turtle.tmLanguage.JSON

// The Turtle syntax is defined here:
//   http://www.w3.org/TR/2012/WD-turtle-20120710/
//
// It looks very much like N-Triples with extra literal forms taken
// from SPARQL, so rather than derive this language file from the
// official grammar, I've copied in patterns from sparql.tmLanguage.
// The only exceptions are for things noted explictly here:
//   http://www.w3.org/TR/2012/WD-turtle-20120710/#sec-diff-sparql
{
	"name": "Turtle",
	"uuid": "12fe59d2-8594-4f33-a77c-e05c4d759325",
 	"scopeName": "source.turtle",
 	"fileTypes": ["ttl"],
 	"patterns": [
 		{
			"name": 	"keyword.turtle",
  			"comment": 	"Keywords",
 			"match": 	"(?:^|\\s+)(?:@base|@prefix)\\b",
		},
 		{
 			"name": 	"string.uri.turtle",
 			"comment": 	"URI",
 			// Best attempt to convert IRI_REF:
 			// http://www.w3.org/TR/rdf-sparql-query/#rIRI_REF
 			"match": 	"<[^<>\"{}|^`\\]\\\\]*>",
 		},
 		{
			"name": 	"constant.complex.blank-node.turtle",
			"comment": 	"Blank node",
 			// Don't bother trying to match the blank node label grammar exactly,
 			// just highlight anything up to a space
			"match": 	"(_:)([^\\s]+)",
	        "captures": {
	            "1": { "name": "constant.language.turtle" },
	            "2": { "name": "entity.name.class.turtle" }
	         }
		},
 		{
			"name": 	"constant.complex.turtle",
			"comment": 	"Prefix / prefixed URI",
			// PN_CHARS_BASE is insane
			// http://www.w3.org/TR/rdf-sparql-query/#rPN_CHARS_BASE
			"match": 	"(\\w*:)([^\\s|/^*?+{}()]*)",
	        "captures": {
	            "1": { "name": "constant.other.turtle" },
	            "2": { "name": "entity.name.class.turtle" }
	         }
		},
		{
			"name":		"entity.name.class.rdfs-type.turtle",
			"comment":	"The special triple predicate 'a'",
			"match":    "\\sa\\s"
		},
		{
			"name": 	"literal.turtle",
			"comment": 	"Typed literal",
			// Note duplication of IRI_REF
			"match":	"(\"[^\"]*\")(\\^\\^)(<[^<>\"{}|^`\\]\\\\]*>|\\w*:[^\\s)]+)",
			"captures":   {
				"1": { "name": "string.turtle" },
				"2": { "name": "keyword.operator.turtle" },
				"3": { "name": "support.type.turtle" }
			}
		},
		{
			"name": 	"string.turtle",
			"comment": 	"String literal",
			// This is an attempt to implement the following BNF combinations:
			// STRING_LITERAL_LONG1 | STRING_LITERAL_LONG2 | STRING_LITERAL1 | STRING_LITERAL2
			// Creating an incorrectly highlighted example of a triple-quoted string
			// turned out to be really hard, so I can't promise the first two options
			// are correct!
			//
			// I didn't bother with multi-line strings - I think you have to use multi-line
			// begin...end captures to do this in Sublime, which is quite a big change from
			// the regex I wrote for the SPARQL syntax.
			"match":	"('''(?:(?:'|'')?(?:[^'\\\\]|\\\\[tbnrf\\\"']))*'''|\"\"\"(?:(?:\"|\"\")?([^\"\\\\]|\\\\[tbnrf\\\"']))*\"\"\"|'(?:(?:[^\\x22\\x5C\\xA\\xD])|\\\\[tbnrf\\\"'])*'|\"(?:(?:[^\\x22\\x5C\\xA\\xD])|\\\\[tbnrf\\\"'])*\"|'''.*'''|'[^']*')(@[a-zA-Z-]+)?",
			"captures":   {
				"1": { "name": "string.turtle" },
				"2": { "name": "keyword.operator.turtle" },
				"3": { "name": "support.type.turtle" }
			}
		},
		{
			"name":		"comment.line.number-sign.turtle",
			"comment":	"Comments",
			"match":	"#.*$"
		},
		{
			"name":		"constant.numeric.turtle",
			"comment":  "Numeric literal",
			"match": 	"\\b[+-]?(?:\\d+|[0-9]+\\.[0-9]*|\\.[0-9]+(?:[eE][+-]?\\d+)?)\\b"
		},
		{
			"name":		"constant.language.turtle",
			"comment":  "Boolean",
			"match":	"\\b(?:true|false)\\b"
		}
 	]
}