Skip to content

Offline, privacy-first grammar checker. Fast, open-source, Rust-powered

Use `Option<NonZeroUsize>` as the return type of `Pattern::matches` (#1124)

Michael Schmidt8b2135b065e5
diff --git a/harper-core/src/linting/back_in_the_day.rs b/harper-core/src/linting/back_in_the_day.rs
index 2ce000507..814a5d335 100644
--- a/harper-core/src/linting/back_in_the_day.rs
+++ b/harper-core/src/linting/back_in_the_day.rs
@@ -36,7 +36,7 @@ impl PatternLinter for BackInTheDay {
fn match_to_lint(&self, matched_tokens: &[Token], source: &[char]) -> Option<Lint> {
if let Some(tail) = matched_tokens.get(8..) {
- if self.exceptions.matches(tail, source) != 0 {
+ if self.exceptions.matches(tail, source).is_some() {
return None;
}
}
diff --git a/harper-core/src/linting/it_is.rs b/harper-core/src/linting/it_is.rs
index eee7a6b98..9c29cf56c 100644
--- a/harper-core/src/linting/it_is.rs
+++ b/harper-core/src/linting/it_is.rs
@@ -44,7 +44,7 @@ impl Default for ItIs {
if !meta.is_adjective() {
return false;
}
- if exceptions.matches(&[tok.clone()], src) > 0 {
+ if exceptions.matches(&[tok.clone()], src).is_some() {
return false;
}
true
diff --git a/harper-core/src/linting/no_oxford_comma.rs b/harper-core/src/linting/no_oxford_comma.rs
index 74634478a..143ef96bb 100644
--- a/harper-core/src/linting/no_oxford_comma.rs
+++ b/harper-core/src/linting/no_oxford_comma.rs
@@ -63,14 +63,14 @@ impl Linter for NoOxfordComma {
.pattern
.matches(&sentence[tok_cursor..], document.get_source());
- if match_len != 0 {
+ if let Some(match_len) = match_len {
let lint = self.match_to_lint(
- &sentence[tok_cursor..tok_cursor + match_len],
+ &sentence[tok_cursor..tok_cursor + match_len.get()],
document.get_source(),
);
lints.extend(lint);
- tok_cursor += match_len;
+ tok_cursor += match_len.get();
} else {
tok_cursor += 1;
}
diff --git a/harper-core/src/linting/oxford_comma.rs b/harper-core/src/linting/oxford_comma.rs
index 2962f4284..a825c597c 100644
--- a/harper-core/src/linting/oxford_comma.rs
+++ b/harper-core/src/linting/oxford_comma.rs
@@ -83,14 +83,14 @@ impl Linter for OxfordComma {
.pattern
.matches(&sentence[tok_cursor..], document.get_source());
- if match_len != 0 {
+ if let Some(match_len) = match_len {
let lint = self.match_to_lint(
- &sentence[tok_cursor..tok_cursor + match_len],
+ &sentence[tok_cursor..tok_cursor + match_len.get()],
document.get_source(),
);
lints.extend(lint);
- tok_cursor += match_len;
+ tok_cursor += match_len.get();
} else {
tok_cursor += 1;
}
diff --git a/harper-core/src/linting/pattern_linter.rs b/harper-core/src/linting/pattern_linter.rs
index da213ac8c..c34296760 100644
--- a/harper-core/src/linting/pattern_linter.rs
+++ b/harper-core/src/linting/pattern_linter.rs
@@ -53,11 +53,12 @@ pub fn run_on_chunk(linter: &impl PatternLinter, chunk: &[Token], source: &[char
let match_len = linter.pattern().matches(&chunk[tok_cursor..], source);
- if match_len != 0 {
- let lint = linter.match_to_lint(&chunk[tok_cursor..tok_cursor + match_len], source);
+ if let Some(match_len) = match_len {
+ let lint =
+ linter.match_to_lint(&chunk[tok_cursor..tok_cursor + match_len.get()], source);
lints.extend(lint);
- tok_cursor += match_len;
+ tok_cursor += match_len.get();
} else {
tok_cursor += 1;
}
diff --git a/harper-core/src/patterns/all.rs b/harper-core/src/patterns/all.rs
index d8342f4e9..7e5122581 100644
--- a/harper-core/src/patterns/all.rs
+++ b/harper-core/src/patterns/all.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::Token;
use super::Pattern;
@@ -22,21 +24,14 @@ impl All {
}
impl Pattern for All {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
