ethinallen / CraigslistScraper

Craigslist Scraper is a web scraper for craigslist, it allows users to define searches then pulls all data from a search and places it neatly inside of a json file.

CraigslistScraper

Note: CraigslistScraper is for personal use and data science only.

CraigslistScraper is a web scraper for craigslist. Users define what they would like to search for then CraigslistScraper pulls ad data from their defined search and places it neatly inside of a JSON file. The file structure might look something like this:

data
    ├── 14-aug-2020
    │   ├── bmw_14:Aug:2020-11:04:01.json
    │   ├── audi_14:Aug:2020-18:03:21.json
    │   └── iphone_14:Aug:2020-11:04:07.json
    ├── 15-aug-2020
    └── 16-aug-2020
        └── iphone_15:Aug:2020-12:06:02.json
       

Where each JSON file might contain data that looks like:

Only tested on macOS with python3.

Table of Contents

• Installation
• Usage
• Analyzing
• License

Installation

MacOS:

To install the package just run:

pip install craigslistscraper

I always recommend the use of Python's virtual enviroments when installing libraries to avoid clutter. The dependencies CraigslistScraper requires to run properly are Beautifulsoup4, Requests, and Pandas. This may change as I continue to update this project.

Not tested on Windows or Linux

Usage

Example:

from craigslistscraper import Searches

if __name__ == '__main__': # Required to run inside of "if name == main"

  filters = ['&postedToday=1'] # Optional | Define your filters here

  cities = ['city1', 'city2'] # Required | Define the cities you want to search

  # car_data=False by default and doesn't need to be defined explicitly.
  foo = Searches("your search", cities, "section", filters, car_data=False)
  
  foo.compile_search()

Note #1: Filters does not have to be defined or passed in as an argument, and will by default be assigned '&postedToday=1'

Note #2: For a list of cities view the craigslistscraper/city_data/craigslist_cities_list.csv file

Note #3: For more filters, check out the FILTERS.md file.

For craigslistscraper to function properly you NEED to run it inside of if __name__ == '__main__' because of a multiprocessing error that occurs if you don't, this will be fixed in a future update.

If your using this scraper to look at car data I recommend putting car_data = True at the end of foo in the example above, as it'll give more complete json data which is easier to work with when converting to csv files.

Analyzing

Data can easily be converted to CSV

This data can then be analyzed, some examples include:

License

Distributed under the MIT License. See LICENSE for more information.