instant-labs
/
instant-segment
mirror of https://github.com/instant-labs/instant-segment.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Use more compact cache key

This commit is contained in:
Dirkjan Ochtman 2021-02-11 11:55:27 +01:00
parent 83aa46593a
commit 9addd3810b
1 changed files with 9 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
src/lib.rs
View File

@ -148,13 +148,18 @@ impl<'a> SegmentState<'a> {
let previous = previous.clone().map(|range| &self.text[range]); let previous = previous.clone().map(|range| &self.text[range]);
let prefix_score = self.data.score(&self.text[start..split], previous).log10(); let prefix_score = self.data.score(&self.text[start..split], previous).log10();
let pair = (split..end, start..split); let key = (
let (suffix_score, suffix_splits) = match self.search.memo.get(&pair) { (start - self.offset) as u8,
(split - self.offset) as u8,
(end - self.offset) as u8,
);
let (suffix_score, suffix_splits) = match self.search.memo.get(&key) {
Some((score, suffix_splits)) => (*score, *suffix_splits), Some((score, suffix_splits)) => (*score, *suffix_splits),
None => { None => {
let suffix_score = self.search(level + 1, split..end, Some(start..split)); let suffix_score = self.search(level + 1, split..end, Some(start..split));
let suffix_splits = self.search.best[level + 1]; let suffix_splits = self.search.best[level + 1];
self.search.memo.insert(pair, (suffix_score, suffix_splits)); self.search.memo.insert(key, (suffix_score, suffix_splits));
(suffix_score, suffix_splits) (suffix_score, suffix_splits)
} }
}; };
@ -175,7 +180,7 @@ impl<'a> SegmentState<'a> {
#[derive(Clone)] #[derive(Clone)]
pub struct Search { pub struct Search {
memo: HashMap<MemoKey, (f64, BitVec)>, memo: HashMap<(u8, u8, u8), (f64, BitVec)>,
best: [BitVec; SEGMENT_SIZE], best: [BitVec; SEGMENT_SIZE],
result: Vec<String>, result: Vec<String>,
} }
@ -264,8 +269,6 @@ impl Iterator for Splits {
} }
} }
type MemoKey = (Range<usize>, Range<usize>);
#[derive(Debug)] #[derive(Debug)]
struct Ascii<'a>(&'a [u8]); struct Ascii<'a>(&'a [u8]);