Merge pull request #8 from ZacharyHampton/fix/zillow-location-validation

- zillow location validation

homeharvest/core/scrapers/zillow/__init__.py

@@ -10,27 +10,26 @@
 class ZillowScraper(Scraper):
     def __init__(self, scraper_input):
         super().__init__(scraper_input)
-        self.listing_type = scraper_input.listing_type
+
         if not self.is_plausible_location(self.location):
             raise NoResultsFound("Invalid location input: {}".format(self.location))
+
         if self.listing_type == ListingType.FOR_SALE:
             self.url = f"https://www.zillow.com/homes/for_sale/{self.location}_rb/"
         elif self.listing_type == ListingType.FOR_RENT:
             self.url = f"https://www.zillow.com/homes/for_rent/{self.location}_rb/"
         else:
             self.url = f"https://www.zillow.com/homes/recently_sold/{self.location}_rb/"
 
-    @staticmethod
-    def is_plausible_location(location: str) -> bool:
-        blocks = location.split()
-        for block in blocks:
-            if (
-                any(char.isdigit() for char in block)
-                and any(char.isalpha() for char in block)
-                and len(block) > 6
-            ):
-                return False
-        return True
+    def is_plausible_location(self, location: str) -> bool:
+        url = ('https://www.zillowstatic.com/autocomplete/v3/suggestions?q={'
+               '}&abKey=6666272a-4b99-474c-b857-110ec438732b&clientId=homepage-render').format(
+            location
+        )
+
+        response = self.session.get(url)
+
+        return response.json()['results'] != []
 
     def search(self):
         resp = self.session.get(self.url, headers=self._get_headers())

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "homeharvest"
-version = "0.2.1"
+version = "0.2.2"
 description = "Real estate scraping library supporting Zillow, Realtor.com & Redfin."
 authors = ["Zachary Hampton <[email protected]>", "Cullen Watson <[email protected]>"]
 homepage = "https://github.com/ZacharyHampton/HomeHarvest"
@@ -10,6 +10,7 @@ readme = "README.md"
 python = "^3.10"
 requests = "^2.31.0"
 pandas = "^2.1.0"
+openpyxl = "^3.1.2"
 
 
 [tool.poetry.group.dev.dependencies]