ebakoba / depper Goto Github PK

I am trying to use to this repo but i have encountered a bug or perhaps i am misusing it.

For example, I have 3 dependers named machine1, machine2 and machine3 and 2 dependees named network1, network2. If i enter the elements in order, dependers before dependees, then it works as expected:

    let mut dependencies_builder = Dependencies::builder()
        .add_element("machine1".to_string(), vec!["network1".to_string()])
        .add_element("machine2".to_string(), vec!["network1".to_string()])
        .add_element("machine3".to_string(), vec!["network2".to_string()])
        .add_element("network1".to_string(), vec![])
        .add_element("network2".to_string(), vec![]);

outputs as expected:
Tranches: [["network2", "network1"], ["machine1", "machine2", "machine3"]]

However, if the elements are entered in any other order the results are unpredictable and erroneous. In this example the dependees are inserted before dependers.

    let mut dependencies_builder = Dependencies::builder()
        .add_element("network1".to_string(), vec![])
        .add_element("network2".to_string(), vec![])
        .add_element("machine1".to_string(), vec!["network1".to_string()])
        .add_element("machine2".to_string(), vec!["network1".to_string()])
        .add_element("machine3".to_string(), vec!["network2".to_string()]);

outputs:
Tranches: [["network2", "network1"], ["network1"], ["machine2", "machine1"]]

If the first example is how it should be used then i would need to start analyzing each entry and their dependencies myself and i feel like that should be covered by the repo instead