def url_builder(self, value, data, response):
system = components(-5, -4)(response.request.url)
year = int(components(-4, -3)(response.request.url))
month = int(components(-3, -2)(response.request.url).lstrip('0'))
return self.pattern.format(system, date(year, month, 1), value,
self.limit)
def parse_list(self, response):
data = response.json()
# The last page returns an empty JSON object.
if not data:
return
for item in data['data']:
url = replace_parameters(response.request.url, offset=None) + item['ocid']
yield self.build_request(url, formatter=components(-2))
url = replace_parameters(response.request.url, offset=data['offset'])
yield self.build_request(url, formatter=join(components(-1), parameters('offset')), callback=self.parse_list)
def parse_list(self, response):
data = json.loads(response.text)
for item in data['data']:
url = item['uri']
if url:
yield self.build_request(url, formatter=components(-1))
if self.sample:
break
else:
next_page_url = data.get('next_page_url')
if next_page_url:
yield self.build_request(next_page_url, formatter=join(components(-1), parameters('page')),
callback=self.parse_list)
def parse_list(self, response):
# remove the last item in the list to fix the str JSON format
urls = json.loads(
response.xpath('//body//text()').getall()[6].replace(
",\r\n\r\nhttps://www.ppra.org.pk", ""))
for url in urls:
yield self.build_request(url, formatter=components(-2))
def parse_list(self, response):
urls = response.css('.fileLink::attr(href)').getall()
json_urls = list(filter(lambda x: '/JSON_DGCP_' in x, urls))
for url in json_urls:
if '/JSON_DGCP_' in url:
yield self.build_request(url, formatter=components(-1))
def parse_list(self, response):
urls = response.xpath('//item/link/text()').getall()
if self.sample:
urls = [urls[0]]
for url in urls:
yield self.build_request(url, formatter=components(-1))
def parse(self, response):
data = json.loads(response.text)
pattern = 'https://datos.hacienda.gov.py:443/odmh-api-v1/rest/api/v1/ocds/release-package/{}'
# If is the first URL, we need to iterate over all the pages to get all the process ids to query
if response.request.meta['first']:
total = data['meta']['totalPages']
for page in range(2, total + 1):
yield self.build_request(self.base_list_url.format(page),
formatter=parameters('page'),
meta={
'meta': True,
'first': False,
},
dont_filter=True)
# if is a meta request it means that is the page that have the process ids to query
if response.request.meta['meta']:
# Now that we have the ids we iterate over them, without duplicate them, and make the
# final requests for the release_package this time
for row in data['results']:
if row['idLlamado'] and row[
'idLlamado'] not in self.release_ids:
self.release_ids.append(row['idLlamado'])
yield self.build_request(pattern.format(row['idLlamado']),
formatter=components(-1),
meta={
'meta': False,
'first': False,
},
dont_filter=True)
else:
yield self.build_file_from_response(response,
data_type=self.data_type)
def start_requests(self):
url = self.url
# date parameter obtained
url = url.format(self.from_date.strftime("%Y%m%d"),
self.until_date.strftime("%Y%m%d"))
# url looks like http://www.gekoware.com/swmp/api/ocds/20190101/20201005
yield self.build_request(url, formatter=components(-2))
def parse_list(self, response):
urls = response.xpath('//a[@class="enlaces_contenido"]/@href').getall()
if self.sample:
urls = [urls[0]]
for url in urls:
# URL looks like https://apiocds.colombiacompra.gov.co:8443/ArchivosSECOP/Archivos/SI2011.zip
yield self.build_request(url, formatter=components(-1))
def parse_list(self, response):
data = json.loads(response.text)
for resource in data['result']['resources']:
if resource['format'].upper() == 'JSON':
# Presently, only one URL matches.
yield self.build_request(resource['url'],
formatter=components(-1))
def start_requests(self):
today = date.today()
if hasattr(self, 'year'):
year = int(self.year)
start = date(year, 1, 1)
stop = date(year, 12, 1)
if year == today.year:
stop = stop.replace(month=today.month)
else:
start = date(2008, 1, 1)
stop = today
if self.sample:
start = stop
for d in date_range_by_month(start, stop):
yield self.build_request(
self.base_list_url.format(d, 0, self.limit),
formatter=components(-4, -1),
meta={
'year': d.year,
'month': d.month,
},
callback=self.parse_list
)
def parse_list(self, response):
for path in response.xpath('//div[@role="rowheader"]/span/a/@href').getall():
if path.endswith('.xlsx'):
yield self.build_request(
f'https://github.com{path}?raw=true',
formatter=components(-1)
)
def parse_list(self, response):
urls = response.xpath('//a[contains(., "[json]")]/@href').getall()
if self.sample:
urls = [urls[0]]
for url in urls:
# URL looks like http://200.13.162.79/datosabiertos/HC1/HC1_datos_2020_json.zip
yield self.build_request(url, formatter=components(-1))
def parse_list(self, response):
urls = json.loads(response.text)
if self.sample:
urls = [urls[0]]
for url in urls:
# URL looks like https://ocds.ageops.net/api/record/5ed2a62c4192f32c8c74a4e5
yield self.build_request(url, formatter=components(-1))
def parse_record_package(self, response):
yield self.build_file_from_response(response,
data_type='record_package')
data = json.loads(response.text)
if 'packages' in data:
for url in data['packages']:
yield self.build_request(url, formatter=components(-1))
def parse_list(self, response):
urls = response.json()['packagesPerMonth']
for url in urls:
# URL looks like https://www.zppa.org.zm/ocds/services/recordpackage/getrecordpackage/2016/7
yield self.build_request(url,
formatter=join(components(-2),
extension='zip'))
def parse_list(self, response):
datas = response.json()
for result in datas['result']['results']:
for resource in result['resources']:
if resource['format'] == 'JSON':
# http://bit.ly/ConcentradoINAI
yield self.build_request(resource['url'], formatter=components(-1), meta={'dont_redirect': True},
callback=self.parse_redirect)
def parse_list(self, response):
data = response.json()
for item in data['data']:
for resource in item['resources']:
description = resource['description']
if description and 'ocds' in description.lower():
