string-width - npm Package Compare versions

Comparing version 8.1.1 to 8.2.0

index.js

@@ -10,3 +10,4 @@ import stripAnsi from 'strip-ansi';
 	2. RGI emoji clusters (\p{RGI_Emoji}) are double-width.
-	3. Otherwise use East Asian Width of the cluster’s first visible code point, and add widths for trailing Halfwidth/Fullwidth Forms within the same cluster (e.g., dakuten/handakuten/prolonged sound mark).
+	3. Minimally-qualified/unqualified emoji clusters (ZWJ sequences with 2+ Extended_Pictographic, or keycap sequences) are double-width.
+	4. Otherwise use East Asian Width of the cluster's first visible code point, and add widths for trailing Halfwidth/Fullwidth Forms within the same cluster (e.g., dakuten/handakuten/prolonged sound mark).
 */
@@ -25,2 +26,25 @@
 
+// Detect minimally-qualified/unqualified emoji sequences (missing VS16 but still render as double-width)
+const unqualifiedKeycapRegex = /^[\d#*]\u20E3$/;
+const extendedPictographicRegex = /\p{Extended_Pictographic}/gu;
+
+function isDoubleWidthNonRgiEmojiSequence(segment) {
+	// Real emoji clusters are < 30 chars; guard against pathological input
+	if (segment.length > 50) {
+		return false;
+	}
+
+	if (unqualifiedKeycapRegex.test(segment)) {
+		return true;
+	}
+
+	// ZWJ sequences with 2+ Extended_Pictographic
+	if (segment.includes('\u200D')) {
+		const pictographics = segment.match(extendedPictographicRegex);
+		return pictographics !== null && pictographics.length >= 2;
+	}
+
+	return false;
+}
+
 function baseVisible(segment) {
@@ -59,3 +83,4 @@ return segment.replace(leadingNonPrintingRegex, '');
 
-	if (!countAnsiEscapeCodes) {
+	// Avoid calling stripAnsi when there are no ANSI escape sequences (ESC = 0x1B, CSI = 0x9B)
+	if (!countAnsiEscapeCodes && (string.includes('\u001B') || string.includes('\u009B'))) {
 		string = stripAnsi(string);
@@ -68,2 +93,7 @@ }
 
+	// Fast path: printable ASCII (0x20–0x7E) needs no segmenter, regex, or EAW lookup — width equals length.
+	if (/^[\u0020-\u007E]*$/.test(string)) {
+		return string.length;
+	}
+
 	let width = 0;
@@ -79,3 +109,3 @@ const eastAsianWidthOptions = {ambiguousAsWide: !ambiguousIsNarrow};
 		// Emoji width logic
-		if (rgiEmojiRegex.test(segment)) {
+		if (rgiEmojiRegex.test(segment) || isDoubleWidthNonRgiEmojiSequence(segment)) {
 			width += 2;
@@ -82,0 +112,0 @@ continue;

package.json

 {
 	"name": "string-width",
-	"version": "8.1.1",
+	"version": "8.2.0",
 	"description": "Get the visual width of a string - the number of columns required to display it",
@@ -57,4 +57,4 @@ "license": "MIT",
 	"dependencies": {
-		"get-east-asian-width": "^1.3.0",
-		"strip-ansi": "^7.1.0"
+		"get-east-asian-width": "^1.5.0",
+		"strip-ansi": "^7.1.2"
 	},
@@ -64,4 +64,4 @@ "devDependencies": {
 		"tsd": "^0.33.0",
-		"xo": "^1.2.2"
+		"xo": "^1.2.3"
 	}
 }

readme.md

@@ -5,3 +5,3 @@ # string-width
 
-Some Unicode characters are [fullwidth](https://en.wikipedia.org/wiki/Halfwidth_and_fullwidth_forms) and use double the normal width. [ANSI escape codes](https://en.wikipedia.org/wiki/ANSI_escape_code) are stripped and doesn't affect the width.
+Some Unicode characters are [fullwidth](https://en.wikipedia.org/wiki/Halfwidth_and_fullwidth_forms) and use double the normal width. [ANSI escape codes](https://en.wikipedia.org/wiki/ANSI_escape_code) are stripped and do not affect the width.
 
@@ -8,0 +8,0 @@ Useful to be able to measure the actual width of command-line output.

New alerts

Long strings

Supply chain risk

Contains long string literals, which may be a sign of obfuscated or packed code.

Found 1 instance in 1 package

Improved metrics

Total package byte prevSize

9253

14.96%

Lines of code

123

24.24%

Worsened metrics

Number of low supply chain risk alerts

2

100%

Dependency changes

• Updated

  get-east-asian-width@^1.5.0

• Updated

  strip-ansi@^7.1.2