Struct MatchScan 

pub struct MatchScan {
    next_start: usize,
    matchless_upto: usize,
}

Cursor for Nfa::next_match: the next start position the scan will try. A fresh scan starts at 0; Nfa::next_match advances it by the AFTER MATCH SKIP mode on every match returned (or by one row past a non-matching start), so pulling repeatedly enumerates exactly the match sequence Nfa::find_matches_dynamic would collect.

Fields

next_start: usize

matchless_upto: usize

End of the contiguous run of starts, from where this scan began, proven MATCHLESS FOREVER: their walks found no accept and never reached the boundary, so every path from them died on a row below it — and those rows are immutable, so no arrival can revive them. The incremental matcher reads this back and begins its next rescan past them.

Implementations

impl MatchScan

pub fn new() -> Self

pub fn starting_at(start: usize) -> Self

A scan beginning at start, with nothing proven yet.

pub fn matchless_upto(&self) -> usize

See the matchless_upto field.

pub fn next_start(&self) -> usize

The next start the pull loop would explore.

This is NOT a “fully scanned prefix” marker, and must not be used as one: on a hit the cursor jumps to skip.next_pos, so under PAST LAST ROW every start strictly inside the match it just returned was never evaluated. What it does guarantee, since the budget fix below, is the narrower property that a start whose walk the budget ABORTED is not advanced past — so the cursor never claims a verdict the walk did not reach.

No production caller reads this today; it exists for the emit-on-update port, which needs to resume a pull, and for the test that pins the abort behaviour.

Trait Implementations

impl Debug for MatchScan

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

Formats the value using the given formatter.

impl Default for MatchScan

fn default() -> MatchScan

Returns the “default value” for a type.