def parse(self, response):
for row in response.css('table.ms-listviewtable tr[class^=building-registry-row]'):
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no", "td:nth-child(1)::text")
l.add_css("number_in_order", "td:nth-child(2)::text")
l.add_css("order_date", "td:nth-child(3)::text")
l.add_css("decree_no", "td:nth-child(4)::text")
l.add_css("customer", "td:nth-child(5)::text, td:nth-child(5) div::text")
l.add_css("obj", "td:nth-child(6)::text, td:nth-child(6) div::text")
l.add_css("address", "td:nth-child(7)::text, td:nth-child(7) div::text")
l.add_css("changes", "td:nth-child(8) div::text")
l.add_css("cancellation", "td:nth-child(9) div::text")
url = row.css("td:nth-child(10) a::attr(href)").extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
# get next page number href next to current inactive with span tag
nextPageJs = response.xpath('//table[@class="ms-listviewtable"]//tr[@class="building-registry-pager"]//table/tr/td[.//span]/following-sibling::td[1]/a/@href')
if len(nextPageJs):
yield FormRequest.from_response(
response,
formname="aspnetForm",
formxpath="//form[@id='aspnetForm']",
dont_click=True,
formdata={
'__EVENTARGUMENT': nextPageJs.re('\',\'(.*)\'\)'),
'__EVENTTARGET': nextPageJs.re('javascript:__doPostBack\(\'(.*)\',')
},
dont_filter=True,
callback=self.parse
)
def parse(self, response):
for i, row in enumerate(response.css("table tbody tr")):
# skip styled as header
if row.css("td:nth-child(1) p strong").get():
self.logger.debug("skipped row : {}".format(row.get()))
continue
self.logger.debug("parsed row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no",
"td:nth-child(1) a.add-google-doc::text",
re=r"№\s?(.+)$")
l.add_css("order_date",
"td:nth-child(1) a.add-google-doc::text",
re=r"([\d\.]*) №")
l.add_css("customer",
"td:nth-child(2)::text, td:nth-child(2) p::text")
l.add_css("obj", "td:nth-child(3)::text, td:nth-child(3) p::text")
l.add_css("address", "td:nth-child(4)::text")
l.add_css("changes", "td:nth-child(5)::text")
l.add_css("cancellation", "td:nth-child(6)::text")
cancellation_url = row.css(
"td:nth-child(6) a.add-google-doc::attr(href)").getall()
if len(cancellation_url):
l.add_value(
"cancellation_url",
[response.urljoin(url) for url in cancellation_url])
url = row.css("td:nth-child(1) a.add-google-doc::attr(href)"
).extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
def parse(self, response):
# only first table with data
for index, row in enumerate(
response.css("div.post-body>table:first-of-type>tbody>tr")):
# first two are headers, skip
if index < 2:
continue
l = StripJoinItemLoader(item=MbuItem(), selector=row)
# because of errors in html, get td from current root only
l.add_xpath(
"number_in_order",
"./td[position()=1]/span/text()|./td[position()=1]/p/span/text()",
re=r"(\d+)\s?")
