def parse_shop(self, response):
"""
解析提取model数据
:param response:
"""
item = Item()
item_loader = ItemLoader(item=item, selector=response)
for field in self.model_xpath:
item.fields[field] = Field()
if 'model_url' in field:
item_loader.add_value(field, response.url)
else:
item_loader.add_xpath(field, self.model_xpath[field])
item = self.format_item(item_loader.load_item())
yield item
# 拼接用户评论链接
user_url = response.xpath(self.user_url).extract()
for uid in user_url:
brand = uid.split('-')[0]
model = uid.split('-')[1]
yield Request(
self.user_review_url.format(brand=brand, model=model),
self.parse_comment)
def parse_forum(self, response):
"""
解析评论内容
"""
item = Item()
itemloader = ItemLoader(item=item, selector=response)
for field in self.forum_xpath:
item.fields[field] = Field()
if 'main_body' in field:
content = re.compile(r'<html><body><.*?>(.*?)</body></html>',
re.S | re.M)
content = content.findall(response.text)
content = re.sub(r'<script>.*?</script>', '', ''.join(content))
content = re.sub(r'[\r\n]', '', content)
content = re.sub(r'<div .*?>.*?</div>', '', content)
content = re.sub(r'<style .*?>.*?</style>', '', content,
re.S | re.M)
content = re.sub(r'&.*?;', '', content)
content = re.sub(r'<.*?>', '', content, re.M | re.I)
content = re.sub(' ', '', content)
itemloader.add_value(field, content)
elif 'content_url' in field:
itemloader.add_value(field, response.url)
else:
itemloader.add_xpath(field, self.forum_xpath[field])
item = self.format_item(itemloader.load_item())
yield item
def test_add_stats_item_scraped_count_by_item_type(spider):
for _ in range(15):
spider.crawler.signals.send_catch_log_deferred(
signal=signals.item_scraped,
item={"_type": "regular_dict"},
response="",
spider=spider,
)
for _ in range(20):
spider.crawler.signals.send_catch_log_deferred(
signal=signals.item_scraped,
item=Item(),
response="",
spider=spider,
)
for _ in range(25):
spider.crawler.signals.send_catch_log_deferred(
signal=signals.item_scraped,
item=TestItem(),
response="",
spider=spider,
)
stats = spider.crawler.stats.get_stats()
assert stats.get("spidermon_item_scraped_count") == 60
assert stats.get("spidermon_item_scraped_count/dict") == 15
assert stats.get("spidermon_item_scraped_count/Item") == 20
assert stats.get("spidermon_item_scraped_count/TestItem") == 25
def PlayerRow(**fields):
item = Item()
for field in all_fields:
item.fields[field] = Field()
for field, value in fields.iteritems():
item[field] = value
return item
def parse_user(self, response):
"""
根据返回的 response 进行数据解析
:param response: scrapy 框架返回的响应
"""
result = {
'user_name': response.meta['user_name'],
'title': response.meta['title'],
'date': response.meta['date'],
'main_body': response.meta['main_body'],
'content_url': response.meta['content_url'],
'brand': response.meta['brand']
}
for content in response.xpath(self.user_list_xpath):
item = Item()
item_loader = ItemLoader(item=item, selector=content)
for field in self.user_xpath:
item.fields[field] = Field()
if 'user_url' in field:
item_loader.add_value(field, response.url)
else:
item_loader.add_xpath(field, self.user_xpath[field])
result.update(item_loader.load_item())
item = self.format_item(result)
yield item
# 用户评论
user_comment = response.xpath('.//ul/li[@class="Comments"]/a/@href').extract()
if user_comment:
yield Request(self.url + user_comment[0], self.parse_comment)
def parse_item(self, response):
item = Item()
item.fields['url'] = Field()
item.fields['url_md5'] = Field()
item.fields['title'] = Field()
item.fields['pubtime'] = Field()
item.fields['content'] = Field()
item.fields['author'] = Field()
item.fields['site_name'] = Field()
l = ItemLoader(item=item, response=response)
url = response.url
md5 = hashlib.md5()
l.add_value(u'url', url)
md5.update(url)
url_md5 = md5.hexdigest()
l.add_value(u'url_md5', url_md5)
