usebin

README

useb.in

Encrypted pastebin stored on Usenet.

Example Config

// $HOME/.config/usebin/config.json
{
    // The interface address to listen to
    "Host": "0.0.0.0",
    // The port to listen to
    "Port": 8080,
    // Your NNTP server connection infos
    "NNTPServers": [
        {
            // The host and port of the NNTP server
            "Host": "news.example.com:119",
            "User": "user",
            "Pass": "pass",
            // Whether the connection should use TLS encryption
            "TLS": false,
            // Whether the server can be used for posting
            "Posting": true,
            // Maximum number of connections for this server
            "Connections": 50,
        }
    ],
    // How long can connections to be idle until being closed, in seconds
    "IdleConnExpiry": 60,
    // The newsgroup to post to if not set explicitly in the request
    "DefaultNewsgroup": "alt.binaries.misc",
    // Max number of bytes an article can have, limited on article get and post
    "ArticleSizeLimit": 4194304,
    // If set, will use the following X509 PEM encoded certificate and key files to enable TLS for the server
    // "CertFile": "./path/to/cert.pem",
    // "KeyFile": "./path/to/key.pem",
}

API

GET /m/<Message-ID>.csv

Get the full article by Message-ID. Any NNTP headers will be prefixed by X-Usenet- and included together in the HTTP response headers. The returned article body is dot-decoded, and <CR> <LF> line endings are converted to a single <LF>. The Content-Length HTTP header is set to be the number of bytes of the dot-decoded article body.

HEAD /m/<Message-ID>.csv

Get the article headers and Content-Length without the article body. However, in order to calculate the Content-Length, the full article is still being download by the Usebin server. If Content-Length is not required, use the GET /h/<Message-ID>.csv API instead, which will return only the article headers without Content-Length.

POST /m/<Message-ID>.csv

Post an article with the specified Message-ID. The HTTP body will be dot-encoded by the Usebin server, also line ending will be normalized to <CR> <LF> before sending to an NNTP server. However the requesting client should keep the article lines under permitted length, usually under 127 bytes per line. Any HTTP header starting with X-Usenet- will be stripped off its prefix and set as an NNTP header and send to the NNTP server.

URL query parameter f, or HTTP header From

If set, will be used to set the From NNTP header. If not set, Usebin will generate a random address that looks like sending from an ngPost client.

URL query parameter g or HTTP header Newsgroups

If set, will be used to set the Newsgroups NNTP header. If not set, DefaultNewsgroup specified in config, or alt.binaries.misc will be used.

URL query parameter s or HTTP header Subject

If set, will be used to set the Subject NNTP header. If not, the part before the @ character from the Message-ID, or the full Message-ID, will be used.

GET /d/<Message-ID>.csv

Just like GET /m/<Message-ID>.csv except the returned HTTP body is the raw article body returned from the NNTP server before doing dot-decoding. All line breaks are untouched including any <CR> <LF> characters, and it will also include the dot-termination sequence <CR> <LF> <DOT> <CR> <LF> at the end of the body.

HEAD /d/<Message-ID>.csv

Same as HEAD /m/<Message-ID>.csv.

POST /d/<Message-ID>.csv

Just like POST /m/<Message-ID>.csv except the request's HTTP body should be dot-encoded and uses proper <CR> <LF> line endings. Although the dot-termination sequence <CR> <LF> <DOT> <CR> <LF> is optional in the request's HTTP body.

GET /h/<Message-ID>.csv

Just like HEAD /m/<Message-ID>.csv without calculating Content-Length. This is implemented as a HEAD NNTP command, which is much faster than reading the full article.

Cloudflare Caching

To better utilize Cloudflare Caching for the SPA program, please add the following settings to your Cloudflare dashboard:

Transform Rule

Add a Transform Rule with the following expression:

(not starts_with(http.request.uri.path, "/m/") and not starts_with(http.request.uri.path, "/d/") and not starts_with(http.request.uri.path, "/h/") and not starts_with(http.request.uri.path, "/assets/"))

And "statically rewrite" it to /.

Cache Rule

Add a Cache Rule with the following expression:

(http.request.uri.path eq "/")

And specify it as "eligible for cache".

Disclaimer

The author of Usebin is not responsible for any legal or economical consequences caused by the act of anyone using the Usebin program, and not bound by any implied promises, warranty or liability from such actions.

License

See LICENSE.