Skip to content

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

Improved documentation on previous commits

Elijah Potter65aeac89bf0a
diff --git a/harper-core/Cargo.toml b/harper-core/Cargo.toml
index 2536af1a5..28737de76 100644
--- a/harper-core/Cargo.toml
+++ b/harper-core/Cargo.toml
@@ -4,6 +4,8 @@ version = "0.2.0"
edition = "2021"
description = "The language checker for artists."
license = "MIT OR Apache-2.0"
+readme = "README.md"
+homepage = "https://github.com/chilipepperhott/harper"
[dependencies]
hashbrown = "0.14.3"
diff --git a/harper-core/README.md b/harper-core/README.md
new file mode 100644
index 000000000..0640d1877
--- /dev/null
+++ b/harper-core/README.md
@@ -0,0 +1,7 @@
+# `harper-core`
+
+This is the core engine behind Harper.
+Right now, it only exists to hold the core functioning of `harper-ls` and `harper-wasm` (and by extension the web interface).
+
+`harper-core` _is_ [available on crates.io](https://crates.io/crates/harper-core), however improving the API is currently not a high priority.
+Feel free to use `harper-core` in your projects, but if you run into issues, create a pull request.
diff --git a/harper-core/dictionary.dict b/harper-core/dictionary.dict
index 1d8d190d1..c27857813 100644
--- a/harper-core/dictionary.dict
+++ b/harper-core/dictionary.dict
@@ -49581,3 +49581,5 @@ zygote/SM
zygotic
zymurgy/M
StackOverflow/M
+CommonMark/M
+lex/D
diff --git a/harper-core/src/document.rs b/harper-core/src/document.rs
index 797901b6e..f3763e0c0 100644
--- a/harper-core/src/document.rs
+++ b/harper-core/src/document.rs
@@ -2,7 +2,7 @@ use std::fmt::Display;
use itertools::Itertools;
-use crate::parsers::{MarkdownParser, Parser, PlainEnglishParser};
+use crate::parsers::{Markdown, Parser, PlainEnglish};
use crate::{
linting::Suggestion,
span::Span,
@@ -34,11 +34,11 @@ impl Document {
}
pub fn new_plain_english(text: &str) -> Self {
- Self::new(text, Box::new(PlainEnglishParser))
+ Self::new(text, Box::new(PlainEnglish))
}
pub fn new_markdown(text: &str) -> Self {
- Self::new(text, Box::new(MarkdownParser))
+ Self::new(text, Box::new(Markdown))
}
/// Re-parse important language constructs.
@@ -295,18 +295,18 @@ fn is_sentence_terminator(punctuation: &Punctuation) -> bool {
mod tests {
use super::Document;
use crate::{
- parsers::{MarkdownParser, PlainEnglishParser},
+ parsers::{Markdown, PlainEnglish},
token::TokenStringExt,
};
fn assert_condensed_contractions(text: &str, final_tok_count: usize) {
- let mut document = Document::new(text, Box::new(PlainEnglishParser));
+ 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(MarkdownParser));
+ let mut document = Document::new(text, Box::new(Markdown));
dbg!(&document.tokens);
document.condense_contractions();
@@ -341,7 +341,7 @@ mod tests {
#[test]
fn parses_sentences_correctly() {
let text = "There were three little pigs. They built three little homes.";
- let document = Document::new(text, Box::new(PlainEnglishParser));
+ let document = Document::new(text, Box::new(PlainEnglish));
let mut sentence_strs = vec![];
diff --git a/harper-core/src/linting/mod.rs b/harper-core/src/linting/mod.rs
index 1f79c82df..94de4d04f 100644
--- a/harper-core/src/linting/mod.rs
+++ b/harper-core/src/linting/mod.rs
@@ -19,10 +19,10 @@ pub trait Linter: Send + Sync {
#[cfg(test)]
mod tests {
- use crate::{parsers::PlainEnglishParser, Document, Linter};
+ use crate::{parsers::PlainEnglish, Document, Linter};
pub fn assert_lint_count(text: &str, mut linter: impl Linter, count: usize) {
- let test = Document::new(text, Box::new(PlainEnglishParser));
+ let test = Document::new(text, Box::new(PlainEnglish));
let lints = linter.lint(&test);
assert_eq!(lints.len(), count);
}
diff --git a/harper-core/src/parsers/markdown.rs b/harper-core/src/parsers/markdown.rs
index 280cc53cc..08b7af3f0 100644
--- a/harper-core/src/parsers/markdown.rs
+++ b/harper-core/src/parsers/markdown.rs
@@ -1,10 +1,14 @@
-use super::{Parser, PlainEnglishParser, StrParser};
+use super::{Parser, PlainEnglish, StrParser};
use crate::Token;
-pub struct MarkdownParser;
-impl Parser for MarkdownParser {
+/// A parser that wraps the [`PlainEnglish`] parser that allows one to parse CommonMark files.
