Struct PositionPostingListReader 

pub struct PositionPostingListReader<'a> { /* private fields */ }

Iterates over a position-aware posting list.

Uses a reusable internal buffer for positions to avoid per-document heap allocation. Supports advance() to skip intermediate documents without decoding their position data.

Implementations

impl<'a> PositionPostingListReader<'a>

pub fn new(data: &'a [u8]) -> Self

Create a reader from position-aware posting list bytes.

pub fn len(&self) -> u32

pub fn is_empty(&self) -> bool

pub fn positions(&self) -> &[u32]

Current document’s positions (valid after next() or advance()).

pub fn first_position(&self) -> u32

The first position for the current document. For TF=1 docs (common in phrase queries), this is the only position. Uses cached value, no Vec access.

pub fn current_tf(&self) -> u32

Term frequency of the current document (valid after next/advance).

pub fn next(&mut self) -> Option<(DocId, Vec<u32>)>

Read the next posting with positions.

pub fn advance(&mut self, target: DocId) -> Option<DocId>

Advance to the first document >= target, decoding its positions. Skips intermediate documents without decoding their position data. Returns the doc_id if found, or None if exhausted. After calling, use positions() to access the decoded positions.

pub fn next_doc(&mut self) -> Option<DocId>

Advance to the next document sequentially (no target, no skip check). Faster than advance() when iterating all docs in order. For TF=1 (most common), skips Vec operations entirely.

pub fn current_doc_id(&self) -> u32

Current document ID (valid after next() or advance()).