l.add_css(
"order_no",
"td:nth-child(2) p span::text, td:nth-child(2) span::text",
re=r"^\s*№ ?(.*)\s?від")
l.add_css(
"order_date",
"td:nth-child(2) p span::text, td:nth-child(2) span::text",
re=r"(\d{1,2}[\. /]?\d{1,2}[\. /]?\d{2,4})[\sр\.]*$")
l.add_css(
"customer",
"td:nth-child(3) p span::text, td:nth-child(3) span::text")
l.add_css(
"obj",
"td:nth-child(4) p span::text, td:nth-child(4) span::text")
l.add_css(
"address",
"td:nth-child(5) p span::text, td:nth-child(5) span::text")
l.add_css(
"changes",
"td:nth-child(6) p span::text, td:nth-child(6) span::text")
l.add_css(
"cancellation",
"td:nth-child(7) p span::text, td:nth-child(7) span::text")
url = row.css(
"td:nth-child(8) p span a::attr(href), td:nth-child(8) span a::attr(href), td:nth-child(8) a::attr(href)"
).extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
def parse(self, response):
for row in response.css("table.table-registry tbody tr"):
self.logger.debug("parse row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no", "td:nth-child(1) strong::text")
l.add_css("order_date", "td:nth-child(1)::text")
l.add_css("prescript_no", "td:nth-child(2) strong::text")
l.add_css("prescript_date", "td:nth-child(2)::text")
l.add_css("customer", "td:nth-child(5)::text")
l.add_css("obj", "td:nth-child(3)::text")
l.add_css("address", "td:nth-child(4)::text")
l.add_css("changes", "td:nth-child(6)::text")
l.add_css("cancellation", "td:nth-child(7)::text")
url = row.css("td:nth-child(8) a::attr(href)").extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
next_page = response.css(
'a.next.page-numbers::attr(href)').extract_first()
if next_page:
self.logger.debug("follow next page : {}".format(next_page))
yield response.follow(next_page)
def parse(self, response):
jsonresponse = json.loads(response.body_as_unicode())
for row in jsonresponse["aaData"]:
self.logger.debug("parsed row : {}".format(row))
l = StripJoinItemLoader(item=MbuItem())
l.add_value("number_in_order", row[0])
l.add_value(
"order_no",
re.search("№ ?(.*) ?(ві|от)", row[1]).group(1)
if row[1] else None)
l.add_value(
"order_date",
re.search("([0-9]{1,2}\.[0-9]{1,2}\. ?[0-9]{1,4})",
row[1]).group(1) if row[1] else None)
l.add_value("customer", row[2])
l.add_value("obj", row[3])
l.add_value("address", row[4])
l.add_value("changes", row[5])
l.add_value("cancellation", row[6])
l.add_value(
"scan_url",
Selector(text=row[7]).css("a::attr(href)").extract_first()
if row[7] else None)
yield l.load_item()
class ZhytomyrSpider(scrapy.spiders.CSVFeedSpider):
location_name = "Житомир"
name = "zhytomyr"
allowed_domains = ["zt-rada.gov.ua"]
start_urls = ["http://zt-rada.gov.ua/?3398[0]=6281"]
custom_settings = {
# specifies exported fields and order
'FEED_EXPORT_FIELDS': [
"location_name", "order_no", "order_date", "customer", "obj",
"address", "changes", "cancellation", "scan_url"
],
}
item_loaders = defaultdict(lambda: StripJoinItemLoader(item=MbuItem()))
def parse_xls_and_flush(self, response):
sheet = xlrd.open_workbook(
file_contents=response.body).sheet_by_index(0)
for index in range(1, sheet.nrows):
row = sheet.row(index)
# if row is not empty
if row[1].value:
order_no = row[0].value.replace('№', '').strip()
l = self.get_item(order_no)
if not l.get_output_value('order_no'):
l.add_value("order_no", order_no)
l.add_value("order_date", row[1].value)
l.add_value("customer", row[6].value)
l.add_value("obj", row[2].value)
l.add_value("address", row[3].value)
l.add_value("changes", row[7].value)
l.add_value("cancellation", row[8].value)
else:
self.logger.debug("skipped index : {}, row : {}".format(