# Presently, only one URL matches.
yield self.build_request(resource['url'], formatter=components(-2))
def parse_list(self, response):
urls = response.json()
for url in urls:
if self.from_date and self.until_date:
date = datetime.strptime(url[-10:], self.date_format)
if not (self.from_date <= date <= self.until_date):
continue
yield self.build_request(url, formatter=components(-2), callback=self.parse_release_list)
def parse_list(self, response):
html_urls = response.xpath('//a/@href').getall()
if html_urls:
# Each link contains different versions of SERCOP's emergency dataset, only the newest should be downloaded
# URL format: ./archivos/ocds-YYYY-MM-DD.json
html_urls.sort(reverse=True)
yield self.build_request(f'{response.request.url}{html_urls[0]}',
formatter=components(-1))
def start_requests(self):
pattern = 'https://birms.bandung.go.id/api/packages/year/{}'
start = 2013
stop = date.today().year
for year in date_range_by_year(start, stop):
yield self.build_request(pattern.format(year), formatter=components(-1), callback=self.parse_list)
def parse_list(self, response):
items = json.loads(response.text)
if self.sample:
items = [items[0]]
for item in items:
url = item['urls']['json']
yield self.build_request(url, formatter=components(-1))
def parse_list(self, response):
urls = json.loads(response.text)
for url in urls:
# A JSON array of URL strings, in reverse chronological order.
# URL looks like https://ocds.ageops.net/api/ocds/releases/2020-05-30
yield self.build_request(url,
formatter=components(-1),
callback=self.parse_release_list)
def parse_list(self, response):
urls = json.loads(response.text)['packagesPerMonth']
if self.sample:
urls = [urls[0]]
for url in urls:
# URL looks like https://www.zppa.org.zm/ocds/services/recordpackage/getrecordpackage/2016/7
yield self.build_request(url, formatter=components(-2))
def parse_list(self, response):
pattern = 'https://admin.ims.susasan.org/ocds/json/dhangadhi-{}.json'
data = response.json()
for item in data['data']['fiscal_years']:
# A URL might redirect to https://admin.ims.susasan.org/login
yield self.build_request(pattern.format(item['name']),
formatter=components(-1),
meta={'dont_redirect': True})
def start_requests(self):
url = 'https://ocds.blob.core.windows.net/ocds/{0.year:d}{0.month:02d}.zip'
start = date(2009, 1, 1)
stop = date.today().replace(day=1)
for d in date_range_by_month(start, stop):
yield self.build_request(url.format(d), formatter=components(-1))
def parse_list(self, response):
pattern = 'https://admin.ims.susasan.org/ocds/json/dhangadhi-{}.json'
data = json.loads(response.text)
for item in data['data']['fiscal_years']:
yield self.build_request(pattern.format(item['name']),
formatter=components(-1))
if self.sample:
break
def parse_list(self, response):
urls = response.json()['packagesPerMonth']
netloc = urlsplit(response.request.url).netloc
for url in urls:
# URL looks like http://malta-demo-server.eurodyn.com/ocds/services/recordpackage/getrecordpackage/2020/1
yield self.build_request(
urlsplit(url)._replace(netloc=netloc).geturl(),
formatter=join(components(-2), extension='zip'))
def parse_list(self, response):
data = json.loads(response.text)
for resource in data['result']['resources']:
if 'ocds' in resource['description']:
# Presently, only one URL matches.
yield scrapy.Request(
resource['url'],
meta={'file_name': components(-1)(resource['url'])},
callback=self.parse_data)
def parse_list(self, response):
datas = json.loads(response.text)
for result in datas['result']['results']:
for resource in result['resources']:
if resource['format'] == 'JSON':
yield self.build_request(resource['url'],
formatter=components(-1),
meta={'dont_redirect': True},
callback=self.parse_redirect)
