add minimal Zed syntax highlighting for Ferro (#2)

* add minimal Ferro tree-sitter grammar * add temporary tree-sitter generation workflow * generate Ferro tree-sitter parser * drop custom grammar in favor of Rust grammar reuse * remove temporary tree-sitter generation workflow * add Zed extension manifest for Ferro * configure Ferro files for Zed * add Ferro syntax highlighting queries * remove generated custom tree-sitter parser * remove generated custom tree-sitter node types * tighten Ferro Zed highlight queries * add minimal Ferro tree-sitter grammar * point Zed at Ferro grammar * refresh Ferro highlights --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
2026-08-16 18:50:19 +09:00
parent deafd27939
commit 83ce26c209
7 changed files with 2051 additions and 0 deletions
@@ -0,0 +1,102 @@
+// Minimal Tree-sitter grammar for Ferro syntax highlighting in Zed.
+// This intentionally models lexical structure rather than full Ferro semantics.
+
+module.exports = grammar({
+  name: "ferro",
+
+  extras: $ => [
+    /[\s\uFEFF\u2060\u200B]/,
+  ],
+
+  rules: {
+    source_file: $ => repeat(choice(
+      $.line_comment,
+      $.block_comment,
+      $.string_literal,
+      $.char_literal,
+      $.integer_literal,
+      $.builtin,
+      $.builtin_type,
+      $.keyword,
+      $.identifier,
+      $.operator,
+      $.punctuation,
+    )),
+
+    line_comment: _ => token(seq("//", /[^\n]*/)),
+
+    // Ferro block comments may nest. Keeping this rule recursive makes syntax
+    // highlighting follow the compiler lexer instead of flattening nested /* */.
+    block_comment: $ => seq(
+      "/*",
+      repeat(choice(
+        $.block_comment,
+        /[^*/]+/,
+        /\*[^/]/,
+        /\/[^*]/,
+      )),
+      "*/",
+    ),
+
+    string_literal: _ => token(seq(
+      '"',
+      repeat(choice(
+        /[^"\\\n\r]/,
+        /\\(x[0-9A-Fa-f]{2}|u[0-9A-Fa-f]{4}|[nrt\\'"0])/,
+      )),
+      '"',
+    )),
+
+    char_literal: _ => token(seq(
+      "'",
+      choice(
+        /[^'\\\n\r]/,
+        /\\(x[0-9A-Fa-f]{2}|u[0-9A-Fa-f]{4}|[nrt\\'"0])/,
+      ),
+      "'",
+    )),
+
+    integer_literal: _ => token(prec(2, choice(
+      /0[xX][0-9A-Fa-f_]+/,
+      /0[bB][01_]+/,
+      /0[oO][0-7_]+/,
+      /[0-9][0-9_]*/,
+    ))),
+
+    builtin: _ => token(prec(3, /@[A-Za-z_][A-Za-z0-9_]*/)),
+
+    builtin_type: _ => token(prec(3, choice(
+      "i8", "i16", "i32",
+      "u8", "u16", "u32",
+      "usize", "isize",
+      "bool", "char", "void", "str",
+    ))),
+
+    keyword: _ => token(prec(3, choice(
+      "unit", "import", "pub", "fn", "struct", "packed", "enum", "error",
+      "const", "static", "var", "let", "mut",
+      "if", "else", "while", "for", "in", "match",
+      "return", "break", "continue", "defer",
+      "unsafe", "critical", "shared", "atomic", "comptime", "asm",
+      "try", "catch", "orelse", "as", "extern",
+      "interrupt", "interrupt_safe", "far",
+      "true", "false", "null", "undefined",
+      "self", "Self", "type",
+      "and", "or", "not",
+    ))),
+
+    identifier: _ => /[A-Za-z_][A-Za-z0-9_]*/,
+
+    operator: _ => token(choice(
+      "<<=", ">>=",
+      "+=", "-=", "*=", "/=", "%=", "&=", "|=", "^=",
+      "==", "!=", "<=", ">=", "<<", ">>", "->", "=>", "..",
+      "+%", "-%", "*%",
+      "+", "-", "*", "/", "%", "=", "<", ">", "&", "|", "^", "~", "!", "?",
+    )),
+
+    punctuation: _ => token(choice(
+      "(", ")", "{", "}", "[", "]", ",", ";", ":", ".",
+    )),
+  },
+});