man_page_parser.py

man_page_parser.py

import os


def parse_man_file(man_filepath):
    with open(man_filepath, "r") as man_file:
        lines = man_file.read().split("\n")
        
        formatted_lines = []
        for line in lines:
            # remove Apple developer comments included on the same line and strip off trailing white space
            formatted_lines.append(line.split('\\"')[0].strip())

        sections = {}
        for n, line in enumerate(formatted_lines):
            section_header = ""
            section_body = ""

            if line.startswith(".Sh "):
                section_header = line.replace(".Sh ", "")
                section_body_line = n + 1
                while True:
                    if section_body_line >= len(formatted_lines):
                        break
                    else:
                        body_line = formatted_lines[section_body_line]
                        if not body_line.startswith(".Sh "):
                            formatted_body_line = ""
                            if body_line.startswith(".Nd"):
                                formatted_body_line = body_line.replace(".Nd", "-")
                            else:
                                for word in body_line.split(" "):
                                    if not word.startswith("."):
                                        replace_list = ["\\fB", "\\fR", "Pa ", "", "", "", ""]
                                        word = word.replace("\\fB", "")
                                        word = word.replace("\\fR", "")
                                        formatted_body_line += word + " "
                                    else:
                                        formatted_body_line += " "
                    
                            section_body += formatted_body_line.replace("  ", " ")
                            section_body_line += 1
                        else:
                            break
                
                # print(current_section_header)
                # print(current_section_body)
                sections[section_header] = section_body.strip().replace("  ", " ")
        
        # print(sections)

        # print(n, ":", line)
        return sections


man_dir = "/usr/share/man/man8/"

service_dirs = [
    os.getenv("HOME") + '/Library/LaunchAgents',    #           Per-user agents provided by the user.
    '/Library/LaunchAgents',                        #           Per-user agents provided by the administrator.
    '/Library/LaunchDaemons',                       #           System wide daemons provided by the administrator.
    '/System/Library/LaunchAgents',                 #           OS X Per-user agents.
    '/System/Library/LaunchDaemons'                 #           OS X System wide daemons.
]

services = []
for service_dir in service_dirs:
    for service_filename in os.listdir(service_dir):
        s = {}
        if service_filename.startswith("com.apple."):
            s["name"] = service_filename.replace(".plist", "")
            s["path"] = service_dir + os.sep + service_filename
            s['has_man_page'] = False
            s["man_page_path"] = None
            
            possible_man_page_paths = [
                man_dir + s["name"].replace("com.apple.", "") + ".8",
                man_dir + s["name"].split(".")[-1] + ".8" 
            ]

            for man_path in possible_man_page_paths:
                if os.path.isfile(man_path) and s['has_man_page'] is False:
                    s['has_man_page'] = True
                    s["man_page_path"] = man_path

            if s['has_man_page']:
                s['man_page'] = parse_man_file(s["man_page_path"])
            else:
                s['man_page'] = None
            
            services.append(s)
            
output = ""

output += "|" + "|".join(services[0].keys()) + "|\n"

for keys in services[0].keys():
    output += '|---'
output += "|\n"

for service in services:
    row = "|"
    for k,v in service.items():
        cell = ""
        if isinstance(v, dict):
            for k2, v2 in v.items():
                cell += f"{k2}: {str(v2)}<br>"
                print("\t", k2, ":", v2)
        else:
            cell = str(v)

        row += cell + "|"
    output += row + "\n"

with open('output.txt', 'w') as outfile:
    outfile.write(output)
    
print(output)