SecT0uch
/
ImmoScrap
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Simplify start_requests: https://docs.scrapy.org/en/latest/intro/tutorial.html#a-shortcut-to-the-start-requests-method

This commit is contained in:
Jordan ERNST 2020-08-11 12:18:24 +02:00
parent 814aea4e66
commit 1e18327a9e
2 changed files with 8 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
ImmoScrap/spiders/PAP.py
View File

@ -13,13 +13,10 @@ class PAPSpider(scrapy.Spider):
next_page_xpath = '//a[@id="pagination-next"]/@href'
def start_requests(self):
urls = [
'https://www.pap.fr/annonce/vente-appartement-immeuble-maison-vienne-38200-g21767-jusqu-a-300000-euros',
'https://www.pap.fr/annonce/vente-appartement-immeuble-maison-saint-etienne-42-g43641-jusqu-a-300000-euros'
]
for url in urls:
yield scrapy.Request(url=url, callback=self.parse)
start_urls = [
'https://www.pap.fr/annonce/vente-appartement-immeuble-maison-vienne-38200-g21767-jusqu-a-300000-euros',
'https://www.pap.fr/annonce/vente-appartement-immeuble-maison-saint-etienne-42-g43641-jusqu-a-300000-euros'
]
def parse(self, response):
ads = response.css(self.ads_css_sel)

11
ImmoScrap/spiders/Seloger.py
View File

@ -35,13 +35,10 @@ class SelogerSpider(scrapy.Spider):
infos_css_sel = ".eJYQQA" # Contains Rooms, bedrooms, area
pagination_xpath = '//div[has-class("ckWPHD")]//text()'
def start_requests(self):
urls = [
f'{self.baseurl}?{urlencode(self.vienne_params)}',
f'{self.baseurl}?{urlencode(self.stetienne_params)}'
]
for url in urls:
yield scrapy.Request(url=url, headers=self.headers, callback=self.parse)
start_urls = [
f'{baseurl}?{urlencode(vienne_params)}',
f'{baseurl}?{urlencode(stetienne_params)}'
]
def parse(self, response):
ads = response.css(self.ads_css_sel)