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_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 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_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 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_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_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 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(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):
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 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 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(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_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_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_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=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):
self.logger.info("parse_item url %s" % response.url)
l = ItemLoader(item=ImgDownloadItem(), response=response)
l.add_xpath('name', '//h1[@class="article-title"]/a/text()')
# l.add_xpath('tags', "//div[@id='maincontent']/div[@class='postmeta clearfix']/div[@class='metaRight']/p")
l.add_xpath('image_urls', "//article[@class='article-content']/p/img/@src", Identity())
l.add_value('url', response.url)
return l.load_item()
def parse_item(self, response):
l = ItemLoader(item=CoserItem(), response=response)
l.add_xpath('name', "//h1[@class='js-post-title']/text()")
l.add_xpath('info', "//div[@class='post__info']/div[@class='post__type post__info-group']/span/text()")
urls = l.get_xpath('//img[@class="detail_std detail_clickable"]/@src')
urls = [url.replace('/w650', '') for url in urls]
l.add_value('image_urls', urls)
l.add_value('url', response.url)
return l.load_item()
def parse_final(self, response):
# sel = Selector(response)
# item = MeiziScrapyItem()
# item["image_name"] = sel.xpath('/html/body/div[2]/div[1]/h2/text()').extract()[0]
# return item
l = ItemLoader(item=MeiziScrapyItem(), response=response)
l.add_xpath('image_name', '/html/body/div[2]/div[1]/h2/text()')
l.add_xpath('image_url', '//*[@id="content"]/a/img/@src')
return l.load_item()
def parse_template(self, response):
"""
Callback used by Scrapy to process downloaded responses
//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[4]/table/tbody/tr[10]/td[2]
"""
response_body = response.body_as_unicode()
# Checking if coffee beans are present in the source, since it shifts down the divs
coffee = True if 'cups of coffee' in response_body else False
prop_xpath = '//div[@class="info_wrapper"]//tr[td[@class="key"]/strong/text() = "{}:"]/td[@class="value"]/text()'
substr_xpath = 'substring-after(normalize-space({}), "{}")'
item_fields = {
'item_hash':
'//*[@id="offer_sku"]/text()',
'title':
'//*[@id="thing_name"]/text()',
'thumbnail':
'//*[@id="thing_image"]/@src',
'description':
'//*[@id="description"]',
'creator':
'//*[@id="product_manufacturer"]/text()',
'when':
prop_xpath.format('Released'),
'bootstrap_version':
substr_xpath.format(prop_xpath.format('Bootstrap'),
'Compatible with '),
'cost_single':
substr_xpath.format(
'//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[{}]/div[1]//span/text()'
.format(3 if coffee else 2), '$'),
'cost_multiple':
substr_xpath.format(
'//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[{}]/div[2]//span/text()'
.format(3 if coffee else 2), '$'),
'cost_extended':
substr_xpath.format(
'//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[{}]/div[3]//span/text()'
.format(3 if coffee else 2), '$'),
'purchases':
'//div[@class="purchases"]/span[@class="count"]/text()',
}
selector = Selector(response)
loader = ItemLoader(WrapBootstrapTemplate(), selector=selector)
