Skip to content

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

feat: created `CharString` and added config for `diagnosticSeverity`

Elijah Potter6e7cbe184481
diff --git a/Cargo.lock b/Cargo.lock
index f9fd90782..8c5d764d1 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -562,6 +562,7 @@ dependencies = [
"once_cell",
"open",
"resolve-path",
+ "serde",
"serde_json",
"tokio",
"tower-lsp",
diff --git a/README.md b/README.md
index faedc4239..7e9ade36c 100644
--- a/README.md
+++ b/README.md
@@ -31,4 +31,3 @@ Harper is even small enough to load via [WebAssembly.](https://writewithharper.c
If you want to use Harper on your machine, you will want to look at the [documentation for
`harper-ls`](./harper-ls/README.md), the Language Server Protocol implementation.
-
diff --git a/harper-core/dictionary.dict b/harper-core/dictionary.dict
index 49966e473..7f66ee172 100644
--- a/harper-core/dictionary.dict
+++ b/harper-core/dictionary.dict
@@ -49622,3 +49622,4 @@ URI/s
suprachiasmatic
cannabinoids
eudaimonia
+whitespace
diff --git a/harper-core/src/char_string.rs b/harper-core/src/char_string.rs
new file mode 100644
index 000000000..26394f629
--- /dev/null
+++ b/harper-core/src/char_string.rs
@@ -0,0 +1,17 @@
+use smallvec::SmallVec;
+
+pub type CharString = SmallVec<[char; 6]>;
+
+pub trait CharStringExt {
+ fn to_lower(&self) -> CharString;
+}
+
+impl CharStringExt for [char] {
+ fn to_lower(&self) -> CharString {
+ let mut out = CharString::with_capacity(self.len());
+
+ out.extend(self.iter().flat_map(|v| v.to_lowercase()));
+
+ out
+ }
+}
diff --git a/harper-core/src/lib.rs b/harper-core/src/lib.rs
index 23152d67b..94fc14c07 100644
--- a/harper-core/src/lib.rs
+++ b/harper-core/src/lib.rs
@@ -1,5 +1,6 @@
#![allow(dead_code)]
+mod char_string;
mod document;
mod lexing;
mod linting;
@@ -8,6 +9,7 @@ mod span;
mod spell;
mod token;
+pub use char_string::{CharString, CharStringExt};
pub use document::Document;
pub use linting::{Lint, LintGroup, LintGroupConfig, LintKind, Linter, Suggestion};
pub use span::Span;
diff --git a/harper-core/src/linting/an_a.rs b/harper-core/src/linting/an_a.rs
index afb9ccf87..089fad481 100644
--- a/harper-core/src/linting/an_a.rs
+++ b/harper-core/src/linting/an_a.rs
@@ -1,6 +1,6 @@
use itertools::Itertools;
-use crate::{Document, Lint, LintKind, Linter, Suggestion, TokenStringExt};
+use crate::{CharStringExt, Document, Lint, LintKind, Linter, Suggestion, TokenStringExt};
#[derive(Debug, Default)]
pub struct AnA;
@@ -23,7 +23,7 @@ impl Linter for AnA {
continue;
};
- let should_be_a_an = !starts_with_vowel(chars_second);
+ let should_be_a_an = !starts_with_vowel(&chars_second.to_lower());
if a_an != should_be_a_an {
let replacement = match a_an {
diff --git a/harper-core/src/linting/matcher.rs b/harper-core/src/linting/matcher.rs
index 03760f48f..8367c431b 100644
--- a/harper-core/src/linting/matcher.rs
+++ b/harper-core/src/linting/matcher.rs
@@ -1,10 +1,20 @@
-use crate::spell::DictWord;
-use crate::{Document, Lint, LintKind, Linter, Punctuation, Span, Suggestion, Token, TokenKind};
+use crate::{
+ CharString,
+ Document,
+ Lint,
+ LintKind,
+ Linter,
+ Punctuation,
+ Span,
+ Suggestion,
+ Token,
+ TokenKind
+};
#[derive(Debug, PartialEq, PartialOrd, Clone)]
struct PatternToken {
kind: TokenKind,
- content: Option<DictWord>
+ content: Option<CharString>
}
impl PatternToken {
diff --git a/harper-core/src/linting/repeated_words.rs b/harper-core/src/linting/repeated_words.rs
index c95c80ae4..2ef59733c 100644
--- a/harper-core/src/linting/repeated_words.rs
+++ b/harper-core/src/linting/repeated_words.rs
@@ -1,14 +1,13 @@
use hashbrown::HashSet;
use super::{Lint, LintKind, Linter};
-use crate::spell::DictWord;
use crate::token::{Token, TokenKind, TokenStringExt};
-use crate::{Document, Span, Suggestion};
+use crate::{CharString, Document, Span, Suggestion};
#[derive(Debug, Clone)]
pub struct RepeatedWords {
/// The set of words that can be considered for repetition checking.
