- finished realtor

homeharvest/__init__.py

@@ -69,9 +69,9 @@ def process_result(result: Union[Building, Property]) -> pd.DataFrame:
     prop_data = result.__dict__
 
     address_data = prop_data["address"]
-    prop_data["site_name"] = prop_data["site_name"].value
+    prop_data["site_name"] = prop_data["site_name"]
     prop_data["listing_type"] = prop_data["listing_type"].value
-    prop_data["property_type"] = prop_data["property_type"].value.lower()
+    prop_data["property_type"] = prop_data["property_type"].value.lower() if prop_data["property_type"] else None
     prop_data["address_one"] = address_data.address_one
     prop_data["city"] = address_data.city
     prop_data["state"] = address_data.state
@@ -90,13 +90,13 @@ def scrape_property(
     location: str,
     site_name: str,
     listing_type: str = "for_sale",  #: for_sale, for_rent, sold
-) -> Union[list[Building], list[Property]]:
+) -> pd.DataFrame:
     validate_input(site_name, listing_type)
 
     scraper_input = ScraperInput(
         location=location,
         listing_type=ListingType[listing_type.upper()],
-        site_name=SiteName[site_name.upper()],
+        site_name=site_name.lower(),
     )
 
     site = _scrapers[site_name.lower()](scraper_input)

homeharvest/core/scrapers/__init__.py

@@ -7,13 +7,15 @@ from .models import Property, ListingType, SiteName
 class ScraperInput:
     location: str
     listing_type: ListingType
-    site_name: SiteName
+    site_name: str
     proxy_url: str | None = None
 
 
 class Scraper:
     def __init__(self, scraper_input: ScraperInput):
         self.location = scraper_input.location
+        self.listing_type = scraper_input.listing_type
+
         self.session = requests.Session()
         self.listing_type = scraper_input.listing_type
         self.site_name = scraper_input.site_name

homeharvest/core/scrapers/models.py

@@ -53,7 +53,7 @@ class Address:
 
 @dataclass()
 class Realty:
-    site_name: SiteName
+    site_name: str
     address: Address
     url: str
     listing_type: ListingType | None = None
@@ -68,7 +68,6 @@ class Property(Realty):
     year_built: int | None = None
     square_feet: int | None = None
     price_per_square_foot: int | None = None
-    year_built: int | None = None
     mls_id: str | None = None
 
     agent_name: str | None = None

homeharvest/core/scrapers/realtor/__init__.py

@@ -1,12 +1,15 @@
 import json
 from ..models import Property, Address
 from .. import Scraper
-from typing import Any
+from typing import Any, Generator
+from ....exceptions import NoResultsFound
+from concurrent.futures import ThreadPoolExecutor, as_completed
 
 
 class RealtorScraper(Scraper):
     def __init__(self, scraper_input):
         super().__init__(scraper_input)
+        self.search_url = "https://www.realtor.com/api/v1/rdc_search_srp?client_id=rdc-search-new-communities&schema=vesta"
 
     def handle_location(self):
         headers = {
@@ -26,7 +29,7 @@ class RealtorScraper(Scraper):
 
         params = {
             "input": self.location,
-            "client_id": "for-sale",
+            "client_id": self.listing_type.value.replace('_', '-'),
             "limit": "1",
             "area_types": "city,state,county,postal_code,address,street,neighborhood,school,school_district,university,park",
         }
@@ -38,14 +41,228 @@ class RealtorScraper(Scraper):
         )
         response_json = response.json()
 
