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

use opensrdk_linear_algebra::sparse::SparseTensor;

use crate::{BracketsLevel, Expression, Size, TensorExpression};
use std::{collections::HashMap, iter::once};

pub trait DirectProduct {
    fn direct_product(self) -> Expression;
}

impl<I> DirectProduct for I
where
    I: Iterator<Item = Expression>,
{
    fn direct_product(self) -> Expression {
        TensorExpression::DirectProduct(self.collect()).into()
    }
}

impl Expression {
    pub fn direct(self, rhs: Expression) -> Expression {
        vec![self, rhs].into_iter().direct_product()
    }
}

impl TensorExpression {
    pub(crate) fn diff_direct_product(
        terms: &Vec<Expression>,
        symbols: &[&str],
    ) -> Vec<Expression> {
        let terms_len = terms.len();
        let symbols_len = symbols.len();

        let result = (0..terms_len)
            .map(|i| {
                let elems_left = (0..i).map(|j| terms[j].clone()).direct_product();
                let elems_right = (i + 1..terms_len)
                    .map(|k| terms[k].clone())
                    .direct_product();
                let elem_diff = terms[i].differential(symbols);

                let elems = (0..symbols_len)
                    .map(|l| {
                        elems_left
                            .clone()
                            .direct(elem_diff[l].clone())
                            .direct(elems_right.clone())
                    })
                    .collect::<Vec<Expression>>();
                elems
            })
            .fold(vec![Expression::from(0f64); symbols_len], |sum, x| {
                let result_orig = (0..symbols_len)
                    .map(|m| sum[m].clone() + x[m].clone())
                    .collect::<Vec<Expression>>();
                result_orig
            });
        result
    }

    pub(crate) fn tex_code_direct_product(
        terms: &Vec<Expression>,
        symbols: &HashMap<&str, &str>,
        brackets_level: BracketsLevel,
    ) -> String {
        let inner = terms
            .into_iter()
            .map(|t| t.tex_code(symbols))
            .collect::<Vec<_>>()
            .join(r" \otimes ");

        match brackets_level {
            BracketsLevel::None => inner,
            BracketsLevel::ForMul | BracketsLevel::ForDiv | BracketsLevel::ForOperation => {
                format!(r"\left({}\right)", inner)
            }
        }
    }

    pub(crate) fn size_direct_product(terms: &Vec<Expression>) -> Vec<Size> {
        terms
            .into_iter()
            .map(|t| t.sizes())
            .fold(vec![], |mut acc, next| {
                if acc.len() < next.len() {
                    for i in 0..acc.len() {
                        if next[i] == Size::Many {
                            acc[i] = next[i];
                        }
                    }
                    acc.extend(next[acc.len()..].iter().copied());
                } else {
                    for i in 0..next.len() {
                        if next[i] == Size::Many {
                            acc[i] = next[i];
                        }
                    }
                }
                acc
            })
    }
}

#[cfg(test)]
mod tests {
    use std::{
        collections::{HashMap, HashSet},
        ops::Add,
    };

    use opensrdk_linear_algebra::{sparse::SparseTensor, Matrix};

    use crate::{new_variable, Expression, MatrixExpression, TensorExpression};

    #[test]
    fn it_works() {
        let mut hash1 = HashMap::new();
        hash1.insert(vec![3usize; 8], 2.0);
        hash1.insert(vec![1usize; 8], 3.0);
        hash1.insert(vec![4usize; 8], 4.0);
        hash1.insert(vec![5usize; 8], 2.0);
        let a = SparseTensor::from(vec![6usize; 8], hash1).unwrap();

        let ea = Expression::from(a);

        let mut hash2 = HashMap::new();
        hash2.insert(vec![3usize; 8], 2.0);
        hash2.insert(vec![2usize; 8], 3.0);
        hash2.insert(vec![4usize; 8], 1.0);
        let b = SparseTensor::from(vec![6usize; 8], hash2).unwrap();

        let eb = Expression::from(b);

        let dp = ea.direct(eb);

        println!("{:?}", dp);
    }

    #[test]
    fn it_works1() {
        let mut hash1 = HashMap::new();
        hash1.insert(vec![3usize; 8], 2.0);
        hash1.insert(vec![1usize; 8], 3.0);
        hash1.insert(vec![4usize; 8], 4.0);
        hash1.insert(vec![5usize; 8], 2.0);
        let a = SparseTensor::from(vec![6usize; 8], hash1).unwrap();

        let ea = Expression::from(a);

        let mut hash2 = HashMap::new();
        hash2.insert(vec![3usize; 8], 2.0);
        hash2.insert(vec![2usize; 8], 3.0);
        hash2.insert(vec![4usize; 8], 1.0);
        let b = SparseTensor::from(vec![6usize; 8], hash2).unwrap();

        let eb = Expression::from(b);

        let id = "x";
        let ec = new_variable((id).to_string());

        let ids = &["x", "y"];

        let diff_dp = TensorExpression::diff_direct_product(&vec![ec.clone(), ea, eb, ec], ids);
        println!("{:?}", diff_dp);

        let tex_symbols = vec![("x", "y")].into_iter().collect();
        let tex_x = diff_dp[0].tex_code(&tex_symbols);
        let tex_y = diff_dp[1].tex_code(&tex_symbols);

        println!("{:?}", tex_x);
        println!("{:?}", tex_y);
    }
}