It scans nginx.conf with included configs and outputs (creates redirects.csv) all redirects in a short, handy form.

scan_nginx.py

import crossplane, re, csv

#  !!! CHANGE include /opt/nginx/conf/conf.d/*.conf;  .  /conf.d/*.conf;

payload = crossplane.parse('nginx.conf')

# In order to understand structure go to https://codebeautify.org/python-formatter-beautifier and paste result of  print(payload['config'])
# print(payload['config'])
# print(payload)
files = payload['config']

# FINDING ALL UPSTREAMS
upstreams ={} #
for file in files: # ITERATING THROUGH EACH FILE
    # print(file['file']) # config file path
    # servers = [] # how did I want to use it?
    for directive1 in file['parsed']: # ITERATING THROUGH FILE LINES
        if directive1['directive'] == 'upstream':  # UPSTREAM CONTEXT
            upstream = directive1
            upstream_name = upstream["args"][0]
            upstreams[upstream_name] = []
            # print(f'upstream {upstream_name}')
            # breakpoint()
            for directive2 in upstream["block"]:
                if directive2['directive'] == 'server':
                    server = directive2
                    server_addr = server["args"][0]
                    # servers.append(server_addr)
                    upstreams[upstream_name].append(server_addr)
                    # print(f'server {server_addr}')



with open('redirects.csv', 'w', newline='') as csvfile:
    fieldnames = ['FROM', 'WHERE']
    csv.register_dialect("tabs",delimiter=",")
    writer = csv.DictWriter(csvfile, fieldnames=fieldnames, dialect='tabs' )
    writer.writeheader()

    for file in files:  # ITERATING THROUGH EACH FILE
        # UNCOMMENT TO FIND LINES OF EACH CONFIG
        # print(file['file']) # config file path
        # if file['file'] == "./conf.d/6056.qbreport.conf": DEBUGGINNG
        #     breakpoint()
        writer.writerow({'FROM': 'config file: ' + file['file'], 'WHERE': '-------------'})
        # DEBUGGING BLOCK
        # if file['file'] == '././conf.d/zabbix.minfin.local.conf':
        #     breakpoint()

        for directive1 in file['parsed']:  # ITERATING
            # THROUGH FILE LINES
            if directive1['directive'] == 'server': # SERVER CONTEXT
                server=directive1
                for directive2 in server['block']:
                    if directive2['directive'] == 'listen':
                        listen = directive2["args"][0]
                        # print(f'listen on {listen}') # listen on
                    if directive2['directive'] == 'location':
                        location = directive2["args"]
                        # print(f'location: {location}') # location specified
                        for directive3 in directive2['block']:
                            if directive3['directive'] == 'proxy_pass':
                                where = directive3["args"][0]
                                # regex to find ip in a proxy pass (?<=http://)[\d|\.|:]+
                                # regex to find upstream in a proxy pass (?<=http://)[a-zA-Z_-]+
                                # print(f'proxy_pass {proxy_pass}') # proxy pass
                                regex=r'(?<=://)[^/:]+'  # !!! WORKS WITH HTTP ONLY !!!
                                if 'https://' in where:
                                    print('UPSTREAM CONTAINS HTTPS! REFACTOR YOUR CODE.', file['file'])

                                    # breakpoint()
                                    # exit(1)
                                # print(proxy_pass)
                                # breakpoint()
                                matches = re.findall(regex, where, re.DOTALL)
                                # print(upstreams)
                                if matches and matches[0] in upstreams: # in case upstream name is present in a proxy pass
                                    # .join may result few servers to pe printed in one line, if there are few servers in a upstream
                                    # print(" ".join(upstreams[matches[0]]))
                                    # print(file['file'])
                                    where = re.sub(regex, " ".join(upstreams[matches[0]]), where, 0, re.MULTILINE) # " ".join(upstreams[matches[0]]) is a list of upstream's servers

                            elif directive3['directive'] == 'fastcgi_pass':
                                where = directive3["args"][0]
                                # breakpoint()
                            elif directive3['directive'] == 'root':
                                where = directive3["args"][0]
                                # breakpoint()

                            try:
                                # .join may result few location args to pe printed in one line, if there are few arags e.g " =  / "
                                map = f'{listen}{" ".join(location)} -> {where}'
                                writer.writerow({'FROM': listen + " ".join(location), 'WHERE': where})
                                # print(map)
                                del where
                            except NameError:
                                 # IF "proxy_pass" , "root", "fastcgi_pass" were not found in "location" it will be skipped
                                 # IF configuration is not based on location context - this script will pass it
                                pass
        # print('\n')

README

This is a handy script to parse large nginx config.

This script scans nginx config (as well as "include" statements) and saves the output in a human readable text file 'redirects.csv'
The format is :

nginx_host:port/location -> upstream:port/location
example:
10.100.174.36:2203/,http://10.100.174.22:2203

Dependencies: crossplane