Merge pull request microsoft#100482 from IllusionMH/search-step-over-surrogates-100134

alexdima · web-flow · commit 60da9d867ac3 · 2020-06-22T22:50:17.000+02:00
Step over surrogate pairs on zero-lenth matches (fixes microsoft#100134)
diff --git a/src/vs/editor/common/model/textModelSearch.ts b/src/vs/editor/common/model/textModelSearch.ts
@@ -548,8 +548,12 @@ export class Searcher {
 			if (matchStartIndex === this._prevMatchStartIndex && matchLength === this._prevMatchLength) {
 				if (matchLength === 0) {
 					// the search result is an empty string and won't advance `regex.lastIndex`, so `regex.exec` will stuck here
-					// we attempt to recover from that by advancing by one
-					this._searchRegex.lastIndex += 1;
+					// we attempt to recover from that by advancing by two if surrogate pair found and by one otherwise
+					if (strings.getNextCodePoint(text, textLength, this._searchRegex.lastIndex) > 0xFFFF) {
+						this._searchRegex.lastIndex += 2;
+					} else {
+						this._searchRegex.lastIndex += 1;
+					}
 					continue;
 				}
 				// Exit early if the regex matches the same range twice
diff --git a/src/vs/editor/test/common/model/textModelSearch.test.ts b/src/vs/editor/test/common/model/textModelSearch.test.ts
@@ -781,4 +781,29 @@ suite('TextModelSearch', () => {
 
 		model.dispose();
 	});
+
+	test('issue #100134. Zero-length matches should properly step over surrogate pairs', () => {
+		// 1[Laptop]1 - there shoud be no matches inside of [Laptop] emoji
+		assertFindMatches('1\uD83D\uDCBB1', '()', true, false, null,
+			[
+				[1, 1, 1, 1],
+				[1, 2, 1, 2],
+				[1, 4, 1, 4],
+				[1, 5, 1, 5],
+
+			]
+		);
+		// 1[Hacker Cat]1 = 1[Cat Face][ZWJ][Laptop]1 - there shoud be matches between emoji and ZWJ
+		// there shoud be no matches inside of [Cat Face] and [Laptop] emoji
+		assertFindMatches('1\uD83D\uDC31\u200D\uD83D\uDCBB1', '()', true, false, null,
+			[
+				[1, 1, 1, 1],
+				[1, 2, 1, 2],
+				[1, 4, 1, 4],
+				[1, 5, 1, 5],
+				[1, 7, 1, 7],
+				[1, 8, 1, 8]
+			]
+		);
+	});
 });