- set: HashSet<DictWord>
+ set: HashSet<CharString>
}
impl RepeatedWords {
diff --git a/harper-core/src/spell/full_dictionary.rs b/harper-core/src/spell/full_dictionary.rs
index 51008242e..434b8cdde 100644
--- a/harper-core/src/spell/full_dictionary.rs
+++ b/harper-core/src/spell/full_dictionary.rs
@@ -4,7 +4,8 @@ use smallvec::{SmallVec, ToSmallVec};
use super::dictionary::Dictionary;
use super::hunspell::{parse_default_attribute_list, parse_default_word_list};
-use super::{seq_to_normalized, DictWord};
+use super::seq_to_normalized;
+use crate::CharString;
/// A full, fat dictionary.
/// All of the elements are stored in-memory.
@@ -16,13 +17,13 @@ pub struct FullDictionary {
/// This is likely due to increased locality :shrug:.
///
/// This list is sorted by word length (i.e. the shortest words are first).
- words: Vec<DictWord>,
+ words: Vec<CharString>,
/// A lookup list for each word length.
/// Each index of this list will return the first index of [`Self::words`]
/// that has a word whose index is that length.
word_len_starts: Vec<usize>,
/// All English words
- word_set: HashSet<DictWord>
+ word_set: HashSet<CharString>
}
fn uncached_inner_new() -> FullDictionary {
@@ -81,7 +82,7 @@ impl FullDictionary {
/// list. NOTE: This function will sort the original word list by its
/// length. If the word list's order is changed after creating the
/// lookup, it will no longer be valid.
- fn create_len_starts(words: &mut [DictWord]) -> Vec<usize> {
+ fn create_len_starts(words: &mut [CharString]) -> Vec<usize> {
words.sort_by_key(|a| a.len());
let mut word_len_starts = vec![0, 0];
diff --git a/harper-core/src/spell/hunspell/attributes.rs b/harper-core/src/spell/hunspell/attributes.rs
index 1bc4bf21a..be101a5fd 100644
--- a/harper-core/src/spell/hunspell/attributes.rs
+++ b/harper-core/src/spell/hunspell/attributes.rs
@@ -6,8 +6,7 @@ use smallvec::ToSmallVec;
use super::matcher::Matcher;
use super::word_list::MarkedWord;
use super::Error;
-use crate::spell::DictWord;
-use crate::Span;
+use crate::{CharString, Span};
#[derive(Debug, Clone)]
struct AffixReplacement {
@@ -111,7 +110,7 @@ impl AttributeList {
/// Will append to the given `dest`;
///
/// In the future, I want to make this function cleaner and faster.
