eComb- Haus and Wohnung Finder

A codebase for scraping house or apartment listings from ebay kleinanzeigen and Ohne-makler from re current day.

This can be run locally (with or without docker) or scheduled to run each day on a server to get the listings emailed to interested parties.

Getting Started

Pre-requisites

Set up a gmail email account to send the listing notifications. Once the account is set up create an app password for later use with the script.

Clone the repository locally onto the machine where it will used:

git clone https://github.com/djwight/ecomb.git

Environment Definition

Variables used at runtime are defined in a .env file.

A dummy .env is shown below and how to determine the base_url and the cat_code.

email_token="secret-token"
sender="bot@gmail.com"
recipients="email1@email.com;email2@email.de"
search_location="hamburg"
search_radius="10"
driver_path="/usr/local/bin/chromedriver"
base_url="https://www.ebay-kleinanzeigen.de/s-haus-kaufen"
cat_code="c208l3331"

To find the base_url and the cat_code go onto ebay kleinanzeigen and perform the search you would like to repeat daily.

For example for "Eigentumswohnung" in Hamburg the url shown would be https://www.ebay-kleinanzeigen.de/s-wohnung-kaufen/hamburg/c196l9409r20, where: https://www.ebay-kleinanzeigen.de/s-wohnung-kaufen = base_url c196l9409 = cat_code r20 = r + radius (in km) for the search

Run with docker (recommended)

Create the .env as described above and note the absolute path to the directory.

# moved into dir and build image
cd ecomb
docker build -t <name> .

# once built the container can be run on command or add to crontab to be run each day
docker run -e dotenv_path=/app/data/.env -v </absolute/path/to/dir/with/.env>:/app/data --shm-size="2g" <name>:latest

Run without docker (not recommended)

To run this script locally you will need chrome and the appropriate version of the chromedriver.

# activate a new environment with python 3.10 installed and move to the local git repository
cd ecomb
pip install -r requirements.txt

# create an .env and fill out as described above
touch .env

# run the script
export dotenv_path="/path/to/.env"
python3 run_search.py

To-do

Ohne-makler: implement the code to scrape adverets from this site
tests: implement more unit and integration tests
documentation: add documentation