1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
#![deny(warnings)]
#![deny(missing_docs)]
#![allow(clippy::needless_doctest_main)]
#![cfg_attr(feature = "unstable", feature(fn_traits))]
#![cfg_attr(feature = "unstable", feature(unboxed_closures))]
//! A simple, ergonomic, idiomatic, macro for generating the boilerplate to use rust futures tasks in a concurrent actor style.
//!
//! # What is GhostActor?
//!
//! GhostActor boils down to a macro that helps you write all the boilerplate
//! needed to treat a Future like an actor. When you "spawn" a GhostActor,
//! you receive a handle called a "Sender", that allows you to make async
//! requests and inline await async responses to/from you actor implementation's
//! driver task.
//!
//! The senders are cheaply clone-able allowing you to easily execute any
//! number of parallel workflows with your task. When all senders are dropped,
//! or if you explicitly call `ghost_actor_shutdown()`, the driver task
//! (a.k.a. your Actor) will end.
//!
//! # Example
//!
//! ```
//! use must_future::*;
//!
//! #[derive(Debug, thiserror::Error)]
//! pub enum MyError {
//!     #[error(transparent)]
//!     GhostError(#[from] ghost_actor::GhostError),
//! }
//!
//! ghost_actor::ghost_actor! {
//!     // Set the visibility of your actor.
//!     // Name your actor.
//!     // Specify the Error type for your actor.
//!     // The error type must implement `From<GhostError>`.
//!
//!     /// Api Docs that should appear on the Sender type for your actor.
//!     pub actor MyActor<MyError> {
//!         // specify your actor api
//!
//!         /// This string will be applied as docs to sender/handler.
//!         fn add_one(
//!             // any api params here
//!             input: u32,
//!         ) -> u32; // return type here
//!     }
//! }
//!
//! /// An example implementation of the example MyActor GhostActor.
//! struct MyActorImpl;
//!
//! // The generics for a handler are:
//! // 1 - the "custom" type you'd like to allow users of your api to send in.
//! // 2 - the "internal" type you'd like your handlers to send in.
//! // It is highly recommended to use a `ghost_chan!` type for these.
//! // However, if you have no use for these capabilities, use `()`.
//! impl MyActorHandler<(), ()> for MyActorImpl {
//!     fn handle_add_one(
//!         &mut self,
//!         input: u32,
//!     ) -> MyActorHandlerResult<u32> {
//!         Ok(async move {
//!             Ok(input + 1)
//!         }.must_box())
//!     }
//! }
//!
//! impl MyActorImpl {
//!     /// Rather than using ghost_actor_spawn directly, use this simple spawn.
//!     pub async fn spawn() -> MyActorSender {
//!         use futures::future::FutureExt;
//!
//!         let (sender, driver) = MyActorSender::ghost_actor_spawn(|_internal_sender| {
//!             async move {
//!                 Ok(MyActorImpl)
//!             }.must_box()
//!         }).await.unwrap();
//!
//!         tokio::task::spawn(driver);
//!
//!         sender
//!     }
//! }
//!
//! #[tokio::main(threaded_scheduler)]
//! async fn main() {
//!     let mut sender = MyActorImpl::spawn().await;
//!
//!     assert_eq!(43, sender.add_one(42).await.unwrap());
//!
//!     sender.ghost_actor_shutdown().await.unwrap();
//!
//!     let res = format!("{:?}", sender.add_one(42).await);
//!     if &res != "Err(GhostError(SendError(SendError { kind: Disconnected })))"
//!         && &res != "Err(GhostError(ResponseError(Canceled)))"
//!     {
//!         panic!("expected send error");
//!     }
//! }
//! ```
//!
//! # Implementing a Handler
//!
//! The `ghost_actor!` macro is going to generate a "[Name]Handler" trait.
//! To provide an implementation for your `ghost_actor!` type, you need an
//! item that implements this trait (see example above).
//!
//! In addition to all the `handle_*` methods that are auto-generated per
//! the `Api` section in the macro, there are also provided implementations
//! for `handle_ghost_actor_custom` and `handle_ghost_actor_internal`.
//!
//! Please see any of the unit tests (or run `cargo doc` on a module containing
//! your `ghost_actor!` macro invocation) for examples on how to implement
//! a handler.
//!
//! # Implementing a Spawn function
//!
//! While you can absolutely require users of your api to call
//! `YourTypeSender::ghost_actor_spawn(...)` and instantiate your handler type
//! inside the callback, it might be polite to provide a function that requires
//! a little less boilerplate.
//!
//! See the example above, however, there may be no need to expose the
//! implemented item type at all, you could, for example:
//!
//! ```
//! # #[derive(Debug, thiserror::Error)]
//! # pub enum MyError {
//! #     #[error(transparent)]
//! #     GhostError(#[from] ghost_actor::GhostError),
//! # }
//! # ghost_actor::ghost_actor! {
//! #     pub actor MyActor<MyError> {
//! #         fn add_one(
//! #             input: u32,
//! #         ) -> u32;
//! #     }
//! # }
//! use must_future::*;
//!
//! /// internal private type
//! struct MyActorImpl;
//!
//! impl MyActorHandler<(), ()> for MyActorImpl {
//!     // ...
//! #    fn handle_add_one(&mut self, input: u32) -> MyActorHandlerResult<u32> {
//! #        unimplemented!();
//! #    }
//! }
//!
//! /// Rather than using ghost_actor_spawn directly, use this simple spawn.
//! /// This spawn makes an assumption that we are in a tokio runtime,
//! /// if we don't want to make that assumption, we can also return the
//! /// driver future here.
//! pub async fn spawn_my_actor() -> MyActorSender {
//!     use futures::future::FutureExt;
//!
//!     let (sender, driver) = MyActorSender::ghost_actor_spawn(|_internal_sender| {
//!         async move {
//!             Ok(MyActorImpl)
//!         }.must_box()
//!     }).await.unwrap();
//!
//!     tokio::task::spawn(driver);
//!
//!     sender
//! }
//! # #[tokio::main(threaded_scheduler)]
//! # async fn main() {
//! # }
//! ```
//!
//! # The `ghost_chan!` macro.
//!
//! The `ghost_chan!` macro has an identical API to the `ghost_actor!` macro.
//! And, in fact, the `ghost_actor!` macro invokes `ghost_chan!` to produce
//! an internal enum for sending messages from your `Sender` struct.
//!
//! When implementing a ghost actor Handler that will make use of Custom
//! and/or Internal types, it is recommended to use a `ghost_chan!` enum as
//! this type.
//!
//! See the unit/integration tests for examples on making use of these.

/// Re-exported dependencies to help with macro references.
pub mod dependencies {
    pub use futures;
    pub use must_future;
    pub use paste;
    pub use thiserror;
    pub use tracing;
}

mod types;
pub use types::*;

pub mod ghost_chan;

mod r#macro;
pub use r#macro::*;

mod tests;