feat: add ink support (#1894)

* feat: add harper_ink crate with impl and tests

* feat: add harper_ink to harper_ls

* docs: add `ink` to the list of supported languages

---------

Co-authored-by: Elijah Potter <me@elijahpotter.dev>

Cargo.lock

@@ -2299,6 +2299,7 @@ dependencies = [
  "dirs 6.0.0",
  "harper-comments",
  "harper-core",
+ "harper-ink",
  "harper-literate-haskell",
  "harper-pos-utils",
  "harper-stats",
@@ -2393,6 +2394,17 @@ dependencies = [
  "tree-sitter-html",
 ]
 
+[[package]]
+name = "harper-ink"
+version = "0.65.0"
+dependencies = [
+ "harper-core",
+ "harper-tree-sitter",
+ "paste",
+ "tree-sitter",
+ "tree-sitter-ink-lbz",
+]
+
 [[package]]
 name = "harper-literate-haskell"
 version = "0.65.0"
@@ -2416,6 +2428,7 @@ dependencies = [
  "harper-comments",
  "harper-core",
  "harper-html",
+ "harper-ink",
  "harper-literate-haskell",
  "harper-stats",
  "harper-typst",
@@ -4836,6 +4849,16 @@ dependencies = [
  "tree-sitter-language",
 ]
 
+[[package]]
+name = "tree-sitter-ink-lbz"
+version = "0.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a44bf578477077d12eb671fd67fc47ab71516a1f3913a5dac96346a8faa33210"
+dependencies = [
+ "cc",
+ "tree-sitter-language",
+]
+
 [[package]]
 name = "tree-sitter-java"
 version = "0.23.5"

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

harper-cli/Cargo.toml

@@ -21,6 +21,7 @@ serde = { version = "1.0.226", features = ["derive"] }
 serde_json = "1.0.145"
 strum = "0.27.2"
 strum_macros = "0.27.2"
+harper-ink = { version = "0.65.0", path = "../harper-ink" }
 
 [features]
 default = []

harper-cli/src/main.rs

@@ -20,6 +20,7 @@ use harper_core::{
     CharStringExt, Dialect, DictWordMetadata, Document, Span, TokenKind, TokenStringExt,
     dict_word_metadata_orthography::OrthFlags, remove_overlaps,
 };
+use harper_ink::InkParser;
 use harper_literate_haskell::LiterateHaskellParser;
 #[cfg(feature = "training")]
 use harper_pos_utils::{BrillChunker, BrillTagger, BurnChunkerCpu};
@@ -818,6 +819,7 @@ fn load_file(
         .map(|v| v.to_str().unwrap())
     {
         Some("md") => Box::new(Markdown::default()),
+        Some("ink") => Box::new(InkParser::default()),
 
         Some("lhs") => Box::new(LiterateHaskellParser::new_markdown(
             MarkdownOptions::default(),

harper-ink/Cargo.toml

@@ -0,0 +1,16 @@
+[package]
+name = "harper-ink"
+version = "0.65.0"
+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 = "0.65.0" }
+harper-tree-sitter = { path = "../harper-tree-sitter", version = "0.65.0" }
+tree-sitter-ink-lbz = "0.0.1"
+tree-sitter = "0.25.8"
+
+[dev-dependencies]
+paste = "1.0.15"

harper-ink/src/lib.rs

@@ -0,0 +1,39 @@
+use harper_core::parsers::{self, Parser, PlainEnglish};
+use harper_core::{Token, TokenKind};
+use harper_tree_sitter::TreeSitterMasker;
+use tree_sitter::Node;
+
+pub struct InkParser {
+    inner: parsers::Mask<TreeSitterMasker, PlainEnglish>,
+}
+
+impl InkParser {
+    fn node_condition(n: &Node) -> bool {
+        matches!(n.kind(), "contentText" | "blockComment" | "lineComment")
+    }
+}
+
+impl Default for InkParser {
+    fn default() -> Self {
+        Self {
+            inner: parsers::Mask::new(
+                TreeSitterMasker::new(tree_sitter_ink_lbz::LANGUAGE.into(), Self::node_condition),
+                PlainEnglish,
+            ),
+        }
+    }
+}
+
+impl Parser for InkParser {
+    fn parse(&self, source: &[char]) -> Vec<Token> {
+        let mut tokens = self.inner.parse(source);
+
+        for token in &mut tokens {
+            if let TokenKind::Space(v) = &mut token.kind {
+                *v = (*v).clamp(0, 1);
+            }
+        }
+
+        tokens
+    }
+}

harper-ink/tests/run_tests.rs

@@ -0,0 +1,41 @@
+use harper_core::linting::{LintGroup, Linter};
+use harper_core::spell::FstDictionary;
+use harper_core::{Dialect, Document};
+use harper_ink::InkParser;
+
+/// Creates a unit test checking that the linting of a Ink document (in
+/// `tests_sources`) produces the expected number of lints.
+macro_rules! create_test {
+    ($filename:ident.ink, $correct_expected:expr) => {
+        paste::paste! {
+            #[test]
+            fn [<lints_ $filename _correctly>](){
+                 let source = include_str!(
+                    concat!(
+                        "./test_sources/",
+                        concat!(stringify!($filename), ".ink")
+                    )
+                 );
+
+                 let dict = FstDictionary::curated();
+                 let document = Document::new(&source, &InkParser::default(),
+                      &FstDictionary::curated()
+                      );
+
+                 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!(good.ink, 0);
+create_test!(bad.ink, 6);

harper-ink/tests/test_sources/bad.ink

@@ -0,0 +1,15 @@
+=== Knot ===
+text here is checked: chungus
+
+= Stitch
+
+~identifiersAreNotchecked = "but strings are: chungus"
+// comments are also checked -> chungus
+
+/*
+chungus
+*/
+
+* choices are checked
++ chungus
++ normal text

harper-ink/tests/test_sources/good.ink

@@ -0,0 +1,15 @@
+=== Knot ===
+    this is a thing
+
+= Stitch
+  * an option
+  + another option
+
+  + normal text
+    + + indented choice
+    + + another indented choice
+
+  - gather
+
+= ThreeWordStitch
+    test

harper-ls/Cargo.toml

@@ -30,6 +30,7 @@ futures = "0.3.31"
 serde = { version = "1.0.226", features = ["derive"] }
 indexmap = { version = "2.11.1", features = ["serde"] }
 globset = "0.4.16"
+harper-ink = { version = "0.65.0", path = "../harper-ink" }
 
 [features]
 default = []

harper-ls/src/backend.rs

@@ -20,6 +20,7 @@ use harper_core::parsers::{
 use harper_core::spell::{Dictionary, FstDictionary, MergedDictionary, MutableDictionary};
 use harper_core::{Dialect, DictWordMetadata, Document, IgnoredLints};
 use harper_html::HtmlParser;
+use harper_ink::InkParser;
 use harper_literate_haskell::LiterateHaskellParser;
 use harper_stats::{Record, Stats};
 use harper_typst::Typst;
@@ -378,6 +379,7 @@ impl Backend {
                     Some(Box::new(parser))
                 }
             }
+            "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)))

packages/web/src/routes/docs/integrations/language-server/+page.md

@@ -301,6 +301,7 @@ These configs are under the `markdown` key:
 | Typst             |            `typst`            |               |
 | Kotlin            |            `kotlin`           |            ✅ |
 | Clojure           |            `clojure`          |            ✅ |
+| Ink               |            `ink`              |               |
 
 Want your language added?
 Let us know by [commenting on this issue](https://github.com/Automattic/harper/issues/79).