let mut max = 0;
for pattern in &self.children {
- let len = pattern.matches(tokens, source);
-
- if len == 0 {
- return 0;
- }
-
- if len > max {
- max = len;
- }
+ let len = pattern.matches(tokens, source)?;
+ max = max.max(len.get());
}
- max
+ NonZeroUsize::new(max)
}
}
diff --git a/harper-core/src/patterns/any_capitalization.rs b/harper-core/src/patterns/any_capitalization.rs
index a66588f19..67cecab25 100644
--- a/harper-core/src/patterns/any_capitalization.rs
+++ b/harper-core/src/patterns/any_capitalization.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::{CharString, Token};
use super::Pattern;
@@ -21,17 +23,15 @@ impl AnyCapitalization {
}
impl Pattern for AnyCapitalization {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let Some(tok) = tokens.first() else {
- return 0;
- };
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ let tok = tokens.first()?;
if !tok.kind.is_word() {
- return 0;
+ return None;
}
if tok.span.len() != self.word.len() {
- return 0;
+ return None;
}
let tok_chars = tok.span.get_content(source);
@@ -41,6 +41,6 @@ impl Pattern for AnyCapitalization {
.zip(&self.word)
.all(|(a, b)| a.eq_ignore_ascii_case(b));
- if partial_match { 1 } else { 0 }
+ NonZeroUsize::new(if partial_match { 1 } else { 0 })
}
}
diff --git a/harper-core/src/patterns/any_pattern.rs b/harper-core/src/patterns/any_pattern.rs
index d7cb454b4..2d68b4516 100644
--- a/harper-core/src/patterns/any_pattern.rs
+++ b/harper-core/src/patterns/any_pattern.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::Token;
use super::Pattern;
@@ -6,7 +8,7 @@ use super::Pattern;
pub struct AnyPattern;
impl Pattern for AnyPattern {
- fn matches(&self, tokens: &[Token], _source: &[char]) -> usize {
- if tokens.is_empty() { 0 } else { 1 }
+ fn matches(&self, tokens: &[Token], _source: &[char]) -> Option<NonZeroUsize> {
+ NonZeroUsize::new(if tokens.is_empty() { 0 } else { 1 })
}
}
diff --git a/harper-core/src/patterns/consumes_remaining_pattern.rs b/harper-core/src/patterns/consumes_remaining_pattern.rs
index ef1359815..b522a1072 100644
--- a/harper-core/src/patterns/consumes_remaining_pattern.rs
+++ b/harper-core/src/patterns/consumes_remaining_pattern.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use super::Pattern;
use crate::Token;
@@ -15,13 +17,9 @@ impl ConsumesRemainingPattern {
}
impl Pattern for ConsumesRemainingPattern {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let match_len = self.inner.matches(tokens, source);
-
- if match_len == tokens.len() {
- match_len
- } else {
- 0
- }
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ self.inner
+ .matches(tokens, source)
+ .filter(|match_len| match_len.get() == tokens.len())
}
}
diff --git a/harper-core/src/patterns/either_pattern.rs b/harper-core/src/patterns/either_pattern.rs
index db6878d3e..512818828 100644
--- a/harper-core/src/patterns/either_pattern.rs
+++ b/harper-core/src/patterns/either_pattern.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::Token;
use super::Pattern;
@@ -19,17 +21,17 @@ impl EitherPattern {
}
impl Pattern for EitherPattern {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
let mut longest = 0;
for pattern in self.patterns.iter() {
- let match_len = pattern.matches(tokens, source);
+ let match_len = pattern.matches(tokens, source).map_or(0, NonZeroUsize::get);
if match_len > longest {
longest = match_len
}
}
- longest
+ NonZeroUsize::new(longest)
}
}
diff --git a/harper-core/src/patterns/exact_phrase.rs b/harper-core/src/patterns/exact_phrase.rs
index f6075962e..9eb543cd2 100644
--- a/harper-core/src/patterns/exact_phrase.rs
