Struct brane_cli::planner::OfflinePlanner

pub struct OfflinePlanner {
    pub results: Arc<Mutex<HashMap<String, String>>>,
    /* private fields */
}

The planner is in charge of assigning locations to tasks in a workflow. This one is very simple, assigning ‘localhost’ to whatever it sees.

Fields

results: Arc<Mutex<HashMap<String, String>>>

The results we planned last time (or whatever).

Implementations

impl OfflinePlanner

pub fn new( data_index: Arc<DataIndex>, results: Arc<Mutex<HashMap<String, String>>>, ) -> Self

Constructor for the OfflinePlanner.

Arguments

• data_index: The DataIndex that is used to resolve datasets at plantime.
• results: A map of results to where they are, which we planned last time around.

Returns

A new OfflinePlanner instance.

pub fn plan_data( name: &DataName, avail: &mut Option<AvailabilityKind>, dindex: &Arc<DataIndex>, results: &HashMap<String, String>, deferred: bool, ) -> Result<(), PlanError>

Plans the given task offline.

Arguments

• name: The name of the dataset or intermediate result, as a DataName (so we can distinguish between the two).
• avail: The availability for this dataset that we will be updating.
• dindex: The DataIndex we use to see what datasets are actually available where.
• results: The map of results that are known in this workflow.
• deferred: If true, then will not error if we failed to find a result yet (its declaration might come later, in that case).

Returns

Nothing, but does change the dataset’s availability.

Trait Implementations

impl Debug for OfflinePlanner

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Planner for OfflinePlanner

fn plan<'life0, 'async_trait>( &'life0 self, workflow: Workflow, ) -> Pin<Box<dyn Future<Output = Result<Workflow, PlanError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Plans the given workflow by: