Struct SegmentedText 

pub struct SegmentedText { /* private fields */ }

Provider for texts partitioned into segments at known cumulative end positions. lim_at(p) binary-searches the sorted ends list and returns the distance from p to the next boundary.

Storage cost is 8 × n_segments bytes (the cumulative-ends Vec<u64>). For a 50 K-junction SA index on a 6 GB genome that is 400 KB total — vs the 750 MB a packed bitmap would need, and the 6 GB an extra-byte-per-symbol u16 text would need.

Lookup is O(log n_segments) — a few cycles for typical segment counts. The merge can cache lim_p/lim_q across LCP calls so the cost amortises to ~one binary search per output record.

Two constructors:

• from_lengths takes per-segment lengths and builds the cumulative-ends list internally. Most ergonomic when the caller has [chr_len_0, chr_len_1, …] already.
• from_ends takes the sorted cumulative ends directly. Useful when the caller already has them — e.g. STAR’s chr_start[] table.

Both constructors require the segments to cover the whole text (sum(lengths) == text_len, or ends.last() == Some(text_len)).

Implementations

impl SegmentedText

pub fn from_lengths(text_len: usize, lengths: &[usize]) -> Self

Build from per-segment lengths. The sum must equal text_len.

pub fn from_ends(text_len: usize, ends: Vec<u64>) -> Self

Build from sorted, strictly-increasing cumulative end positions. ends.last() must equal text_len.

pub fn text_len(&self) -> usize

Total text length in symbols.

pub fn n_segments(&self) -> usize

Number of segments.

pub fn ends(&self) -> &[u64]

Cumulative end positions, sorted, strictly increasing. ends()[i] is the position one past the last symbol of segment i.

Trait Implementations

impl Clone for SegmentedText

fn clone(&self) -> SegmentedText

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SegmentedText

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

Formats the value using the given formatter.

impl LimitProvider for SegmentedText

fn lim_at(&self, p: usize) -> usize

Logical length of the suffix starting at position p in symbols — i.e. the number of comparable symbols before the next segment boundary or end-of-text. Must be at most text.len() - p.

fn boundary_order( &self, p_a: usize, lim_a: usize, p_b: usize, lim_b: usize, ) -> Ordering

Order to resolve when one or both suffixes hit their boundary before any byte of their shared prefix differs. The default is lim_a.cmp(&lim_b) — “shorter-suffix-is-smaller”, the standard generalised-SA / multi-string-SA convention, what a Vec<&str> sort with &str ordering produces.