Version bump to 0.8.8

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

homeharvest/core/scrapers/__init__.py

@@ -76,26 +76,24 @@ class Scraper:
                 total=3, backoff_factor=4, status_forcelist=[429, 403], allowed_methods=frozenset(["GET", "POST"])
             )
 
-            adapter = HTTPAdapter(max_retries=retries)
+            adapter = HTTPAdapter(max_retries=retries, pool_connections=10, pool_maxsize=20)
             Scraper.session.mount("http://", adapter)
             Scraper.session.mount("https://", adapter)
             Scraper.session.headers.update(
                 {
-                    "accept": "application/json, text/javascript",
-                    "accept-language": "en-US,en;q=0.9",
-                    "cache-control": "no-cache",
-                    "content-type": "application/json",
-                    "origin": "https://www.realtor.com",
-                    "pragma": "no-cache",
-                    "priority": "u=1, i",
-                    "rdc-ab-tests": "commute_travel_time_variation:v1",
-                    "sec-ch-ua": '"Not)A;Brand";v="99", "Google Chrome";v="127", "Chromium";v="127"',
-                    "sec-ch-ua-mobile": "?0",
-                    "sec-ch-ua-platform": '"Windows"',
-                    "sec-fetch-dest": "empty",
-                    "sec-fetch-mode": "cors",
-                    "sec-fetch-site": "same-origin",
-                    "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.0.0 Safari/537.36",
+                    'Host': 'api.frontdoor.realtor.com',
+                    'rdc-ab-test-client': 'ios_for_sale',
+                    'Content-Type': 'application/json',
+                    'apollographql-client-version': '26.9.25-26.9.25.0774600',
+                    'Accept': '*/*',
+                    'Accept-Language': 'en-US,en;q=0.9',
+                    'rdc-client-version': '26.9.25',
+                    'X-APOLLO-OPERATION-TYPE': 'query',
+                    'rdc-client-name': 'RDC_NATIVE_MOBILE-iPhone-com.move.Realtor',
+                    'apollographql-client-name': 'com.move.Realtor-apollo-ios',
+                    'newrelic': '',
+                    'transparent': '',
+                    'User-Agent': 'Realtor.com/26.9.25.0774600 CFNetwork/3860.200.71 Darwin/25.1.0',
                 }
             )
 

homeharvest/core/scrapers/realtor/__init__.py

@@ -26,7 +26,7 @@ from ..models import (
     ListingType,
     ReturnType
 )
-from .queries import GENERAL_RESULTS_QUERY, SEARCH_HOMES_DATA, HOMES_DATA, HOME_FRAGMENT
+from .queries import GENERAL_RESULTS_QUERY, SEARCH_HOMES_DATA, HOMES_DATA, HOME_FRAGMENT, SEARCH_RESULTS_FRAGMENT
 from .processors import (
     process_property,
     process_extra_property_details,
@@ -35,64 +35,128 @@ from .processors import (
 
 
 class RealtorScraper(Scraper):
-    SEARCH_GQL_URL = "https://www.realtor.com/api/v1/rdc_search_srp?client_id=rdc-search-new-communities&schema=vesta"
-    PROPERTY_URL = "https://www.realtor.com/realestateandhomes-detail/"
-    PROPERTY_GQL = "https://graph.realtor.com/graphql"
-    ADDRESS_AUTOCOMPLETE_URL = "https://parser-external.geo.moveaws.com/suggest"
+    SEARCH_GQL_URL = "https://api.frontdoor.realtor.com/graphql"
     NUM_PROPERTY_WORKERS = 20
     DEFAULT_PAGE_SIZE = 200
 
     def __init__(self, scraper_input):
         super().__init__(scraper_input)
 
-    def handle_location(self):
-        # Get client_id from listing_type
-        if self.listing_type is None:
-            client_id = "for-sale"
-        elif isinstance(self.listing_type, list):
-            client_id = self.listing_type[0].value.lower().replace("_", "-") if self.listing_type else "for-sale"
-        else:
-            client_id = self.listing_type.value.lower().replace("_", "-")
+    def _graphql_post(self, query: str, variables: dict, operation_name: str) -> dict:
+        """
+        Execute a GraphQL query with operation-specific headers.
 
