graphblas_sparse_linear_algebra

Rust wrapper for SuiteSparse:GraphBLAS

Minimum example

use graphblas_sparse_linear_algebra::collections::Collection;
use graphblas_sparse_linear_algebra::context::{Context, Mode};
use graphblas_sparse_linear_algebra::operators::apply::{BinaryOperatorApplier, ApplyBinaryOperator};
use graphblas_sparse_linear_algebra::operators::binary_operator::{Assignment, First};
use graphblas_sparse_linear_algebra::operators::{binary_operator::BinaryOperator, options::OperatorOptions};
use graphblas_sparse_linear_algebra::operators::mask::{MatrixMask, SelectEntireMatrix};
use graphblas_sparse_linear_algebra::collections::sparse_matrix::{
    FromMatrixElementList, GetMatrixElementValue, MatrixElementList, Size, SparseMatrix
};
use graphblas_sparse_linear_algebra::collections::sparse_scalar::SparseScalar;
use graphblas_sparse_linear_algebra::collections::sparse_vector::{
    FromVectorElementList, GetVectorElementValue, VectorElementList,
};

fn main() {
    let context = Context::init_ready(Mode::NonBlocking).unwrap();

    let element_list = MatrixElementList::<u8>::from_element_vector(vec![
        (1, 1, 1).into(),
        (2, 1, 2).into(),
        (4, 2, 4).into(),
        (5, 2, 5).into(),
    ]);

    let matrix_size: Size = (10, 15).into();
    let matrix = SparseMatrix::<u8>::from_element_list(
        &context.to_owned(),
        &matrix_size,
        &element_list,
        &First::<u8>::new(),
    )
    .unwrap();

    let mut product_matrix = SparseMatrix::<u8>::new(&context, &matrix_size).unwrap();

    let operator = BinaryOperatorApplier::new(
    );
    let first_agrument = 10;
    operator
        .apply_with_matrix_as_left_argument(
            &matrix, &First::<u8>::new(), 
            &first_agrument, 
            &Assignment::new(), 
            &mut product_matrix,
            &SelectEntireMatrix::new(&context),
            &OperatorOptions::new_default())
        .unwrap();

    println!("{}", product_matrix);

    assert_eq!(product_matrix.number_of_stored_elements().unwrap(), 4);
    assert_eq!(product_matrix.get_element_value(&(2, 1).into()).unwrap(), Some(2));
    assert_eq!(product_matrix.get_element_value(&(9, 1).into()).unwrap(), None);

    let operator = BinaryOperatorApplier::new();
    let second_agrument = 10;
    operator
        .apply_with_matrix_as_right_argument(
            &second_agrument,
            &First::<u8>::new(), 
            &matrix, 
            &Assignment::new(),
            &mut product_matrix,
            &SelectEntireMatrix::new(&context),
            &OperatorOptions::new_default())
        .unwrap();

    println!("{}", matrix);
    println!("{}", product_matrix);

    assert_eq!(product_matrix.number_of_stored_elements().unwrap(), 4);
    assert_eq!(
        product_matrix.get_element_value(&(2, 1).into()).unwrap(),
        Some(10)
    );
    assert_eq!(product_matrix.get_element_value(&(9, 1).into()).unwrap(), None);
 }

Dependencies

graphblas_sparse_linear_algebra uses the SuiteSparse:GraphBLAS implementation developed by Timothy A. Davis.

Building from source

Please make sure to meet the requirements for building suitesparse_graphblas_sys.

Compatibility

graphblas_sparse_linear_algebra is mostly compatible with the GraphBLAS specification version 2.0 and uses SuiteSparse:GraphBLAS v7.3.0.

Contributing

Awesome, contributions are welcome. Graphblas_sparse_linear_algebra and your contribution may be relicensed and integrated into commercial software in the future. Therefore, you will be asked to agree to the Contributor License Agreement when you make a pull request.

Licensing

graphblas_sparse_linear_algebra is licensed under Creative Commons Attribution Non Commercial 4.0 International. For other licensing options, please contact Sam Dekker.