Skip to content

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

fix: handle another false positive in "adjective of a"

hippietrail8771e43b7447
diff --git a/harper-core/src/linting/adjective_of_a.rs b/harper-core/src/linting/adjective_of_a.rs
index cc7e11a45..9cd2d694c 100644
--- a/harper-core/src/linting/adjective_of_a.rs
+++ b/harper-core/src/linting/adjective_of_a.rs
@@ -15,37 +15,39 @@ impl Linter for AdjectiveOfA {
let word_of = document.get_token(i + 2);
let space_2 = document.get_token(i + 3);
let a_or_an = document.get_token(i + 4);
- // Ensure the adjective adjective is positive form, not comparative or superlative.
- // Rightly prevents flagging: "for the better of a day"
- // And "might not be the best of a given run"
- // And "Which brings me to my best of a bad situation."
- //
- // But wrongly prevents flagging: "I don't think that's too much better of an idea."
- // And: "see if you can give us a little better of an answer"
- // And: "hopefully it won't be too much worse of a problem"
- // And: "seems far worse of a result to me"
- let word: &[char] = document.get_span_content(&adjective.span);
- // Avoid common false positives where the word is more common as a
- // noun than an adjective in this context.
- if word == ['k', 'i', 'n', 'd']
- || word == ['m', 'e', 'a', 'n', 'i', 'n', 'g']
- || word == ['p', 'a', 'r', 't']
+ let adj_chars: &[char] = document.get_span_content(&adjective.span);
+
+ // Rule out false positives
+
+ // "much of a" is a different and valid construction.
+ if adj_chars == ['m', 'u', 'c', 'h']
+ // The word is used more as a noun in this context.
+ // (using .kind.is_likely_homograph() here is too strict)
+ || adj_chars == ['k', 'i', 'n', 'd']
+ || adj_chars == ['m', 'e', 'a', 'n', 'i', 'n', 'g']
+ || adj_chars == ['p', 'a', 'r', 't']
{
continue;
}
- let len = word.len();
+
+ // Rule out comparatives and superlatives.
+
+ // Pros:
+ // "for the better of a day"
+ // "might not be the best of a given run"
+ // "Which brings me to my best of a bad situation."
+ //
+ // Cons:
+ // "see if you can give us a little better of an answer"
+ // "hopefully it won't be too much worse of a problem"
+ // "seems far worse of a result to me"
+ let len = adj_chars.len();
if len > 2 {
- let ending = &word[len - 2..len];
+ let ending = &adj_chars[len - 2..len];
if ending == ['e', 'r'] || ending == ['s', 't'] {
continue;
}
}
- // Check if it might also be valid as other parts of speech.
- // This stops us flagging "meaning of a".
- // But it also stops "good of a".
- // if adjective.kind.is_likely_homograph() {
- // continue;
- // }
if space_1.is_none() || word_of.is_none() || space_2.is_none() || a_or_an.is_none() {
continue;
}
@@ -57,8 +59,8 @@ impl Linter for AdjectiveOfA {
if !word_of.kind.is_word() {
continue;
}
- let w2 = document.get_span_content(&word_of.span);
- if w2 != ['o', 'f'] {
+ let word_of = document.get_span_content(&word_of.span);
+ if word_of != ['o', 'f'] {
continue;
}
let space_2 = space_2.unwrap();
@@ -69,8 +71,8 @@ impl Linter for AdjectiveOfA {
if !a_or_an.kind.is_word() {
continue;
}
- let w4 = document.get_span_content(&a_or_an.span);
- if w4 != ['a'] && w4 != ['a', 'n'] {
+ let a_or_an_chars = document.get_span_content(&a_or_an.span);
+ if a_or_an_chars != ['a'] && a_or_an_chars != ['a', 'n'] {
continue;
}
@@ -165,4 +167,13 @@ mod tests {
0,
);
}
+
+ #[test]
+ fn dont_flag_much() {
+ assert_lint_count(
+ "How much of a performance impact when switching from rails to rails-api ?",
+ AdjectiveOfA,
+ 0,
+ );
+ }
}