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

//! Offline Algorithms.

pub mod multi_dimensional;
pub mod uni_dimensional;

mod graph_search;

pub use graph_search::{Cache, CachedPath};

use crate::algorithms::Options;
use crate::model::{ModelOutputFailure, ModelOutputSuccess};
use crate::problem::Problem;
use crate::result::Result;
use crate::schedule::Schedule;
use pyo3::prelude::*;

/// Implementation of an offline algorithm.
///
/// * `R` - result
/// * `P` - problem
/// * `O` - options
///
/// Receives the arguments:
/// * `p` - problem instance
/// * `options` - algorithm options
/// * `offline_options` - general configuration of the offlien setting
pub trait OfflineAlgorithm<T, R, P, O, C, D>:
    Fn(P, O, OfflineOptions) -> Result<R>
where
    R: OfflineResult<T>,
    P: Problem<T, C, D>,
    O: Options<T, P, C, D>,
    C: ModelOutputSuccess,
    D: ModelOutputFailure,
{
    fn solve(
        &self,
        p: P,
        options: O,
        offline_options: OfflineOptions,
    ) -> Result<R> {
        self(p, options, offline_options)
    }

    fn solve_with_default_options(
        &self,
        p: P,
        offline_options: OfflineOptions,
    ) -> Result<R> {
        let options = O::default(&p);
        self.solve(p, options, offline_options)
    }
}
impl<T, R, P, O, C, D, F> OfflineAlgorithm<T, R, P, O, C, D> for F
where
    R: OfflineResult<T>,
    P: Problem<T, C, D>,
    O: Options<T, P, C, D>,
    C: ModelOutputSuccess,
    D: ModelOutputFailure,
    F: Fn(P, O, OfflineOptions) -> Result<R>,
{
}

/// Configuration of offline algorithms.
#[pyclass]
#[derive(Clone, Debug)]
pub struct OfflineOptions {
    /// Compute inverted movement costs (SSCO only).
    #[pyo3(get, set)]
    pub inverted: bool,
    /// Compute the $\alpha$-unfair offline optimum.
    #[pyo3(get, set)]
    pub alpha: f64,
    /// Compute the $L$-constrained offline optimum (`convex_optimization` only).
    #[pyo3(get, set)]
    pub l: Option<f64>,
}
impl Default for OfflineOptions {
    fn default() -> Self {
        OfflineOptions {
            inverted: false,
            alpha: 1.,
            l: None,
        }
    }
}
impl OfflineOptions {
    pub fn inverted() -> Self {
        Self {
            inverted: true,
            ..Self::default()
        }
    }

    pub fn alpha_unfair(alpha: f64) -> Self {
        Self {
            alpha,
            ..Self::default()
        }
    }

    pub fn l_constrained(l: f64) -> Self {
        Self {
            l: Some(l),
            ..Self::default()
        }
    }
}
#[pymethods]
impl OfflineOptions {
    #[new]
    pub fn new(inverted: bool, alpha: f64, l: Option<f64>) -> Self {
        OfflineOptions { inverted, alpha, l }
    }
}

/// Result of an offline algorithm.
pub trait OfflineResult<T> {
    /// Resulting schedule.
    fn xs(self) -> Schedule<T>;
}

/// Result of an offline algorithm which only returns the obtained schedule.
pub struct PureOfflineResult<T> {
    /// Schedule.
    pub xs: Schedule<T>,
}
impl<T> OfflineResult<T> for PureOfflineResult<T> {
    fn xs(self) -> Schedule<T> {
        self.xs
    }
}