- pending date, property type fields (#45)

- alt photos bug fix (#57)

README.md

@@ -11,8 +11,6 @@
 - **Source**: Fetches properties directly from **Realtor.com**.
 - **Data Format**: Structures data to resemble MLS listings.
 - **Export Flexibility**: Options to save as either CSV or Excel.
-- **Usage Modes**:
-  - **Python**: For those who'd like to integrate scraping into their Python scripts.
 
 [Video Guide for HomeHarvest](https://youtu.be/J1qgNPgmSLI) - _updated for release v0.3.4_
 
@@ -21,7 +19,7 @@
 ## Installation
 
 ```bash
-pip install homeharvest
+pip install -U homeharvest
 ```
   _Python version >= [3.10](https://www.python.org/downloads/release/python-3100/) required_ 
 
@@ -126,6 +124,7 @@ Property
 │ ├── days_on_mls
 │ ├── list_price
 │ ├── list_date
+│ ├── pending_date
 │ ├── sold_price
 │ ├── last_sold_date
 │ ├── price_per_sqft

homeharvest/core/scrapers/models.py

@@ -23,6 +23,27 @@ class ListingType(Enum):
     SOLD = "SOLD"
 
 
+class PropertyType(Enum):
+    APARTMENT = "APARTMENT"
+    BUILDING = "BUILDING"
+    COMMERCIAL = "COMMERCIAL"
+    CONDO_TOWNHOME = "CONDO_TOWNHOME"
+    CONDO_TOWNHOME_ROWHOME_COOP = "CONDO_TOWNHOME_ROWHOME_COOP"
+    CONDO = "CONDO"
+    CONDOS = "CONDOS"
+    COOP = "COOP"
+    DUPLEX_TRIPLEX = "DUPLEX_TRIPLEX"
+    FARM = "FARM"
+    INVESTMENT = "INVESTMENT"
+    LAND = "LAND"
+    MOBILE = "MOBILE"
+    MULTI_FAMILY = "MULTI_FAMILY"
+    RENTAL = "RENTAL"
+    SINGLE_FAMILY = "SINGLE_FAMILY"
+    TOWNHOMES = "TOWNHOMES"
+    OTHER = "OTHER"
+
+
 @dataclass
 class Address:
     street: str | None = None
@@ -36,7 +57,7 @@ class Address:
 class Description:
     primary_photo: str | None = None
     alt_photos: list[str] | None = None
-    style: str | None = None
+    style: PropertyType | None = None
     beds: int | None = None
     baths_full: int | None = None
     baths_half: int | None = None
@@ -58,6 +79,7 @@ class Property:
 
     list_price: int | None = None
     list_date: str | None = None
+    pending_date: str | None = None
     last_sold_date: str | None = None
     prc_sqft: int | None = None
     hoa_fee: int | None = None

homeharvest/core/scrapers/realtor/__init__.py

@@ -9,7 +9,7 @@ from typing import Dict, Union, Optional
 from concurrent.futures import ThreadPoolExecutor, as_completed
 
 from .. import Scraper
-from ..models import Property, Address, ListingType, Description
+from ..models import Property, Address, ListingType, Description, PropertyType
 
 
 class RealtorScraper(Scraper):
@@ -84,13 +84,12 @@ class RealtorScraper(Scraper):
                             garage
                             permalink
                         }
-                        primary_photo {
-                            href
-                        }
+                        media {
                             photos {
                                 href
                             }
                         }
+                    }
                 }"""
 
         variables = {"listing_id": listing_id}
@@ -120,9 +119,11 @@ class RealtorScraper(Scraper):
             "list_date") else None
         last_sold_date_str = property_info["basic"]["sold_date"].split("T")[0] if property_info["basic"].get(
             "sold_date") else None
+        pending_date_str = property_info["pending_date"].split("T")[0] if property_info.get("pending_date") else None
 
         list_date = datetime.strptime(list_date_str, "%Y-%m-%d") if list_date_str else None
         last_sold_date = datetime.strptime(last_sold_date_str, "%Y-%m-%d") if last_sold_date_str else None
+        pending_date = datetime.strptime(pending_date_str, "%Y-%m-%d") if pending_date_str else None
         today = datetime.now()
 
