Kirkland scraper

ca_qc_kirkland/people.py

@@ -8,9 +8,18 @@
 
 class KirklandPersonScraper(CanadianScraper):
     def scrape(self):
-        page = self.lxmlize(COUNCIL_PAGE)
+        def decode_email(e):
+            de = ""
+            k = int(e[:2], 16)
 
-        councillors = page.xpath('//div[@class="container_content"]//tbody/tr')
+            for i in range(2, len(e) - 1, 2):
+                de += chr(int(e[i : i + 2], 16) ^ k)
+
+            return de
+
+        page = self.lxmlize(COUNCIL_PAGE, "iso-8859-1")
+
+        councillors = page.xpath('//table/tbody[not(@id)]/tr/td[@valign="top"]')
         assert len(councillors), "No councillors found"
         for councillor in councillors:
             if councillor == councillors[0]:
@@ -23,19 +32,24 @@ def scrape(self):
 
             name = councillor.xpath(".//strong/text()")[0]
 
+            # Using self.get_phone does not include the extension #
             phone = (
                 councillor.xpath('.//div[contains(text(), "#")]/text()')[0]
                 .replace("T ", "")
                 .replace(" ", "-")
-                .replace(".", ",")  # correcting a typo
+                .replace(".", ",")
                 .replace(",-#-", " x")
             )
-            email = self.get_email(councillor)
+            encrypted_email = councillor.xpath('.//@href[contains(., "email")]')[0].split("#")[1]
+            email = decode_email(encrypted_email)
 
+            # cloudflare encrypts the email data
+            email = councillor.xpath(".//div/*/*/@href | .//div/*/@href | .//@href")[0]
+            decoded_email = decode_email(email.split("#", 1)[1])
             p = Person(primary_org="legislature", name=name, district=district, role=role)
             p.add_source(COUNCIL_PAGE)
             p.add_contact("voice", phone, "legislature")
-            p.add_contact("email", email)
+            p.add_contact("email", decoded_email)
             image = councillor.xpath(".//img/@src")
             if image:
                 p.image = image[0]