chore: clippy

Cargo.lock

@@ -32,6 +32,12 @@ dependencies = [
  "alloc-no-stdlib",
 ]
 
+[[package]]
+name = "allocator-api2"
+version = "0.2.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "683d7910e743518b0e34f1186f92494becacb047c7b6bf616c96772180fef923"
+
 [[package]]
 name = "anstream"
 version = "1.0.0"
@@ -100,6 +106,15 @@ dependencies = [
  "object",
 ]
 
+[[package]]
+name = "arc-swap"
+version = "1.9.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c049c0be4daef0b145cb3555416b3b8ef5b7888a38aea1a3a155801fe7b0810b"
+dependencies = [
+ "rustversion",
+]
+
 [[package]]
 name = "arrayref"
 version = "0.3.9"
@@ -585,6 +600,12 @@ version = "0.5.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7cd812cc2bc1d69d4764bd80df88b4317eaef9e773c75226407d9bc0876b211c"
 
+[[package]]
+name = "diff"
+version = "0.1.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "56254986775e3233ffa9c4d7d3faaf6d36a2c09d30b20687e9f88bc8bafc16c8"
+
 [[package]]
 name = "displaydoc"
 version = "0.2.6"
@@ -733,6 +754,12 @@ version = "1.0.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
 
+[[package]]
+name = "foldhash"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d9c4f5dac5e15c24eb999c26181a6ca40b39fe946cbe4c263c7209467bc83af2"
+
 [[package]]
 name = "foldhash"
 version = "0.2.0"
@@ -842,8 +869,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d99fc21d493812643aae86d53b7bbd02f376434a90317e8a790bc209fdd6605e"
 dependencies = [
  "bytemuck",
- "foldhash",
+ "foldhash 0.2.0",
- "hashbrown",
+ "hashbrown 0.17.1",
  "log",
  "peniko",
  "png",
@@ -872,13 +899,24 @@ dependencies = [
  "zerocopy",
 ]
 
+[[package]]
+name = "hashbrown"
+version = "0.15.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9229cfe53dfd69f0609a49f65461bd93001ea1ef889cd5529dd176593f5338a1"
+dependencies = [
+ "allocator-api2",
+ "equivalent",
+ "foldhash 0.1.5",
+]
+
 [[package]]
 name = "hashbrown"
 version = "0.17.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ed5909b6e89a2db4456e54cd5f673791d7eca6732202bbf2a9cc504fe2f9b84a"
 dependencies = [
- "foldhash",
+ "foldhash 0.2.0",
 ]
 
 [[package]]
@@ -1341,7 +1379,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d466e9454f08e4a911e14806c24e16fba1b4c121d1ea474396f396069cf949d9"
 dependencies = [
  "equivalent",
- "hashbrown",
+ "hashbrown 0.17.1",
  "rayon",
  "serde",
  "serde_core",
@@ -1385,18 +1423,32 @@ dependencies = [
  "clap",
  "flate2",
  "itertools",
+ "ropey",
  "tar",
  "time",
  "tokio",
  "tower-http",
  "tracing",
  "tracing-subscriber",
+ "tree-house",
+ "tree-house-bindings",
  "typst",
  "typst-html",
  "typst-kit",
+ "typst-library",
+ "typst-utils",
  "ureq",
 ]
 
+[[package]]
+name = "kstring"
+version = "2.0.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b609e7ca5ea38f093c20a4a102335b247221c9643b7a6bc3510f196f99499a9e"
+dependencies = [
+ "static_assertions",
+]
+
 [[package]]
 name = "kurbo"
 version = "0.13.1"
@@ -1421,6 +1473,16 @@ version = "0.2.186"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "68ab91017fe16c622486840e4c83c9a37afeff978bd239b5293d61ece587de66"
 
+[[package]]
+name = "libloading"
+version = "0.8.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d7c4b02199fee7c5d21a5ae7d8cfa79a6ef5bb2fc834d6e9058e89c825efdc55"
+dependencies = [
+ "cfg-if",
+ "windows-link",
+]
+
 [[package]]
 name = "libm"
 version = "0.2.16"
@@ -1822,6 +1884,16 @@ dependencies = [
  "zerocopy",
 ]
 
