autoindex
Lightweight go web server that provides a searchable directory index. Optimized for handling large numbers of files (100k+) and remote file systems (with high latency) through a continously updated directory cache.
Features:
- Lightweight single-page application (
~8KB html/css/js) - Responsive design
- Recursive file search
- Directory cache (
sqlite) - Sitemap support
Usage
./autoindex [options]
| Flag | Type | Description |
|---|---|---|
-a |
string |
TCP network address to listen for connections |
-d |
string |
Database location |
-r |
string |
Root directory to serve |
-i |
string |
Refresh interval |
-l |
int |
Request rate limit (req/sec per IP) |
-t |
duration |
Request timeout |
-forwarded |
bool |
Trust X-Real-IP and X-Forwarded-For headers |
-cached |
bool |
Serve everything from cache (rather than search/recursive queries only) |
Example
./autoindex -a=":4000" -i=5m -d=/tmp/autoindex.db -cached -r=/mnt/storage
Behind nginx
Example configuration for running autoindex behind an nginx proxy.
upstream autoindex {
server 127.0.0.1:4000;
keepalive 8;
}
map $request_uri $request_basename {
~/(?[^/?]*)(?:\?|$) $captured_request_basename;
}
map $request_uri $idx_path {
~/(?\?.*)?$ $captured_request_args;
~/(?[^?]*)(?\?.*)?$ $captured_request_path/$captured_request_args;
}
server {
listen 443 ssl http2;
listen [::]:443 ssl http2;
server_name _;
root /opt/autoindex/public;
location / {
rewrite ^/(.*)/$ /$1 permanent;
try_files $uri /index.html;
expires 1y;
}
location = /index.html {
http2_push /idx/$idx_path;
expires 1d;
}
location ^~ /dl/ {
limit_rate 1m;
add_header Content-Disposition 'attachment; filename="$request_basename"';
add_header X-Robots-Tag "noindex, nofollow, nosnippet, noarchive";
}
location ~ ^(/idx/|/urllist.txt) {
proxy_pass http://autoindex;
}
}