l.add_xpath('title', "//h2/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('pubtime', "//p[@class='time']/span[1]/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('content', "//div[@class='content']//text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('author', "//p[@class='time']/span[last()]/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_value("site_name", u"网易号")
yield l.load_item()
def parse(self, response):
"""
根据返回的 response 进行数据解析
:param response: scrapy 框架返回的响应
"""
keyword = response.meta['keyword']
for complaint in response.xpath(self.list_xpath):
item = Item()
item_loader = ItemLoader(item=item, selector=complaint)
for field in self.content_xpath:
item.fields[field] = Field()
if 'content_url' in field:
item_loader.add_value(field, response.url)
elif 'brand' in field:
item_loader.add_value(field, keyword)
else:
item_loader.add_xpath(field, self.content_xpath[field])
# 用户链接
user_id = complaint.xpath('.//span[@class="Author"]/a/@href').extract()
for uid in user_id:
yield Request(self.url + uid, self.parse_user, meta=dict(item_loader.load_item()))
# 下一页
next_page = response.xpath('//div[@id="PagerBefore"]/a[last()]/@href').extract()
if next_page:
yield Request(self.url + next_page[0], meta={'keyword': keyword}, callback=self.parse)
def parse(self, response):
item = Item()
sel = Selector(response)
fields = json.loads(self.conf.get("fields"))
rules = json.loads(self.conf.get("rules"))
loops = rules.get("rules").get("rules_listxpath")
if fields.get("fields", "") == "":
logging.error(u"内容解析未得到!!!")
yield item
item.fields["url"] = Field()
item.fields["spider_jobid"] = Field()
item["spider_jobid"] = self.spider_jobid
item.fields['word'] = Field()
item['word'] = response.meta.get("word")
# 加载动态库字段建立Field,xpath规则 (方法一)
for loop in sel.xpath("{}".format(loops)):
item['url'] = loop.xpath(u"{}".format(fields.get("fields").get("url").get("xpath"))).extract()
for k in loadMySQL(self.conf.get("spider_name")):
if fields.get("fields").get(k[2]) != None:
item.fields[k[2]] = Field()
if fields.get("fields").get(k[2]).keys()[0] == "xpath":
item[k[2]] = loop.xpath(u"{}".format(fields.get("fields").get(k[2]).get("xpath"))).extract()
elif fields.get("fields").get(k[2]).keys()[0] == "value":
item[k[2]] = u"{}".format(fields.get("fields").get(k[2]).get("value"))
yield item
class PostItem(Item):
title=Field()
content=Field()
url=Field()
user=Item()
fromurl=Field()
postdate=Field()
def describe_request_result_handling():
single_request = Request("http://test.com")
mixed_requests = [ Request("http://test.com"), FormRequest("http://test2.com") ]
complete_mix = mixed_requests + [ Item() ]
def it_can_extract_request_objects():
assert requests_in_parse_result([single_request]) == [single_request]
def it_tolerates_None():
assert requests_in_parse_result(None) == []
assert items_in_parse_result(None) == []
assert count_requests_in_parse_result(None) == 0
assert count_items_in_parse_result(None) == 0
def it_tolerates_single_elements():
assert requests_in_parse_result(single_request) == [single_request]
assert items_in_parse_result(single_request) == []
def it_tolerates_and_sorts_out_items_mixed_in_between():
assert requests_in_parse_result(complete_mix) == mixed_requests
def it_tolerates_different_request_types():
assert requests_in_parse_result(mixed_requests) == mixed_requests
def it_extracts_urls_from_requests():
urls = urls_from_requests(complete_mix)
assert len(urls) == 2
assert "http://test.com" in urls
assert "http://test2.com" in urls
def it_counts_the_requests_and_other_results():
assert count_requests_in_parse_result(complete_mix) == 2
assert count_items_in_parse_result(complete_mix) == 1
def parse(self, response):
item = Item()
fields = json.loads(self.conf.get("fields"))
l = ItemLoader(item, response)
item.fields["url"] = Field()
item.fields["spider_jobid"] = Field()
l.add_value("url", response.url)
l.add_value("spider_jobid", self.spider_jobid)
# 加载动态库字段建立Field,xpath规则 (方法一)
for k in self.keys:
if fields.get("fields", "") == "":
logging.error(u"内容解析未得到!!!")
