[MIRROR] Python package for scraping real estate property data
Go to file
Zachary Hampton 088088ae51 - last x days param 2023-10-03 15:05:17 -07:00
.github/workflows - scrapers & pypi init 2023-09-15 13:06:14 -07:00
homeharvest - last x days param 2023-10-03 15:05:17 -07:00
tests - last x days param 2023-10-03 15:05:17 -07:00
.gitignore feat(redfin): rental support 2023-09-19 11:58:20 -05:00
HomeHarvest_Demo.ipynb docs:add guide 2023-09-18 23:53:10 -05:00
LICENSE Create LICENSE 2023-09-16 10:39:36 -07:00
README.md - realtor.com default 2023-10-02 10:28:13 -07:00
poetry.lock [fix] zilow tls client 2023-09-28 19:34:01 -05:00
pyproject.toml - version bump 2023-10-01 09:13:37 -07:00

README.md

HomeHarvest is a simple, yet comprehensive, real estate scraping library.

Try with Replit


Not technical? Try out the web scraping tool on our site at tryhomeharvest.com.

Looking to build a data-focused software product? Book a call to work with us.

Check out another project we wrote: JobSpy a Python package for job scraping

Features

  • Scrapes properties from Zillow, Realtor.com & Redfin simultaneously
  • Aggregates the properties in a Pandas DataFrame

Video Guide for HomeHarvest - updated for release v0.2.7

homeharvest

Installation

pip install homeharvest

Python version >= 3.10 required

Usage

Python

from homeharvest import scrape_property
import pandas as pd

properties: pd.DataFrame = scrape_property(
    location="85281",
    listing_type="for_rent" # for_sale / sold
)

#: Note, to export to CSV or Excel, use properties.to_csv() or properties.to_excel().
print(properties)

CLI

homeharvest "San Francisco, CA" -s zillow realtor.com redfin -l for_rent -o excel -f HomeHarvest

This will scrape properties from the specified sites for the given location and listing type, and save the results to an Excel file named HomeHarvest.xlsx.

By default:

  • If -s or --site_name is not provided, it will scrape from all available sites.
  • If -l or --listing_type is left blank, the default is for_sale. Other options are for_rent or sold.
  • The -o or --output default format is excel. Options are csv or excel.
  • If -f or --filename is left blank, the default is HomeHarvest_<current_timestamp>.
  • If -p or --proxy is not provided, the scraper uses the local IP.
  • Use -k or --keep_duplicates to keep duplicate properties based on address. If not provided, duplicates will be removed.

Output

>>> properties.head()
                                        property_url site_name listing_type  apt_min_price  apt_max_price   ...  
0  https://www.redfin.com/AZ/Tempe/1003-W-Washing...    redfin     for_rent         1666.0         2750.0   ... 
1  https://www.redfin.com/AZ/Tempe/VELA-at-Town-L...    redfin     for_rent         1665.0         3763.0   ...  
2  https://www.redfin.com/AZ/Tempe/Camden-Tempe/a...    redfin     for_rent         1939.0         3109.0   ...  
3  https://www.redfin.com/AZ/Tempe/Emerson-Park/a...    redfin     for_rent         1185.0         1817.0   ... 
4  https://www.redfin.com/AZ/Tempe/Rio-Paradiso-A...    redfin     for_rent         1470.0         2235.0   ...   
[5 rows x 41 columns]

Parameters for scrape_properties()

Required
├── location (str): address in various formats e.g. just zip, full address, city/state, etc.
└── listing_type (enum): for_rent, for_sale, sold
Optional
├── site_name (list[enum], default=all three sites): zillow, realtor.com, redfin
├── proxy (str): in format 'http://user:pass@host:port' or [https, socks]
└── keep_duplicates (bool, default=False): whether to keep or remove duplicate properties based on address

Property Schema

Property
├── Basic Information:
│   ├── property_url (str)
│   ├── site_name (enum): zillow, redfin, realtor.com
│   ├── listing_type (enum): for_sale, for_rent, sold
│   └── property_type (enum): house, apartment, condo, townhouse, single_family, multi_family, building

├── Address Details:
│   ├── street_address (str)
│   ├── city (str)
│   ├── state (str)
│   ├── zip_code (str)
│   ├── unit (str)
│   └── country (str)

├── House for Sale Features:
│   ├── tax_assessed_value (int)
│   ├── lot_area_value (float)
│   ├── lot_area_unit (str)
│   ├── stories (int)
│   ├── year_built (int)
│   └── price_per_sqft (int)

├── Building for Sale and Apartment Details:
│   ├── bldg_name (str)
│   ├── beds_min (int)
│   ├── beds_max (int)
│   ├── baths_min (float)
│   ├── baths_max (float)
│   ├── sqft_min (int)
│   ├── sqft_max (int)
│   ├── price_min (int)
│   ├── price_max (int)
│   ├── area_min (int)
│   └── unit_count (int)

├── Miscellaneous Details:
│   ├── mls_id (str)
│   ├── agent_name (str)
│   ├── img_src (str)
│   ├── description (str)
│   ├── status_text (str)
│   └── posted_time (str)

└── Location Details:
    ├── latitude (float)
    └── longitude (float)

Supported Countries for Property Scraping

  • Zillow: contains listings in the US & Canada
  • Realtor.com: mainly from the US but also has international listings
  • Redfin: listings mainly in the US, Canada, & has expanded to some areas in Mexico

Exceptions

The following exceptions may be raised when using HomeHarvest:

  • InvalidSite - valid options: zillow, redfin, realtor.com
  • InvalidListingType - valid options: for_sale, for_rent, sold
  • NoResultsFound - no properties found from your input
  • GeoCoordsNotFound - if Zillow scraper is not able to derive geo-coordinates from the location you input

Frequently Asked Questions


Q: Encountering issues with your queries?
A: Try a single site and/or broaden the location. If problems persist, submit an issue.


Q: Received a Forbidden 403 response code?
A: This indicates that you have been blocked by the real estate site for sending too many requests. Currently, Zillow is particularly aggressive with blocking. We recommend:

  • Waiting a few seconds between requests.
  • Trying a VPN to change your IP address.