+[[package]]
+name = "pretty_assertions"
+version = "1.4.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3ae130e2f271fbc2ac3a40fb1d07180839cdbbe443c7a27e1e3c13c5cac0116d"
+dependencies = [
+ "diff",
+ "yansi",
+]
+
 [[package]]
 name = "proc-macro-hack"
 version = "0.5.20+deprecated"
@@ -1974,6 +2046,19 @@ dependencies = [
  "regex-syntax",
 ]
 
+[[package]]
+name = "regex-cursor"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0497c781d2f982ae8284d2932aee6a877e58a4541daa5e8fadc18cc75c23a61d"
+dependencies = [
+ "log",
+ "memchr",
+ "regex-automata",
+ "regex-syntax",
+ "ropey",
+]
+
 [[package]]
 name = "regex-syntax"
 version = "0.8.11"
@@ -2000,6 +2085,16 @@ version = "3.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c85cd47a33a4510b1424fe796498e174c6a9cf94e606460ef022a19f3e4ff85e"
 
+[[package]]
+name = "ropey"
+version = "1.6.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "93411e420bcd1a75ddd1dc3caf18c23155eda2c090631a85af21ba19e97093b5"
+dependencies = [
+ "smallvec",
+ "str_indices",
+]
+
 [[package]]
 name = "roxmltree"
 version = "0.21.1"
@@ -2073,6 +2168,12 @@ dependencies = [
  "untrusted",
 ]
 
+[[package]]
+name = "rustversion"
+version = "1.0.23"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cf54715a573b99ac80df0bc206da022bcd442c974952c7b9720069370852e21f"
+
 [[package]]
 name = "rustybuzz"
 version = "0.20.1"
@@ -2308,6 +2409,18 @@ dependencies = [
  "windows-sys 0.61.2",
 ]
 
+[[package]]
+name = "static_assertions"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f"
+
+[[package]]
+name = "str_indices"
+version = "0.4.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d08889ec5408683408db66ad89e0e1f93dff55c73a4ccc71c427d5b277ee47e6"
+
 [[package]]
 name = "strict-num"
 version = "0.1.1"
@@ -2715,6 +2828,38 @@ dependencies = [
  "tracing-log",
 ]
 
+[[package]]
+name = "tree-house"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "333476442882205ab249a8ced263ae53db3d0797c727cc3424836a6fb221d7b1"
+dependencies = [
+ "arc-swap",
+ "hashbrown 0.15.5",
+ "kstring",
+ "once_cell",
+ "pretty_assertions",
+ "regex",
+ "regex-cursor",
+ "ropey",
+ "slab",
+ "tree-house-bindings",
+ "unicode-width",
+]
+
+[[package]]
+name = "tree-house-bindings"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6f5d0eed0db98578618e598b8f6f413f7c172bade1f12f8242d71c38dd5deb0f"
+dependencies = [
+ "cc",
+ "libloading",
+ "regex-cursor",
+ "ropey",
+ "thiserror",
+]
+
 [[package]]
 name = "ttf-parser"
 version = "0.25.1"
@@ -3155,6 +3300,12 @@ version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b1d386ff53b415b7fe27b50bb44679e2cc4660272694b7b6f3326d8480823a94"
 
+[[package]]
+name = "unicode-width"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "68f5e5f3158ecfd4b8ff6fe086db7c8467a2dfdac97fe420f2b7c4aa97af66d6"
+
 [[package]]
 name = "unsafe-libyaml"
 version = "0.2.11"
@@ -3282,7 +3433,7 @@ dependencies = [
  "bytemuck",
  "fearless_simd",
  "guillotiere",
- "hashbrown",
+ "hashbrown 0.17.1",
  "log",
  "peniko",
  "png",
@@ -3299,7 +3450,7 @@ dependencies = [
  "bytemuck",
  "fearless_simd",
  "guillotiere",
- "hashbrown",
+ "hashbrown 0.17.1",
  "log",
  "peniko",
  "png",
@@ -3315,7 +3466,7 @@ checksum = "6d8ded630e8316bb94a55881256506d1f3b9947b5f66db8a7d32ca7ba02decd0"
 dependencies = [
  "bytemuck",
  "glifo",
- "hashbrown",
+ "hashbrown 0.17.1",
  "png",
  "vello_common 0.0.8",
 ]
@@ -3546,6 +3697,12 @@ dependencies = [
  "linked-hash-map",
 ]
 