return l.load_item()
if fields.get("fields").get(k) != None:
item.fields[k] = Field()
if fields.get("fields").get(k).keys()[0] == "xpath":
l.add_xpath(
k,
u"{}".format(fields.get("fields").get(k).get("xpath")),
MapCompose(unicode.strip))
elif fields.get("fields").get(k).keys()[0] == "value":
if fields.get("fields").get(k).get("value") == u"{TODAY}":
l.add_value(k, u"{}".format(datetime.now()))
l.add_value(
k,
u"{}".format(fields.get("fields").get(k).get("value")))
return l.load_item()
def parse_item(self, response):
item = Item()
word = response.meta['word']
fields = json.loads(self.conf.get("fields"))
l = ItemLoader(item, response)
if fields.get("fields", "") == "":
logging.error(u"内容解析未得到!!!")
return l.load_item()
item.fields["url"] = Field()
item.fields["spider_jobid"] = Field()
l.add_value("url", response.url)
l.add_value("spider_jobid", self.spider_jobid)
item.fields['word'] = Field()
l.add_value('word', word)
# 加载动态库字段建立Field,xpath规则 (方法一)
for k in loadMySQL(self.name_spider)['fields'].keys():
if fields.get("fields", "") == "":
logging.error(u"内容解析未得到!!!")
return l.load_item()
if fields.get("fields").get(k) != None:
item.fields[k] = Field()
if fields.get("fields").get(k).keys()[0] == "xpath":
l.add_xpath(
k,
u"{}".format(fields.get("fields").get(k).get("xpath")),
MapCompose(unicode.strip))
elif fields.get("fields").get(k).keys()[0] == "value":
l.add_value(
k,
u"{}".format(fields.get("fields").get(k).get("value")))
return l.load_item()
def parse(self, response):
"""
解析网页获取评论
:param response:响应内容
"""
search = response.meta['keyword']
# 评论页
for complaint in response.xpath(self.list_xpath):
item = Item()
item_loader = ItemLoader(item=item, selector=complaint)
for field in self.content_xpath:
item.fields[field] = Field()
if 'content_url' in field:
item_loader.add_value(field, response.url)
elif 'brand' in field:
item_loader.add_value(field, search)
else:
item_loader.add_xpath(field, self.content_xpath[field])
# 拼接user_url链接
uid = complaint.xpath('.//tr/td[@class="small"]/a/@href').extract()
yield Request(self.urls + uid[0],
self.parse_user,
meta=dict(item_loader.load_item()),
dont_filter=True)
# 内容下一页
next_page = response.xpath(
'//div[@class="pagelinks"]/a[last()]/@href').extract()
for page in next_page:
yield Request(self.urls + page,
self.parse,
meta={'keyword': search})
def parse_item(self, response):
item = Item()
item.fields['url'] = Field()
item.fields['url_md5'] = Field()
item.fields['title'] = Field()
item.fields['pubtime'] = Field()
item.fields['content'] = Field()
item.fields['author'] = Field()
item.fields['site_name'] = Field()
l = ItemLoader(item=item, response=response)
url = response.url
url = url.replace("http://rym.quwenge.com/baidu_tiaozhuan.php?url=",
"")
md5 = hashlib.md5()
l.add_value(u'url', url)
md5.update(url)
url_md5 = md5.hexdigest()
l.add_value(u'url_md5', url_md5)
l.add_xpath('title', "//h1/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('pubtime', "//span[@class='read']/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('content', "//div[@id='content']//text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('author', "//div[@class='name']/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_value("site_name", u"百家号")
yield l.load_item()
def parse_item(self, response):
item = Item()
item.fields['url'] = Field()
item.fields['url_md5'] = Field()
item.fields['title'] = Field()
item.fields['pubtime'] = Field()
item.fields['content'] = Field()
item.fields['author'] = Field()
item.fields['author_url'] = Field()
item.fields['site_name'] = Field()
l = ItemLoader(item=item, response=response)
url = response.url
md5 = hashlib.md5()
l.add_value(u'url', url)
md5.update(url)