-        return response_json["autocomplete"][0]
+        result = response_json["autocomplete"]
+
+        if result is None:
+            raise NoResultsFound("No results found for location: " + self.location)
+
+        return result[0]
+
+    def handle_address(self, property_id: str) -> list[Property]:
+        query = """query Property($property_id: ID!) {
+                    property(id: $property_id) {
+                        property_id
+                        details {
+                            date_updated
+                            garage
+                            permalink
+                            year_built
+                            stories
+                        }
+                        address {
+                            address_validation_code
+                            city
+                            country
+                            county
+                            line
+                            postal_code
+                            state_code
+                            street_direction
+                            street_name
+                            street_number
+                            street_suffix
+                            street_post_direction
+                            unit_value
+                            unit
+                            unit_descriptor
+                            zip
+                        }
+                        basic {
+                            baths
+                            beds
+                            price
+                            sqft
+                            lot_sqft
+                            type
+                            sold_price
+                        }
+                        public_record {
+                            lot_size
+                            sqft
+                            stories
+                            units
+                            year_built
+                        }
+                    }
+                }"""
+
+        variables = {
+            'property_id': property_id
+        }
+
+        payload = {
+            'query': query,
+            'variables': variables,
+        }
+
+        response = self.session.post(self.search_url, json=payload)
+        response_json = response.json()
+
+        property_info = response_json['data']['property']
+
+        return [Property(
+            site_name=self.site_name,
+            address=Address(
+                address_one=property_info['address']['line'],
+                city=property_info['address']['city'],
+                state=property_info['address']['state_code'],
+                zip_code=property_info['address']['postal_code'],
+            ),
+            url="https://www.realtor.com/realestateandhomes-detail/" + property_info['details']['permalink'],
+            beds=property_info['basic']['beds'],
+            baths=property_info['basic']['baths'],
+            stories=property_info['details']['stories'],
+            year_built=property_info['details']['year_built'],
+            square_feet=property_info['basic']['sqft'],
+            price_per_square_foot=property_info['basic']['price'] / property_info['basic']['sqft']
+            if property_info['basic']['sqft'] is not None and
+            property_info['basic']['price'] is not None
+            else None,
+            price=property_info['basic']['price'],
+            mls_id=property_id,
+            listing_type=self.listing_type,
+            lot_size=property_info['public_record']['lot_size'] if property_info['public_record'] is not None else None,
+        )]
+
+    def handle_area(self, variables: dict, return_total: bool = False) -> list[Property] | int:
+        query = """query Home_search(
+                            $city: String,
+                            $county: [String],
+                            $state_code: String,
+                            $postal_code: String
+                            $offset: Int,
+                        ) {
+                            home_search(
+                                query: {
+                                    city: $city
+                                    county: $county
+                                    postal_code: $postal_code
+                                    state_code: $state_code
+                                    status: %s
+                                }
+                                limit: 200
+                                offset: $offset
+                            ) {
+                                count
+                                total
+                                results {
+                                    property_id
+                                    description {
+                                        baths
+                                        beds
+                                        lot_sqft
+                                        sqft
+                                        text
+                                        sold_price
+                                        stories
+                                        year_built
+                                        garage
+                                        unit_number
+                                        floor_number
+                                    }
+                                    location {
+                                        address {
+                                            city
+                                            country
+                                            line
+                                            postal_code
+                                            state_code
+                                            state
+                                            street_direction
+                                            street_name
+                                            street_number
+                                            street_post_direction
+                                            street_suffix
+                                            unit
+                                        }
+                                    }
+                                    list_price
+                                    price_per_sqft
+                                    source {
+                                        id
+                                    }
+                                }
+                            }
+                        }""" % self.listing_type.value
+
+        payload = {
+            'query': query,
+            'variables': variables,
+        }
+
+        response = self.session.post(self.search_url, json=payload)
+        response_json = response.json()
+
+        if return_total:
+            return response_json['data']['home_search']['total']
+
+        properties: list[Property] = []
+
+        for result in response_json['data']['home_search']['results']:
+            realty_property = Property(
+                address=Address(
+                    address_one=result['location']['address']['line'],
+                    city=result['location']['address']['city'],
+                    state=result['location']['address']['state_code'],
+                    zip_code=result['location']['address']['postal_code'],
+                    address_two=result['location']['address']['unit'],
+                ),
+                site_name=self.site_name,
+                url="https://www.realtor.com/realestateandhomes-detail/" + result['property_id'],
+                beds=result['description']['beds'],
+                baths=result['description']['baths'],
+                stories=result['description']['stories'],
+                year_built=result['description']['year_built'],
+                square_feet=result['description']['sqft'],
+                price_per_square_foot=result['price_per_sqft'],
+                price=result['list_price'],
+                mls_id=result['property_id'],
+                listing_type=self.listing_type,
+                lot_size=result['description']['lot_sqft'],
+            )
+
+            properties.append(realty_property)
+
+        return properties
 
     def search(self):
         location_info = self.handle_location()
         location_type = location_info["area_type"]
 
-        """
+        if location_type == 'address':
-        property types:
+            property_id = location_info['mpr_id']
-        apartment + building + commercial + condo_townhome + condo_townhome_rowhome_coop + condos + coop + duplex_triplex + farm + investment + land + mobile + multi_family + rental + single_family + townhomes
+            return self.handle_address(property_id)
-        """
+
-        print("a")
+        offset = 0
+        search_variables = {
+            'city': location_info.get('city'),
+            'county': location_info.get('county'),
+            'state_code': location_info.get('state_code'),
+            'postal_code': location_info.get('postal_code'),
+            'offset': offset,
+        }
+
+        total = self.handle_area(search_variables, return_total=True)
+
+        homes = []
+        with ThreadPoolExecutor(max_workers=10) as executor:
+            futures = [
+                executor.submit(
+                    self.handle_area, variables=search_variables | {'offset': i}, return_total=False
+                ) for i in range(0, total, 200)
+            ]
+
+            for future in as_completed(futures):
+                homes.extend(future.result())
+
+        return homes

tests/test_realtor.py

@@ -3,6 +3,9 @@ from homeharvest import scrape_property
 
 def test_realtor():
     results = [
+        scrape_property(location="2530 Al Lipscomb Way", site_name="realtor.com"),
+        scrape_property(location="Phoenix, AZ", site_name="realtor.com"),  #: does not support "city, state, USA" format
+        scrape_property(location="Dallas, TX", site_name="realtor.com"),  #: does not support "city, state, USA" format
         scrape_property(location="85281", site_name="realtor.com"),
     ]