index, row))
for item in self.item_loaders.values():
yield item.load_item()
# clear in case of several xls files found
self.item_loaders.clear()
def parse(self, response):
for row in response.css(".docrowcontainer"):
document_url = row.css(
"div:nth-child(2) a.docdownload::attr(href)").extract_first()
if document_url.endswith('.xls'):
self.logger.info(
"xls document found : {}".format(document_url))
yield response.follow(document_url,
callback=self.parse_xls_and_flush,
priority=10) # big priority to run last
else:
continue
# self.logger.debug("parse site row : {}".format(row.get()))
# order_no = "".join(row.css("div:nth-child(1)::text").re(r"№ ?(.*)")).strip()
# l = self.get_item(order_no)
# l.selector = row
# l.add_value("order_no", order_no)
# l.add_xpath("order_date", "./@data-year")
# l.add_value("scan_url", response.urljoin(document_url))
next_page_link = response.xpath(
'//*[@id="tp6"]//ul/li[@class="active"]/following-sibling::li[1]/a/@href'
).get()
if next_page_link:
self.logger.debug("next page link : {}".format(next_page_link))
yield response.follow(next_page_link, callback=self.parse)
def get_item(self, order_no):
# tries to find similar order number with different separating signs
splitter = '[\.\s/\-\|]+'
order_id_list = re.split(splitter, order_no.strip())
filtered_order = list(
filter(
lambda ord: order_id_list == re.split(splitter, ord.strip()),
self.item_loaders.keys()))
if len(filtered_order):
self.logger.debug(
'similar to order_no : {} found in loaders : {}'.format(
order_no, filtered_order))
return self.item_loaders[filtered_order[0] if len(filtered_order
) else order_no]
def parse_filtered(self, response):
jsonresponse = json.loads(response.body_as_unicode())
table = Selector(text=jsonresponse['table'])
for row in table.css("tr"):
self.logger.debug("parse row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no", "td:nth-child(1) strong::text")
l.add_css("order_date", "td:nth-child(1)::text", re=r"([\d\.])+\s*№")
l.add_css("customer", "td:nth-child(2)::text")
l.add_css("obj", "td:nth-child(3) small::text")
l.add_css("address", "td:nth-child(4)::text")
changes_url = row.xpath("./td[position()=5]//a[.//span[contains(@class, 'glyphicon glyphicon-edit')]]/@href").extract_first()
if changes_url:
l.add_value("changes", response.urljoin(changes_url))
cancellation_url = row.xpath("./td[position()=5]//a[.//span[contains(@class, 'glyphicon glyphicon-ban-circle')]]/@href").extract_first()
if cancellation_url:
l.add_value("cancellation", response.urljoin(cancellation_url))
scan_url = row.xpath("./td[position()=5]//a[.//span[contains(@class, 'glyphicon glyphicon-info-sign')]]/@href").extract_first()
if scan_url:
l.add_value("scan_url", response.urljoin(scan_url))
yield l.load_item()
def parse(self, response):
jsonresponse = json.loads(response.body_as_unicode())
for attributes in jsonresponse['features']:
self.logger.debug("parse row : {}".format(attributes))
attributes = attributes['attributes']
l = StripJoinItemLoader(item=MbuItem())
l.add_value("number_in_order", str(attributes['Kadastr2016.DBO.MBU.OBJECTID']))
l.add_value("order_no", attributes['Kadastr2016.DBO.MBU.NomKancel'] if attributes['Kadastr2016.DBO.MBU.NomKancel'] else '-')
l.add_value("order_date",
str(datetime.fromtimestamp(attributes['Kadastr2016.DBO.MBU.Data'] / 1000).date()) if attributes[
'Kadastr2016.DBO.MBU.Data'] else None)
l.add_value("customer", attributes['Kadastr2016.DBO.MBU.Zamovnuk_MBO'])
l.add_value("obj", attributes['Kadastr2016.DBO.MBU.Nazobekty'])
l.add_value("address", self.get_address(attributes))