+///
+/// Will ignore code blocks and tables.
+pub struct Markdown;
+
+impl Parser for Markdown {
fn parse(&mut self, source: &[char]) -> Vec<Token> {
- let mut english_parser = PlainEnglishParser;
+ let mut english_parser = PlainEnglish;
let source_str: String = source.iter().collect();
let md_parser =
diff --git a/harper-core/src/parsers/mod.rs b/harper-core/src/parsers/mod.rs
index 5c8e46d86..69a3b3dc6 100644
--- a/harper-core/src/parsers/mod.rs
+++ b/harper-core/src/parsers/mod.rs
@@ -2,8 +2,8 @@ mod markdown;
mod plain_english;
pub use crate::token::{Quote, Token, TokenKind, TokenStringExt};
-pub use markdown::MarkdownParser;
-pub use plain_english::PlainEnglishParser;
+pub use markdown::Markdown;
+pub use plain_english::PlainEnglish;
pub trait Parser: Send + Sync {
fn parse(&mut self, source: &[char]) -> Vec<Token>;
@@ -25,7 +25,7 @@ where
#[cfg(test)]
mod tests {
- use super::{MarkdownParser, Parser, PlainEnglishParser};
+ use super::{Markdown, Parser, PlainEnglish};
use crate::{
Punctuation,
TokenKind::{self, *},
@@ -44,12 +44,12 @@ mod tests {
}
fn assert_tokens_eq_plain(test_str: impl AsRef<str>, expected: &[TokenKind]) {
- let mut parser = PlainEnglishParser;
+ let mut parser = PlainEnglish;
assert_tokens_eq(test_str, expected, &mut parser);
}
fn assert_tokens_eq_md(test_str: impl AsRef<str>, expected: &[TokenKind]) {
- let mut parser = MarkdownParser;
+ let mut parser = Markdown;
assert_tokens_eq(test_str, expected, &mut parser)
}
diff --git a/harper-core/src/parsers/plain_english.rs b/harper-core/src/parsers/plain_english.rs
index c81652e99..c58e0cb19 100644
--- a/harper-core/src/parsers/plain_english.rs
+++ b/harper-core/src/parsers/plain_english.rs
@@ -4,9 +4,11 @@ use crate::{
Span, Token,
};
-pub struct PlainEnglishParser;
+/// A parser that will attempt to lex as many tokens a possible,
+/// without discrimination and until the end of input.
+pub struct PlainEnglish;
-impl Parser for PlainEnglishParser {
+impl Parser for PlainEnglish {
fn parse(&mut self, source: &[char]) -> Vec<Token> {
let mut cursor = 0;
let mut tokens = Vec::new();
diff --git a/harper-ls/src/backend.rs b/harper-ls/src/backend.rs
index 0c247251a..1e50edebc 100644
--- a/harper-ls/src/backend.rs
+++ b/harper-ls/src/backend.rs
@@ -1,7 +1,7 @@
use std::{collections::HashMap, fs};
use harper_core::{
- parsers::{MarkdownParser, Parser},
+ parsers::{Markdown, Parser},
Dictionary, Document, LintSet, Linter,
};
use tokio::sync::Mutex;
@@ -37,12 +37,12 @@ impl Backend {
}
async fn update_document(&self, url: &Url, text: &str) {
- let mut parser: Box<dyn Parser> = Box::new(MarkdownParser);
+ let mut parser: Box<dyn Parser> = Box::new(Markdown);
if let Some(extension) = url.to_file_path().unwrap().extension() {
parser = TreeSitterParser::new_from_extension(&extension.to_string_lossy())
.map::<Box<dyn Parser>, _>(|v| Box::new(v))
- .unwrap_or(Box::new(MarkdownParser));
+ .unwrap_or(Box::new(Markdown));
}
let doc = Document::new(text, parser);
diff --git a/harper-ls/src/tree_sitter_parser.rs b/harper-ls/src/tree_sitter_parser.rs
index 2c522b4c6..a4730b277 100644
--- a/harper-ls/src/tree_sitter_parser.rs
+++ b/harper-ls/src/tree_sitter_parser.rs
@@ -1,9 +1,11 @@
use harper_core::{
- parsers::{Parser, PlainEnglishParser},
+ parsers::{Markdown, Parser},
Span,
};
use tree_sitter::{Language, TreeCursor};
+/// A Harper parser that wraps the standard [`Markdown`] parser that exclusively parses
+/// comments in any language supported by [`tree_sitter`].
pub struct TreeSitterParser {
language: Language,
}
@@ -39,7 +41,7 @@ impl Parser for TreeSitterParser {
fn parse(&mut self, source: &[char]) -> Vec<harper_core::Token> {
let text: String = source.iter().collect();
- let mut english_parser = PlainEnglishParser;
+ let mut markdown_parser = Markdown;
let mut parser = tree_sitter::Parser::new();
parser.set_language(self.language).unwrap();
@@ -56,7 +58,7 @@ impl Parser for TreeSitterParser {
let mut tokens = Vec::new();
for span in comments_spans {
- let mut new_tokens = english_parser.parse(&source[span.start..span.end]);
+ let mut new_tokens = markdown_parser.parse(&source[span.start..span.end]);
new_tokens
.iter_mut()