fix: normalize unit num

homeharvest/__init__.py

@@ -120,7 +120,7 @@ def _scrape_single_site(
 
 def scrape_property(
     location: str,
-    site_name: Union[str, list[str]],
+    site_name: Union[str, list[str]] = list(_scrapers.keys()),
     listing_type: str = "for_sale",
 ) -> pd.DataFrame:
     """
@@ -139,7 +139,9 @@ def scrape_property(
         site_name = [site_name]
 
     if len(site_name) == 1:
-        return _scrape_single_site(location, site_name[0], listing_type)
+        final_df = _scrape_single_site(location, site_name[0], listing_type)
+        final_df = final_df.drop_duplicates(subset="street_address", keep="first")
+        return final_df
 
     results = []
     with ThreadPoolExecutor() as executor:

homeharvest/core/scrapers/realtor/__init__.py

@@ -240,7 +240,7 @@ class RealtorScraper(Scraper):
                     city=result["location"]["address"]["city"],
                     state=result["location"]["address"]["state_code"],
                     zip_code=result["location"]["address"]["postal_code"],
-                    unit=result["location"]["address"]["unit"],
+                    unit=parse_address_two(result["location"]["address"]["unit"]),
                     country="USA",
                 ),
                 site_name=self.site_name,

homeharvest/core/scrapers/zillow/__init__.py

@@ -130,7 +130,7 @@ class ZillowScraper(Scraper):
                 home_info = result["hdpData"]["homeInfo"]
                 address_data = {
                     "street_address": home_info["streetAddress"],
-                    "unit": home_info.get("unit"),
+                    "unit": parse_address_two(home_info['unit']) if 'unit' in home_info else None,
                     "city": home_info["city"],
                     "state": home_info["state"],
                     "zip_code": home_info["zipcode"],
@@ -213,22 +213,6 @@ class ZillowScraper(Scraper):
 
         return properties_list
 
-    def _extract_units(self, result: dict):
-        units = {}
-        if "units" in result:
-            num_units = result.get("availabilityCount", len(result["units"]))
-            prices = [
-                int(unit["price"].replace("$", "").replace(",", "").split("+")[0])
-                for unit in result["units"]
-            ]
-            units["apt_availability_count"] = num_units
-            units["apt_min_unit_price"] = min(prices)
-            units["apt_max_unit_price"] = max(prices)
-            units["apt_avg_unit_price"] = (
-                sum(prices) // num_units if num_units else None
-            )
-        return units
-
     def _get_single_property_page(self, property_data: dict):
         """
         This method is used when a user enters the exact location & zillow returns just one property
@@ -239,10 +223,9 @@ class ZillowScraper(Scraper):
             else property_data["hdpUrl"]
         )
         address_data = property_data["address"]
-        unit = parse_address_two(address_data["streetAddress"])
         address = Address(
             street_address=address_data["streetAddress"],
-            unit=unit,
+            unit=parse_address_two(address_data["streetAddress"]),
             city=address_data["city"],
             state=address_data["state"],
             zip_code=address_data["zipcode"],
@@ -301,11 +284,10 @@ class ZillowScraper(Scraper):
         else:
             raise ValueError(f"Unexpected state/zip format in address: {address_str}")
 
-        unit = parse_address_two(street_address)
         return Address(
             street_address=street_address,
             city=city,
-            unit=unit,
+            unit=parse_address_two(street_address),
             state=state,
             zip_code=zip_code,
             country="USA",

homeharvest/utils.py

@@ -1,6 +1,20 @@
 import re
 
 
-def parse_address_two(address_one: str):
-    apt_match = re.search(r"(APT\s*.+|#[\s\S]+)$", address_one, re.I)
-    return apt_match.group().strip() if apt_match else None
+def parse_address_two(street_address: str):
+    if not street_address:
+        return None
+    apt_match = re.search(r"(APT\s*[\dA-Z]+|#[\dA-Z]+|UNIT\s*[\dA-Z]+|LOT\s*[\dA-Z]+)$", street_address, re.I)
+
+    if apt_match:
+        apt_str = apt_match.group().strip()
+        apt_str = re.sub(r"(APT\s*|UNIT\s*|LOT\s*)", "#", apt_str, flags=re.I)
+        return apt_str
+    else:
+        return None
+
+if __name__ == "__main__":
+    print(parse_address_two("810 E Colter St APT 32"))
+    print(parse_address_two("1234 Elm Street apt 2B"))
+    print(parse_address_two("1234 Elm Street UNIT 3A"))
+    print(parse_address_two("1234 Elm Street unit 3A"))