l.add_value("changes", attributes['Kadastr2016.DBO.MBU.Zmini'])
l.add_value("cancellation", attributes['Kadastr2016.DBO.MBU.Skasuvannia'])
# all original fields to be used later
additional_fields = dict([[item[len('Kadastr2016.DBO.'):],attributes[item]] for item in attributes.keys()])
l.add_value('additional_fields', json.dumps(additional_fields, ensure_ascii=False))
scan_urls = []
if attributes['Kadastr2016.DBO.MBU.Link_1']:
scan_urls.append(response.urljoin(attributes['Kadastr2016.DBO.MBU.Link_1']))
if attributes['Kadastr2016.DBO.MBU.Link_2']:
scan_urls.append(response.urljoin(attributes['Kadastr2016.DBO.MBU.Link_2']))
if scan_urls:
l.add_value("scan_url", ",".join(scan_urls))
yield l.load_item()
def parse(self, response):
for row in response.xpath(
'//table/tbody/tr[count(td)=8 and not(./td//span/strong)]'):
self.logger.debug("parse row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("number_in_order",
"td:nth-child(1) p::text, td:nth-child(1)::text")
l.add_css(
"order_no",
"td:nth-child(2) p:nth-child(1)::text, td:nth-child(2) p:nth-child(1) span::text",
re=r"№\s?(\d+)")
l.add_css(
"order_date",
"td:nth-child(2) p:nth-child(1)::text, td:nth-child(2) p:nth-child(1) span::text",
re=r"(\d{1,2}[\. /]?\d{1,2}[\. /]?\d{2,4})")
l.add_css("remarks", "td:nth-child(2) p:nth-child(2)::text")
l.add_css("customer", "td:nth-child(3) p::text")
l.add_css("obj", "td:nth-child(4) p::text")
l.add_css("address", "td:nth-child(5) p::text")
l.add_css("changes",
"td:nth-child(6) p::text, td:nth-child(6) a::text")
l.add_css("cancellation",
"td:nth-child(7) p::text, td:nth-child(7) a::text")
l.add_css("scan_text",
"td:nth-child(8) a::text, td:nth-child(8) p::text")
l.add_css("changes_url", "td:nth-child(6) a::attr(href)")
l.add_css("cancellation_url", "td:nth-child(7) a::attr(href)")
l.add_css("scan_url", "td:nth-child(8) a::attr(href)")
yield l.load_item()
def parse(self, response):
for row in response.css("table#droptablesTbl4 tbody tr"):
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no", "td:nth-child(2)::text", re=r"([\d]+)$")
l.add_css("order_date", "td:nth-child(2)::text", re=r"^[\d.]*")
l.add_css("customer", "td:nth-child(3)::text")
l.add_css("obj", "td:nth-child(4)::text")
l.add_css("address", "td:nth-child(5)::text")
l.add_css("changes", "td:nth-child(6)::text")
l.add_css("cancellation", "td:nth-child(7)::text")
url = row.css("td:nth-child(8) a::attr(href)").extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
def parse(self, response):
loaded_order_numbers = []
for index, row in enumerate(response.css("table tbody tr")):
# first and second are headers, skip
if index == 0 or index == 1:
continue
orders_in_row = len(row.css("td:nth-child(3) p").getall())
if orders_in_row == 0:
orders_in_row = len(row.css("td:nth-child(3) span"))
# each row is sub divided for main order and it's changes
for order_in_row in range(orders_in_row):
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_value(
"number_in_order",
self.get_first_existed(row, "td:nth-child(1) p span::text",
"td:nth-child(1) span::text"))
order_no = row.css(
"td:nth-child(3) p:nth-child({}) span::text, td:nth-child(3) span:nth-child({})::text"
.format(order_in_row + 1, order_in_row + 1)).get()
if not order_no:
continue
l.add_value("order_no", order_no)
l.add_value(
"order_date",
self.get_first_existed(
row, "td:nth-child(2) p:nth-child(" +
str(order_in_row + 1) + ") span::text",
"td:nth-child(2) p:nth-child(1) span::text",
"td:nth-child(2) span:nth-child(" +
str(order_in_row + 1) + ")::text",
"td:nth-child(2) span:nth-child(1) span::text"))