# define processors
loader.default_input_processor = MapCompose(unicode.strip)
loader.default_output_processor = Join()
# iterate over fields and add xpaths to the loader
for field, xpath in item_fields.iteritems():
loader.add_xpath(field, xpath)
yield loader.load_item()
def parse(self, response):
l = ItemLoader(item=CoserItem(), response=response)
l.add_xpath('name', "//h1[@class='js-post-title']/text()")
l.add_xpath('info', "//div[@class='post__info']/div[@class='post__type post__info-group']/span/text()")
urls = l.get_xpath('//img[@class="detail_std detail_clickable"]/@src')
urls = [url.replace('/w650', '') for url in urls]
l.add_value('image_urls', urls)
l.add_value('url', response.url)
return l.load_item()
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()
def parse(self, response):
for e in response.xpath(
'//table[@id="tbl_proxy_list"]//tr[count(td)=6]'):
l = ItemLoader(ProxyHunterItem(), selector=e)
l.add_value('prot', 'http')
l.add_xpath('ip', 'td[1]', TakeFirst(), remove_tags, unicode.strip)
l.add_xpath('port', 'td[2]', TakeFirst(), remove_tags,
unicode.strip)
yield l.load_item()
def parse_item2(self, response):
l = ItemLoader(item=DmozItem(), response=response)
l.add_xpath(
'type',
'//div[@class="location ask_main_location"]/span[@class="fl"]/a[last()]/text()'
)
l.add_xpath('type', '//div[@class="question"]/h2/text()')
l.add_xpath('answer', '//div[@class="anwser"]/h2/text()')
l.add_value('answer', '牛逼')
yield l.load_item()
def parse(self, response):
sel = Selector(response)
preguntas = sel.xpath(
'//div[@id="question-mini-list"]/div') #es una lista
for i, elem in enumerate(preguntas):
item = ItemLoader(Pregunta(), elem) #elem tiene el xpath
item.add_xpath('pregunta', './/h3/a/text()')
item.add_value('id', i)
yield item.load_item()
def Loader_index(self, item_selector):
l = ItemLoader(item={}, selector=item_selector)
conver_img = l.get_xpath('.//*[@class="lz_img"]/img/@src')
l.add_xpath('title', './/*[@class="k_list-lb-2"]/div[1]/a[1]/text()')
l.add_xpath('url', './/*[@class="k_list-lb-2"]/div[1]/a/@href')
l.add_value('preview', conver_img)
l.add_css('date', '#k_list-lb-2-f::text', re=r'(\d{4}-\d{2}-\d{2})')
l.add_value('image_urls', conver_img)
return l.load_item()
def parse_item(self, response):
t_cl_item=response.xpath('//table[@class="item"]')
for t in t_cl_item:
l = ItemLoader(item=SSDItem(), selector=t)
l.add_xpath('name', './/td[@class="l"]/a/text()')
l.add_xpath('usual_price', './/div[@class="price"]/text()')
l.add_xpath('club_price', './/div[@class="price club_price"]/text()')
l.add_xpath('img_url', './/div[@class="photo"]/span/img/@src')
l.add_xpath('url', './/td[@class="l"]/a/@href')
yield l.load_item()
def parse(self, response):
sel = Selector(response)
lugares = sel.xpath('//div[@id=hotellist_inner]/div')
#vamos a iterar sobre todas las preguntas
for i, elem in enumerate(lugares):
item = ItemLoader(Lugar(), elem)
item.add_xpath('lugar', './/h3/a/span/text()')
item.add_value('id', i)
yield item.load_item()
def parse(self, response):
direction = response.xpath('//li[@class="btn-schedules-active"][1]/text()').extract()
day = response.xpath('//li[@class="btn-schedules-active"][2]/text()').extract()
for sel in response.xpath('//tr'):
loader = ItemLoader(item = RtdItem(), selector = sel)
loader.default_output_processor = TakeFirst()
loader.add_value('day', day)
loader.add_value('direction', direction)
loader.add_xpath('route', 'th/a/text()')
loader.add_xpath('depart_time', 'td[1]/text()')
loader.add_xpath('arrive_time', 'td[2]/text()')
yield loader.load_item()
def parse_detail(self, response):
print("response.url===", response.url)
#具体值
url = response.url
#使用ItemLoader类
item = ItemLoader(item=Meizitu2Item(), response=response)
item.add_xpath("tilte", "//h2/a/text()")
item.add_xpath("image_urls", '//div[@id="picture"]//img/@src')
