Completely overhaul the grammar

Mango0x45 · Sep 14, 2024 · 642d526 · 642d526
1 parent 291cb06
commit 642d526
Show file tree

Hide file tree

Showing 6 changed files with 1,037 additions and 1,486 deletions.
diff --git a/.exrc b/.exrc
@@ -0,0 +1 @@
+let g:editorconfig = v:false
diff --git a/example b/example
@@ -0,0 +1,38 @@
+html lang="en" {
+	head {
+		meta charset="UTF-8" {}
+		meta
+			name="viewport"
+			content="width=device-width, initial-scale=1.0"
+		{}
+		link href="/favicon.svg" rel="shortcut icon" type="image/svg" {}
+		link href="/style.svg" rel="stylesheet" {}
+		title {-My Website Title}
+	}
+
+	body {
+		p #my-id  {= This is a paragraph with the id ‘my-id’     }
+		p .my-cls {= This is a paragraph with the class ‘my-cls’ }
+
+		/ div {
+			p {- This entire div is commented out. }
+			p {- Isn’t that neat? }
+		}
+
+		p
+			#some-id
+			.class-1
+			.class-2
+			key-1="value-1"
+			key-2 = "value-2"
+		{-
+			This paragraph has an ID, two classes, and two additional
+			attributes.  GSP allows us to use the ‘#ident’ and ‘.ident’
+			syntaxes as shorthands for applying IDs, and classes.  This
+			is a text node, so nothing is being interpreted as GSP nodes,
+			but we can include them inline if we want.  As an example,
+			here is some @em {-emphatic} text.  Your inline nodes can
+			also have attributes @em #id {-just like a regular node}.
+		}
+	}
+}
diff --git a/grammar.js b/grammar.js
@@ -1,60 +1,38 @@
+const IDENT = /\p{XID_Start}[-\p{XID_Continue}]*/u
+
 module.exports = grammar({
 	name: 'gsp',
-	extras: $ => [$._S],
+	extras: $ => [/\p{Pattern_White_Space}+/u],
 	rules: {
-		source_file: $ => repeat(choice($.comment, $.node)),
+		document: $ => repeat($._toplevel),
+		_toplevel: $ => choice($.comment, $.node),
 
-		comment: $ => seq(
-      '/',
-			optional($.node_name),
-			optional($.attribute_list),
-			'{',
-			optional($.node_body),
-			'}',
-		),
+		comment: $ => seq('/', $.node),
+		node: $ => seq($.node_name, optional($.attrs), $.node_body),
 
-		node: $ => seq(
-			optional('>'),
-			$.node_name,
-			optional($.attribute_list),
+		node_name: $ => IDENT,
+		node_body: $ => seq(
 			'{',
-			optional($.node_body),
-			'}',
-		),
-
-		node_body: $ => choice(
-			repeat1(choice($.comment, $.node)),
-			seq(choice('-', '='), optional($.text)),
-		),
-
-		node_name: $ => /[a-zA-Z:_\u{000C0}-\u{000D6}\u{000D8}-\u{000F6}\u{000F8}-\u{002FF}\u{00370}-\u{0037D}\u{0037F}-\u{01FFF}\u{0200C}-\u{0200D}\u{02070}-\u{0218F}\u{02C00}-\u{02FEF}\u{03001}-\u{0D7FF}\u{0F900}-\u{0FDCF}\u{0FDF0}-\u{0FFFD}\u{10000}-\u{EFFFF}][a-zA-Z0-9:_\-.·\u{00300}-\u{0036F}\u{0203F}-\u{02040}\u{000C0}-\u{000D6}\u{000D8}-\u{000F6}\u{000F8}-\u{002FF}\u{00370}-\u{0037D}\u{0037F}-\u{01FFF}\u{0200C}-\u{0200D}\u{02070}-\u{0218F}\u{02C00}-\u{02FEF}\u{03001}-\u{0D7FF}\u{0F900}-\u{0FDCF}\u{0FDF0}-\u{0FFFD}\u{10000}-\u{EFFFF}]*/u,
-
-		text: $ => repeat1(
 			choice(
-				$.literal_text,
-				seq('@', choice($.comment, $.node)),
-			),
-		),
-
-		literal_text: $ => /(\\[@}\\]|[^@}\\])+/,
-
-		attribute_list: $ => repeat1($.attribute),
-
-		attribute: $ => choice(
-			$.class_shorthand,
-			$.id_shorthand,
-			seq(
-				$.attribute_name,
-				optional(seq('=', $.attribute_value)),
+				repeat($._toplevel),
+				seq(choice('-', '='), optional($.text)),
 			),
+			'}',
 		),
 
-		class_shorthand: $ => /\.\P{White_Space}+/u,
-		id_shorthand: $ => /#\P{White_Space}+/u,
+		attrs: $ => repeat1(choice(
+			$.attr,
+			$.id_attr,
+			$.class_attr,
+		)),
 
-		attribute_name: $ => /[a-zA-Z0-9_-]+/,
-		attribute_value: $ => /"(\\.|[^"\\])*"/,
+		attr: $ => seq(IDENT, '=', /"(\\["\\]|[^"\\])+"/),
+		id_attr: $ => seq('#', token.immediate(IDENT)),
+		class_attr: $ => seq('.', token.immediate(IDENT)),
 
-		_S: $ => /\p{White_Space}+/u,
+		text: $ => repeat1(choice(
+			/(\\[@}\\]|[^@}\\])+/,
+			seq('@', $._toplevel),
+		)),
 	},
 })