- None address bug fix

[enh] days_on_mls attr

README.md

@@ -6,9 +6,9 @@
 
 **Not technical?** Try out the web scraping tool on our site at [tryhomeharvest.com](https://tryhomeharvest.com).
 
-*Looking to build a data-focused software product?* **[Book a call](https://calendly.com/zachary-products/15min)** *to work with us.*
+*Looking to build a data-focused software product?* **[Book a call](https://calendly.com/bunsly/15min)** *to work with us.*
 
-Check out another project we wrote: ***[JobSpy](https://github.com/cullenwatson/JobSpy)** – a Python package for job scraping*
+Check out another project we wrote: ***[JobSpy](https://github.com/Bunsly/JobSpy)** – a Python package for job scraping*
 
 ## HomeHarvest Features
 
@@ -20,7 +20,7 @@ Check out another project we wrote: ***[JobSpy](https://github.com/cullenwatson/
   - **CLI**: For users who prefer command-line operations.
 
 
-[Video Guide for HomeHarvest](https://youtu.be/JnV7eR2Ve2o) - _updated for release v0.2.7_
+[Video Guide for HomeHarvest](https://youtu.be/J1qgNPgmSLI) - _updated for release v0.3.4_
 
 ![homeharvest](https://github.com/ZacharyHampton/HomeHarvest/assets/78247585/b3d5d727-e67b-4a9f-85d8-1e65fd18620a)
 
@@ -147,6 +147,7 @@ Property
 │ └── lot_sqft
 
 ├── Property Listing Details:
+│ ├── days_on_mls
 │ ├── list_price
 │ ├── list_date
 │ ├── sold_price

examples/HomeHarvest_Demo.ipynb

@@ -4,7 +4,9 @@
    "cell_type": "code",
    "execution_count": null,
    "id": "cb48903e-5021-49fe-9688-45cd0bc05d0f",
-   "metadata": {},
+   "metadata": {
+    "is_executing": true
+   },
    "outputs": [],
    "source": [
     "from homeharvest import scrape_property\n",
@@ -84,10 +86,34 @@
    "outputs": [],
    "source": [
     "# check sold properties\n",
-    "scrape_property(\n",
+    "properties = scrape_property(\n",
     "    location=\"90210\",\n",
-    "    listing_type=\"sold\"\n",
+    "    listing_type=\"sold\",\n",
-    ")"
+    "    past_days=10\n",
+    ")\n",
+    "display(properties)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "628c1ce2",
+   "metadata": {
+    "collapsed": false,
+    "is_executing": true,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
+   "outputs": [],
+   "source": [
+    "# display clickable URLs\n",
+    "from IPython.display import display, HTML\n",
+    "properties['property_url'] = '<a href=\"' + properties['property_url'] + '\" target=\"_blank\">' + properties['property_url'] + '</a>'\n",
+    "\n",
+    "html = properties.to_html(escape=False)\n",
+    "truncate_width = f'<style>.dataframe td {{ max-width: 200px; overflow: hidden; text-overflow: ellipsis; white-space: nowrap; }}</style>{html}'\n",
+    "display(HTML(truncate_width))"
    ]
   }
  ],

homeharvest/core/scrapers/models.py

@@ -59,6 +59,7 @@ class Property:
     last_sold_date: str | None = None
     prc_sqft: int | None = None
     hoa_fee: int | None = None
+    days_on_mls: int | None = None
     description: Description | None = None
 
     latitude: float | None = None

homeharvest/core/scrapers/realtor/__init__.py

@@ -4,6 +4,7 @@ homeharvest.realtor.__init__
 
 This module implements the scraper for realtor.com
 """
+from datetime import datetime
 from typing import Dict, Union, Optional
 from concurrent.futures import ThreadPoolExecutor, as_completed
 
@@ -104,11 +105,29 @@ class RealtorScraper(Scraper):
         )
 
         able_to_get_lat_long = (
-            property_info
+                property_info
-            and property_info.get("address")
+                and property_info.get("address")
-            and property_info["address"].get("location")
+                and property_info["address"].get("location")
-            and property_info["address"]["location"].get("coordinate")
+                and property_info["address"]["location"].get("coordinate")
         )
+        list_date_str = property_info["basic"]["list_date"].split("T")[0] if property_info["basic"].get(
+            "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
+
+        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
+        today = datetime.now()
+
+        days_on_mls = None
+        status = property_info["basic"]["status"].lower()
+        if list_date:
+            if status == "sold" and last_sold_date:
+                days_on_mls = (last_sold_date - list_date).days
+            elif status in ('for_sale', 'for_rent'):
+                days_on_mls = (today - list_date).days
+            if days_on_mls and days_on_mls < 0:
+                days_on_mls = None
 
         listing = Property(
             mls=mls,
@@ -118,17 +137,13 @@ class RealtorScraper(Scraper):
             property_url=f"{self.PROPERTY_URL}{property_info['details']['permalink']}",
             status=property_info["basic"]["status"].upper(),
             list_price=property_info["basic"]["price"],
-            list_date=property_info["basic"]["list_date"].split("T")[0]
+            list_date=list_date,
-            if property_info["basic"].get("list_date")
-            else None,
             prc_sqft=property_info["basic"].get("price")
-            / property_info["basic"].get("sqft")
+                     / property_info["basic"].get("sqft")
             if property_info["basic"].get("price")
-            and property_info["basic"].get("sqft")
+               and property_info["basic"].get("sqft")
-            else None,
-            last_sold_date=property_info["basic"]["sold_date"].split("T")[0]
-            if property_info["basic"].get("sold_date")
             else None,
+            last_sold_date=last_sold_date,
             latitude=property_info["address"]["location"]["coordinate"].get("lat")
             if able_to_get_lat_long
             else None,
@@ -148,6 +163,7 @@ class RealtorScraper(Scraper):
                 garage=property_info["details"].get("garage"),
                 stories=property_info["details"].get("stories"),
             ),
+            days_on_mls=days_on_mls
         )
 
         return [listing]
@@ -455,6 +471,9 @@ class RealtorScraper(Scraper):
 
             is_pending = result["flags"].get("is_pending") or result["flags"].get("is_contingent")
 