url_md5 = md5.hexdigest()
l.add_value(u'url_md5', url_md5)
l.add_xpath('title', "//h1/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath(
'pubtime',
"//p[@class='clearfix']//span/text() | //div[@class='titleLine-gY7DniPB']//span/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('content', "//div[@class='text-3zQ3cZD4']//text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath(
'author',
"//p[@class='clearfix']/a[2]/text() | //div[@class='titleLine-gY7DniPB']/p/a[2]/text()",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_xpath('author_url',
"//div[@class='titleLine-gY7DniPB']/p/a[2]/@href",
MapCompose(unicode.lstrip, unicode.rstrip))
l.add_value("site_name", u"大风号")
yield l.load_item()
def test_mpm_middleware(self):
# create fake response
a = MagicMock()
a.meta = {'key1': 'value1', 'key2': 'value2'}
yield_count = 0
# test all types of results from a spider
# dicts, items, or requests
test_list = [{}, Item(), Request('http://istresearch.com')]
for item in self.mpm.process_spider_output(a, test_list, MagicMock()):
if isinstance(item, Request):
self.assertEquals(a.meta, item.meta)
yield_count += 1
self.assertEquals(yield_count, 3)
# 1 debug for the method, 1 debug for the request
self.assertEquals(self.mpm.logger.debug.call_count, 2)
# test meta unchanged if already exists
r = Request('http://aol.com')
r.meta['key1'] = 'othervalue'
for item in self.mpm.process_spider_output(a, [r], MagicMock()):
# key1 value1 did not pass through, since it was already set
self.assertEquals(item.meta['key1'], 'othervalue')
# key2 was not set, therefor it passed through
self.assertEquals(item.meta['key2'], 'value2')
def test_process_not_trackable(self):
pipeline = ItemTrackerPipeline.from_crawler(self.crawler)
pipeline.storage = mock.Mock()
expected = Item()
found = pipeline.process_item(expected, self.spider)
self.assertEqual(expected, found)
pipeline.storage.assert_not_called()
def second_parse(self, response):
href_list = response.xpath('//ul[@class="wp100"]/li//div/p[@class="fs22"]/a/@href')
print(len(href_list), response.url)
for href in href_list:
item = Item()
next_url = 'http://www.sinyi.com.cn/' + href.extract().split('/', 1)[1].split('&cookieuid=')[0]
item.fields['HouseUrl'] = Field()
item['HouseUrl'] = next_url
yield scrapy.Request(next_url, callback=self.third_parse, meta={'item': item})
def housing_handle(self, response):
item = response.meta.get("item")
item1 = Item()
PropertyAddress = response.xpath(
"//div[@class='detailDesc']/text()").extract_first()
PriceUnit = response.xpath(
"//span[@class='xiaoquUnitPrice']/text()").extract_first()
detail_community = response.xpath("//div[@class='xiaoquInfo']")
BuildedTime = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][1]/span[@class='xiaoquInfoContent']/text()"
).extract_first()
BuildingType = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][2]/span[@class='xiaoquInfoContent']/text()"
).extract_first()
PropertyFee = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][3]/span[@class='xiaoquInfoContent']/text()"
).extract_first()
PropertyCompany = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][4]/span[@class='xiaoquInfoContent']/text()"
).extract_first()
Developers = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][5]/span[@class='xiaoquInfoContent']/text()"
).extract_first()
TotalBuilding = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][6]/span[@class='xiaoquInfoContent']/text()"
).extract_first()
TotalHouseholds = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][7]/span[@class='xiaoquInfoContent']/text()"
).extract_first()
NearbyStores = detail_community.xpath(
"./div[@class='xiaoquInfoItem'][8]/span[@class='xiaoquInfoContent']"