l.add_value(
"customer",
self.get_first_existed(
row, "td:nth-child(4) p:nth-child(" +
str(order_in_row + 1) + ") span::text",
"td:nth-child(4) p:nth-child(1) span::text",
"td:nth-child(4) span:nth-child(" +
str(order_in_row + 1) + ")::text",
"td:nth-child(4) span:nth-child(1)::text"))
l.add_value(
"obj",
self.get_first_existed(
row, "td:nth-child(5) p:nth-child(" +
str(order_in_row + 1) + ") span::text",
"td:nth-child(5) p:nth-child(1) span::text",
"td:nth-child(5) span:nth-child(" +
str(order_in_row + 1) + ")::text",
"td:nth-child(5) span:nth-child(1)::text"))
l.add_value(
"address",
self.get_first_existed(
row, "td:nth-child(6) p:nth-child(" +
str(order_in_row + 1) + ") span::text",
"td:nth-child(6) p:nth-child(1) span::text",
"td:nth-child(6) span:nth-child(" +
str(order_in_row + 1) + ")::text",
"td:nth-child(6) span:nth-child(1)::text"))
l.add_value(
"changes",
self.get_first_existed(
row, "td:nth-child(7) p:nth-child(" +
str(order_in_row + 1) + ") span::text",
"td:nth-child(7) p:nth-child(1) span::text",
"td:nth-child(7) span:nth-child(" +
str(order_in_row + 1) + ")::text",
"td:nth-child(7) span:nth-child(1)::text"))
l.add_value(
"cancellation",
self.get_first_existed(
row, "td:nth-child(8) p:nth-child(" +
str(order_in_row + 1) + ") span::text",
"td:nth-child(8) p:nth-child(1) span::text",
"td:nth-child(8) span:nth-child(" +
str(order_in_row + 1) + ")::text",
"td:nth-child(8) span:nth-child(1)::text"))
url = self.get_first_existed(
row, "td:nth-child(9) p:nth-child(" +
str(order_in_row + 1) + ") span a::attr(href)",
"td:nth-child(9) p:nth-child(1) span a::attr(href)",
"td:nth-child(9) a:nth-child(" + str(order_in_row + 1) +
")::attr(href)",
"td:nth-child(9) a:nth-child(1)::attr(href)")
# TODO add decree found in 357th row
l.add_value("scan_url", response.urljoin(url))
address_assign_url = self.get_first_existed(
row, "td:nth-child(10) a::attr(href)")
if address_assign_url:
l.add_value("address_assign_url",
response.urljoin(address_assign_url))
l.add_css("address_assign_no",
"td:nth-child(10) a span::text",
re=r"№(.*) ?від")
l.add_css("address_assign_date",
"td:nth-child(10) a span::text",
re=r"від ?(.*)$")
loaded_order_numbers.append(
int(l.get_collected_values("number_in_order")[0]))
yield l.load_item()
# check if all consecutive orders where loaded
missed_order_numbers = {*range(1, max(loaded_order_numbers))
}.difference(loaded_order_numbers)
if missed_order_numbers:
self.logger.warning("Missed order numbers: %s",
missed_order_numbers)
else:
self.logger.info("All order numbers processed")
def parse_row(self, response, row):
self.logger.debug("parse row : {}".format(row))
l = StripJoinItemLoader(item=MbuItem())
l.add_value("order_no",
row['restrictionNumber ']) # space in the end is needed
l.add_value("order_date", row['restrictionDate'])
l.add_value("customer", row['objectOwner'])
l.add_value("obj", row['objectDescription'])
l.add_value("address", row['objectAddress'])
l.add_value("changes", row['objectChanges'])
l.add_value("cancellation", row['objectCancel'])
l.add_value("status", row['objectStatus'])
yield l.load_item()
def parse(self, response):
for index, row in enumerate(response.css("table>tbody>tr")):
# skip headers and rows with empty lines
if index < 5 and not "".join(
row.css("td::text, td span::text").getall()).strip():
self.logger.debug("skipped index : {}, row : {}".format(
index, row.get()))
continue
self.logger.debug("parse index : {}, row : {}".format(
index, row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("decree", "td:nth-child(1)::text")