+[[package]]
+name = "yansi"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cfe53a6657fd280eaa890a3bc59152892ffa3e30101319d168b781ed6529b049"
+
 [[package]]
 name = "yoke"
 version = "0.8.3"

Cargo.toml

@@ -23,3 +23,8 @@ typst-kit = "0.15.1"
 typst = "0.15.1"
 ureq = "3.3.0"
 clap = { version = "4.6.2", features = ["derive"] }
+typst-library = "0.15.1"
+typst-utils = "0.15.1"
+tree-house = "0.4.0"
+tree-house-bindings = { version = "0.3.2", features = ["ropey"] }
+ropey = "1.6.1"

flake.lock

@@ -18,7 +18,42 @@
         "type": "github"
       }
     },
+    "helix": {
+      "inputs": {
+        "nixpkgs": "nixpkgs",
+        "rust-overlay": "rust-overlay"
+      },
+      "locked": {
+        "lastModified": 1784219418,
+        "narHash": "sha256-yfNBfV6HWAQU/xZXwrFoedsSYsDmCDIlTHRxAZI6teY=",
+        "owner": "helix-editor",
+        "repo": "helix",
+        "rev": "f6f3eb1fe4a73933466f30175843059a4381a985",
+        "type": "github"
+      },
+      "original": {
+        "owner": "helix-editor",
+        "repo": "helix",
+        "type": "github"
+      }
+    },
     "nixpkgs": {
+      "locked": {
+        "lastModified": 1775036866,
+        "narHash": "sha256-ZojAnPuCdy657PbTq5V0Y+AHKhZAIwSIT2cb8UgAz/U=",
+        "owner": "nixos",
+        "repo": "nixpkgs",
+        "rev": "6201e203d09599479a3b3450ed24fa81537ebc4e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nixos",
+        "ref": "nixos-unstable",
+        "repo": "nixpkgs",
+        "type": "github"
+      }
+    },
+    "nixpkgs_2": {
       "locked": {
         "lastModified": 1784120854,
         "narHash": "sha256-KesHgItiZPgGX740axSiQLcIQ8D24MDqNpkKYWIek8k=",
@@ -37,7 +72,29 @@
     "root": {
       "inputs": {
         "flake-utils": "flake-utils",
-        "nixpkgs": "nixpkgs"
+        "helix": "helix",
+        "nixpkgs": "nixpkgs_2"
+      }
+    },
+    "rust-overlay": {
+      "inputs": {
+        "nixpkgs": [
+          "helix",
+          "nixpkgs"
+        ]
+      },
+      "locked": {
+        "lastModified": 1775358767,
+        "narHash": "sha256-f2eC+WIfhjevCPQILuV08i/kmKZzYZpUvkom/33VxCA=",
+        "owner": "oxalica",
+        "repo": "rust-overlay",
+        "rev": "20fd44bc663daa53a2575e01293e24e681d62244",
+        "type": "github"
+      },
+      "original": {
+        "owner": "oxalica",
+        "repo": "rust-overlay",
+        "type": "github"
       }
     },
     "systems": {

flake.nix

@@ -2,6 +2,7 @@
   inputs = {
     nixpkgs.url = "github:nixos/nixpkgs/nixos-unstable";
     flake-utils.url = "github:numtide/flake-utils";
+    helix.url = "github:helix-editor/helix";
   };
 
   outputs =
@@ -9,6 +10,7 @@
       self,
       nixpkgs,
       flake-utils,
+      helix,
       ...
     }:
     flake-utils.lib.eachDefaultSystem (
@@ -19,6 +21,31 @@
         };
       in
       rec {
+        packages.syntax =
+          let
+            grammars = pkgs.callPackage "${helix}/grammars.nix" { };
+          in
+          pkgs.stdenv.mkDerivation {
+            name = "syntax";
+            phases = [
+              "buildPhase"
+              "installPhase"
+            ];
+            buildPhase = ''
+              cp -r ${helix}/runtime/queries/* ./
+              chmod u+w ./* -R
+              for so in ${grammars}/*.so; do
+                name=$(basename "$so" .so)
+                if [ -d "./$name" ]; then
+                  cp "$so" "./$name/$name.so"
+                fi
+              done
+            '';
+            installPhase = ''
+              mkdir -p $out
+              cp -r ./* $out/
+            '';
+          };
         packages.karnaugh = pkgs.rustPlatform.buildRustPackage (final: {
           pname = "karnaugh";
           version = "0.1.5";

src/main.rs

@@ -25,6 +25,7 @@ use tracing_subscriber::FmtSubscriber;
 
 use crate::typst::TypstContext;
 