+++ b/harper-core/src/patterns/exact_phrase.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::{Document, Token, TokenKind};
use super::{AnyCapitalization, Pattern, SequencePattern};
@@ -44,7 +46,7 @@ impl ExactPhrase {
}
impl Pattern for ExactPhrase {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
self.inner.matches(tokens, source)
}
}
diff --git a/harper-core/src/patterns/implies_quantity.rs b/harper-core/src/patterns/implies_quantity.rs
index be2ef7b25..06a96cf27 100644
--- a/harper-core/src/patterns/implies_quantity.rs
+++ b/harper-core/src/patterns/implies_quantity.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::{Token, TokenKind};
use super::Pattern;
@@ -36,12 +38,12 @@ impl ImpliesQuantity {
}
impl Pattern for ImpliesQuantity {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- if Self::implies_plurality(tokens, source).is_some() {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ NonZeroUsize::new(if Self::implies_plurality(tokens, source).is_some() {
1
} else {
0
- }
+ })
}
}
diff --git a/harper-core/src/patterns/indefinite_article.rs b/harper-core/src/patterns/indefinite_article.rs
index db89765f0..6ef2bb597 100644
--- a/harper-core/src/patterns/indefinite_article.rs
+++ b/harper-core/src/patterns/indefinite_article.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::Token;
use super::{Pattern, WordSet};
@@ -15,7 +17,7 @@ impl Default for IndefiniteArticle {
}
impl Pattern for IndefiniteArticle {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
self.inner.matches(tokens, source)
}
}
diff --git a/harper-core/src/patterns/invert.rs b/harper-core/src/patterns/invert.rs
index 870bbef7d..b5d4febde 100644
--- a/harper-core/src/patterns/invert.rs
+++ b/harper-core/src/patterns/invert.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::Token;
use super::Pattern;
@@ -16,11 +18,11 @@ impl Invert {
}
impl Pattern for Invert {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- if self.inner.matches(tokens, source) != 0 {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ NonZeroUsize::new(if self.inner.matches(tokens, source).is_some() {
0
} else {
1
- }
+ })
}
}
diff --git a/harper-core/src/patterns/is_not_title_case.rs b/harper-core/src/patterns/is_not_title_case.rs
index 3fbac500a..a06333e70 100644
--- a/harper-core/src/patterns/is_not_title_case.rs
+++ b/harper-core/src/patterns/is_not_title_case.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::{Dictionary, Token, TokenStringExt, make_title_case};
use super::Pattern;
@@ -16,18 +18,17 @@ impl<D: Dictionary> IsNotTitleCase<D> {
}
impl<D: Dictionary> Pattern for IsNotTitleCase<D> {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let inner_match = self.inner.matches(tokens, source);
-
- if inner_match == 0 {
- return 0;
- }
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ let inner_match = self.inner.matches(tokens, source)?;
- let matched_chars = tokens[0..inner_match].span().unwrap().get_content(source);
- if make_title_case(&tokens[0..inner_match], source, &self.dict) != matched_chars {
- inner_match
+ let matched_chars = tokens[0..inner_match.get()]
+ .span()
+ .unwrap()
+ .get_content(source);
+ if make_title_case(&tokens[0..inner_match.get()], source, &self.dict) != matched_chars {
+ Some(inner_match)
} else {
- 0
+ None
}
}
}
diff --git a/harper-core/src/patterns/mod.rs b/harper-core/src/patterns/mod.rs
index 6c60b4f97..db16f6682 100644
--- a/harper-core/src/patterns/mod.rs
+++ b/harper-core/src/patterns/mod.rs
@@ -5,7 +5,7 @@
//!
//! See the page about [`SequencePattern`] for a concrete example of their use.