l.add_css("order_no",
"td:nth-child(2)::text, td:nth-child(2) span::text",
re=r"^\s?(.*)від")
l.add_css("order_date",
"td:nth-child(2)::text, td:nth-child(2) span::text",
re=r"від\s?([\d\.]*)\s*$")
l.add_css("customer",
"td:nth-child(3)::text, td:nth-child(3) span::text")
l.add_css("obj",
"td:nth-child(4)::text, td:nth-child(4) span::text")
l.add_css("address",
"td:nth-child(5)::text, td:nth-child(5) span::text")
l.add_css("changes",
"td:nth-child(6)::text, td:nth-child(6) span::text")
l.add_css(
"cancellation",
"td:nth-child(7)::text, td:nth-child(7) a::text, td:nth-child(7) span::text"
)
cancellation_url = row.css(
"td:nth-child(7) a::attr(href)").extract_first()
if cancellation_url:
l.add_value("cancellation_url",
response.urljoin(cancellation_url))
scan_url = row.css("td:nth-child(8) a::attr(href)").extract_first()
if scan_url:
l.add_value("scan_url", response.urljoin(scan_url))
yield l.load_item()
def parse(self, response):
for row in response.css("table#tabledataMto tbody tr"):
self.logger.debug("parse row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no", "td:nth-child(1)::text")
l.add_css("order_date", "td:nth-child(4)::text")
l.add_css("customer", "td:nth-child(5)::text")
l.add_css("obj", "td:nth-child(6)::text")
l.add_css("obj_purpose", "td:nth-child(7)::text")
l.add_css("address_street", "td:nth-child(2)::text")
l.add_css("address_street_number", "td:nth-child(3)::text")
l.add_css("address", "td:nth-child(2)::text")
l.add_css("address", "td:nth-child(3)::text")
l.add_css("cancellation", "td:nth-child(8)::text")
url = row.css("td:nth-child(9) a::attr(href)").extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
def parse(self, response):
for row in response.css("table tbody tr"):
# first is header, skip
if row.css("td:nth-child(1)::text").get() == "№ з/п":
self.logger.debug("skiped row : {}".format(row.get()))
continue
self.logger.debug("parse row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("number_in_order", "td:nth-child(1)::text")
l.add_css("order_no",
"td:nth-child(2) p::text, td:nth-child(2)::text",
re=r"№ ?(.*)\s?$")
l.add_css(
"order_date",
"td:nth-child(2) p:nth-child(1)::text, td:nth-child(2)::text",
re=r"(\d{1,2}[\. /]?\d{1,2}[\. /]?\d{2,4})[\sр\.]*")
l.add_css("customer", "td:nth-child(3)::text")
l.add_css("obj", "td:nth-child(4)::text")
l.add_css("address",
"td:nth-child(5)::text, td:nth-child(5) a::text")
l.add_css("changes", "td:nth-child(6)::text")
l.add_css("cancellation", "td:nth-child(7)::text")
url = row.css("td:nth-child(8) a::attr(href)").extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
l.add_css("scan_no", "td:nth-child(8) a::text", re=r"№(.*) ?від")
l.add_css("scan_date", "td:nth-child(8) a::text", re=r"від ?(.*)")
item = l.load_item()
if not item.get("number_in_order"):
continue
yield item
def parse(self, response):
jsonresponse = json.loads(response.body_as_unicode())
for row in jsonresponse["aaData"]:
self.logger.debug("parse row : {}".format(row))
l = StripJoinItemLoader(item=MbuItem())
l.add_value("number_in_order", row[0])
l.add_value("order_no", row[2])
l.add_value("order_date", row[1])
l.add_value("customer", row[7])
l.add_value("obj", row[3])
l.add_value("address", row[4])
l.add_value("changes", row[5])
l.add_value("cancellation", row[6])
l.add_value("scan_url",
response.urljoin(row[8]) if row[8] else None)
yield l.load_item()
def parse(self, response):
for row in response.css(
"div.table-content-container div#orders div.one_order"):
self.logger.debug("parsed row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_date", "li.order_date::text")
