Skip to content

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

Fixed issues with inline code blocks in Markdown

Elijah Potter6dbcebd50bba
diff --git a/harper-core/src/document.rs b/harper-core/src/document.rs
index f3763e0c0..3716c92af 100644
--- a/harper-core/src/document.rs
+++ b/harper-core/src/document.rs
@@ -301,13 +301,11 @@ mod tests {
fn assert_condensed_contractions(text: &str, final_tok_count: usize) {
let mut document = Document::new(text, Box::new(PlainEnglish));
- dbg!(&document.tokens);
document.condense_contractions();
assert_eq!(document.tokens.len(), final_tok_count);
let mut document = Document::new(text, Box::new(Markdown));
- dbg!(&document.tokens);
document.condense_contractions();
assert_eq!(document.tokens.len(), final_tok_count);
diff --git a/harper-core/src/linting/mod.rs b/harper-core/src/linting/mod.rs
index 94de4d04f..13bfb8988 100644
--- a/harper-core/src/linting/mod.rs
+++ b/harper-core/src/linting/mod.rs
@@ -19,11 +19,12 @@ pub trait Linter: Send + Sync {
#[cfg(test)]
mod tests {
- use crate::{parsers::PlainEnglish, Document, Linter};
+ use crate::{parsers::Markdown, Document, Linter};
pub fn assert_lint_count(text: &str, mut linter: impl Linter, count: usize) {
- let test = Document::new(text, Box::new(PlainEnglish));
+ let test = Document::new(text, Box::new(Markdown));
let lints = linter.lint(&test);
+ dbg!(&lints);
assert_eq!(lints.len(), count);
}
}
diff --git a/harper-core/src/linting/sentence_capitalization.rs b/harper-core/src/linting/sentence_capitalization.rs
index a140dc6d2..6d239b616 100644
--- a/harper-core/src/linting/sentence_capitalization.rs
+++ b/harper-core/src/linting/sentence_capitalization.rs
@@ -14,11 +14,7 @@ impl Linter for SentenceCapitalization {
let mut lints = Vec::new();
for sentence in document.sentences() {
- if let Some(first_word) = sentence.first() {
- if !first_word.kind.is_word() {
- break;
- }
-
+ if let Some(first_word) = sentence.first_sentence_word() {
let letters = document.get_span_content(first_word.span);
if let Some(first_letter) = letters.first() {
@@ -64,4 +60,22 @@ mod tests {
2,
)
}
+
+ #[test]
+ fn ignores_unlintable() {
+ assert_lint_count(
+ "[`misspelled_word`] is assumed to be quite small (n < 100). ",
+ SentenceCapitalization,
+ 0,
+ )
+ }
+
+ #[test]
+ fn unphased_unlintable() {
+ assert_lint_count(
+ "the linter should not be affected by `this` unlintable.",
+ SentenceCapitalization,
+ 1,
+ )
+ }
}
diff --git a/harper-core/src/parsers/markdown.rs b/harper-core/src/parsers/markdown.rs
index 2f3e38bd6..bf749c656 100644
--- a/harper-core/src/parsers/markdown.rs
+++ b/harper-core/src/parsers/markdown.rs
@@ -1,3 +1,6 @@
+use itertools::Itertools;
+use pulldown_cmark::LinkType;
+
use super::{Parser, PlainEnglish, StrParser};
use crate::{Span, Token, TokenKind};
@@ -29,10 +32,23 @@ impl Parser for Markdown {
pulldown_cmark::Event::End(_) => {
stack.pop();
}
+ pulldown_cmark::Event::Code(code) => {
+ traversed_chars += source_str[traversed_bytes..range.start].chars().count();
+ traversed_bytes = range.start;
+
+ let chunk_len = code.chars().count();
+
+ tokens.push(Token {
+ span: Span::new(traversed_chars, chunk_len),
+ kind: TokenKind::Unlintable,
+ });
+ }
pulldown_cmark::Event::Text(text) => {
traversed_chars += source_str[traversed_bytes..range.start].chars().count();
traversed_bytes = range.start;
+ let chunk_len = text.chars().count();
+
if let Some(tag) = stack.last() {
use pulldown_cmark::Tag;
@@ -45,19 +61,20 @@ impl Parser for Markdown {
}
if !(matches!(tag, Tag::Paragraph)
- || matches!(tag, Tag::Heading(_, _, _))
+ || matches!(tag, Tag::Link(..))
+ || matches!(tag, Tag::Heading(..))
|| matches!(tag, Tag::Item)
|| matches!(tag, Tag::TableCell)
|| matches!(tag, Tag::Emphasis)
|| matches!(tag, Tag::Strong)
- || matches!(tag, Tag::Link(..))
|| matches!(tag, Tag::Strikethrough))
{
continue;
}
}
- let mut new_tokens = english_parser.parse_str(text);
+ let mut new_tokens =
+ english_parser.parse(&source[traversed_chars..traversed_chars + chunk_len]);
new_tokens
.iter_mut()
diff --git a/harper-core/src/spell/hunspell/attributes.rs b/harper-core/src/spell/hunspell/attributes.rs
index 899327937..95e5bbd27 100644
--- a/harper-core/src/spell/hunspell/attributes.rs
+++ b/harper-core/src/spell/hunspell/attributes.rs
@@ -245,7 +245,6 @@ impl<'a> AttributeArgParser<'a> {
.char_indices()
.find(|(_i, c)| !c.is_whitespace())
else {
- dbg!(self.cursor);
return Err(Error::UnexpectedEndOfLine);
};
diff --git a/harper-core/src/token.rs b/harper-core/src/token.rs
index c8054a043..35e9893c6 100644
--- a/harper-core/src/token.rs
+++ b/harper-core/src/token.rs
@@ -139,7 +139,7 @@ impl TokenStringExt for [Token] {
return Some(*word);
};
- if w_idx > u_idx {
+ if w_idx < u_idx {
Some(*word)
} else {
None
diff --git a/harper-ls/src/tree_sitter_parser.rs b/harper-ls/src/tree_sitter_parser.rs
index a4730b277..9981fcc0e 100644
--- a/harper-ls/src/tree_sitter_parser.rs
+++ b/harper-ls/src/tree_sitter_parser.rs
@@ -58,11 +58,22 @@ impl Parser for TreeSitterParser {
let mut tokens = Vec::new();
for span in comments_spans {
- let mut new_tokens = markdown_parser.parse(&source[span.start..span.end]);
+ // Skip over the comment start characters
+ let actual_start = source[span.start..span.end]
+ .iter()
+ .position(|c| !is_comment_character(*c))
+ .unwrap_or(0)
+ + span.start;
+
+ if span.end <= actual_start {
+ continue;
+ }
+
+ let mut new_tokens = markdown_parser.parse(&source[actual_start..span.end]);
new_tokens
.iter_mut()
- .for_each(|t| t.span.offset(span.start));
+ .for_each(|t| t.span.offset(actual_start));
tokens.append(&mut new_tokens);
}
@@ -71,6 +82,10 @@ impl Parser for TreeSitterParser {
}
}
+fn is_comment_character(c: char) -> bool {
+ matches!(c, '#' | '-' | '/')
+}
+
/// Converts a set of byte-indexed [`Span`]s to char-index Spans, in-place.
/// NOTE: Will sort the given array by their [`Span::start`].
///