fix: xml legal characters regex (#12)

qq15725 · Mar 14, 2023 · e6f1e0b · e6f1e0b · vercel · Mar 14, 2023
1 parent dfa9e24
commit e6f1e0b
Show file tree

Hide file tree

Showing 2 changed files with 10 additions and 5 deletions.
diff --git a/src/utils.ts b/src/utils.ts
@@ -96,8 +96,9 @@ export function createSvg(width: number, height: number, ownerDocument?: Documen
 export function svgToDataUrl(svg: SVGElement) {
   const xhtml = new XMLSerializer()
     .serializeToString(svg)
+    // https://www.w3.org/TR/xml/#charsets
     // eslint-disable-next-line no-control-regex
-    .replace(/[\u0000-\u0008\u000B-\u000C\u000E-\u001F]/g, '')
+    .replace(/[\u0000-\u0008\u000B\u000C\u000E-\u001F\uD800-\uDFFF\uFFFE\uFFFF]/ug, '')
   return `data:image/svg+xml;charset=utf-8,${ encodeURIComponent(xhtml) }`
 }
 

diff --git a/test/fixtures/xml.ascii-null-character.html b/test/fixtures/xml.ascii-null-character.html
@@ -7,11 +7,15 @@
 
   ;[
     // \u0000-\u0008
-    '\u0000', '\u0001', '\u0002', '\u0003', '\u0004', '\u0005', '\u0006', '\u0007', '\u0008',
-    // \u000B
-    '\u000B',
+    ...Array.from({ length: 0x0008 }, (_, i) => String.fromCharCode(i)),
+    // \u000B-\u000C
+    '\u000B', '\u000C',
     // \u000E-\u001F
-    '\u000E', '\u000F', '\u0010', '\u0011', '\u0012', '\u0013', '\u0014', '\u0015', '\u0016', '\u0017', '\u0018', '\u0019', '\u001A', '\u001B', '\u001C', '\u001D', '\u001E',
+    ...Array.from({ length: 0x001F - 0x000E }, (_, i) => String.fromCharCode(0x000E + i)),
+    // \uD800-\uDFFF
+    ...Array.from({ length: 0xDFFF - 0xD800 }, (_, i) => String.fromCharCode(0xD800 + i)),
+    // \uFFFE-\uFFFF
+    '\uFFFE', '\uFFFF',
   ].forEach(unicode => {
     root.appendChild(document.createTextNode(unicode))
   })