il.add_value("fax_number", fax_number)
return il.load_item()
def parsePage(self, response):
rentHouse = ItemLoader(item = RentItem(), response = response)
rentHouse.add_value('id', self.name + '-' +
response.url.split('/')[-1].split('.')[0])
rentHouse.add_value('link', response.url)
rentHouse.add_xpath('title', "//dl[@class = 'title']/dt/p/text()")
return rentHouse.load_item()
def make_bot_task_error(self, task_id, code, message=''):
t = ItemLoader(BotTaskError())
t.message_in = lambda slist: [s[:1023] for s in slist]
t.add_value('task_id', task_id)
t.add_value('bot', self.name)
t.add_value('code', code)
t.add_value('message', message)
return t.load_item()
def parse_item(self, response):
l = ItemLoader(item=CrawlpictureItem(), response=response)
l.add_xpath('name', '//h2/a/text()')
l.add_css('tags', 'div.metaRight p::text')
#l.add_xpath('image_urls','//div[@id="picture"]/p/img/@src' or '//img[@class="scrollLoading"]/@src',Identity())
l.add_css('image_urls', 'div.postContent img::attr(src)', Identity())
l.add_value('url', response.url)
return l.load_item()
def parse_item(self, response):
for e in response.xpath('//table[@id="basic"]/tbody/tr'):
l = ItemLoader(ProxyHunterItem(), selector=e)
l.add_xpath('ip', 'td[2]/a/text()')
l.add_xpath('port', 'td[3]/text()')
l.add_xpath('prot', 'td[4]/a/text()')
yield l.load_item()
def parse_item(self, response):
l = ItemLoader(item=MeizituItem(), response=response)
# l.add_xpath('name', '//div[@class="postContent"]/div[@id="picture"]/p/a/text()')
# l.add_xpath('tags', '//div[@class="postContent"]')
l.add_xpath('img_url', '//div[@class="text"]/p/br/img/@src', Identity())
l.add_value('url', response.url)
return l.load_item()
def parse_item(self, response):
l = ItemLoader(item=MeizituItem(), response=response)
l.add_xpath('name', '//h2/a/text()')
l.add_xpath('tags', "//div[@id='maincontent']/div[@class='postmeta clearfix']/div[@class='metaRight']/p")
l.add_xpath('image_urls', "//div[@id='picture']/p/img/@src", Identity())
l.add_value('url', response.url)
return l.load_item()
def parse(self, response):
item_list = []
for a in response.css(".menu_box .menu_main h2"):
l = ItemLoader(item=CategoryItem(), response=response)
# l.add_css('category', ".menu_box .menu_main h2")
l.add_value("category", a.extract(), self.get_text)
item_list.append(l.load_item())
return item_list
def parse2(self, response):
item = json.loads(response.body_as_unicode())
for i in range(len(item['list'])):
data_tmp = item['list'][i]
loader = ItemLoader(item=XqtestItem())
loader.add_value('title', data_tmp['data'])
org = loader.load_item()
yield org
def test_load_item_using_default_loader(self):
i = TestItem()
i['summary'] = u'lala'
il = ItemLoader(item=i)
il.add_value('name', u'marta')
item = il.load_item()
assert item is i
self.assertEqual(item['summary'], u'lala')
self.assertEqual(item['name'], [u'marta'])
def parse(self, response):
sel = Selector(response)
articulos = sel.xpath('/html/body/div[2]/div/div/div/div[1]/div[3]/div')
for i, elem in enumerate(articulos):
item = ItemLoader(Articulos(), elem)
item.add_xpath('title', './/h3/text()')
item.add_value('id', i)
yield item.load_item()
