Struct SparseBitMatrix

pub struct SparseBitMatrix<R, C>
where
    R: Idx,
    C: Idx,
{ /* private fields */ }

A fixed-column-size, variable-row-size 2D bit matrix with a moderately sparse representation.

Initially, every row has no explicit representation. If any bit within a row is set, the entire row is instantiated as Some(<DenseBitSet>). Furthermore, any previously uninstantiated rows prior to it will be instantiated as None. Those prior rows may themselves become fully instantiated later on if any of their bits are set.

R and C are index types used to identify rows and columns respectively; typically newtyped usize wrappers, but they can also just be usize.

Implementations

impl<R: Idx, C: Idx> SparseBitMatrix<R, C>

pub fn new(num_columns: usize) -> Self

Creates a new empty sparse bit matrix with no rows or columns.

pub fn insert(&mut self, row: R, column: C) -> bool

Sets the cell at (row, column) to true. Put another way, insert column to the bitset for row.

Returns true if this changed the matrix.

pub fn remove(&mut self, row: R, column: C) -> bool

Sets the cell at (row, column) to false. Put another way, delete column from the bitset for row. Has no effect if row does not exist.

Returns true if this changed the matrix.

pub fn clear(&mut self, row: R)

Sets all columns at row to false. Has no effect if row does not exist.

pub fn contains(&self, row: R, column: C) -> bool

Do the bits from row contain column? Put another way, is the matrix cell at (row, column) true? Put yet another way, if the matrix represents (transitive) reachability, can row reach column?

pub fn union_rows(&mut self, read: R, write: R) -> bool

Adds the bits from row read to the bits from row write, and returns true if anything changed.

This is used when computing transitive reachability because if you have an edge write -> read, because in that case write can reach everything that read can (and potentially more).

pub fn insert_all_into_row(&mut self, row: R)

Insert all bits in the given row.

pub fn rows(&self) -> impl Iterator<Item = R>

pub fn iter(&self, row: R) -> impl Iterator<Item = C> + '_

Iterates through all the columns set to true in a given row of the matrix.

pub fn row(&self, row: R) -> Option<&DenseBitSet<C>>

pub fn intersect_row<Set>(&mut self, row: R, set: &Set) -> bool

where DenseBitSet<C>: BitRelations<Set>,

Intersects row with set. set can be either DenseBitSet or ChunkedBitSet. Has no effect if row does not exist.

Returns true if the row was changed.

pub fn subtract_row<Set>(&mut self, row: R, set: &Set) -> bool

where DenseBitSet<C>: BitRelations<Set>,

Subtracts set from row. set can be either DenseBitSet or ChunkedBitSet. Has no effect if row does not exist.

Returns true if the row was changed.

pub fn union_row<Set>(&mut self, row: R, set: &Set) -> bool

where DenseBitSet<C>: BitRelations<Set>,

Unions row with set. set can be either DenseBitSet or ChunkedBitSet.

Returns true if the row was changed.

Trait Implementations

impl<R, C> Clone for SparseBitMatrix<R, C>

where R: Idx + Clone, C: Idx + Clone,

fn clone(&self) -> SparseBitMatrix<R, C>

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<R, C> Debug for SparseBitMatrix<R, C>

where R: Idx + Debug, C: Idx + Debug,

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

Formats the value using the given formatter.