).xpath("string(.)").extract_first()
item1.fields["PropertyAddress"] = Field()
item1["PropertyAddress"] = PropertyAddress
item1.fields["PriceUnit"] = Field()
item1["PriceUnit"] = PriceUnit
item1.fields["BuildedTime"] = Field()
item1["BuildedTime"] = BuildedTime
item1.fields["BuildingType"] = Field()
item1["BuildingType"] = BuildingType
item1.fields["PropertyFee"] = Field()
item1["PropertyFee"] = PropertyFee
item1.fields["PropertyCompany"] = Field()
item1["PropertyCompany"] = PropertyCompany
item1.fields["Developers"] = Field()
item1["Developers"] = Developers
item1.fields["TotalBuilding"] = Field()
item1["TotalBuilding"] = TotalBuilding
item1.fields["TotalHouseholds"] = Field()
item1["TotalHouseholds"] = TotalHouseholds
item1.fields["NearbyStores"] = Field()
item1["NearbyStores"] = NearbyStores
item1.update(item)
yield item1
def setUp(self):
self.url = 'http://localhost'
self.kwargs = {'url': self.url, 'dont_filter': True}
self.crawler = MagicMock()
self.spider = MetaSpider.from_crawler(self.crawler)
self.crawler.spider = self.spider
self.crawl_manager = self.create_crawl_manager()
self.crawl_manager.crawler = self.crawler
self.item = Item()
self.response = Response('http://localhost')
self.another_spider = MetaSpider.from_crawler(self.crawler)
def describe_item_result_handling():
single_item = Item()
mixed_items = [ Item(), MyItem() ]
complete_mix = mixed_items + [ Request("http://test.com") ]
def it_can_extract_item_objects():
assert items_in_parse_result([single_item]) == [single_item]
def it_tolerates_single_elements():
assert items_in_parse_result(single_item) == [single_item]
assert requests_in_parse_result(single_item) == []
def it_tolerates_items_mixed_in_between():
assert items_in_parse_result(complete_mix) == mixed_items
def it_tolerates_different_item_types():
assert items_in_parse_result(mixed_items) == mixed_items
def it_counts_the_items_and_other_results():
assert count_requests_in_parse_result(complete_mix) == 1
assert count_items_in_parse_result(complete_mix) == 2
def handle_1(self, response):
item1 = response.meta.get("item1")
req = response.meta.get("req")
fragments = re.findall("url \+= '(.*?)'", response.text, re.S)
detail_url = ''
for j in fragments:
detail_url += j
item2 = Item()
item2.fields["NewUrl"] = Field()
item2["NewUrl"] = detail_url
item2.update(item1)
yield scrapy.Request(url=detail_url, callback=self.parse, meta={"req": req, "item2": item2, "last_page": True})
def handle_1(self, response):
item = response.meta.get("item")
item1 = Item()
PropertyAddress = response.xpath(
"//div[contains(@class,'rent-top')]/a/text()").extract_first()
PriceUnit = response.xpath(
"//div[contains(@class,'junjia')]/span/text()").extract_first()
ls_detail = response.xpath(
"//div[@class='xqfangs detail_bor_bottom']/ul[@class='clear']/li/text()"
).extract()
BuildedTime = None
BuildingType = None
for detail in ls_detail:
if "年" in detail:
BuildedTime = detail
else:
BuildingType = detail
PropertyCompany = response.xpath(
"//ul/li[@class='wuyes']/em/text()").extract_first()
Developers = response.xpath(
"//ul/li[@class='kaifas']/em/text()").extract_first()
TotalBuilding = response.xpath(
"//div[@class='xqsaleinfo']/ul/li[1]/span/text()").extract_first()
TotalHouseholds = response.xpath(
"//div[@class='xqsaleinfo']/ul/li[2]/span/text()").extract_first()
NearbyStores = response.xpath(
"//div[@class='xqsaleinfo']/ul/li[6]/span/text()").extract_first()
AroundTraffic = response.xpath(
"//div[@class='xqsaleinfo']/ul/li[5]/span/text()").extract_first()
item1.fields["PropertyAddress"] = Field()
item1["PropertyAddress"] = PropertyAddress
item1.fields["PriceUnit"] = Field()