#添加值的方式
item.add_value("url", url)
return item.load_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_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')
# instantiate parsing variables
MONTHS = {'Jan': '01',
'Feb': '02',
'Mar': '03',
'Apr': '04',
'May': '05',
'Jun': '06',
'Jul': '07',
'Aug': '08',
'Sep': '09',
'Oct': '10',
'Nov': '11',
'Dec': '12'}
# loop through players
for row in rows:
loader = ItemLoader(GoalBioItem(), 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)
# parse the name
name = row.xpath('td[2]/a/text()').extract()
sName = name[0].split(' ', 1)
loader.add_value('first_name', sName[0])
loader.add_value('last_name', sName[1])
# collect birth year
bDate = row.xpath('td[4]/text()').extract()[0]
bYear = "19" + bDate[-2:]
bMonth = MONTHS[bDate[:3]]
bDay = bDate[4:6]
loader.add_value('birthday', "%s-%s-%s" % (bYear, bMonth, bDay))
# add other data points
loader.add_value('position', 'G')
loader.add_xpath('draft_year', './/td[12]/text()')
loader.add_xpath('draft_position', './/td[14]/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(SkatPMItem(), 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("team_goals_for", ".//td[14]/text()")
loader.add_xpath("team_pp_goals_for", ".//td[15]/text()")
loader.add_xpath("team_goals_against", ".//td[16]/text()")
loader.add_xpath("team_pp_goals_against", ".//td[17]/text()")
# feed item to pipeline
yield loader.load_item()
def parse_item(self, response):
# l=用ItemLoader载入MeizituItem()
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)
return l.load_item()
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):
l = ItemLoader(item=CoserItem(), response=response)
l.add_xpath('name', "//div[@class='mb10 dib']/a/text()")
l.add_xpath('info', "//div/p[@class='mb20']/text()")
#l.add_xpath('image_urls',"//div[@class='content-img-wrap-inner']/img[@src]")
l.add_value('url', response.url)
# //div[@class='content-img-wrap']//img/@src
# 抓不到，正则还是牛逼
urls = l.selector.re(r'src="(.+?.jpg)/w650')
# urls = l.get_xpath("//div[@class='content-img-wrap']//img/@src")
# urls = [url.replace('/w650', '') for url in urls]
l.add_value('image_urls', urls)
# l.add_xpath('image_urls',"//div/p[@class='mb20']/text()")
yield l.load_item()
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(self, response):
items = []
for everyday in response.xpath('//ul/li/strong/a'):
loader = ItemLoader(ProductItem(), everyday)
loader.default_input_processor = MapCompose(unicode.strip)
loader.default_output_processor = Join()
loader.add_xpath('name', 'text()')
loader.add_xpath('price', '@href')
loader.add_xpath('stock', '@mon')
loader.add_value('last_updated', 'today') # you can also use literal values
item = self.to_utf8(loader.load_item(), *['name', 'price', 'stock', 'last_updated'])
self.log(item['name'], log.INFO)
items.append(item)
return items
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 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(self, response): selector = response.selector.xpath(view) #iterate over titles for page in selector.select(self.view): loader = ItemLoader(AmazonItem(), page) #define processors loader.default_input_processor = MapCompose(unicode.strip) loader.default_output_processor = Join() #iterate over fields and add xpaths to the loader for field, xpath in self.iem_fields.iteritems(): loader.add_xpath(field, xpath) return loader.load_item()
def parse_item(self, response):
# sel = Selector(response)
#
# name = sel.xpath("//div[@class='main-image']/p/a/img/@alt").extract()[0]
# print(name)
l = ItemLoader(item=MzituItem(), response=response)
l.add_xpath('image_urls', "//div[@class='main-image']/p/a/img/@src",
Identity())
l.add_xpath('name', "//div[@class='main-image']/p/a/img/@alt",
Identity())
# l.add_value('name', name)
return l.load_item()