+            if is_pending and self.listing_type != ListingType.PENDING:
+                continue
+
             realty_property = Property(
                 mls=mls,
                 mls_id=result["source"].get("listing_id")
@@ -478,8 +497,8 @@ class RealtorScraper(Scraper):
                 if able_to_get_lat_long
                 else None,
                 address=self._parse_address(result, search_type="general_search"),
-                #: neighborhoods=self._parse_neighborhoods(result),
                 description=self._parse_description(result),
+                days_on_mls=self.calculate_days_on_mls(result)
             )
             properties.append(realty_property)
 
@@ -568,21 +587,28 @@ class RealtorScraper(Scraper):
         return ", ".join(neighborhoods_list) if neighborhoods_list else None
 
     @staticmethod
-    def _parse_address(result: dict, search_type):
+    def handle_none_safely(address_part):
+        if address_part is None:
+            return ""
+
+        return address_part
+
+    def _parse_address(self, result: dict, search_type):
         if search_type == "general_search":
-            return Address(
+            address = result['location']['address']
-                street=f"{result['location']['address']['street_number']} {result['location']['address']['street_name']} {result['location']['address']['street_suffix']}",
+        else:
-                unit=result["location"]["address"]["unit"],
+            address = result["address"]
-                city=result["location"]["address"]["city"],
+
-                state=result["location"]["address"]["state_code"],
-                zip=result["location"]["address"]["postal_code"],
-            )
         return Address(
-            street=f"{result['address']['street_number']} {result['address']['street_name']} {result['address']['street_suffix']}",
+            street=" ".join([
-            unit=result["address"]["unit"],
+                self.handle_none_safely(address.get('street_number')),
-            city=result["address"]["city"],
+                self.handle_none_safely(address.get('street_name')),
-            state=result["address"]["state_code"],
+                self.handle_none_safely(address.get('street_suffix')),
-            zip=result["address"]["postal_code"],
+            ]).strip(),
+            unit=address["unit"],
+            city=address["city"],
+            state=address["state_code"],
+            zip=address["postal_code"],
         )
 
     @staticmethod
@@ -590,7 +616,6 @@ class RealtorScraper(Scraper):
         description_data = result.get("description", {})
 
         if description_data is None or not isinstance(description_data, dict):
-            print("Warning: description_data is invalid!")
             description_data = {}
 
         style = description_data.get("type", "")
@@ -609,3 +634,22 @@ class RealtorScraper(Scraper):
             garage=description_data.get("garage"),
             stories=description_data.get("stories"),
         )
+
+    @staticmethod
+    def calculate_days_on_mls(result: dict) -> Optional[int]:
+        list_date_str = result.get("list_date")
+        list_date = datetime.strptime(list_date_str.split("T")[0], "%Y-%m-%d") if list_date_str else None
+        last_sold_date_str = result.get("last_sold_date")
+        last_sold_date = datetime.strptime(last_sold_date_str, "%Y-%m-%d") if last_sold_date_str else None
+        today = datetime.now()
+
+        if list_date:
+            if result["status"] == 'sold':
+                if last_sold_date:
+                    days = (last_sold_date - list_date).days
+                    if days >= 0:
+                        return days
+            elif result["status"] in ('for_sale', 'for_rent'):
+                days = (today - list_date).days
+                if days >= 0:
+                    return days

homeharvest/utils.py

@@ -18,6 +18,7 @@ ordered_properties = [
     "half_baths",
     "sqft",
     "year_built",
+    "days_on_mls",
     "list_price",
     "list_date",
     "sold_price",

pyproject.toml

@@ -1,9 +1,9 @@
 [tool.poetry]
 name = "homeharvest"
-version = "0.3.4"
+version = "0.3.7"
 description = "Real estate scraping library supporting Zillow, Realtor.com & Redfin."
 authors = ["Zachary Hampton <zachary@zacharysproducts.com>", "Cullen Watson <cullen@cullen.ai>"]
-homepage = "https://github.com/ZacharyHampton/HomeHarvest"
+homepage = "https://github.com/Bunsly/HomeHarvest"
 readme = "README.md"
 
 [tool.poetry.scripts]

tests/test_realtor.py

@@ -7,14 +7,10 @@ from homeharvest.exceptions import (
 
 def test_realtor_pending_or_contingent():
     pending_or_contingent_result = scrape_property(
-        location="Surprise, AZ",
+        location="Surprise, AZ", listing_type="pending"
-        listing_type="pending"
     )
 
-    regular_result = scrape_property(
+    regular_result = scrape_property(location="Surprise, AZ", listing_type="for_sale")
-        location="Surprise, AZ",
-        listing_type="for_sale"
-    )
 
     assert all(
         [
@@ -54,6 +50,16 @@ def test_realtor_pending_comps():
     assert len(set([len(result) for result in results])) == len(results)
 
 
+def test_realtor_sold_past():
+    result = scrape_property(
+        location="San Diego, CA",
+        past_days=30,
+        listing_type="sold",
+    )
+
+    assert result is not None and len(result) > 0
+
+
 def test_realtor_comps():
     result = scrape_property(
         location="2530 Al Lipscomb Way",