2 releases

0.1.4 Jan 3, 2023
0.1.0 Nov 23, 2022

#18 in #workflow-engine

MIT license

250KB
6.5K SLoC

Yao workflow engine

yao is a fast, tiny, extensiable workflow engine, which provides the abilities to execute workflow based on simple yml model.

The yml workflow model is not as same as the tranditional flow. such as bpmn. It is inspired by Github actions. As a contrast, it added branch defination for more complex flow, for the purpose of business approval flow, it defines the subject property in step to support the top absolute rules for user, org and role.

Fast

Uses rust to create the lib, there is no virtual machine, no db dependencies. The default store uses rocksdb for local storage.

Tiny

The lib size is only 3mb (no local db)

Extensiable

Supports the plugin to extend the functions

Examples

Here are some examples:

How to start

First, you should load a ymal workflow model, and call engine.start to start and call engine.close to stop it.

use yao::{Engine, Vars, Workflow};

#[tokio::main]
async fn main() {
    let engine = Engine::new();

    let text = include_str!("../examples/simple/model.yml");
    let mut workflow = Workflow::from_str(text).unwrap();
    let mut vars = Vars::new();
    vars.insert("input".into(), 3.into());
    workflow.set_env(vars);

    engine.push(&workflow);

    let e = engine.clone();
    engine.on_workflow_complete(move |w: &Workflow| {
        println!(
            "on_workflow_complete: {:?}, cost={}ms",
            w.outputs(),
            w.cost()
        );
        e.close();
    });
    engine.start().await;
}

How to create model

Notices the struct of the yaml, there are different type of node, which is constructed by Workflow, [Job], Branch and Step. Every workflow can have more jobs, every job can have more steps, a step can have more branches and a branch can have if property to judge the condition.

The env property can be set in each node, in the run scripts, you can use env moudle to get(env.get) or set(env.set) the value

The run property is the script based on rhai script

name: model name
jobs:
  - id: job1
    env:
      value: 0
    steps:
      - name: step 1
        run: |
          print("step 1")

      - name: step 2
        branches:
          - name: branch 1
            if: ${ env.get("value") > 100 }
            run: |
                print("branch 1");

          - name: branch 2
            if: ${ env.get("value") <= 100 }
            steps:
                - name: step 3
                  run: |
                    print("branch 2")
            

In the Workflow, you can set the outputs to output the env to use.

name: model name
outputs:
  output_key:
jobs:
  - id: job1
    steps:
      - name: step1
        run: |
          env.set("output_key", "output value");

The subject is used to create the user [Act], which can wait util calling the post_message to complete by user.

name: model name
jobs:
  - id: job1
    steps:
      - name: step1
        subject: 
            matcher: any
            users: |
                let a = ["u1"];
                let b = ["u2"];
                a.union(b)

It will generate the user act and send message automationly according to the sub users. The matcher tells the workflow how to pass the step act when you post_message to workflow, there are several match rules.

  • matcher
  1. one to generate or check only one user

  2. any to match any of the users

  3. some(rule_name) to match some users by giving rule name, which can be registed by register_some_rule function.

  4. ord or ord(rule_name) to generate the act message one by one by giving rule name, which can be registed by register_ord_rule function

  • users Used to generate the step participants, the final results are the user ids, which can be generated by org or role rule.

The code role("test_role") uses the role rule to get the users from the role test_role

users: |
    let users = role("test_role");
    users

The following code uses the user("test_user") the get the user and then throght the relate rule to find the user's owner of the department (d.owner).

users: |
    let users = user("test_user").relate("d.owner");
    users

Use builder to create model

use yao::{Workflow};

let mut workflow = Workflow::new()
        .with_name("workflow builder")
        .with_output("result", 0.into())
        .with_job(|job| {
            job.with_id("job1")
                .with_env("index", 0.into())
                .with_env("result", 0.into())
                .with_step(|step| {
                    step.with_id("cond")
                        .with_branch(|branch| {
                            branch
                                .with_if(r#"env.get("index") <= env.get("count")"#)
                                .with_step(|step| {
                                    step.with_id("c1")
                                        .with_action(|env| {
                                            let result =
                                                env.get("result").unwrap().as_i64().unwrap();
                                            let index = env.get("index").unwrap().as_i64().unwrap();
                                            env.set("result", (result + index).into());
                                            env.set("index", (index + 1).into());
                                        })
                                        .with_next("cond")
                                })
                        })
                        .with_branch(|branch| {
                            branch.with_if(r#"env.get("index") > env.get("count")"#)
                        })
                })
                .with_step(|step| {
                    step.with_name("step2")
                        .with_action(|env| println!("result={:?}", env.get("result").unwrap()))
                })
        });

Register plugin

See the example of ActPlugin.

Register store adapter

See the example of StoreAdapter.

Register org adapter

See the example of OrgAdapter.

Register role adapter

See the example of RoleAdapter.

Register rule adapter

See the example of RuleAdapter.

Dependencies

~78MB
~1.5M SLoC