-use std::collections::VecDeque;
+use std::{collections::VecDeque, num::NonZeroUsize};
use crate::{Document, Span, Token, VecExt};
@@ -55,16 +55,13 @@ pub use whitespace_pattern::WhitespacePattern;
pub use word_pattern_group::WordPatternGroup;
pub use word_set::WordSet;
-#[cfg(not(feature = "concurrent"))]
-#[blanket(derive(Rc, Arc))]
-pub trait Pattern {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize;
-}
-
-#[cfg(feature = "concurrent")]
-#[blanket(derive(Arc))]
+#[cfg_attr(feature = "concurrent", blanket(derive(Arc)))]
+#[cfg_attr(not(feature = "concurrent"), blanket(derive(Rc, Arc)))]
pub trait Pattern: Send + Sync {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize;
+ /// Check if the pattern matches at the start of the given token slice.
+ ///
+ /// Returns the length of the match if successful, or `None` if not.
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize>;
}
pub trait PatternExt {
@@ -82,8 +79,8 @@ where
for i in 0..tokens.len() {
let len = self.matches(&tokens[i..], source);
- if len > 0 {
- found.push(Span::new_with_len(i, len));
+ if let Some(len) = len {
+ found.push(Span::new_with_len(i, len.get()));
}
}
@@ -127,14 +124,8 @@ where
F: Fn(&Token, &[char]) -> bool,
F: Send + Sync,
{
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- if tokens.is_empty() {
- return 0;
- }
-
- let tok = &tokens[0];
-
- if self(tok, source) { 1 } else { 0 }
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ NonZeroUsize::new(if self(tokens.first()?, source) { 1 } else { 0 })
}
}
@@ -143,14 +134,8 @@ impl<F> Pattern for F
where
F: Fn(&Token, &[char]) -> bool,
{
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- if tokens.is_empty() {
- return 0;
- }
-
- let tok = &tokens[0];
-
- if self(tok, source) { 1 } else { 0 }
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ NonZeroUsize::new(if self(tokens.first()?, source) { 1 } else { 0 })
}
}
diff --git a/harper-core/src/patterns/naive_pattern_group.rs b/harper-core/src/patterns/naive_pattern_group.rs
index 1629330b1..0c08564e6 100644
--- a/harper-core/src/patterns/naive_pattern_group.rs
+++ b/harper-core/src/patterns/naive_pattern_group.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use super::Pattern;
use crate::Token;
@@ -15,14 +17,10 @@ impl NaivePatternGroup {
}
impl Pattern for NaivePatternGroup {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
self.patterns
.iter()
- .find_map(|p| {
- let res = p.matches(tokens, source);
-
- if res != 0 { Some(res) } else { None }
- })
- .unwrap_or_default()
+ .filter_map(|p| p.matches(tokens, source))
+ .next()
}
}
diff --git a/harper-core/src/patterns/nominal_phrase.rs b/harper-core/src/patterns/nominal_phrase.rs
index 155368add..a655f5bef 100644
--- a/harper-core/src/patterns/nominal_phrase.rs
+++ b/harper-core/src/patterns/nominal_phrase.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::Token;
use super::Pattern;
@@ -6,21 +8,17 @@ use super::Pattern;
pub struct NominalPhrase;
impl Pattern for NominalPhrase {
- fn matches(&self, tokens: &[Token], _source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], _source: &[char]) -> Option<NonZeroUsize> {
let mut cursor = 0;
loop {
- let Some(tok) = tokens.get(cursor) else {
- return 0;
- };
+ let tok = tokens.get(cursor)?;
if tok.kind.is_adjective() || tok.kind.is_determiner() {
- let Some(next) = tokens.get(cursor + 1) else {
- return 0;
- };
+ let next = tokens.get(cursor + 1)?;
if !next.kind.is_whitespace() {
- return 0;
+ return None;
}
cursor += 2;
@@ -28,10 +26,10 @@ impl Pattern for NominalPhrase {
}
if tok.kind.is_nominal() {
- return cursor + 1;