- pub fn expand_marked_word(&self, word: MarkedWord, dest: &mut Vec<DictWord>) {
+ pub fn expand_marked_word(&self, word: MarkedWord, dest: &mut Vec<CharString>) {
dest.reserve(word.attributes.len() + 1);
let start_len = dest.len();
@@ -173,7 +172,7 @@ impl AttributeList {
pub fn expand_marked_words(
&self,
words: impl IntoIterator<Item = MarkedWord>,
- dest: &mut Vec<DictWord>
+ dest: &mut Vec<CharString>
) {
for word in words {
self.expand_marked_word(word, dest);
@@ -184,7 +183,7 @@ impl AttributeList {
replacement: &AffixReplacement,
letters: &[char],
suffix: bool
- ) -> Option<DictWord> {
+ ) -> Option<CharString> {
if replacement.condition.len() > letters.len() {
return None;
}
@@ -199,7 +198,7 @@ impl AttributeList {
if replacement.condition.matches(target_segment) {
let mut replaced_segment = letters.to_smallvec();
- let mut remove: DictWord = replacement.remove.to_smallvec();
+ let mut remove: CharString = replacement.remove.to_smallvec();
if !suffix {
replaced_segment.reverse();
diff --git a/harper-core/src/spell/hunspell/mod.rs b/harper-core/src/spell/hunspell/mod.rs
index 27e954a8a..fd8c5bbbb 100644
--- a/harper-core/src/spell/hunspell/mod.rs
+++ b/harper-core/src/spell/hunspell/mod.rs
@@ -22,7 +22,7 @@ mod tests {
use super::attributes::AttributeList;
use super::word_list::parse_word_list;
use super::{parse_default_attribute_list, parse_default_word_list};
- use crate::spell::DictWord;
+ use crate::CharString;
pub const TEST_WORD_LIST: &str = "3\nhello\ntry/B\nwork/AB";
pub const ATTR_LIST: &str =
@@ -65,7 +65,7 @@ mod tests {
assert!(expanded.contains(&split("giants")))
}
- fn build_expanded() -> Vec<DictWord> {
+ fn build_expanded() -> Vec<CharString> {
let words = parse_default_word_list().unwrap();
let attributes = parse_default_attribute_list().unwrap();
@@ -91,7 +91,7 @@ mod tests {
assert!(build_expanded().contains(&split("deallocate")));
}
- fn split(text: &str) -> DictWord {
+ fn split(text: &str) -> CharString {
text.chars().collect()
}
}
diff --git a/harper-core/src/spell/hunspell/word_list.rs b/harper-core/src/spell/hunspell/word_list.rs
index b83204a07..21b915b53 100644
--- a/harper-core/src/spell/hunspell/word_list.rs
+++ b/harper-core/src/spell/hunspell/word_list.rs
@@ -1,9 +1,9 @@
use super::Error;
-use crate::spell::DictWord;
+use crate::CharString;
#[derive(Debug, Clone)]
pub struct MarkedWord {
- pub letters: DictWord,
+ pub letters: CharString,
pub attributes: Vec<char>
}
diff --git a/harper-core/src/spell/mod.rs b/harper-core/src/spell/mod.rs
index 6b09c56bd..af51ee5ac 100644
--- a/harper-core/src/spell/mod.rs
+++ b/harper-core/src/spell/mod.rs
@@ -1,7 +1,6 @@
use std::borrow::Cow;
use itertools::{Itertools, MinMaxResult};
-use smallvec::SmallVec;
pub use self::dictionary::Dictionary;
pub use self::full_dictionary::FullDictionary;
@@ -12,9 +11,6 @@ mod full_dictionary;
mod hunspell;
mod merged_dictionary;
-/// A word from a dictionary or other similar structure.
-pub type DictWord = SmallVec<[char; 6]>;
-
/// Suggest a correct spelling for a given misspelled word.
/// [`misspelled_word`] is assumed to be quite small (n < 100).
/// [`max_edit_dist`] relates to an optimization that allows the search
diff --git a/harper-ls/Cargo.toml b/harper-ls/Cargo.toml
index 0eb146a8d..218d77254 100644
--- a/harper-ls/Cargo.toml
+++ b/harper-ls/Cargo.toml
@@ -36,3 +36,4 @@ resolve-path = "0.1.0"
open = "5.1.1"
futures = "0.3.30"
unicode-width = "0.1.11"
+serde = { version = "1.0.197", features = ["derive"] }
diff --git a/harper-ls/README.md b/harper-ls/README.md
index c3b1ce471..df870071a 100644
--- a/harper-ls/README.md
+++ b/harper-ls/README.md
@@ -78,6 +78,20 @@ lspconfig.harper_ls.setup {
}
```
+By default, `harper-ls` will mark all diagnostics with HINT.
+If you want to configure this, refer below:
+
+```lua
+lspconfig.harper_ls.setup {
+ settings = {
+ ["harper-ls"] = {
+ diagnosticSeverity = "hint" -- Can also be "information", "warning", or "error"
+ }
+ },
+}
+```
+
+
### File-Local Dictionary
Sometimes, you'll encounter a word (or name) that is only valid within the context of a specific file.
