fix(spiders): use date filters as part of file names

kingfisher_scrapy/spiders/australia.py

@@ -32,7 +32,8 @@ class Australia(LinksSpider):
     formatter = staticmethod(parameters('cursor'))
 
     def start_requests(self):
-        url = f'https://api.tenders.gov.au/ocds/findByDates/contractPublished/' \
-              f'{self.from_date.strftime(self.date_format)}Z/{self.until_date.strftime(self.date_format)}Z'
+        from_date = self.from_date.strftime(self.date_format)
+        until_date = self.until_date.strftime(self.date_format)
+        url = f'https://api.tenders.gov.au/ocds/findByDates/contractPublished/{from_date}Z/{until_date}Z'
 
-        yield scrapy.Request(url, meta={'file_name': 'start.json'})
+        yield scrapy.Request(url, meta={'file_name': f'{until_date}.json'})  # reverse chronological order

kingfisher_scrapy/spiders/colombia_api.py

@@ -32,7 +32,8 @@ class ColombiaAPI(LinksSpider):
     formatter = staticmethod(parameters('_id'))
 
     def start_requests(self):
-        url = 'https://apiocds.colombiacompra.gov.co/apiCCE2.0/rest/releases/dates/' \
-              f'{self.from_date.strftime(self.date_format)}/{self.until_date.strftime(self.date_format)}'
+        from_date = self.from_date.strftime(self.date_format)
+        until_date = self.until_date.strftime(self.date_format)
+        url = f'https://apiocds.colombiacompra.gov.co/apiCCE2.0/rest/releases/dates/{from_date}/{until_date}'
 
-        yield scrapy.Request(url, meta={'file_name': 'page-1.json'})
+        yield scrapy.Request(url, meta={'file_name': f'{from_date}.json'})

kingfisher_scrapy/spiders/kyrgyzstan.py

@@ -33,6 +33,10 @@ class Kyrgyzstan(LinksSpider):
     def start_requests(self):
         url = 'http://ocds.zakupki.gov.kg/api/tendering'
         if self.from_date:
+            from_date = self.from_date.strftime(self.date_format)
             # The API requires the timezone and seconds in the since parameter.
-            url = f'{url}?since={self.from_date.strftime(self.date_format)}.00%2B06:00'
-        yield scrapy.Request(url, meta={'file_name': 'start.json'})
+            url = f'{url}?since={from_date}.00%2B06:00'
+            self.formatter = staticmethod(parameters('offset', 'since'))
+        else:
+            from_date = '1970-01-01T00:00:00'
+        yield scrapy.Request(url, meta={'file_name': f'{from_date}.json'})

kingfisher_scrapy/spiders/portugal_base.py

@@ -20,10 +20,13 @@ class PortugalBase(LinksSpider):
     def start_requests(self):
         url = self.start_url
         if self.from_date and self.until_date:
-            url = f'{url}?contractStartDate={self.from_date.strftime(self.date_format)}' \
-                  f'&contractEndDate={self.until_date.strftime(self.date_format)}'
-
-        yield scrapy.Request(url, meta={'file_name': 'offset-1.json'})
+            from_date = self.from_date.strftime(self.date_format)
+            until_date = self.until_date.strftime(self.date_format)
+            url = f'{url}?contractStartDate={from_date}&contractEndDate={until_date}'
+            self.formatter = staticmethod(parameters('offset',  'contractStartDate'))
+        else:
+            from_date = self.default_from_date
+        yield scrapy.Request(url, meta={'file_name': f'{from_date}.json'})
 
     def is_http_retryable(self, response):
         return response.status != 404

kingfisher_scrapy/spiders/south_africa_national_treasury_api.py

@@ -27,8 +27,10 @@ class SouthAfricaNationalTreasuryAPI(LinksSpider):
     data_type = 'release_package'
 
     # LinksSpider
-    formatter = staticmethod(parameters('PageNumber'))
+    formatter = staticmethod(parameters('PageNumber', 'dateFrom'))
 
     def start_requests(self):
+
         yield scrapy.Request('https://ocds-api.etenders.gov.za/api/OCDSReleases?PageNumber=1&PageSize=50&'
-                             f'dateFrom={self.from_date}&dateTo={self.until_date}', meta={'file_name': 'start.json'})
+                             f'dateFrom={self.from_date}&dateTo={self.until_date}',
+                             meta={'file_name': f'{self.from_date}.json'})

kingfisher_scrapy/spiders/united_kingdom_contracts_finder_base.py

@@ -1,3 +1,5 @@
+from datetime import datetime
+
 import scrapy
 
 from kingfisher_scrapy.base_spiders import LinksSpider
@@ -12,7 +14,6 @@ class UnitedKingdomContractsFinderBase(LinksSpider):
 
     # BaseSpider
     date_format = 'datetime'
-    date_required = True
     default_from_date = '2014-01-01T00:00:00'
     encoding = 'iso-8859-1'
     max_attempts = 5
@@ -32,8 +33,10 @@ def start_requests(self):
             from_date = self.from_date.strftime(self.date_format)
             until_date = self.until_date.strftime(self.date_format)
             url = f'{url}&publishedFrom={from_date}&publishedTo={until_date}'
-
-        yield scrapy.Request(url, meta={'file_name': 'page-1.json'}, callback=self.parse_page)
+        else:
+            until_date = datetime.utcnow().strftime(self.date_format)
+        yield scrapy.Request(url, meta={'file_name': f'{until_date}.json'},  # reverse chronological order
+                             callback=self.parse_page)
 
     @handle_http_error
     def parse(self, response):

kingfisher_scrapy/spiders/united_kingdom_fts.py

@@ -1,3 +1,5 @@
+from datetime import datetime
+
 import scrapy
 
 from kingfisher_scrapy.base_spiders import LinksSpider
@@ -36,8 +38,10 @@ def start_requests(self):
             from_date = self.from_date.strftime(self.date_format)
             until_date = self.until_date.strftime(self.date_format)
             url = f'{url}?updatedFrom={from_date}&updatedTo={until_date}'
-
-        yield scrapy.Request(url, meta={'file_name': 'start.json'}, headers={'Accept': 'application/json'})
+        else:
+            until_date = datetime.utcnow().strftime(self.date_format)
+        yield scrapy.Request(url, meta={'file_name': f'{until_date}.json'},  # reverse chronological order
+                             headers={'Accept': 'application/json'})
 
     @handle_http_error
     def parse(self, response):