+ return NonZeroUsize::new(cursor + 1);
}
- return 0;
+ return None;
}
}
}
@@ -72,7 +70,11 @@ mod tests {
#[test]
fn simplest_banana() {
let doc = Document::new_markdown_default_curated("a banana");
- assert!(NominalPhrase.matches(doc.get_tokens(), doc.get_source()) != 0);
+ assert!(
+ NominalPhrase
+ .matches(doc.get_tokens(), doc.get_source())
+ .is_some()
+ );
}
#[test]
diff --git a/harper-core/src/patterns/pattern_map.rs b/harper-core/src/patterns/pattern_map.rs
index 1de3b5c2b..3a6b19ad1 100644
--- a/harper-core/src/patterns/pattern_map.rs
+++ b/harper-core/src/patterns/pattern_map.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::LSend;
use crate::Token;
@@ -54,7 +56,7 @@ where
for row in &self.rows {
let len = row.key.matches(tokens, source);
- if len != 0 {
+ if len.is_some() {
return Some(&row.element);
}
}
@@ -67,15 +69,10 @@ impl<T> Pattern for PatternMap<T>
where
T: LSend,
{
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- for row in &self.rows {
- let len = row.key.matches(tokens, source);
-
- if len != 0 {
- return len;
- }
- }
-
- 0
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ self.rows
+ .iter()
+ .filter_map(|row| row.key.matches(tokens, source))
+ .next()
}
}
diff --git a/harper-core/src/patterns/repeating_pattern.rs b/harper-core/src/patterns/repeating_pattern.rs
index ffc32e2c0..514551c4c 100644
--- a/harper-core/src/patterns/repeating_pattern.rs
+++ b/harper-core/src/patterns/repeating_pattern.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use super::Pattern;
use crate::Token;
@@ -19,22 +21,20 @@ impl RepeatingPattern {
}
impl Pattern for RepeatingPattern {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
let mut tok_cursor = 0;
let mut repetition = 0;
loop {
let match_len = self.inner.matches(&tokens[tok_cursor..], source);
- if match_len == 0 {
- if repetition >= self.required_repetitions {
- return tok_cursor;
- } else {
- return 0;
- }
- } else {
- tok_cursor += match_len;
+ if let Some(match_len) = match_len {
+ tok_cursor += match_len.get();
repetition += 1;
+ } else if repetition >= self.required_repetitions {
+ return NonZeroUsize::new(tok_cursor);
+ } else {
+ return None;
}
}
}
@@ -42,6 +42,8 @@ impl Pattern for RepeatingPattern {
#[cfg(test)]
mod tests {
+ use std::num::NonZeroUsize;
+
use super::RepeatingPattern;
use crate::Document;
use crate::patterns::{AnyPattern, Pattern};
@@ -55,7 +57,7 @@ mod tests {
assert_eq!(
pat.matches(doc.get_tokens(), doc.get_source()),
- doc.get_tokens().len()
+ NonZeroUsize::new(doc.get_tokens().len())
)
}
@@ -64,6 +66,6 @@ mod tests {
let doc = Document::new_plain_english_curated("No match");
let pat = RepeatingPattern::new(Box::new(AnyPattern), 4);
- assert_eq!(pat.matches(doc.get_tokens(), doc.get_source()), 0)
+ assert_eq!(pat.matches(doc.get_tokens(), doc.get_source()), None)
}
}
diff --git a/harper-core/src/patterns/sequence_pattern.rs b/harper-core/src/patterns/sequence_pattern.rs
index 84e368246..bb08b7c1b 100644
--- a/harper-core/src/patterns/sequence_pattern.rs
+++ b/harper-core/src/patterns/sequence_pattern.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use paste::paste;
use super::whitespace_pattern::WhitespacePattern;
@@ -181,26 +183,23 @@ impl SequencePattern {
}
impl Pattern for SequencePattern {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
let mut tok_cursor = 0;
for pat in self.token_patterns.iter() {
- let match_length = pat.matches(&tokens[tok_cursor..], source);
-
- if match_length == 0 {
- return 0;
- }
-
- tok_cursor += match_length;
+ let match_length = pat.matches(&tokens[tok_cursor..], source)?;
+ tok_cursor += match_length.get();
}