item1["PriceUnit"] = PriceUnit
item1.fields["BuildedTime"] = Field()
item1["BuildedTime"] = BuildedTime
item1.fields["BuildingType"] = Field()
item1["BuildingType"] = BuildingType
item1.fields["PropertyCompany"] = Field()
item1["PropertyCompany"] = PropertyCompany
item1.fields["Developers"] = Field()
item1["Developers"] = Developers
item1.fields["TotalBuilding"] = Field()
item1["TotalBuilding"] = TotalBuilding
item1.fields["TotalHouseholds"] = Field()
item1["TotalHouseholds"] = TotalHouseholds
item1.fields["NearbyStores"] = Field()
item1["NearbyStores"] = NearbyStores
item1.fields["AroundTraffic"] = Field()
item1["AroundTraffic"] = AroundTraffic
item1.update(item)
yield item1
def start_requests(self):
for kw in self.kw_list:
start_page = 1
for page in range(start_page, 11):
item = Item()
item.fields["SearchWord"] = Field()
item.fields["Page"] = Field()
item["SearchWord"] = kw
item["Page"] = page
start_url = 'https://weixin.sogou.com/weixin?query={}&type=2&page={}&ie=utf8'.format(
parse.quote(kw), str(page))
yield scrapy.Request(url=start_url, callback=self.parse,
meta={"start_url": start_url, "item": item})
def page_handle(self, response):
region = response.meta.get("region")
plate = response.meta.get("plate")
plate_url = response.meta.get("plate_url")
try:
housing_num_flag = response.xpath(
"//h2[contains(@class,'total')]/span/text()").extract_first(
).strip()
except:
housing_num_flag = '0'
if housing_num_flag != "0":
housing_list = response.xpath(
"//div[@class='leftContent']/ul[@class='sellListContent']/li[contains(@class, 'clear')]/div[@class='info clear']"
)
for housing in housing_list:
# 列表页面字段获取
item = Item()
item.fields["AreaName"] = Field()
item["AreaName"] = region
item.fields["PlateName"] = Field()
item["PlateName"] = plate
housing_url = housing.xpath(
"./div[@class='title']/a/@href").extract_first()
HouseDesc = housing.xpath(
"./div[@class='title']/a/text()").extract_first()
item.fields["HouseDesc"] = Field()
item["HouseDesc"] = HouseDesc
item.fields["HouseUrl"] = Field()
item["HouseUrl"] = housing_url
yield scrapy.Request(url=housing_url,
callback=self.housing_handle,
meta={"item": deepcopy(item)},
headers=self.get_headers())
page_dict_handle = response.xpath(
"//div[contains(@class, 'page-box')]/@page-data"
).extract_first()
page_dict = json.loads(page_dict_handle)
total_page = page_dict.get("totalPage")
current_page = page_dict.get("curPage")
if current_page < total_page:
next_page = plate_url + "pg" + str(current_page + 1) + "/"
yield scrapy.Request(
url=next_page,
callback=self.page_handle,
meta={
"plate_url": plate_url,
'region': region,
"plate": plate
},
headers=self.get_headers(),
)
def parse(self, response):
req = response.meta.get("req")
item = response.meta.get("item")
url_list_handle = Selector(text=response.text)
url_list = url_list_handle.xpath("//div[@class='txt-box']/h3/a/@href").extract()
for index, url in enumerate(url_list):
item1 = Item()
item1.fields["Located"] = Field()
item1["Located"] = index
url = self.base_url + url
url = self.get_real_url_handle(url)
item1.update(item)
yield scrapy.Request(url=url, callback=self.handle_1, meta={"req": req, "item1": item1})
def test_hs_middlewares(hs_downloader_middleware, hs_spider_middleware):
assert hs_spider_middleware._seen_requests == WeakKeyDictionary()
assert hs_downloader_middleware._seen_requests == WeakKeyDictionary()
assert hs_spider_middleware._seen_requests is hs_downloader_middleware._seen_requests
spider = Spider('test')
url = 'http://resp-url'
request_0 = Request(url)
response_0 = Response(url)
hs_downloader_middleware.process_request(request_0, spider)
assert HS_REQUEST_ID_KEY not in request_0.meta
