def confirm_parse(self, response):
html_response = HtmlResponse(response.url, body=response.body)
body = html_response.body_as_unicode().encode('utf8')
if 'Esta solicitação de transferência será gratuita' in body:
data = {
'acsrfToken':
response.xpath('//input[@name="acsrfToken"]/@value')
[0].extract()
}
yield FormRequest(
url=
'https://pagseguro.uol.com.br/operations/startWithdraw.jhtml',
dont_filter=True,
callback=self.success_parse,
formdata=data,
meta=response.meta)
else:
yield Request(
url=
'https://pagseguro.uol.com.br/operations/changeWithdraw.jhtml',
dont_filter=True,
callback=self.logout_parse,
meta=response.meta)
def parse_station_information(self, response: HtmlResponse):
data = json.loads(response.body_as_unicode())
model: StationInformationModel = StationInformationModel.parse(data)
# chain station status request
# see https://stackoverflow.com/questions/13910357/how-can-i-use-multiple-requests-and-pass-items-in-between-them-in-scrapy-python
# TODO is there RxJS forkJoin mechanism for Twisted? could then dispatch multiple requests and await responses
yield scrapy.Request(self.feeds['station_status'],
callback=self.parse_station_status,
meta={'station_information': model})
def parse(self, response: HtmlResponse):
data = json.loads(response.body_as_unicode())
model: GbfsModel = GbfsModel.parse(data)
self.feeds = {feed.name: feed.url for feed in model.feeds}
for name, url in self.feeds.items():
if name == 'system_information':
yield scrapy.Request(url,
callback=self.parse_system_information)
elif name == 'station_information':
yield scrapy.Request(url,
callback=self.parse_station_information)
def process_response(self, request, response, spider):
#log.msg('%s is type %s' % (response.url, type(response)), level=log.DEBUG)
if type(response) is Response and not _file_pattern.match(response.url):
response = HtmlResponse(response.url, body=response.body)
if hasattr(response, 'body_as_unicode'):
hdoc = html.fromstring(response.body_as_unicode())
links = hdoc.xpath('//a')
for link in links:
href = link.get('href')
link.set('href', urlparse.urljoin(get_base_url(response), href) )
return response.replace(body=html.tostring(hdoc, encoding='unicode'))
return response
def parse_station_status(self, response: HtmlResponse):
data = json.loads(response.body_as_unicode())
status_model: StationStatusModel = StationStatusModel.parse(data)
statuses = {
station.station_id: station
for station in status_model.stations
}
# Get previous model
information_model: StationInformationModel = response.meta[
'station_information']
for station in information_model.stations:
station_id = station.station_id
status: Optional[
StationStatusModel.DataModel.StationModel] = statuses.get(
station_id, None)
bikes_available = None
docks_available = None
bikes_disabled = None
docks_disabled = None
is_open = None
if status:
bikes_available = status.num_bikes_available
docks_available = status.num_docks_available
bikes_disabled = status.num_bikes_disabled
docks_disabled = status.num_docks_disabled
is_open = status.is_renting
yield {
'item_type':
'station',
'data':
attr.asdict(
StationItem(scraper_id=self.scraper.id,
source_id=station_id,
name=station.name,
address=station.address,
latitude=station.lat,
longitude=station.lon,
capacity=station.capacity,
bikes_available=bikes_available,
docks_available=docks_available,
bikes_disabled=bikes_disabled,
docks_disabled=docks_disabled,
open=is_open))
}
def after_post(self, response):
item = []
# body = response.body.decode('cp949')
# html_response = HtmlResponse(url=response.url, body=response.body, encoding='cp949')
html_response = HtmlResponse(url=response.url, body=response.body, encoding='cp949')
# xxs = HtmlXPathSelector(html_response)
# prices = xxs.select('//td[@class="border7_2 text_align1"]/text()').extract()
doc = html_response.body_as_unicode()
# parser = etree.HTMLParser(encoding='utf-8')
# tree = etree.fromstring(doc, parser)
# xresult = tree.xpath('//td[@class="border7_2 text_align1 dataItem1"]/text()')
# print len(xresult)
# for x in xresult:
# print x
# return
pattern = re.compile(r'var chartData = ([^;]*);')
results = pattern.findall(doc)
a = json.loads(results[0])
print a
def parse_system_information(self, response: HtmlResponse):
data = json.loads(response.body_as_unicode())
model: SystemInformationModel = SystemInformationModel.parse(data)
system = model.data
# Note - not passing source_id through as gbfs sources only contain a single system.
# In addition, the station feeds do not contain the system_id, so it makes it easier
# for pipeline to find the system when its source_id is null.
yield {
'item_type':
'system',
'data':
attr.asdict(
SystemItem(
scraper_id=self.scraper.id,
name=system.name,
# source_id=system.system_id,
phone_number=system.phone_number,
email=system.email,
timezone=system.timezone,
url=system.url,
language=system.language))
}
def after_post(self, response):
item = []
# body = response.body.decode('cp949')
# html_response = HtmlResponse(url=response.url, body=response.body, encoding='cp949')
html_response = HtmlResponse(url=response.url,
body=response.body,
encoding='cp949')
# xxs = HtmlXPathSelector(html_response)
# prices = xxs.select('//td[@class="border7_2 text_align1"]/text()').extract()
doc = html_response.body_as_unicode()
# parser = etree.HTMLParser(encoding='utf-8')
# tree = etree.fromstring(doc, parser)
# xresult = tree.xpath('//td[@class="border7_2 text_align1 dataItem1"]/text()')
# print len(xresult)
# for x in xresult:
# print x
# return
pattern = re.compile(r'var chartData = ([^;]*);')
results = pattern.findall(doc)
a = json.loads(results[0])
print a
def parse_VIN_db(self, response: HtmlResponse, item, page_id):
json_VIN_db = json.loads(response.body_as_unicode())
item.add_value('json_VIN_db', json_VIN_db)
yield item.load_item()
def parse_car_spec(self, response: HtmlResponse, item, page_id):
json_car_spec = json.loads(response.body_as_unicode())
item.add_value('json_car_spec', json_car_spec)
url_VIN_db = f'https://www.avito.ru/web/1/swaha/v1/autoteka/teaser/{page_id}'
yield response.follow(url_VIN_db, callback=self.parse_VIN_db, cb_kwargs={'item': item, 'page_id': page_id})