+mod syntax;
 mod typst;
 
 #[derive(Clone)]
@@ -59,6 +60,15 @@ struct AppConfig {
     #[arg(short = 'd', long, value_name = "DIR", default_value = "/tmp/karnaugh")]
     packages_root: PathBuf,
 
+    /// The directory where the syntax files are stored.
+    #[arg(short = 'g', long, value_name = "DIR", default_value = None)]
+    syntax_root: Option<PathBuf>,
+
+    /// Grammars to load, separated by commas.
+    /// Useful to avoid the memory required to load all languages.
+    #[arg(short = 'L', long = "lang", value_name = "NAME", value_delimiter = ',')]
+    languages: Vec<String>,
+
     /// The log level to use
     #[arg(short = 'l', long, value_name = "LEVEL", default_value_t = Level::INFO)]
     log_level: Level,
@@ -94,6 +104,8 @@ async fn main() {
         .with_max_level(state.config.log_level)
         .init();
 
+    syntax::init(&state.config);
+
     // TODO: Replace with config option
     let mut favicon = state.config.get_full_assets_path();
     favicon.push(PathBuf::from("favicon.svg"));
@@ -117,7 +129,7 @@ async fn main() {
     info!("Serving Karnaugh on socket {}", socket);
     let listener = tokio::net::TcpListener::bind(socket)
         .await
-        .expect(&format!("Failed to bind to socket {}", socket));
+        .unwrap_or_else(|_| panic!("Failed to bind to socket {}", socket));
     axum::serve(listener, app)
         .await
         .expect("Failed to serve site");

