Add support for Hack

CHANGELOG.md

@@ -5,6 +5,8 @@
 Fixed an issue with changes being ignored in OCaml's `{||}` string
 literals.
 
+Added support for Hack programs.
+
 ### Display
 
 Fixed an issue where larger additions were not lined up with removals.

README.md

@@ -65,6 +65,7 @@ Difftastic supports the following languages:
 * Elixir
 * Emacs Lisp
 * Go
+* Hack
 * Haskell
 * Java
 * JavaScript (and JSX)

build.rs

@@ -104,6 +104,11 @@ fn main() {
             src_dir: "vendor/tree-sitter-go-src",
             extra_files: vec![],
         },
+        TreeSitterParser {
+            name: "tree-sitter-hack",
+            src_dir: "vendor/tree-sitter-hack-src",
+            extra_files: vec!["scanner.cc"],
+        },
         TreeSitterParser {
             name: "tree-sitter-haskell",
             src_dir: "vendor/tree-sitter-haskell-src",

manual/src/introduction.md

@@ -13,6 +13,7 @@ tool that understands syntax. It supports the following languages:
 * Elixir
 * Emacs Lisp
 * Go
+* Hack
 * Haskell
 * Java
 * JavaScript (and JSX)

manual/src/upstream_parsers.md

@@ -14,6 +14,7 @@ Difftastic uses the following tree-sitter parsers:
 | Elixir          | [elixir-lang/tree-sitter-elixir](https://github.com/elixir-lang/tree-sitter-elixir)                                                     |
 | Emacs Lisp      | [wilfred/tree-sitter-elisp](https://github.com/Wilfred/tree-sitter-elisp)                                                               |
 | Go              | [tree-sitter/tree-sitter-go](https://github.com/tree-sitter/tree-sitter-go)                                                             |
+| Hack            | [slackhq/tree-sitter-hack](https://github.com/slackhq/tree-sitter-hack)                                                                 |
 | Haskell         | [tree-sitter/tree-sitter-haskell](https://github.com/tree-sitter/tree-sitter-haskell)                                                   |
 | Java            | [tree-sitter/tree-sitter-java](https://github.com/tree-sitter/tree-sitter-java)                                                         |
 | JavaScript, JSX | [tree-sitter/tree-sitter-javascript](https://github.com/tree-sitter/tree-sitter-javascript)                                             |

sample_files/hack_after.php

@@ -0,0 +1,5 @@
+<?hh
+
+function foo(): vec<?int> {
+  return vec[1, null];
+}

sample_files/hack_before.php

@@ -0,0 +1,5 @@
+<?hh
+
+function foo(): vec<int> {
+  return vec[1];
+}

src/guess_language.rs

@@ -28,6 +28,7 @@ pub enum Language {
     Elixir,
     EmacsLisp,
     Go,
+    Hack,
     Haskell,
     Java,
     JavaScript,
@@ -76,6 +77,7 @@ fn from_shebang(src: &str) -> Option<Language> {
                     "lisp" | "sbc" | "ccl" | "clisp" | "ecl" => return Some(CommonLisp),
                     "elixir" => return Some(Elixir),
                     "runghc" | "runhaskell" | "runhugs" => return Some(Haskell),
+                    "hhvm" => return Some(Hack),
                     "chakra" | "d8" | "gjs" | "js" | "node" | "nodejs" | "qjs" | "rhino" | "v8"
                     | "v8-shell" => return Some(JavaScript),
                     "ocaml" | "ocamlrun" | "ocamlscript" => return Some(OCaml),
@@ -85,7 +87,12 @@ fn from_shebang(src: &str) -> Option<Language> {
                     _ => {}
                 }
             }
-        };
+        }
+
+        // Hack can use <?hh in files with a .php extension.
+        if first_line.starts_with("<?hh") {
+            return Some(Hack);
+        }
     }
 
     None
@@ -130,6 +137,7 @@ fn from_extension(extension: &OsStr) -> Option<Language> {
         "el" => Some(EmacsLisp),
         "ex" | "exs" => Some(Elixir),
         "go" => Some(Go),
+        "hack" | "hck" => Some(Hack),
         "hs" => Some(Haskell),
         "java" => Some(Java),
         "cjs" | "js" | "mjs" => Some(JavaScript),

src/tree_sitter_parser.rs

@@ -53,6 +53,7 @@ extern "C" {
     fn tree_sitter_elisp() -> ts::Language;
     fn tree_sitter_elixir() -> ts::Language;
     fn tree_sitter_go() -> ts::Language;
+    fn tree_sitter_hack() -> ts::Language;
     fn tree_sitter_haskell() -> ts::Language;
     fn tree_sitter_java() -> ts::Language;
     fn tree_sitter_javascript() -> ts::Language;
@@ -239,6 +240,16 @@ pub fn from_language(language: guess::Language) -> TreeSitterConfig {
                 .unwrap(),
             }
         }
+        Hack => {
+            let language = unsafe { tree_sitter_hack() };
+            TreeSitterConfig {
+                name: "Hack",
+                language: unsafe { tree_sitter_hack() },
+                atom_nodes: (vec![]).into_iter().collect(),
+                delimiter_tokens: (vec![("[", "]"), ("(", ")"), ("{", "}")]),
+                highlight_query: ts::Query::new(language, "").unwrap(),
+            }
+        }
         Haskell => {
             let language = unsafe { tree_sitter_haskell() };
             TreeSitterConfig {

vendor/tree-sitter-hack-src

@@ -0,0 +1 @@
+tree-sitter-hack/src