-        params = {
-            "input": self.location,
-            "client_id": client_id,
-            "limit": "1",
-            "area_types": "city,state,county,postal_code,address,street,neighborhood,school,school_district,university,park",
+        Args:
+            query: GraphQL query string (must include operationName matching operation_name param)
+            variables: Query variables dictionary
+            operation_name: Name of the GraphQL operation for Apollo headers
+
+        Returns:
+            Response JSON dictionary
+        """
+        # Set operation-specific header (must match query's operationName)
+        self.session.headers['X-APOLLO-OPERATION-NAME'] = operation_name
+
+        payload = {
+            "operationName": operation_name,  # Include in payload
+            "query": query,
+            "variables": variables,
         }
 
-        response = self.session.get(
-            self.ADDRESS_AUTOCOMPLETE_URL,
-            params=params,
-        )
-        response_json = response.json()
+        response = self.session.post(self.SEARCH_GQL_URL, json=payload)
+        return response.json()
 
-        result = response_json["autocomplete"]
+    @retry(
+        retry=retry_if_exception_type(Exception),
+        wait=wait_exponential(multiplier=1, min=1, max=4),
+        stop=stop_after_attempt(3),
+    )
+    def handle_location(self):
+        query = """
+        fragment SuggestionFragment on SearchSuggestionGeoResult {
+            type
+            text
+            geo {
+                _id
+                area_type
+                city
+                state_code
+                postal_code
+                county
+                centroid { lat lon }
+                slug_id
+                geo_id
+            }
+        }
+        query SearchSuggestions($searchInput: SearchSuggestionsInput!) {
+            search_suggestions(search_input: $searchInput) {
+                geo_results {
+                    ...SuggestionFragment
+                }
+            }
+        }"""
 
-        if not result:
+        variables = {
+            "searchInput": {
+                "search_term": self.location
+            }
+        }
+
+        response_json = self._graphql_post(query, variables, "SearchSuggestions")
+
+        if (
+            response_json is None
+            or "data" not in response_json
+            or response_json["data"] is None
+            or "search_suggestions" not in response_json["data"]
+            or response_json["data"]["search_suggestions"] is None
+            or "geo_results" not in response_json["data"]["search_suggestions"]
+            or not response_json["data"]["search_suggestions"]["geo_results"]
+        ):
+            # If we got a 400 error with "Required parameter is missing", raise to trigger retry
+            if response_json and "errors" in response_json:
+                error_msgs = [e.get("message", "") for e in response_json.get("errors", [])]
+                if any("Required parameter is missing" in msg for msg in error_msgs):
+                    raise Exception(f"Transient API error: {error_msgs}")
             return None
 
-        return result[0]
+        geo_result = response_json["data"]["search_suggestions"]["geo_results"][0]
+        geo = geo_result.get("geo", {})
+
+        result = {
+            "text": geo_result.get("text"),
+            "area_type": geo.get("area_type"),
+            "city": geo.get("city"),
+            "state_code": geo.get("state_code"),
+            "postal_code": geo.get("postal_code"),
+            "county": geo.get("county"),
+            "centroid": geo.get("centroid"),
+        }
+
+        if geo.get("area_type") == "address":
+            geo_id = geo.get("_id", "")
+            if geo_id.startswith("addr:"):
+                result["mpr_id"] = geo_id.replace("addr:", "")
+
+        return result
 
     def get_latest_listing_id(self, property_id: str) -> str | None:
-        query = """query Property($property_id: ID!) {
+        query = """
+                fragment ListingFragment on Listing {
+                    listing_id
+                    primary
+                }
+                query GetPropertyListingId($property_id: ID!) {
                     property(id: $property_id) {
                         listings {
-                            listing_id
-                            primary
+                            ...ListingFragment
                         }
                     }
                 }
                 """
 
         variables = {"property_id": property_id}
-        payload = {
-            "query": query,
-            "variables": variables,
-        }
-
-        response = self.session.post(self.SEARCH_GQL_URL, json=payload)
-        response_json = response.json()
+        response_json = self._graphql_post(query, variables, "GetPropertyListingId")
 
         property_info = response_json["data"]["property"]
         if property_info["listings"] is None:
@@ -108,31 +172,43 @@ class RealtorScraper(Scraper):
             return property_info["listings"][0]["listing_id"]
 
     def handle_home(self, property_id: str) -> list[Property]:
+        """Fetch single home with proper error handling."""
         query = (
-            """query Home($property_id: ID!) {
-                    home(property_id: $property_id) %s
+            """%s
+                query GetHomeDetails($property_id: ID!) {
+                    home(property_id: $property_id) {
+                        ...HomeDetailsFragment
+                    }
                 }"""
-            % HOMES_DATA
+            % HOME_FRAGMENT
         )
 
         variables = {"property_id": property_id}
-        payload = {
-            "query": query,
-            "variables": variables,
-        }
 