src/syntax.rs

@@ -0,0 +1,189 @@
+use std::{
+    borrow::Cow,
+    collections::{HashMap, HashSet},
+    error::Error,
+    fs,
+    path::Path,
+    sync::OnceLock,
+};
+
+use tracing::{info, warn};
+use tree_house::{
+    InjectionLanguageMarker, Language, LanguageConfig, LanguageLoader, highlighter::Highlight,
+    read_query,
+};
+use tree_house_bindings::Grammar;
+use typst::foundations::{Func, NativeFunc, func};
+
+use crate::AppConfig;
+
+static REGISTRY: OnceLock<LanguageRegistry> = OnceLock::new();
+
+/// Class names used in highlighted code.
+const HIGHLIGHT_NAMES: &[&str] = &[
+    "attribute",
+    "comment",
+    "comment.block",
+    "comment.line",
+    "constant",
+    "constant.builtin",
+    "constructor",
+    "function",
+    "function.builtin",
+    "function.macro",
+    "function.method",
+    "keyword",
+    "keyword.control",
+    "keyword.operator",
+    "label",
+    "namespace",
+    "number",
+    "operator",
+    "property",
+    "punctuation",
+    "punctuation.bracket",
+    "punctuation.delimiter",
+    "special",
+    "string",
+    "string.special",
+    "tag",
+    "type",
+    "type.builtin",
+    "variable",
+    "variable.builtin",
+    "variable.parameter",
+];
+
+/// Get the longest matching highlight name for a given scope.
+fn highlight_for_scope(scope: &str) -> Option<Highlight> {
+    let mut candidate = scope;
+    loop {
+        if let Some(i) = HIGHLIGHT_NAMES.iter().position(|&n| n == candidate) {
+            return Some(Highlight::new(i as u32));
+        }
+        match candidate.rfind('.') {
+            Some(dot) => candidate = &candidate[..dot],
+            None => return None,
+        }
+    }
+}
+
+struct LanguageRegistry {
+    by_name: HashMap<String, Language>,
+    configs: Vec<LanguageConfig>,
+}
+
+impl LanguageRegistry {
+    /// Loads grammars from `syntax_root`. When `languages` is non-empty, only
+    /// grammars with those names are loaded; otherwise every grammar found is.
+    fn new(syntax_root: &Path, languages: &[String]) -> Self {
+        let filter: Option<HashSet<&str>> =
+            (!languages.is_empty()).then(|| languages.iter().map(String::as_str).collect());
+
+        let mut by_name = HashMap::new();
+        let mut configs = Vec::new();
+
+        for entry in fs::read_dir(syntax_root).into_iter().flatten().flatten() {
+            let name = entry.file_name().to_string_lossy().into_owned();
+            if filter.as_ref().is_some_and(|f| !f.contains(name.as_str())) {
+                continue;
+            }
+            let so_path = entry.path().join(format!("{name}.so"));
+            if !so_path.exists() {
+                continue;
+            }
+
+            match load_language(&name, &entry.path(), syntax_root) {
+                Ok(config) => {
+                    by_name.insert(name, Language::new(configs.len() as u32));
+                    configs.push(config);
+                }
+                Err(e) => warn!("Failed to load grammar {name}: {e}"),
+            }
+        }
+
+        // Warn about explicitly requested grammars that were not found.
+        for lang in languages {
+            if !by_name.contains_key(lang) {
+                warn!("Requested grammar {lang} was not found in the syntax root");
+            }
+        }
+
+        info!("Loaded {} syntax highlighting grammars", configs.len());
+        Self { by_name, configs }
+    }
+}
+
+fn load_language(
+    name: &str,
+    lang_dir: &Path,
+    syntax_root: &Path,
+) -> Result<LanguageConfig, Box<dyn Error>> {
+    let grammar = unsafe { Grammar::new(name, &lang_dir.join(format!("{name}.so")))? };
+
+    let read = |kind: &str| {
+        read_query(name, |lang| {
+            fs::read_to_string(syntax_root.join(lang).join(format!("{kind}.scm")))
+                .unwrap_or_default()
+        })
+    };
+
+    let config = LanguageConfig::new(
+        grammar,
+        &read("highlights"),
+        &read("injections"),
+        &read("locals"),
+    )?;
+    config.configure(highlight_for_scope);
+
+    Ok(config)
+}
+
+impl LanguageLoader for LanguageRegistry {
+    fn language_for_marker(&self, marker: InjectionLanguageMarker<'_>) -> Option<Language> {
+        match marker {
+            InjectionLanguageMarker::Name(name) => self.by_name.get(name).copied(),
+            InjectionLanguageMarker::Match(text) => {
+                let name: Cow<str> = text.into();
+                self.by_name.get(name.as_ref()).copied()
+            }
+            _ => None,
+        }
+    }
+
+    fn get_config(&self, lang: Language) -> Option<&LanguageConfig> {
+        self.configs.get(lang.idx())
+    }
+}
+
+pub fn init(config: &AppConfig) {
+    match &config.syntax_root {
+        Some(syntax_root) => {
+            info!("Loading syntax files from {}", syntax_root.display());
+            REGISTRY.get_or_init(|| LanguageRegistry::new(syntax_root, &config.languages));
+        }
+        None => {
+            warn!(
+                "No syntax root has been configured, improved tree-sitter highlighting will not be possible."
+            )
+        }
+    }
+}
+
+pub fn highlight_func() -> Func {
+    highlight::func()
+}
+
+#[func]
+fn highlight(
+    text: String,
+    #[named]
+    #[default(false)]
+    block: bool,
+    #[named]
+    #[default(None)]
+    lang: Option<String>,
+) -> String {
+    let rev = text.chars().rev();
+    rev.collect()
+}

src/typst.rs

@@ -5,7 +5,7 @@ use tracing::info;
 use typst::{
     Library, LibraryExt, World,
     diag::{FileError, FileResult, PackageError, PackageResult, SourceDiagnostic},
-    foundations::{Bytes, Datetime, Duration},
+    foundations::{Bytes, Datetime, Dict, Duration, Value},
     syntax::{FileId, RootedPath, Source, VirtualPath, VirtualRoot, package::PackageSpec},
     text::{Font, FontBook},
     utils::LazyHash,
@@ -23,7 +23,15 @@ pub struct TypstContext {
 
 impl TypstContext {
     pub fn new(config: AppConfig) -> Self {
+        let mut inputs = Dict::new();
+        if config.syntax_root.is_some() {
+            inputs.insert(
+                "highlight".into(),
+                Value::Func(crate::syntax::highlight_func()),
+            );
+        }
         let library = Library::builder()
+            .with_inputs(inputs)
             .with_features([typst::Feature::Html].into_iter().collect())
             .build();
         let library = LazyHash::new(library);