Python script to parse a Requirements.txt file and get a list of Package Name, Version, and Dependency Source URL

parse_python_requirements.py

import os
import re
import subprocess

packages = {
    'package_name': re.compile(r'Name: (?P<package_name>.*)\n'),
    'version': re.compile(r'Version: (?P<version>.*)\n'),
    'url': re.compile(r'Home-page: (?P<url>.*)\n'),
}

def _parse_line(line):
    for key, val in packages.items():
        match = val.search(line)
        if match:
            return key, match
    # no matches
    return None, None

def parse_file(filepath):
    with open(filepath, 'r') as file:
        lines = file.readlines()
        for line in lines:
            key, match = _parse_line(line)
            if key == 'package_name':
                package_name = match.group('package_name')
                print(package_name)
            if key == 'version':
                version = match.group('version')
                print(version)
            if key == 'url':
                url = match.group('url')
                print(url)

def generate_package_info(filepath):
    with open(filepath, 'r') as file:
        lines = file.readlines()
        for line in lines:
            end = line.find('==')
            match = line[0:end]
            cmd = ['pip', 'show', match]
            proc = subprocess.Popen(
                cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
            o, e = proc.communicate()
            print(o.decode('ascii'))

if __name__ == '__main__':
    parse_file('./pip_info.txt')
    
"""
Attribution: Regex and Line Matching inspired by https://www.vipinajayakumar.com/parsing-text-with-python/
"""