feat: use tree-sitter for gitcommit

Cargo.toml

@@ -1,5 +1,5 @@
 [workspace]
-members = ["harper-cli", "harper-core", "harper-ls", "harper-comments", "harper-wasm", "harper-tree-sitter", "harper-html", "harper-literate-haskell", "harper-typst", "harper-stats", "harper-pos-utils", "harper-brill", "harper-ink", "harper-python"]
+members = ["harper-cli", "harper-core", "harper-ls", "harper-comments", "harper-wasm", "harper-tree-sitter", "harper-html", "harper-literate-haskell", "harper-typst", "harper-stats", "harper-pos-utils", "harper-brill", "harper-ink", "harper-python", "harper-git-commit"]
 resolver = "2"
 
 # Comment out the below lines if you plan to use a debugger.

harper-git-commit/Cargo.toml

@@ -0,0 +1,16 @@
+[package]
+name = "harper-git-commit"
+version = "0.68.0"
+edition = "2024"
+description = "The language checker for developers."
+license = "Apache-2.0"
+repository = "https://github.com/automattic/harper"
+
+[dependencies]
+harper-core = { path = "../harper-core", version = "0.68.0" }
+harper-tree-sitter = { path = "../harper-tree-sitter", version = "0.68.0" }
+tree-sitter-gitcommit = { git = "https://github.com/gbprod/tree-sitter-gitcommit", rev = "a716678c0f00645fed1e6f1d0eb221481dbd6f6d" }
+tree-sitter = "0.25.10"
+
+[dev-dependencies]
+paste = "1.0.15"

harper-git-commit/src/lib.rs

@@ -0,0 +1,30 @@
+use harper_core::Token;
+use harper_core::parsers::{self, Markdown, MarkdownOptions, Parser};
+use harper_tree_sitter::TreeSitterMasker;
+use tree_sitter::Node;
+
+pub struct GitCommitParser {
+    /// Used to grab the text nodes, and parse them as markdown.
+    inner: parsers::Mask<TreeSitterMasker, Markdown>,
+}
+
+impl GitCommitParser {
+    fn node_condition(n: &Node) -> bool {
+        matches!(n.kind(), "subject" | "message_line" | "breaking_change")
+    }
+
+    pub fn new(markdown_options: MarkdownOptions) -> Self {
+        Self {
+            inner: parsers::Mask::new(
+                TreeSitterMasker::new(tree_sitter_gitcommit::language(), Self::node_condition),
+                Markdown::new(markdown_options),
+            ),
+        }
+    }
+}
+
+impl Parser for GitCommitParser {
+    fn parse(&self, source: &[char]) -> Vec<Token> {
+        self.inner.parse(source)
+    }
+}

harper-git-commit/tests/run_tests.rs

@@ -0,0 +1,41 @@
+use harper_core::linting::{LintGroup, Linter};
+use harper_core::parsers::MarkdownOptions;
+use harper_core::spell::FstDictionary;
+use harper_core::{Dialect, Document};
+use harper_git_commit::GitCommitParser;
+
+/// Creates a unit test checking that the linting of a git commit document (in
+/// `tests_sources`) produces the expected number of lints.
+macro_rules! create_test {
+    ($filename:ident.txt, $correct_expected:expr) => {
+        paste::paste! {
+            #[test]
+            fn [<lints_ $filename _correctly>](){
+                 let source = include_str!(
+                    concat!(
+                        "./test_sources/",
+                        concat!(stringify!($filename), ".txt")
+                    )
+                 );
+
+                 let dict = FstDictionary::curated();
+                 let document = Document::new(source, &GitCommitParser::new(MarkdownOptions::default()), &dict);
+
+                 let mut linter = LintGroup::new_curated(dict, Dialect::American);
+                 let lints = linter.lint(&document);
+
+                 dbg!(&lints);
+                 assert_eq!(lints.len(), $correct_expected);
+
+                 // Make sure that all generated tokens span real characters
+                 for token in document.tokens(){
+                     assert!(token.span.try_get_content(document.get_source()).is_some());
+                 }
+            }
+        }
+    };
+}
+
+create_test!(simple_commit.txt, 1);
+create_test!(complex_verbose_commit.txt, 2);
+create_test!(conventional_commit.txt, 3);