assert HS_PARENT_ID_KEY not in request_0.meta
assert len(hs_spider_middleware._seen_requests) == 0
assert len(hs_downloader_middleware._seen_requests) == 0
hs_downloader_middleware.process_response(request_0, response_0, spider)
assert request_0.meta[HS_REQUEST_ID_KEY] == 0
assert request_0.meta[HS_PARENT_ID_KEY] is None
assert hs_spider_middleware._seen_requests[request_0] == 0
response_0.request = request_0
request_1 = Request(url)
request_2 = Request(url)
item1 = {}
item2 = Item()
output = [request_1, request_2, item1, item2]
processed_output = list(
hs_spider_middleware.process_spider_output(response_0, output, spider))
assert processed_output[0] is request_1
assert request_1.meta[HS_PARENT_ID_KEY] == 0
assert processed_output[1] is request_2
assert request_2.meta[HS_PARENT_ID_KEY] == 0
assert processed_output[2] is item1
assert processed_output[3] is item2
response_1 = Response(url)
hs_downloader_middleware.process_request(request_1, spider)
hs_downloader_middleware.process_response(request_1, response_1, spider)
assert request_1.meta[HS_REQUEST_ID_KEY] == 1
assert request_1.meta[HS_PARENT_ID_KEY] == 0
response_2 = Response(url)
hs_downloader_middleware.process_request(request_2, spider)
hs_downloader_middleware.process_response(request_2, response_2, spider)
assert request_2.meta[HS_REQUEST_ID_KEY] == 2
assert request_2.meta[HS_PARENT_ID_KEY] == 0
def start_requests(self):
item = Item()
for url in self.start_urls:
url_slit = url.split(":")
site_name = url_slit[0]
url = ":".join(url_slit[1:-1])
item.fields["site_name"] = Field()
item['site_name'] = site_name
item.fields["source_url"] = Field()
item['source_url'] = url_slit[-1]
yield scrapy.Request(url=url,
meta={"item": item},
dont_filter=True,
callback=self.parse_url)
def handle_2(self, response):
item2 = response.meta.get("item2")
item3 = Item()
item3.fields["Title"] = Field()
item3.fields["Content"] = Field()
res_text = response.text
res3_handle = Selector(text=res_text)
# title
title = res3_handle.xpath('//meta[@property="og:title"]/@content').extract_first()
try:
content = res3_handle.xpath("//div[@id='js_content']").xpath("string(.)").extract_first()
content = content.strip()
except Exception as e:
content = None
item3["Title"] = title
item3["Content"] = content
item3.update(item2)
yield item3
def second_parse(self, response):
obj = json.loads(response.text)
data = self.get_home_data(obj)
for index, url in enumerate(data['house_url']):
# TODO 新增判断网页重复
item = Item()
item.fields["AreaName"] = Field()
item.fields["PlateName"] = Field()
item.fields["HouseUrl"] = Field()
item.fields["BuildedTime"] = Field()
item.fields["TimeToRelease"] = Field()
item.fields["PropertyAddress"] = Field()
# item.fields["Floor"] = Field()
item.fields["TotalPrice"] = Field()
item.fields["BuildingSquare"] = Field()
item.fields["PropertyCommunity"] = Field()
item.fields["HasElevator"] = Field()
item['AreaName'] = data['area'][index]
item['PlateName'] = data['road'][index]
flag = data['house_url']
if flag:
house_url = flag[index]
if house_url:
item['HouseUrl'] = self.base_url + house_url
item['BuildedTime'] = data['build_time'][index]
item['TimeToRelease'] = time.strftime(
"%Y-%m-%d",
time.localtime(
int(str(data['release_time'][index])[:-3])))
item['PropertyAddress'] = data['addr'][index]
# item['Floor'] = data['floor'][index]
item['TotalPrice'] = data['total_price'][index]
item['BuildingSquare'] = data['build_size'][index]
item['PropertyCommunity'] = data['community'][index]
item['HasElevator'] = data['elevator'][index]
try:
next_url = "https://www.dafangya.com" + url
except:
continue
yield scrapy.Request(url=next_url,
callback=self.third_parse,
meta={'item': item},
dont_filter=True)
