Struct IndexSchurData 

pub struct IndexSchurData { /* private fields */ }

Specialization for B matrices whose non-zero entries are ±1 (the parametric / reduced-Hessian common case). Storage is two parallel arrays mirroring upstream’s idx_ and val_ (SensIndexSchurData.hpp:127-128).

Implementations

impl IndexSchurData

pub fn new() -> Self

Empty / uninitialized instance. Must be populated via one of the set_from_* methods before being read.

pub fn from_parts( idx: Vec<Index>, val: Vec<Index>, ) -> Result<Self, SchurDataError>

Construct directly from pre-built (idx, val) arrays. val must contain only ±1 entries. Mirrors upstream’s two-arg constructor (SensIndexSchurData.cpp:24-36).

pub fn col_indices(&self) -> &[Index]

Column indices the rows refer to (one index per row). Upstream GetColIndices() (SensIndexSchurData.hpp:114).

pub fn signs(&self) -> &[Index]

Per-row ±1 sign carried by the single non-zero entry in that row. Pounce-specific accessor; upstream exposes the data only through the multiplying_* / multiply APIs.

Trait Implementations

impl Clone for IndexSchurData

fn clone(&self) -> IndexSchurData

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for IndexSchurData

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

Formats the value using the given formatter.

impl Default for IndexSchurData

fn default() -> IndexSchurData

Returns the “default value” for a type.

impl PartialEq for IndexSchurData

fn eq(&self, other: &IndexSchurData) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl SchurData for IndexSchurData

fn nrows(&self) -> Index

Number of rows the schur matrix has, i.e. the row count of B. Upstream GetNRowsAdded() (SensSchurData.hpp:77-80).

fn is_initialized(&self) -> bool

true if one of the set_* methods has been called and the rest of the surface is safe to call. Upstream Is_Initialized() (SensSchurData.hpp:82-85).

fn set_from_flags( &mut self, flags: &[Index], v: Number, ) -> Result<(), SchurDataError>

Set rows from a 0/1 flag array of length dim. For each i with flags[i] == 1, add a row whose only non-zero column is i with sign sign(v). The magnitude of v is collapsed to ±1 — this trait only ever stores signs, mirroring upstream’s SetData_Flag(dim, flags, v) (SensSchurData.hpp:45-49, SensIndexSchurData.cpp:51-78).

fn set_from_list( &mut self, cols: &[Index], v: Number, ) -> Result<(), SchurDataError>

Set rows from a list of column indices. Each cols[k] becomes row k of B, with the single non-zero entry at column cols[k] carrying sign sign(v). Upstream SetData_List (SensSchurData.hpp:64-67, SensIndexSchurData.cpp:149-167).

fn multiplying_row( &self, i: Index, ) -> Result<(Vec<Index>, Vec<Number>), SchurDataError>

Row-i access: return the parallel arrays (indices, factors) such that row i of B has non-zero column entries factors[j] at columns indices[j]. For IndexSchurData indices has length 1 and factors[0] == ±1.0. Upstream GetMultiplyingVectors (SensSchurData.hpp:93-104, SensIndexSchurData.cpp:199-212).

fn multiply(&self, v: &[Number], u: &mut [Number]) -> Result<(), SchurDataError>

Apply u = B v for a v of length n_full and pre-sized u buffer (length must equal self.nrows()). Upstream Multiply(IteratesVector, Vector) (SensSchurData.hpp:106-110, SensIndexSchurData.cpp:214-251).

fn trans_multiply( &self, u: &[Number], v: &mut [Number], ) -> Result<(), SchurDataError>

Apply v = Bᵀ u for a u of length self.nrows() and a pre-sized v buffer (length n_full). Upstream TransMultiply (SensSchurData.hpp:112-116, SensIndexSchurData.cpp:253-307).

impl Eq for IndexSchurData

impl StructuralPartialEq for IndexSchurData