Improve TextMate syntax performance

RedCMD · Dec 7, 2024 · ca75bcc · ca75bcc
1 parent b2ecc87
commit ca75bcc
Show file tree

Hide file tree

Showing 3 changed files with 56 additions and 19 deletions.
diff --git a/package.json b/package.json
@@ -158,7 +158,7 @@
 		],
 		"configurationDefaults": {
 			"[json-textmate]": {
-				"editor.maxTokenizationLineLength": 500000,
+				"editor.maxTokenizationLineLength": 600000,
 				"editor.suggest.insertMode": "replace",
 				"editor.suggest.preview": true,
 				"editor.quickSuggestions": {

diff --git a/syntaxes/json.tmLanguage.json b/syntaxes/json.tmLanguage.json
@@ -785,15 +785,15 @@
 				{ "match": "\\w+" },
 				{
 					"comment": "An invalid capture group number does not convert, nor becomes null",
-					"match": "(\\$)(0*)(\\d{1,3})(?!\\d)",
+					"match": "(\\$)(0*)(\\d{3,1})(?!\\d)",
 					"captures": {
 						"1": { "name": "meta.embedded constant.character.escape.json.tm markup.italic" },
 						"2": { "name": "punctuation.definition.tag.json.tm" },
 						"3": { "name": "constant.numeric.json.tm" }
 					}
 				},
 				{
-					"match": "(\\$)({)(0*)(\\d{1,3})(:/)(upcase|downcase)(})",
+					"match": "(\\$)({)(0*)(\\d{3,1})(:/)(upcase|downcase)(})",
 					"captures": {
 						"1": { "name": "meta.embedded constant.character.escape.json.tm markup.italic.tm" },
 						"2": { "name": "meta.embedded constant.character.escape.json.tm" },
@@ -1246,7 +1246,7 @@
 					"name": "constant.character.escape.json.tm"
 				},
 				{
-					"match": "(\\\\u)(?>(\\h{4})|([^\"].{,3}))",
+					"match": "(\\\\u)(\\h{4,0})",
 					"captures": {
 						"1": { "name": "constant.character.escape.tm" },
 						"2": { "name": "constant.numeric.json.tm" },
@@ -1423,7 +1423,15 @@
 					]
 				},
 				{
-					"match": "\\b(?>true|false|null)\\b",
+					"match": "\\btrue\\b",
+					"name": "constant.language.json.tm"
+				},
+				{
+					"match": "\\bfalse\\b",
+					"name": "constant.language.json.tm"
+				},
+				{
+					"match": "\\bnull\\b",
 					"name": "constant.language.json.tm"
 				},
 				{