-        response = self.session.post(self.SEARCH_GQL_URL, json=payload)
-        response_json = response.json()
+        try:
+            data = self._graphql_post(query, variables, "GetHomeDetails")
 
-        property_info = response_json["data"]["home"]
+            # Check for errors or missing data
+            if "errors" in data or "data" not in data:
+                return []
 
-        if self.return_type != ReturnType.raw:
-            return [process_property(property_info, self.mls_only, self.extra_property_data, 
-                                   self.exclude_pending, self.listing_type, get_key, process_extra_property_details)]
-        else:
-            return [property_info]
+            if data["data"] is None or "home" not in data["data"]:
+                return []
 
+            property_info = data["data"]["home"]
+            if property_info is None:
+                return []
 
+            # Process based on return type
+            if self.return_type != ReturnType.raw:
+                return [process_property(property_info, self.mls_only, self.extra_property_data,
+                                       self.exclude_pending, self.listing_type, get_key,
+                                       process_extra_property_details)]
+            else:
+                return [property_info]
+
+        except Exception:
+            return []
 
     def general_search(self, variables: dict, search_type: str) -> Dict[str, Union[int, Union[list[Property], list[dict]]]]:
         """
@@ -329,12 +405,13 @@ class RealtorScraper(Scraper):
             is_foreclosure = "foreclosure: false"
 
         if search_type == "comps":  #: comps search, came from an address
-            query = """query Property_search(
+            query = """%s
+                query GetHomeSearch(
                     $coordinates: [Float]!
                     $radius: String!
                     $offset: Int!,
                     ) {
-                        home_search(
+                        homeSearch: home_search(
                             query: {
                                 %s
                                 nearby: {
@@ -352,6 +429,7 @@ class RealtorScraper(Scraper):
                             offset: $offset
                     ) %s
                 }""" % (
+                SEARCH_RESULTS_FRAGMENT,
                 is_foreclosure,
                 status_param,
                 date_param,
@@ -362,20 +440,15 @@ class RealtorScraper(Scraper):
                 GENERAL_RESULTS_QUERY,
             )
         elif search_type == "area":  #: general search, came from a general location
-            query = """query Home_search(
-                                $city: String,
-                                $county: [String],
-                                $state_code: String,
-                                $postal_code: String
+            query = """%s
+                            query GetHomeSearch(
+                                $search_location: SearchLocation,
                                 $offset: Int,
                             ) {
-                                home_search(
+                                homeSearch: home_search(
                                     query: {
                                         %s
-                                        city: $city
-                                        county: $county
-                                        postal_code: $postal_code
-                                        state_code: $state_code
+                                        search_location: $search_location
                                         %s
                                         %s
                                         %s
@@ -388,6 +461,7 @@ class RealtorScraper(Scraper):
                                     offset: $offset
                                 ) %s
                             }""" % (
+                SEARCH_RESULTS_FRAGMENT,
                 is_foreclosure,
                 status_param,
                 date_param,
@@ -400,11 +474,12 @@ class RealtorScraper(Scraper):
             )
         else:  #: general search, came from an address
             query = (
-                """query Property_search(
+                """%s
+                    query GetHomeSearch(
                         $property_id: [ID]!
                         $offset: Int!,
                     ) {
-                        home_search(
+                        homeSearch: home_search(
                             query: {
                                 property_id: $property_id
                             }
@@ -412,17 +487,11 @@ class RealtorScraper(Scraper):
                             offset: $offset
                         ) %s
                     }"""
-                % GENERAL_RESULTS_QUERY
+                % (SEARCH_RESULTS_FRAGMENT, GENERAL_RESULTS_QUERY)
             )
 
-        payload = {
-            "query": query,
-            "variables": variables,
-        }
-
-        response = self.session.post(self.SEARCH_GQL_URL, json=payload)
-        response_json = response.json()
-        search_key = "home_search" if "home_search" in query else "property_search"
+        response_json = self._graphql_post(query, variables, "GetHomeSearch")
+        search_key = "homeSearch"
 
         properties: list[Union[Property, dict]] = []
 
@@ -511,24 +580,16 @@ class RealtorScraper(Scraper):
                 if not location_info.get("centroid"):
                     return []
 