harper-git-commit/tests/test_sources/complex_verbose_commit.txt

@@ -0,0 +1,26 @@
+This is the the subject
+
+This is a first line without typos
+# This is a comment with a typooo that should be ignored
+This is a line below the comment with typooos
+
+# Please enter the commit message for your changes. Lines starting
+# with '#' will be ignored, and an empty message aborts the commit.
+#
+# On branch main
+#
+# Initial commit
+#
+# Changes to be committed:
+#new file:   myfile.txt
+#
+# ------------------------ >8 ------------------------
+# Do not modify or remove the line above.
+# Everything below it will be ignored.
+diff --git a/myfile.txt b/myfile.txt
+new file mode 100644
+index 0000000..485c415
+--- /dev/null
++++ b/myfile.txt
+@@ -0,0 +1 @@
++some textt in the file that was added (the typo should be ignored)

harper-git-commit/tests/test_sources/conventional_commit.txt

@@ -0,0 +1,17 @@
+feat(stuff): use session-based authentiation
+
+BREAKING CHANGE: JWT authentication removed. API clients mustt now use
+session cookies instead of Authorization headers with bearer tokens.
+
+Sessions expire after 24 hours of inactvity.
+
+Closes: #247
+Reviewed-by: John Doe <[email protected]>
+
+# Please enter the commit message for your changes. Lines starting
+# with '#' will be ignored, and an empty message aborts the commit.
+#
+# On branch main
+# Changes to be committed:
+#	modified:   myfile.txt
+#

harper-git-commit/tests/test_sources/simple_commit.txt

@@ -0,0 +1 @@
+A simple commit with a typo: comit

harper-ls/Cargo.toml

@@ -12,6 +12,7 @@ harper-stats = { path = "../harper-stats", version = "0.68.0" }
 harper-literate-haskell = { path = "../harper-literate-haskell", version = "0.68.0" }
 harper-core = { path = "../harper-core", version = "0.68.0", features = ["concurrent"] }
 harper-comments = { path = "../harper-comments", version = "0.68.0" }
+harper-git-commit = { path = "../harper-git-commit", version = "0.68.0" }
 harper-typst = { path = "../harper-typst", version = "0.68.0" }
 harper-html = { path = "../harper-html", version = "0.68.0" }
 harper-python = { path = "../harper-python", version = "0.68.0" }

harper-ls/src/backend.rs

@@ -7,7 +7,6 @@ use std::sync::Arc;
 use crate::config::Config;
 use crate::dictionary_io::{load_dict, save_dict};
 use crate::document_state::DocumentState;
-use crate::git_commit_parser::GitCommitParser;
 use crate::ignored_lints_io::{load_ignored_lints, save_ignored_lints};
 use crate::io_utils::fileify_path;
 use anyhow::{Context, Result, anyhow};
@@ -19,6 +18,7 @@ use harper_core::parsers::{
 };
 use harper_core::spell::{Dictionary, FstDictionary, MergedDictionary, MutableDictionary};
 use harper_core::{Dialect, DictWordMetadata, Document, IgnoredLints};
+use harper_git_commit::GitCommitParser;
 use harper_html::HtmlParser;
 use harper_ink::InkParser;
 use harper_literate_haskell::LiterateHaskellParser;
@@ -382,9 +382,7 @@ impl Backend {
             }
             "ink" => Some(Box::new(InkParser::default())),
             "markdown" => Some(Box::new(Markdown::new(markdown_options))),
-            "git-commit" | "gitcommit" => {
-                Some(Box::new(GitCommitParser::new_markdown(markdown_options)))
-            }
+            "git-commit" | "gitcommit" => Some(Box::new(GitCommitParser::new(markdown_options))),
             "html" => Some(Box::new(HtmlParser::default())),
             "mail" | "plaintext" | "text" => Some(Box::new(PlainEnglish)),
             "typst" => Some(Box::new(Typst)),

harper-ls/src/main.rs

@@ -9,7 +9,7 @@ mod config;
 mod diagnostics;
 mod dictionary_io;
 mod document_state;
-mod git_commit_parser;
+
 mod ignored_lints_io;
 mod io_utils;
 mod pos_conv;