         days_on_mls = None
@@ -150,6 +151,7 @@ class RealtorScraper(Scraper):
                and property_info["basic"].get("sqft")
             else None,
             last_sold_date=last_sold_date,
+            pending_date=pending_date,
             latitude=property_info["address"]["location"]["coordinate"].get("lat")
             if able_to_get_lat_long
             else None,
@@ -158,8 +160,7 @@ class RealtorScraper(Scraper):
             else None,
             address=self._parse_address(property_info, search_type="handle_listing"),
             description=Description(
-                primary_photo=property_info["primary_photo"].get("href", "").replace("s.jpg", "od-w480_h360_x2.webp?w=1080&q=75"),
-                alt_photos=self.process_alt_photos(property_info.get("photos", [])),
+                alt_photos=self.process_alt_photos(property_info.get("media", {}).get("photos", [])),
                 style=property_info["basic"].get("type", "").upper(),
                 beds=property_info["basic"].get("beds"),
                 baths_full=property_info["basic"].get("baths_full"),
@@ -297,6 +298,7 @@ class RealtorScraper(Scraper):
                             count
                             total
                             results {
+                                pending_date
                                 property_id
                                 list_date
                                 status
@@ -309,6 +311,7 @@ class RealtorScraper(Scraper):
                                     is_pending
                                 }
                                 description {
+                                    type
                                     sqft
                                     beds
                                     baths_full
@@ -385,10 +388,9 @@ class RealtorScraper(Scraper):
         listing_type = ListingType.FOR_SALE if self.listing_type == ListingType.PENDING else self.listing_type
         is_foreclosure = ""
 
-        if 'foreclosure' in variables and variables['foreclosure'] == True:
+        if variables.get('foreclosure') is True:
             is_foreclosure = "foreclosure: true"
-        
-        if 'foreclosure' in variables and variables['foreclosure'] == False:
+        elif variables.get('foreclosure') is False:
             is_foreclosure = "foreclosure: false"
 
         if search_type == "comps":  #: comps search, came from an address
@@ -578,6 +580,9 @@ class RealtorScraper(Scraper):
                     return gql_results["properties"]
 
             else:  #: general search, comps (radius)
+                if not location_info.get("centroid"):
+                    return []
+
                 coordinates = list(location_info["centroid"].values())
                 search_variables |= {
                     "coordinates": coordinates,
@@ -660,7 +665,6 @@ class RealtorScraper(Scraper):
 
     @staticmethod
     def _parse_description(result: dict) -> Description:
-
         description_data = result.get("description", {})
 
         if description_data is None or not isinstance(description_data, dict):
@@ -680,7 +684,7 @@ class RealtorScraper(Scraper):
         return Description(
             primary_photo=primary_photo,
             alt_photos=RealtorScraper.process_alt_photos(result.get("photos")),
-            style=style,
+            style=PropertyType(style) if style else None,
             beds=description_data.get("beds"),
             baths_full=description_data.get("baths_full"),
             baths_half=description_data.get("baths_half"),
@@ -692,7 +696,6 @@ class RealtorScraper(Scraper):
             stories=description_data.get("stories"),
         )
 
-
     @staticmethod
     def calculate_days_on_mls(result: dict) -> Optional[int]:
         list_date_str = result.get("list_date")

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "homeharvest"
-version = "0.3.12"
+version = "0.3.14"
 description = "Real estate scraping library"
 authors = ["Zachary Hampton <zachary@bunsly.com>", "Cullen Watson <cullen@bunsly.com>"]
 homepage = "https://github.com/Bunsly/HomeHarvest"

tests/test_realtor.py

@@ -131,6 +131,15 @@ def test_realtor():
     assert all([result is not None for result in results])
 
 
+def test_realtor_city():
+    results = scrape_property(
+        location="Atlanta, GA",
+        listing_type="for_sale",
+    )
+
+    assert results is not None and len(results) > 0
+
+
 def test_realtor_bad_address():
     bad_results = scrape_property(
             location="abceefg ju098ot498hh9",