Struct IncrementalPageRank 

pub struct IncrementalPageRank { /* private fields */ }

Stateful incremental PageRank engine bound to a mutable graph.

Implementations

impl IncrementalPageRank

pub fn new( graph: DynamicGraph, damping: f64, max_iter: usize, tol: f64, ) -> GraphalgResult<Self>

Create an engine over graph and converge an initial cold-start solve.

• damping — the teleport-complement d ∈ [0, 1] (typically 0.85).
• max_iter — power-iteration cap per (re)solve.
• tol — L1 convergence tolerance between successive iterates.

pub fn ranks(&self) -> &[f64]

Current (normalised) rank vector.

pub fn rank_of(&self, v: usize) -> GraphalgResult<f64>

PageRank of a single vertex.

pub fn last_iterations(&self) -> usize

Iterations consumed by the most recent (re)solve — strictly informational, useful for demonstrating the warm-start speed-up in tests/benchmarks.

pub fn graph(&self) -> &DynamicGraph

Borrow the underlying graph.

pub fn apply_insert(&mut self, u: usize, v: usize) -> GraphalgResult<()>

Insert a directed edge and re-converge from the warm start.

pub fn apply_remove(&mut self, u: usize, v: usize) -> GraphalgResult<()>

Remove a directed edge and re-converge from the warm start.

pub fn apply_batch( &mut self, updates: &[(usize, usize, bool)], ) -> GraphalgResult<()>

Apply a batch of (u, v, is_insert) mutations atomically, re-converging once at the end (cheaper than re-solving per edge).

pub fn recompute(&mut self)

Resume the power iteration from the current rank vector. Public so callers can force a re-solve after mutating the graph through other means.

Trait Implementations

impl Clone for IncrementalPageRank

fn clone(&self) -> IncrementalPageRank

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for IncrementalPageRank

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

Formats the value using the given formatter.