l.add_css("order_no", "li.order_number::text")
l.add_css("customer", "li.cust::text")
l.add_css("obj", "li.order_name::text")
l.add_css("address", "li.addr::text")
l.add_css("changes", "li.changes_info::text")
l.add_css("cancellation", "li.reason_canc::text")
url = row.css("li.download a::attr(href)").extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
def parse(self, response):
for i, row in enumerate(response.css("table tr")):
# skip first as header
if i == 0:
self.logger.debug("skipped row : {}".format(row.get()))
continue
self.logger.debug("parse row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no", "td:nth-child(1)::text", re=r" (.*)$")
l.add_css("order_date", "td:nth-child(1)::text", re=r"^[\d-]*")
l.add_css("customer", "td:nth-child(2)::text")
l.add_css("obj", "td:nth-child(3)::text")
l.add_css("address", "td:nth-child(4)::text")
l.add_css("changes", "td:nth-child(5)::text")
l.add_css("cancellation", "td:nth-child(6)::text")
l.add_css("scan_url", "td:nth-child(7) a::attr(href)")
yield l.load_item()
def parse(self, response):
for row in response.css("table.table.table-striped.small tbody tr"):
self.logger.debug("parse row : {}".format(row.get()))
l = StripJoinItemLoader(item=MbuItem(), selector=row)
l.add_css("order_no", "td:nth-child(1)::text")
l.add_css("order_date", "td:nth-child(2)::text")
l.add_css("customer", "td:nth-child(3)::text")
l.add_css("obj", "td:nth-child(4)::text")
l.add_css("address", "td:nth-child(5)::text")
l.add_css("cadastre_number", "td:nth-child(6)::text")
l.add_css("document_status", "td:nth-child(7) span::text")
l.add_css("changes", "td:nth-child(8)::text")
document_url = row.css(
"td:nth-child(9) a::attr(href)").extract_first()
l.add_value("scan_url", response.urljoin(document_url))
map_url = row.css("td:nth-child(1) a::attr(href)").extract_first()
l.add_value("map_url", response.urljoin(map_url))
yield l.load_item()
# if 'Next' page label present continue crawling
if response.css("ul.pagination li a[aria-label=Next]").get():
yield scrapy.Request(
self.get_next_page(response),
callback=self.parse,
meta={"proxy": self.get_random_proxy()},
)
def parse(self, response):
for row in response.css(
"article table tbody tr:not([align=\"center\"])"):
l = StripJoinItemLoader(item=MbuItem(), selector=row)
# skip empty lines
date_order = ''.join(
row.css(
"td:nth-child(2) p span::text, td:nth-child(2) span::text, td:nth-child(2)::text"
).getall())
if not date_order or not date_order.strip():
self.logger.debug("skipped row : {}".format(row.get()))
continue
self.logger.debug("parse row : {}".format(row.get()))
# number_in_order is unique only per year
l.add_css(
"number_in_order",
"td:nth-child(1) p span::text, td:nth-child(1) span::text, td:nth-child(1)::text"
)
l.add_value(
"order_no",
re.search('^\\s?№? ?(.*)\\s?(в?ід|dsl)', date_order).group(1))
l.add_css(
"order_date",
"td:nth-child(2) p span::text, td:nth-child(2) span::text, td:nth-child(2)::text",
re=r"(\d{1,2}[\. /]?\d{1,2}[\. /]?\d{2,4})\s*$")
l.add_css(
"customer",
"td:nth-child(3) span::text, td:nth-child(3)::text, td:nth-child(3) p::text"
)
l.add_css("obj",
"td:nth-child(4) span::text, td:nth-child(4)::text")
l.add_css(
"address",
"td:nth-child(5) span::text, td:nth-child(5) p::text, td:nth-child(5)::text"
)
l.add_css("changes",
"td:nth-child(6) span::text, td:nth-child(6)::text")
l.add_css("cancellation",
"td:nth-child(7) span::text, td:nth-child(7)::text")
url = row.css("td:nth-child(8) a::attr(href)").extract_first()
if url:
l.add_value("scan_url", response.urljoin(url))
yield l.load_item()