- tok_cursor
+ NonZeroUsize::new(tok_cursor)
}
}
#[cfg(test)]
mod tests {
+ use std::num::NonZeroUsize;
+
use super::SequencePattern;
use crate::Document;
use crate::patterns::Pattern;
@@ -215,7 +214,7 @@ mod tests {
assert_eq!(
pat.matches(doc.get_tokens(), doc.get_source()),
- doc.get_tokens().len()
+ NonZeroUsize::new(doc.get_tokens().len())
);
}
@@ -229,7 +228,7 @@ mod tests {
assert_eq!(
pat.matches(doc.get_tokens(), doc.get_source()),
- doc.get_tokens().len()
+ NonZeroUsize::new(doc.get_tokens().len())
);
}
@@ -240,7 +239,7 @@ mod tests {
assert_eq!(
pat.matches(doc.get_tokens(), doc.get_source()),
- doc.get_tokens().len()
+ NonZeroUsize::new(doc.get_tokens().len())
);
}
}
diff --git a/harper-core/src/patterns/similar_to_phrase.rs b/harper-core/src/patterns/similar_to_phrase.rs
index d464d63bf..10fdd1c31 100644
--- a/harper-core/src/patterns/similar_to_phrase.rs
+++ b/harper-core/src/patterns/similar_to_phrase.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use crate::{Document, Token, TokenKind};
use super::{
@@ -59,14 +61,10 @@ impl SimilarToPhrase {
}
impl Pattern for SimilarToPhrase {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let exact_match = self.phrase.matches(tokens, source);
- let fuzzy_match = self.fuzzy_phrase.matches(tokens, source);
-
- if (exact_match == 0) && fuzzy_match > 0 {
- exact_match.max(fuzzy_match)
- } else {
- 0
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ if self.phrase.matches(tokens, source).is_some() {
+ return None;
}
+ self.fuzzy_phrase.matches(tokens, source)
}
}
diff --git a/harper-core/src/patterns/split_compound_word.rs b/harper-core/src/patterns/split_compound_word.rs
index de5666749..836d788ec 100644
--- a/harper-core/src/patterns/split_compound_word.rs
+++ b/harper-core/src/patterns/split_compound_word.rs
@@ -1,4 +1,4 @@
-use std::sync::Arc;
+use std::{num::NonZeroUsize, sync::Arc};
use crate::{CharString, Dictionary, FstDictionary, Token, WordMetadata};
@@ -57,20 +57,20 @@ impl SplitCompoundWord {
}
impl Pattern for SplitCompoundWord {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let inner_match = self.inner.matches(tokens, source);
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ let inner_match = self.inner.matches(tokens, source)?;
- if inner_match != 3 {
- return 0;
+ if inner_match.get() != 3 {
+ return None;
}
let a = &tokens[0];
let b = &tokens[2];
if self.get_merged_word(a, b, source).is_some() {
- return inner_match;
+ return Some(inner_match);
}
- 0
+ None
}
}
diff --git a/harper-core/src/patterns/token_kind_pattern_group.rs b/harper-core/src/patterns/token_kind_pattern_group.rs
index bfcf178bb..5025616b2 100644
--- a/harper-core/src/patterns/token_kind_pattern_group.rs
+++ b/harper-core/src/patterns/token_kind_pattern_group.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use hashbrown::HashMap;
use super::Pattern;
@@ -9,15 +11,9 @@ pub struct TokenKindPatternGroup {
}
impl Pattern for TokenKindPatternGroup {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let Some(first_kind) = tokens.first().map(|t| &t.kind) else {
- return 0;
- };
-
- let Some(pattern) = self.strict_patterns.get(first_kind) else {
- return 0;
- };
-
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ let first_kind = tokens.first().map(|t| &t.kind)?;
+ let pattern = self.strict_patterns.get(first_kind)?;
pattern.matches(tokens, source)
}
}
diff --git a/harper-core/src/patterns/whitespace_pattern.rs b/harper-core/src/patterns/whitespace_pattern.rs
index b3348abe6..2ead95e6f 100644
--- a/harper-core/src/patterns/whitespace_pattern.rs
+++ b/harper-core/src/patterns/whitespace_pattern.rs