-                coordinates = list(location_info["centroid"].values())
+                centroid = location_info["centroid"]
+                coordinates = [centroid["lon"], centroid["lat"]]  # GeoJSON order: [lon, lat]
                 search_variables |= {
                     "coordinates": coordinates,
                     "radius": "{}mi".format(self.radius),
                 }
 
-        elif location_type == "postal_code":
+        else:  #: general search (city, county, postal_code, etc.)
             search_variables |= {
-                "postal_code": location_info.get("postal_code"),
-            }
-
-        else:  #: general search, location
-            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"),
-
+                "search_location": {"location": location_info.get("text")},
             }
 
         if self.foreclosure:
@@ -1050,8 +1111,8 @@ class RealtorScraper(Scraper):
 
 
     @retry(
-        retry=retry_if_exception_type(JSONDecodeError),
-        wait=wait_exponential(min=4, max=10),
+        retry=retry_if_exception_type((JSONDecodeError, Exception)),
+        wait=wait_exponential(multiplier=1, min=1, max=10),
         stop=stop_after_attempt(3),
     )
     def get_bulk_prop_details(self, property_ids: list[str]) -> dict:
@@ -1064,21 +1125,24 @@ class RealtorScraper(Scraper):
 
         property_ids = list(set(property_ids))
 
-        # Construct the bulk query
         fragments = "\n".join(
-            f'home_{property_id}: home(property_id: {property_id}) {{ ...HomeData }}'
+            f'home_{property_id}: home(property_id: {property_id}) {{ ...HomeDetailsFragment }}'
             for property_id in property_ids
         )
         query = f"""{HOME_FRAGMENT}
-        
-        query GetHomes {{
-            {fragments}
-        }}"""
 
-        response = self.session.post(self.SEARCH_GQL_URL, json={"query": query})
-        data = response.json()
+query GetHomeDetails {{
+    {fragments}
+}}"""
+
+        data = self._graphql_post(query, {}, "GetHomeDetails")
 
         if "data" not in data:
+            # If we got a 400 error with "Required parameter is missing", raise to trigger retry
+            if data and "errors" in data:
+                error_msgs = [e.get("message", "") for e in data.get("errors", [])]
+                if any("Required parameter is missing" in msg for msg in error_msgs):
+                    raise Exception(f"Transient API error: {error_msgs}")
             return {}
 
         properties = data["data"]

homeharvest/core/scrapers/realtor/queries.py

@@ -1,3 +1,200 @@
+SEARCH_RESULTS_FRAGMENT = """
+fragment SearchFragment on SearchHome {
+    __typename
+    pending_date
+    listing_id
+    property_id
+    href
+    permalink
+    list_date
+    status
+    mls_status
+    last_sold_price
+    last_sold_date
+    last_status_change_date
+    last_update_date
+    list_price
+    list_price_max
+    list_price_min
+    price_per_sqft
+    tags
+    open_houses {
+        start_date
+        end_date
+        description
+        time_zone
+        dst
+        href
+        methods
+    }
+    details {
+        category
+        text
+        parent_category
+    }
+    pet_policy {
+        cats
+        dogs
+        dogs_small
+        dogs_large
+        __typename
+    }
+    units {
+        availability {
+          date
+          __typename
+        }
+        description {
+          baths_consolidated
+          baths
+          beds
+          sqft
+          __typename
+        }
+        photos(https: true) {
+            title
+            href
+            tags {
+                label
+            }
+        }
+        list_price
+        __typename
+    }
+    flags {
+        is_contingent
+        is_pending
+        is_new_construction
+    }
+    description {
+        type
+        sqft
+        beds
+        baths_full
+        baths_half
+        lot_sqft
+        year_built
+        garage
+        type
+        name
+        stories
+        text
+    }
+    source {
+        id
+        listing_id
+    }
+    hoa {
+        fee
+    }
+    location {
+        address {
+            street_direction
+            street_number
+            street_name
+            street_suffix
+            line
+            unit
+            city
+            state_code
+            postal_code
+            coordinate {
+                lon
+                lat
+            }
+        }
+        county {
+            name
+            fips_code
+        }
+        neighborhoods {
+            name
+        }
+    }
+    tax_record {
+        cl_id
+        public_record_id
+        last_update_date
+        apn
+        tax_parcel_id
+    }
+    primary_photo(https: true) {
+        href
+    }
+    photos(https: true) {
+        title
+        href
+        tags {
+            label
+        }
+    }
+    advertisers {
+        email
+        broker {
+            name
+            fulfillment_id
+        }
+        type
+        name
+        fulfillment_id
+        builder {
+            name
+            fulfillment_id
+        }
+        phones {
+            ext
+            primary
+            type
+            number
+        }
+        office {
+            name
+            email
+            fulfillment_id
+            href
+            phones {
+                number
+                type
+                primary
+                ext
+            }
+            mls_set
+        }
+        corporation {
+            specialties
+            name
+            bio
+            href
+            fulfillment_id
+        }
+        mls_set
+        nrds_id
+        state_license
+        rental_corporation {
+            fulfillment_id
+        }
+        rental_management {
+            name
+            href
+            fulfillment_id
+        }
+    }
+    current_estimates {
+        __typename
+        source {
+            __typename
+            type
+            name
+        }
+        estimate
+        estimateHigh: estimate_high
+        estimateLow: estimate_low
+        date
+        isBestHomeValue: isbest_homevalue
+    }
+}
+"""
+
 _SEARCH_HOMES_DATA_BASE = """{
     pending_date
     listing_id
@@ -181,7 +378,7 @@ _SEARCH_HOMES_DATA_BASE = """{
 
