caddy-nats
is a caddy module that allows the caddy server to interact with a
NATS server. This extension supports multiple patterns:
publish/subscribe, fan in/out, and request reply.
The purpose of this project is to better bridge HTTP based services with NATS in a pragmatic and straightforward way. If you've been wanting to use NATS, but have some use cases that still need to use HTTP, this may be a really good option for you.
To use caddy-nats
, simply run the xcaddy build tool to create a caddy-nats
compatible caddy server.
xcaddy build --with github.com/codegangsta/caddy-nats
Getting up and running with caddy-nats
is pretty simple:
First install NATS and make sure the NATS server is running:
nats-server
Then create your Caddyfile:
{
nats {
reply hello GET https://localhost/hello
}
}
localhost {
route /hello {
respond "Hello, world"
}
}
After running your custom built caddy
server with this Caddyfile, you can
then run the nats
cli tool to test it out:
nats req hello ""
To connect to nats
, simply use the nats
global option in your Caddyfile:
{
nats
}
This will connect to the currently selected nats
context that the nats
CLI
uses. If there is no nats
context available, it will connect to the default
nats
server URL.
You can view your nats
context information via the nats
cli:
nats context list
We recommend connecting to a specific context:
{
nats <context>
}
caddy-nats
supports subscribing to NATS subjects in a few different flavors,
depending on your needs.
All subscribe
based directives (subscribe
, reply
, queue_subscribe
, queue_reply
) support the following caddy
placeholders in the method
and url
arguments:
{nats.path}
: The subject of this message, with dots "." replaced with a slash "/" to make it easy to map to a URL.{nats.path.*}
: You can also select a segment of a path ex:{nats.path.0}
or a subslice of a path:{nats.path.2:}
or{nats.path.2:7}
to have ultimate flexibility in what to forward onto the URL path.
subscribe <subject> <method> <url>
subscribe
will subscribe to the specific NATS subject (wildcards are
supported) and forward the NATS payload to the specified URL inside the caddy
web server. This directive does not care about the HTTP response, and is
generally fire and forget.
Subscribe to an event stream in NATS and call an HTTP endpoint:
{
nats {
subscribe events.> POST https://localhost/nats_events/{nats.path.1:}
}
}
reply <subject> <method> <url>
reply
will subscribe to the specific NATS subject and forward the NATS
payload to the specified URL inside the caddy web server. This directive will
then respond back to the nats message with the response body of the HTTP
request.
Respond to the hello.world
NATS subject with the response of the /hello/world
endpoint.
{
nats {
reply hello.world GET https://localhost/hello/world
}
}
queue_subscribe <subject> <queue> <method> <url>
queue_subscribe
operates the same way as subscribe
, but subscribes under a NATS queue group
Subscribe to a worker queue:
{
nats {
queue_subscribe jobs.* workers_queue POST https://localhost/{nats.path}
}
}
queue_reply <subject> <queue> <method> <url>
queue_reply
operates the same way as reply
, but subscribes under a NATS queue group
Subscribe to a worker queue, and respond to the NATS message:
{
nats {
queue_reply jobs.* workers_queue POST https://localhost/{nats.path}
}
}
caddy-nats
also supports publishing to NATS subjects when an HTTP call is
matched within caddy
, this makes for some very powerful bidirectional
patterns.
All publish
based directives (nats_publish
, nats_request
) support the following caddy
placeholders in the subject
argument:
{nats.subject}
: The path of the http request, with slashes "/" replaced with dots "." to make it easy to map to a NATS subject.{nats.subject.*}
: You can also select a segment of a subject ex:{nats.subject.0}
or a subslice of the subject:{nats.subject.2:}
or{nats.subject.2:7}
to have ultimate flexibility in what to forward onto the URL path.
Additionally, since publish
based directives are caddy http handlers, you also get access to all caddy http placeholders.
nats_publish [<matcher>] <subject> {
timeout <timeout-ms>
}
nats_publish
publishes the request body to the specified NATS subject. This
http handler is not a terminal handler, which means it can be used as
middleware (Think logging and events for specific http requests).
Publish an event before responding to the http request:
localhost {
route /hello {
nats_publish events.hello
respond "Hello, world"
}
}
nats_request [<matcher>] <subject> {
timeout <timeout-ms>
}
nats_request
publishes the request body to the specified NATS subject, and
writes the response of the NATS reply to the http response body.
Publish an event before responding to the http request:
localhost {
route /hello/* {
nats_request hello_service.{nats.subject.1}
}
}
While this is currently functional and useful as is, here are the things I'd like to add next:
- Add more examples in the /examples directory
- Add Validation for all caddy modules in this package
- Add godoc comments
- Support mapping nats headers and http headers for upstream and downstream
- Customizable error handling
- Jetstream support (for all that persistence babyyyy)
- nats KV for storage