@@ -1,12 +1,16 @@
+use std::num::NonZeroUsize;
+
use super::Pattern;
pub struct WhitespacePattern;
impl Pattern for WhitespacePattern {
- fn matches(&self, tokens: &[crate::Token], _source: &[char]) -> usize {
- tokens
- .iter()
- .position(|t| !t.kind.is_whitespace())
- .unwrap_or(tokens.len())
+ fn matches(&self, tokens: &[crate::Token], _source: &[char]) -> Option<NonZeroUsize> {
+ NonZeroUsize::new(
+ tokens
+ .iter()
+ .position(|t| !t.kind.is_whitespace())
+ .unwrap_or(tokens.len()),
+ )
}
}
diff --git a/harper-core/src/patterns/within_edit_distance.rs b/harper-core/src/patterns/within_edit_distance.rs
index 19795efd3..0b5fb6b5b 100644
--- a/harper-core/src/patterns/within_edit_distance.rs
+++ b/harper-core/src/patterns/within_edit_distance.rs
@@ -1,7 +1,8 @@
use std::cell::RefCell;
+use std::num::NonZeroUsize;
use super::Pattern;
-use crate::{CharString, CharStringExt, Token, TokenKind};
+use crate::{CharString, CharStringExt, Token};
use crate::edit_distance::edit_distance_min_alloc;
@@ -32,29 +33,22 @@ thread_local! {
}
impl Pattern for WithinEditDistance {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let Some(first) = tokens.first() else {
- return 0;
- };
-
- let TokenKind::Word(_) = first.kind else {
- return 0;
- };
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ let first = tokens.first()?;
+ if !first.kind.is_word() {
+ return None;
+ }
let content = first.span.get_content(source);
BUFFERS.with_borrow_mut(|(buffer_a, buffer_b)| {
- if edit_distance_min_alloc(
+ let distance = edit_distance_min_alloc(
&content.to_lower(),
&self.word.to_lower(),
buffer_a,
buffer_b,
- ) <= self.max_edit_dist
- {
- 1
- } else {
- 0
- }
+ );
+ NonZeroUsize::new(if distance <= self.max_edit_dist { 1 } else { 0 })
})
}
}
diff --git a/harper-core/src/patterns/word_pattern_group.rs b/harper-core/src/patterns/word_pattern_group.rs
index 15b3fa85f..1c3f5399d 100644
--- a/harper-core/src/patterns/word_pattern_group.rs
+++ b/harper-core/src/patterns/word_pattern_group.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use hashbrown::HashMap;
use super::naive_pattern_group::NaivePatternGroup;
@@ -40,19 +42,14 @@ impl<P> Pattern for WordPatternGroup<P>
where
P: Pattern,
{
- fn matches(&self, tokens: &[crate::Token], source: &[char]) -> usize {
- let Some(first) = tokens.first() else {
- return 0;
- };
-
+ fn matches(&self, tokens: &[crate::Token], source: &[char]) -> Option<NonZeroUsize> {
+ let first = tokens.first()?;
if !first.kind.is_word() {
- return 0;
+ return None;
}
let word_chars = first.span.get_content(source);
- let Some(inner_pattern) = self.patterns.get(word_chars) else {
- return 0;
- };
+ let inner_pattern = self.patterns.get(word_chars)?;
inner_pattern.matches(tokens, source)
}
diff --git a/harper-core/src/patterns/word_set.rs b/harper-core/src/patterns/word_set.rs
index 0e5225312..cbb892004 100644
--- a/harper-core/src/patterns/word_set.rs
+++ b/harper-core/src/patterns/word_set.rs
@@ -1,3 +1,5 @@
+use std::num::NonZeroUsize;
+
use super::Pattern;
use smallvec::SmallVec;
@@ -38,13 +40,10 @@ impl WordSet {
}
impl Pattern for WordSet {
- fn matches(&self, tokens: &[Token], source: &[char]) -> usize {
- let Some(tok) = tokens.first() else {
- return 0;
- };
-
+ fn matches(&self, tokens: &[Token], source: &[char]) -> Option<NonZeroUsize> {
+ let tok = tokens.first()?;
if !tok.kind.is_word() {
- return 0;
+ return None;
}
let tok_chars = tok.span.get_content(source);
@@ -60,11 +59,11 @@ impl Pattern for WordSet {
.all(|(a, b)| a.eq_ignore_ascii_case(b));
if partial_match {
- return 1;
+ return NonZeroUsize::new(1);
}
}
- 0
+ None
}
}