def parse_detail(self, response):
l = ItemLoader(response.meta['item'], response)
# l.add_xpath('fanhao','//span[@class="list_text"]/em/b/a/text()')
l.add_xpath('image_name', '//span[@class="list_text"]/em/b/a/text()')
photo = response.xpath(
'//span[@class="list_img"]/a/img/@data-original').extract()
# item = response.meta['item']
# item['fanhao'] = selector.xpath('//span[@class="list_text"]/em/b/a/text()').extract()
# photo = selector.xpath('//span[@class="list_img"]/a/img/@data-original').extract()
img = []
for p in photo:
img.append('http://www.nh87.cn' + p)
l.add_value('image_urls', img)
# 返回item
return l.load_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 parse(self, response):
items = ItemLoader(item=LiveItem(), response=response)
for content in response.xpath('//*[@id="sortdetail-container"]/li/a'):
i = ItemLoader(item=LiveItem(), selector=content)
#标题
i.add_xpath('title', 'div[2]/span[1]/text()')
#用户名
i.add_xpath('username', 'div[2]/span[2]/@title')
#热度
i.add_xpath('num', 'div[2]/span[4]/i/text()')
#图片的地址
i.add_xpath('pic_addr', 'div[1]/img/@data-original')
#直播间的相对地址
i.add_xpath('addr', '@href')
#直播平台
i.add_value('platform', 'panda')
yield i.load_item()
def parse_item(self, response):
"""Fields not populated by this method:
locality, country, mobile_number, category.
"""
il = ItemLoader(item=UKBusinessItem(), response=response)
il.add_value('url', unicode(response.url))
il.add_xpath('name', '//div[@class="company_details"]/h3/text()')
address_p =response.xpath(
'//div[@class="fl company_details_inset1"]/p')
address_first_part = address_p.xpath(
'text()[normalize-space()]').extract()
address_first_part = [ad.strip().replace('\\n\\r', '') for ad
in address_first_part]
address_second_part = cond_set_value(
address_p.xpath(
'.//a[contains(@href, "town")]/text()').extract(), '')
address = ', '.join(address_first_part) + ', '+ address_second_part
if address == ', ':
return
il.add_value('address', address)
postal_code = ''.join(
address_p.xpath(
'.//a[not(contains(@href, "town"))]/text()').extract())
il.add_value('postal_code', postal_code)
il.add_xpath('website', './/div[@class="company-website "]/a/text()')
il.add_xpath('fax_number', './/span[@class="company-fax"]/text()')
il.add_xpath('phone_number', './/span[@class="company-phno"]/@phone')
mail_a = response.xpath('.//a[@value="Contact us"]/text()').extract()
try:
mail_a.remove('Contact Us')
except:
pass
mail_a = [a.strip() for a in mail_a]
mail = '@'.join(mail_a)
il.add_value('email', mail)
latitude = ''
longitude = ''
map_data = cond_set_value(response.xpath(
'.//p[@class="direction_map"]/iframe/@src').extract())
if map_data:
try:
qs = parse_qs(map_data)
coordinates = qs['ll'][0].split(',')
latitude = coordinates[0]
longitude = coordinates[1]
except:
pass
il.add_value('latitude', latitude)
il.add_value('longitude', longitude)
return il.load_item()
def parse(self, response):
"""
Default callback used by Scrapy to process downloaded responses
Testing contracts:
@url http://odds.500.com/index_jczq_2014-08-29.shtml
"""
selector = Selector(response)
# iterate over matchs
for match in selector.select(self.match_list_xpath):
loader = ItemLoader(Match(), selector=match)
# define processors
loader.default_input_processor = MapCompose(unicode.strip)
loader.default_output_processor = Join()
# iterate over fields and add xpaths to the loader
for field, xpath in self.match_fields.iteritems():
loader.add_xpath(field, xpath)
match_item = loader.load_item()
match_item["game_date"] = self.game_date
match_item["season_id"] = match_item["season_id"].split('-')[-1]
match_item["teama_id"] = match_item["teama_id"].split('-')[-1]
match_item["teamb_id"] = match_item["teamb_id"].split('-')[-1]
if "score" in match_item:
sa, sb = match_item["score"].split(':')
