feat: add jjdescription parser (#2082)

jjdescription file types are used by jujutsu to write messages in,
similar to git commits. We use tree-sitter-jjdescription and the
existing markdown parser to parse these kinds of files.

Author: Tijs-B

Cargo.lock

@@ -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-jjdescription"]
 resolver = "2"
 
 # Comment out the below lines if you plan to use a debugger.

Cargo.toml

@@ -0,0 +1,16 @@
+[package]
+name = "harper-jjdescription"
+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-jjdescription = "0.0.1"
+tree-sitter = "0.25.10"
+
+[dev-dependencies]
+paste = "1.0.15"

harper-jjdescription/Cargo.toml

@@ -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 JJDescriptionParser {
+    /// Used to grab the text nodes, and parse them as markdown.
+    inner: parsers::Mask<TreeSitterMasker, Markdown>,
+}
+
+impl JJDescriptionParser {
+    fn node_condition(n: &Node) -> bool {
+        n.kind() == "text"
+    }
+
+    pub fn new(markdown_options: MarkdownOptions) -> Self {
+        Self {
+            inner: parsers::Mask::new(
+                TreeSitterMasker::new(tree_sitter_jjdescription::language(), Self::node_condition),
+                Markdown::new(markdown_options),
+            ),
+        }
+    }
+}
+
+impl Parser for JJDescriptionParser {
+    fn parse(&self, source: &[char]) -> Vec<Token> {
+        self.inner.parse(source)
+    }
+}

harper-jjdescription/src/lib.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_jjdescription::JJDescriptionParser;
+
+/// 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, &JJDescriptionParser::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_description.txt, 1);
+create_test!(complex_verbose_description.txt, 2);
+create_test!(conventional_description.txt, 3);

harper-jjdescription/tests/run_tests.rs

@@ -0,0 +1,20 @@
+This is the the subject
+
+This is a first line without typos
+JJ: This is a comment with a typoo that should be ignored
+This is a line below the comment with typooos
+
+JJ: This commit contains the following changes:
+JJ:     myfile.txt | 1 +
+JJ:     1 file changed, 1 insertion(+), 0 deletions(-)
+
+JJ: ignore-rest
+diff --git a/myfile.txt b/myfile.txt
+new file mode 100644
+index 0000000000..54f266d2db
+--- /dev/null
++++ b/myfile.txt
+@@ -0,0 +1,1 @@
++typooo in the file
+
+JJ: Lines starting with "JJ:" (like this one) will be removed.

harper-jjdescription/tests/test_sources/complex_verbose_description.txt

@@ -0,0 +1,25 @@
+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 <john@example.com>
+
+JJ: Change ID: qrutlxlw
+JJ: This commit contains the following changes:
+JJ:     M Cargo.lock
+JJ:     M Cargo.toml
+JJ:     A harper-jjdescription/Cargo.toml
+JJ:     A harper-jjdescription/src/lib.rs
+JJ:     A harper-jjdescription/tests/run_tests.rs
+JJ:     A harper-jjdescription/tests/test_sources/complex_verbose_description.txt
+JJ:     A harper-jjdescription/tests/test_sources/conventional_description.txt
+JJ:     A harper-jjdescription/tests/test_sources/simple_description.txt
+JJ:     M harper-ls/Cargo.toml
+JJ:     M harper-ls/src/backend.rs
+JJ:
+JJ: Lines starting with "JJ:" (like this one) will be removed.
+

harper-jjdescription/tests/test_sources/conventional_description.txt

@@ -0,0 +1 @@
+A simple description with a typo: descrption

harper-jjdescription/tests/test_sources/simple_description.txt

@@ -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-jjdescription = { path = "../harper-jjdescription", 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/Cargo.toml

@@ -21,6 +21,7 @@ use harper_core::spell::{Dictionary, FstDictionary, MergedDictionary, MutableDic
 use harper_core::{Dialect, DictWordMetadata, Document, IgnoredLints};
 use harper_html::HtmlParser;
 use harper_ink::InkParser;
+use harper_jjdescription::JJDescriptionParser;
 use harper_literate_haskell::LiterateHaskellParser;
 use harper_python::PythonParser;
 use harper_stats::{Record, Stats};
@@ -385,6 +386,7 @@ impl Backend {
             "git-commit" | "gitcommit" => {
                 Some(Box::new(GitCommitParser::new_markdown(markdown_options)))
             }
+            "jjdescription" => Some(Box::new(JJDescriptionParser::new(markdown_options))),
             "html" => Some(Box::new(HtmlParser::default())),
             "mail" | "plaintext" | "text" => Some(Box::new(PlainEnglish)),
             "typst" => Some(Box::new(Typst)),