tokio-beanstalkd

This crate provides a client for working with Beanstalkd, a simple fast work queue.

About Beanstalkd

Beanstalkd is a simple fast work queue. It works at the TCP connection level, considering each TCP connection individually. A worker may have multiple connections to the Beanstalkd server and each connection will be considered separate.

The protocol is ASCII text based but the data itself is just a bytestream. This means that the application is responsible for interpreting the data.

Operation

This library can serve as a client for both the application and the worker. The application would put jobs on the queue and the workers can reserve them. Once they are done with the job, they have to delete job. This is required for every job, or else Beanstalkd will not remove it from its internal datastructres.

If a worker cannot finish the job in it's TTR (Time To Run), then it can release the job. The application can use the using method to put jobs in a specific tube, and workers can use watch to only reserve jobs from the specified tubes.

Interaction with Tokio

The futures in this crate expect to be running under a tokio::Runtime. In the common case, you cannot resolve them solely using .wait(), but should instead use tokio::run or explicitly create a tokio::Runtime and then use Runtime::block_on.

An simple example client could look something like this:

let mut rt = tokio::runtime::Runtime::new().unwrap();
let bean = rt.block_on(
    Beanstalkd::connect(&"127.0.0.1:11300".parse().unwrap()).and_then(|bean| {
        bean.put(0, 1, 100, &b"update:42"[..])
            .inspect(|(_, response)| {
                response.as_ref().unwrap();
            })
            .and_then(|(bean, _)| {
                // Use a particular tube
                bean.using("notifications")
            }).and_then(|(bean, _)| bean.put(0, 1, 100, &b"notify:100"[..]))
    }),
);
# }

And a worker could look something like this:

 let mut rt = tokio::runtime::Runtime::new().unwrap();
 let bean = rt.block_on(
     Beanstalkd::connect(&"127.0.0.1:11300".parse().unwrap()).and_then(|bean| {
         bean.reserve()
             .inspect(|(_, response)| {
                 // Do something with the response
             }).and_then(|(bean, response)| {
                 // Delete the job once it is done
                 bean.delete(response.as_ref().unwrap().id)
             })
     }),
 );
 rt.shutdown_on_idle();