match_item["score_a"] = sa
match_item["score_b"] = sb
match_item["result"] = "win" if sa > sb else "draw" if sa == sb else "lost"
else:
match_item["score_a"] = match_item["score_b"] = -1
match_item["result"] = "none"
yield match_item
#scrap asia odds
#id=454359&ctype=1&start=60&r=1&style=0&guojia=0
for i in xrange(3):
url = self.asia_odds_url % (match_item["match_id"], i * 30)
request = scrapy.Request(url, callback=self.parse_asia_odds)
request.meta['match_item'] = match_item
yield request
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 parse(self,response):
l = ItemLoader(item = googleItem(),response = response)
l.add_xpath('news','//span[contains(@class,"titletext")]/text()')
x = l.load_item()
#print(len(x['date']),len(x['topnews']),len(x['sectionnews']))
nytdict = dict()
datelist = []
datalist = datetime.date.today()
newslist = []
nytdict['date'] = str(datalist)
for t in x['news']:
newslist.append(str(t.encode('ascii','ignore')))
nytdict['news']=newslist
filename = datetime.date.today()
f=open('{}.json'.format(filename),'w')
json.dump(nytdict,f)
return 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 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_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(GoalSOItem(), 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('so_wins', './/td[14]/text()')
loader.add_xpath('so_losses', './/td[15]/text()')
loader.add_xpath('so_shots_against', './/td[16]/text()')
loader.add_xpath('so_goals_against', './/td[17]/text()')
# feed item to pipeline
yield loader.load_item()
def parse_template(self, response):
"""
Callback used by Scrapy to process downloaded responses
//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[4]/table/tbody/tr[10]/td[2]
"""
response_body = response.body_as_unicode()
# Checking if coffee beans are present in the source, since it shifts down the divs
coffee = True if 'cups of coffee' in response_body else False
prop_xpath = '//div[@class="info_wrapper"]//tr[td[@class="key"]/strong/text() = "{}:"]/td[@class="value"]/text()'
substr_xpath = 'substring-after(normalize-space({}), "{}")'
item_fields = {
'item_hash': '//*[@id="offer_sku"]/text()',
'title': '//*[@id="thing_name"]/text()',
'thumbnail': '//*[@id="thing_image"]/@src',
'description': '//*[@id="description"]',
'creator': '//*[@id="product_manufacturer"]/text()',
'when': prop_xpath.format('Released'),
'bootstrap_version': substr_xpath.format(prop_xpath.format('Bootstrap'), 'Compatible with '),
'cost_single': substr_xpath.format('//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[{}]/div[1]//span/text()'.format(3 if coffee else 2), '$'),
'cost_multiple': substr_xpath.format('//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[{}]/div[2]//span/text()'.format(3 if coffee else 2), '$'),
'cost_extended': substr_xpath.format('//*[@id="page_theme"]/div[2]/div/div/div/div[2]/div[{}]/div[3]//span/text()'.format(3 if coffee else 2), '$'),
'purchases': '//div[@class="purchases"]/span[@class="count"]/text()',
}
selector = Selector(response)
loader = ItemLoader(WrapBootstrapTemplate(), selector=selector)
# define processors
loader.default_input_processor = MapCompose(unicode.strip)
loader.default_output_processor = Join()
# iterate over fields and add xpaths to the loader
for field, xpath in item_fields.iteritems():
loader.add_xpath(field, xpath)
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_content(self, response):
bbsItem_loader = ItemLoader(item=BbsItem(), response=response)
url = str(response.url)
bbsItem_loader.add_value("url", url)
bbsItem_loader.add_xpath("forum", self._x_query["forum"])
bbsItem_loader.add_xpath("poster", self._x_query["poster"])
bbsItem_loader.add_xpath("content", self._x_query["page_content"])
return bbsItem_loader.load_item()
