Struct ComponentScalarRow 

pub struct ComponentScalarRow { /* private fields */ }

One deterministic scalar residual row lowered from a shaped root.

Implementations

§

impl ComponentScalarRow

Signature links:ComponentScalarRow

pub const fn root_index(&self) -> usize

Signature links:root_indexusize

Original Relation root index.

pub const fn is_imaginary(&self) -> bool

Signature links:is_imaginarybool

Whether this is the imaginary residual coordinate of the original root.

pub const fn component_index(&self) -> &[u32]

Signature links:component_indexu32

Row-major component multi-index within the original shaped root.

pub fn symbols(&self) -> &[ScalarSymbolCoordinate]

Signature links:symbolsScalarSymbolCoordinate

Dense shaped-symbol coordinates expected by Self::evaluate.

pub fn input_slots(&self) -> &[ScalarInputSlot]

Signature links:input_slotsScalarInputSlot

Typed IR-local reads corresponding one-for-one with Self::symbols.

These dense slots are evaluator plumbing. Their source coordinates retain real Semantic identities, but the slots are not Parameters or any other Semantic symbol kind.

pub fn constant_coordinate_jacobian( &self, selected: &[ScalarSymbolCoordinate], ) -> Result<ConstantSymbolJacobian, Diagnostic>

Signature links:constant_coordinate_jacobianScalarSymbolCoordinateResultConstantSymbolJacobianDiagnostic

Prove a row Jacobian constant with respect to every other source coordinate. This uses the scalar SSA proof, retaining real/imaginary component identity. Unselected inputs remain symbolic, so state-dependent mass cannot be frozen accidentally by evaluation at a particular point.

Errors

Rejects duplicate selections, nonlinear dependence, variable coefficients, unsupported instructions, and nonfinite coefficient arithmetic.

pub fn bind_affine( &self, selected: &[ScalarSymbolCoordinate], bindings: &[(ScalarSymbolCoordinate, f64)], ) -> Result<BoundAffineScalarIr<ScalarSymbolCoordinate>, Diagnostic>

Signature links:bind_affineScalarSymbolCoordinateScalarSymbolCoordinatef64ResultBoundAffineScalarIrScalarSymbolCoordinateDiagnostic

Structurally bind a real affine row in exact symbol/channel/part coordinates. This uses the ordinary scalar SSA affine proof, without numerical probing.

Errors

Rejects nonlinear dependence, duplicate coordinates, missing or nonfinite bindings, selected coordinates bound as constants, and invalid arithmetic.

pub fn bind_polynomial_pencil( &self, selected: &[ScalarSymbolCoordinate], spectral: &[ScalarSymbolCoordinate], maximum_degree: u32, term_budget: usize, bindings: &[(ScalarSymbolCoordinate, f64)], ) -> Result<BTreeMap<Vec<u32>, BoundAffineScalarIr<ScalarSymbolCoordinate>>, Diagnostic>

Signature links:bind_polynomial_pencilScalarSymbolCoordinateScalarSymbolCoordinateu32usizeScalarSymbolCoordinatef64ResultBTreeMapVecu32BoundAffineScalarIrScalarSymbolCoordinateDiagnostic

Prove and bind a homogeneous polynomial pencil in real spectral coordinates.

Each key lists spectral powers in the supplied coordinate order; each value is the corresponding homogeneous mode map. Missing keys are zero. Structural degree admission precedes coefficient evaluation. Coefficients are propagated algebraically, never inferred from sampled values. Complex analyticity, physical dimensions and eigenproblem classes remain with the mathematical owner. term_budget limits retained coefficient resources, not the dimension or mathematical degree of admitted finite spaces.

pub fn linearize( &self, inputs: &[f64], roles: &[DifferentiationRole], ) -> Result<ScalarLinearization<'_>, Diagnostic>

Signature links:linearizef64DifferentiationRoleResultScalarLinearizationDiagnostic

Bind first-order products in the exact real component-coordinate order. Complex symbols retain two coordinates per component. The derivative is real-linear: conj, real, imag, and abs2 do not require holomorphy. The VJP uses the Euclidean pairing on these real coordinates, equivalent to Re(sum(conj(a) * b)) on complex components and the ordinary product on real components. An abs2(z) cotangent therefore includes the factor two from its real differential; it is not a Wirtinger partial.

Numerical regularity checks are replayed by the shared primal/JVP/VJP evaluator.

Errors

Rejects wrong cardinality, non-finite points and unsupported numerical instructions.

pub fn evaluate(&self, inputs: &[f64]) -> Result<f64, Diagnostic>

Signature links:evaluatef64Resultf64Diagnostic

Evaluate this scalar row using dense inputs matching Self::symbols.

Errors

Returns the ordinary scalar Operator IR diagnostics for invalid input cardinality or non-finite arithmetic.

Trait Implementations

§

impl Clone for ComponentScalarRow

Signature links:CloneComponentScalarRow

§

fn clone(&self) -> ComponentScalarRow

Signature links:cloneComponentScalarRow

Returns a duplicate of the value.

§

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

Signature links:unstableSourceclone_from

Performs copy-assignment from source.

§

impl Debug for ComponentScalarRow

Signature links:DebugComponentScalarRow

§

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

Signature links:fmtFormatterResult()Error

Formats the value using the given formatter.

§

impl PartialEq for ComponentScalarRow

Signature links:PartialEqComponentScalarRow

§

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

Signature links:eqComponentScalarRowbool

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

§

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

Signature links:unstableSourcene&Rhsbool

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

impl StructuralPartialEq for ComponentScalarRow