Struct ProjectedBlockOps 

pub struct ProjectedBlockOps { /* private fields */ }

A direct sum of sector operators acting in a shared parent Hilbert space.

Each projector maps one sector coordinate vector into the parent space. The combined projector is validated as an isometry, so this operator represents P (⊕ H_sector) P† without materializing either the direct sum or the parent-space matrix.

Implementations

impl ProjectedBlockOps

pub fn new( blocks: impl IntoIterator<Item = Arc<dyn LinearOperator>>, projectors: impl IntoIterator<Item = Arc<dyn LinearOperator>>, tolerance: f64, ) -> Result<Self>

pub fn blocks(&self) -> usize

pub fn full_dimension(&self) -> usize

pub fn block_dimension(&self) -> usize

pub fn project( &self, parent: &[Complex64], blocks: &mut [Complex64], ) -> Result<()>

pub fn lift(&self, blocks: &[Complex64], parent: &mut [Complex64]) -> Result<()>

pub fn completeness_residual(&self) -> Result<f64>

Maximum entrywise residual of P P† - I in the parent basis.

An isometric projector collection need not be complete: selected sectors can intentionally span only a subspace. Callers which require a full decomposition can enforce their own tolerance on this residual.

pub fn materialize(&self, format: MatrixFormat) -> Result<Operator>

Trait Implementations

impl Clone for ProjectedBlockOps

fn clone(&self) -> ProjectedBlockOps

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl LinearOperator for ProjectedBlockOps

fn shape(&self) -> (usize, usize)

Return (rows, columns).

fn format(&self) -> MatrixFormat

Report the public materialization format.

fn apply(&self, input: &[Complex64], output: &mut [Complex64]) -> Result<()>

Compute output = A * input.

fn is_real(&self) -> bool

Whether this operator preserves real vectors exactly.

fn apply_real(&self, input: &[f64], output: &mut [f64]) -> Result<()>

Apply this operator to a real vector.

fn apply_transpose( &self, input: &[Complex64], output: &mut [Complex64], ) -> Result<()>

Apply the algebraic transpose without conjugating either operand.

fn apply_adjoint( &self, input: &[Complex64], output: &mut [Complex64], ) -> Result<()>

Apply the conjugate transpose.

fn stored_triplets(&self) -> Result<Option<Vec<(usize, usize, Complex64)>>>

Return canonical stored nonzeros when the representation already owns them.

fn shifted_solver( &self, _shift: f64, ) -> Result<Option<Box<dyn ShiftedLinearSolver>>>

Prepare a reusable solver for (A - shift I) x = b, when supported.