diff --git a/syntaxes/regex.tmLanguage.json b/syntaxes/regex.tmLanguage.json
@@ -424,7 +424,7 @@
 				},
 				{
 					"comment": "(?y{g}-imxWDSP) (?y{w}imxWDSP-x: ...)",
-					"begin": "(?>(\\((?=[^:)]+:))|(\\())(\\?)(?=[-IimxWDSPy])(?=[^x:)]++[:)]|[^-:)]*+-[^x:)]*+x[^:)]*+[:)])",
+					"begin": "(?=\\(\\?[-IimxWDSPy])(?>(\\((?=[^:)]++:))|(\\())(\\?)(?=[^x:)]++[:)]|[^-:)]*+-[^x:)]*+x[^:)]*+[:)])",
 					"end": "(?((?!\\1))\\)|(?=[\")]))(?#Condition is true only when Backreference \\1 from capture group 1 in \"begin\" is *not* empty.)|(?=\")",
 					"beginCaptures": {
 						"1": { "name": "constant.character.set.regexp.tm" },
@@ -451,12 +451,12 @@
 		"capture-group-modify-extended": {
 			"patterns": [
 				{
-					"match": "\\(\\?-+:?\\)",
+					"match": "\\(\\?-++:?+\\)",
 					"name": "punctuation.definition.tag.tm"
 				},
 				{
 					"comment": "(?y{g}imxWDSP-) (?y{w}x-imWDSP:...)",
-					"begin": "(?>(\\((?=[^:)]++:))|(\\())(\\?)(?=[-IimxWDSPy])(?=[^-:)]*+[^x:)]*+[:)])",
+					"begin": "(?=\\(\\?[-IimxWDSPy])(?>(\\((?=[^:)]++:))|(\\())(\\?)(?=[^-:)]*+[^x:)]*+[:)])",
 					"end": "(?((?!\\1))\\)|(?=[\")]))(?#Condition is true only when Backreference \\1 from capture group 1 in \"begin\" is *not* empty.)|(?=\")",
 					"beginCaptures": {
 						"1": { "name": "constant.character.set.regexp.tm" },
@@ -670,7 +670,7 @@
 		},
 		"absent": {
 			"comment": "(?~|...|...) (?~|...) (?~...) (?~|)",
-			"begin": "(\\()(\\?~\\|?)",
+			"begin": "(\\()(\\?~\\|?+)",
 			"end": "\\)|(?=\")",
 			"beginCaptures": {
 				"1": { "name": "constant.character.set.regexp.tm" },
@@ -788,7 +788,7 @@
 				},
 				{
 					"comment": "(*name[tag]{args...})",
-					"match": "(\\()(\\*)([A-Za-z_]\\w*)([^\\[{)]*)(?>(\\[)([^]]*)(]))?([^{)]*)(?>({)([^}]*)(}))?([^)]*)(\\))([+*?]|(?D){(?>\\d+,?\\d*|,\\d+)})?",
+					"match": "(\\()(\\*)([A-Za-z_]\\w*+)([^\\[{)]*+)(?>(\\[)([^]]*+)(]))?+([^{)]*+)(?>({)([^}]*+)(}))?+([^)]*+)(\\))([+*?]|(?D){(?>\\d++,?+\\d*+|,\\d++)})?+",
 					"captures": {
 						"1": { "name": "constant.character.set.regexp.tm markup.italic" },
 						"2": { "name": "support.function.tm" },
@@ -941,7 +941,7 @@
 					}
 				},
 				{
-					"match": "(\\*(-)\\+)(\\\\{2}(?=-))?",
+					"match": "(\\*(-)\\+)(\\\\{2}(?=-))?+",
 					"captures": {
 						"1": { "name": "character-class-range.tm support.class.tm" },
 						"2": { "name": "punctuation.definition.tag.tm strong" },
@@ -1231,7 +1231,8 @@
 			"comment": "\\777 \\xFF \\cZ \\o{0 4777777} \\x{0 13FFFF} \\u0000 \\p{L} \\C-\\M-]",
 			"patterns": [
 				{
-					"match": "(\\\\{2})([Ppox])(\\\\{2})({)",
+					"comment": "The look-ahead increases performance. No clue why :shrug:",
+					"match": "(\\\\{2})([Ppox])(?=\\\\\\\\{)(\\\\{2})({)",
 					"captures": {
 						"1": { "name": "punctuation.definition.tag.tm" },
 						"2": { "name": "entity.name.label.tm strong" },
@@ -1280,21 +1281,21 @@
 					]
 				},
 				{
-					"match": "(\\\\{2})([0-7]{1,3})",
+					"match": "(\\\\{2})([0-7]{3,1})",
 					"captures": {
 						"1": { "name": "punctuation.definition.list.begin.markdown.tm" },
 						"2": { "name": "constant.numeric.tm" }
 					}
 				},
 				{
-					"match": "(\\\\{2}x)(?!{|\")(\\h{,2})",
+					"match": "(\\\\{2}x)(?!{|\")(\\h{2,0})",
 					"captures": {
 						"1": { "name": "punctuation.definition.list.begin.markdown.tm" },
 						"2": { "name": "constant.numeric.tm" }
 					}
 				},
 				{
-					"match": "(?>(\\\\{2}u)|(\\\\u))(?>(\\h{4})|([^\"].{3,}))",
+					"match": "(?>(\\\\{2}u)|(\\\\u))(?>(\\h{4})|([^\"].{3,0}))",
 					"captures": {
 						"1": { "name": "variable.language.tm" },
 						"2": { "name": "constant.character.escape.tm" },
@@ -1311,7 +1312,7 @@
 					}
 				},
 				{
-					"match": "\\\\{2}(?>c|[CM](?>-|(.))?+)(?>(\\\\{3})(\\\\)|\\g<0>|(\\\\{2})?+(\\\\u.{4,}|\\\\?+.))?+",
+					"match": "\\\\{2}(?>c|[CM](?>-|(.))?+)(?>(\\\\{3})(\\\\)|\\g<0>|(\\\\{2})?+(\\\\u.{4,0}|\\\\?+.))?+",
 					"captures": {
 						"0": { "name": "punctuation.definition.list.begin.markdown.tm" },
 						"1": { "name": "invalid.illegal.tm markup.underline regex" },
@@ -1325,8 +1326,35 @@
 					}
 				},
 				{
+					"//": "[pP] degrades performance :shrug:",
+					"comment": "https://github.com/kkos/oniguruma/blob/v6.9.8/doc/UNICODE_PROPERTIES",
+					"begin": "(\\\\{2}p{)(\\^)?+",
+					"end": "}|$|(?=[(){|\"])",
+					"beginCaptures": {
+						"1": { "name": "variable.language.tm" },
+						"2": { "name": "support.function.tm" }
+					},
+					"endCaptures": { "0": { "name": "variable.language.tm" } },
+					"name": "unicode.properties.tm",
+					"patterns": [
+						{
+							"match": "[a-zA-Z1]+",
+							"name": "variable.other.enummember.tm"
+						},
+						{
+							"match": "[ _-]+",
+							"name": "comment.tm"
+						},
+						{
+							"match": "[^\"(){}|a-zA-Z1 _-]+",
+							"name": "invalid.illegal.regex.tm"
+						}
+					]
+				},
+				{
+					"//": "[pP] degrades performance :shrug:",
 					"comment": "https://github.com/kkos/oniguruma/blob/v6.9.8/doc/UNICODE_PROPERTIES",
-					"begin": "(\\\\{2}[pP]{)(\\^)?",
+					"begin": "(\\\\{2}P{)(\\^)?+",
 					"end": "}|$|(?=[(){|\"])",
 					"beginCaptures": {
 						"1": { "name": "variable.language.tm" },
@@ -1396,7 +1424,7 @@
 			"patterns": [
 				{
 					"comment": "\\1",
-					"match": "(\\\\{2})([1-9][0-9]{2,})([0-9]*+)",
+					"match": "(\\\\{2})([1-9][0-9]{2,0})([0-9]*+)",
 					"captures": {
 						"1": { "name": "string.regexp.tm" },
 						"2": { "name": "constant.numeric.tm" },
@@ -1433,7 +1461,8 @@
 					"name": "invalid.illegal.tm markup.underline"
 				},
 				{
-					"match": "^(?>[+*?]|(?D){(?>\\d+,?\\d*|,\\d+)})",
+					"comment": "For some reason it isn't very performant using ^",
+					"match": "\\A(?>[+*?]|(?D){(?>\\d+,?\\d*|,\\d+)})",
 					"name": "invalid.illegal.tm markup.underline"
 				},
 				{