 
 HOME_FRAGMENT = """
-fragment HomeData on Home {
+fragment HomeDetailsFragment on Home {
     property_id
     nearbySchools: nearby_schools(radius: 5.0, limit_per_level: 3) {
         __typename schools { district { __typename id name } }
@@ -300,8 +497,19 @@ current_estimates {
 }
 }""" % _SEARCH_HOMES_DATA_BASE
 
-GENERAL_RESULTS_QUERY = """{
+# Query body using inline fields (kept for backward compatibility)
+GENERAL_RESULTS_QUERY_BODY = """{
                             count
                             total
                             results %s
                         }""" % SEARCH_HOMES_DATA
+
+GENERAL_RESULTS_QUERY = """{
+                            __typename
+                            count
+                            total
+                            results {
+                                __typename
+                                ...SearchFragment
+                            }
+                        }"""

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "homeharvest"
-version = "0.8.5"
+version = "0.8.8"
 description = "Real estate scraping library"
 authors = ["Zachary Hampton <zachary@bunsly.com>", "Cullen Watson <cullen@bunsly.com>"]
 homepage = "https://github.com/ZacharyHampton/HomeHarvest"

tests/test_realtor.py

@@ -1,4 +1,5 @@
 import pytz
+from concurrent.futures import ThreadPoolExecutor, as_completed
 
 from homeharvest import scrape_property, Property
 import pandas as pd
@@ -87,6 +88,25 @@ def test_realtor_date_range_sold():
     )
 
 
+def test_listing_type_none_includes_sold():
+    """Test that listing_type=None includes sold listings (issue #142)"""
+    # Get properties with listing_type=None (should include all common types)
+    result_none = scrape_property(
+        location="Warren, MI",
+        listing_type=None
+    )
+
+    # Verify we got results
+    assert result_none is not None and len(result_none) > 0
+
+    # Verify sold listings are included
+    status_types = set(result_none['status'].unique())
+    assert 'SOLD' in status_types, "SOLD listings should be included when listing_type=None"
+
+    # Verify we get multiple listing types (not just one)
+    assert len(status_types) > 1, "Should return multiple listing types when listing_type=None"
+
+
 def test_realtor_single_property():
     results = [
         scrape_property(
@@ -288,6 +308,30 @@ def test_phone_number_matching():
     assert row["agent_phones"].values[0] == matching_row["agent_phones"].values[0]
 
 
+def test_parallel_search_consistency():
+    """Test that the same search executed 3 times in parallel returns consistent results"""
+    def search_task():
+        return scrape_property(
+            location="Phoenix, AZ",
+            listing_type="for_sale",
+            limit=100
+        )
+
+    with ThreadPoolExecutor(max_workers=3) as executor:
+        futures = [executor.submit(search_task) for _ in range(3)]
+        results = [future.result() for future in as_completed(futures)]
+
+    # Verify all results are valid
+    assert all([result is not None for result in results])
+    assert all([isinstance(result, pd.DataFrame) for result in results])
+    assert all([len(result) > 0 for result in results])
+
+    # Verify all results have the same length (primary consistency check)
+    lengths = [len(result) for result in results]
+    assert len(set(lengths)) == 1, \
+        f"All parallel searches should return same number of results, got lengths: {lengths}"
+
+
 def test_return_type():
     results = {
         "pandas": [scrape_property(location="Surprise, AZ", listing_type="for_rent", limit=100)],