An example crawler for ads.txt files given a list of URLs or domains etc and saves them to a SQLite DB table.

Usage: adstxt_crawler.py [options]

Options:
  -h, --help            show this help message and exit
  -t FILE, --targets=FILE
                        list of domains to crawler ads.txt from
  -d FILE, --database=FILE
                        Database to dump crawlered data into
  -v, --verbose         Increase verbosity (specify multiple times for more)

The targets file can be a list of domains, URLs etc. For each, line the crawler will extract the full hostname, validate it, and cause a request to http://HOSTNAME/ads.txt

$ cat target_domains.txt 
#https://chicagotribune.com
#http://latimes.com/sports
#washingtonpost.com
#http://nytimes.com/index.html
localhosttribune.com

The project depends on these libraries and programs installed

• Python 2 or better
• sqlite3
• See requirements.txt for all Python packages to install

Execute this command to install the DB table

$sqlite3 adstxt.db < adstxt_crawler.sql 

The usual usage would be to pass a filename of target URLs and a filename of the SQLite DB.

$ ./adstxt_crawler.py -t target_domains.txt -d adstxt.db
Wrote 3 records from 1 URLs to adstxt.db

Upon each run a sequence of entries in adstxt_crawler.log is created.

You can examine the DB records created as follows:

$echo "select * from adstxt;" | sqlite3 adstxt.db

You can clear the DB records as follows:

$echo "delete from adstxt;" | sqlite3 adstxt.db

This is an example prototype crawler and would be suitable only for a very modest production usage. It doesn't contain a lot of niceties of a production crawler, such as parallel HTTP download and parsing of the data files, stateful recovery of target servers being down, usage of a real production DB server etc.

Maintainer: Neal Richter, [email protected] or [email protected]

Contributors (GitHub.com account names) iantri jhpacker brk212 bradlucas nag4 AntoineJac markparolisi sean-mcmann Breza miyaichi

The open source license used is the 2-clause BSD license