diff --git a/harper-ls/src/backend.rs b/harper-ls/src/backend.rs
index d15a1115e..6cc93420d 100644
--- a/harper-ls/src/backend.rs
+++ b/harper-ls/src/backend.rs
@@ -318,8 +318,13 @@ impl Backend {
};
let lints = doc_state.linter.lint(&doc_state.document);
+ let config = self.config.read().await;
- lints_to_diagnostics(doc_state.document.get_full_content(), &lints)
+ lints_to_diagnostics(
+ doc_state.document.get_full_content(),
+ &lints,
+ config.diagnostic_severity
+ )
}
#[instrument(skip(self))]
@@ -519,6 +524,8 @@ impl LanguageServer for Backend {
}
};
+ dbg!(new_config.diagnostic_severity);
+
{
let mut config = self.config.write().await;
*config = new_config;
@@ -530,6 +537,7 @@ impl LanguageServer for Backend {
doc_state.keys().cloned().collect()
};
+ // Update documents with new config
futures::future::join_all(
keys.iter()
.map(|key| self.update_document_from_file(key, None))
diff --git a/harper-ls/src/config.rs b/harper-ls/src/config.rs
index 19d01ea81..8454e552b 100644
--- a/harper-ls/src/config.rs
+++ b/harper-ls/src/config.rs
@@ -3,13 +3,38 @@ use std::path::PathBuf;
use dirs::{config_dir, data_local_dir};
use harper_core::LintGroupConfig;
use resolve_path::PathResolveExt;
+use serde::{Deserialize, Serialize};
use serde_json::Value;
+#[derive(Debug, Serialize, Deserialize, Clone, Copy)]
+#[serde(rename_all = "camelCase")]
+pub enum DiagnosticSeverity {
+ Error,
+ Warning,
+ Information,
+ Hint
+}
+
+impl DiagnosticSeverity {
+ /// Converts `self` to the equivalent LSP type.
+ pub fn to_lsp(self) -> tower_lsp::lsp_types::DiagnosticSeverity {
+ match self {
+ DiagnosticSeverity::Error => tower_lsp::lsp_types::DiagnosticSeverity::ERROR,
+ DiagnosticSeverity::Warning => tower_lsp::lsp_types::DiagnosticSeverity::WARNING,
+ DiagnosticSeverity::Information => {
+ tower_lsp::lsp_types::DiagnosticSeverity::INFORMATION
+ }
+ DiagnosticSeverity::Hint => tower_lsp::lsp_types::DiagnosticSeverity::HINT
+ }
+ }
+}
+
#[derive(Debug, Clone)]
pub struct Config {
pub user_dict_path: PathBuf,
pub file_dict_path: PathBuf,
- pub lint_config: LintGroupConfig
+ pub lint_config: LintGroupConfig,
+ pub diagnostic_severity: DiagnosticSeverity
}
impl Config {
@@ -38,6 +63,10 @@ impl Config {
base.lint_config = serde_json::from_value(v.clone())?;
}
+ if let Some(v) = value.get("diagnosticSeverity") {
+ base.diagnostic_severity = serde_json::from_value(v.clone())?;
+ }
+
Ok(base)
}
}
@@ -49,7 +78,8 @@ impl Default for Config {
file_dict_path: data_local_dir()
.unwrap()
.join("harper-ls/file_dictionaries/"),
- lint_config: LintGroupConfig::default()
+ lint_config: LintGroupConfig::default(),
+ diagnostic_severity: DiagnosticSeverity::Hint
}
}
}
diff --git a/harper-ls/src/diagnostics.rs b/harper-ls/src/diagnostics.rs
index 60ec154dd..190d0b0e9 100644
--- a/harper-ls/src/diagnostics.rs
+++ b/harper-ls/src/diagnostics.rs
@@ -12,12 +12,17 @@ use tower_lsp::lsp_types::{
WorkspaceEdit
};
+use crate::config::DiagnosticSeverity;
use crate::pos_conv::span_to_range;
-pub fn lints_to_diagnostics(source: &[char], lints: &[Lint]) -> Vec<Diagnostic> {
+pub fn lints_to_diagnostics(
+ source: &[char],
+ lints: &[Lint],
+ severity: DiagnosticSeverity
+) -> Vec<Diagnostic> {
lints
.iter()
- .map(|lint| lint_to_diagnostic(lint, source))
+ .map(|lint| lint_to_diagnostic(lint, source, severity))
.collect()
}
@@ -82,12 +87,12 @@ pub fn lint_to_code_actions<'a>(
results
}
-fn lint_to_diagnostic(lint: &Lint, source: &[char]) -> Diagnostic {
+fn lint_to_diagnostic(lint: &Lint, source: &[char], severity: DiagnosticSeverity) -> Diagnostic {
let range = span_to_range(source, lint.span);
Diagnostic {
range,
- severity: None,
+ severity: Some(severity.to_lsp()),
code: None,
code_description: None,
source: Some("Harper".to_string()),