Add DataURI encoding table

tdewolff · Jul 17, 2020 · 4ab57ac · 4ab57ac
1 parent f642368
commit 4ab57ac
Showing 1 changed file with 81 additions and 33 deletions.
diff --git a/util.go b/util.go
@@ -339,39 +339,14 @@ func ReplaceMultipleWhitespaceAndEntities(b []byte, entitiesMap map[string][]byt
 	return b[:j]
 }
 
-func DecodeURL(b []byte) []byte {
-	for i := 0; i < len(b); i++ {
-		if b[i] == '%' && i+2 < len(b) {
-			j := i + 1
-			c := 0
-			for ; j < i+3 && (b[j] >= '0' && b[j] <= '9' || b[j] >= 'a' && b[j] <= 'z' || b[j] >= 'A' && b[j] <= 'Z'); j++ {
-				if b[j] <= '9' {
-					c = c<<4 + int(b[j]-'0')
-				} else if b[j] <= 'F' {
-					c = c<<4 + int(b[j]-'A') + 10
-				} else if b[j] <= 'f' {
-					c = c<<4 + int(b[j]-'a') + 10
-				}
-			}
-			if j == i+3 && c < 128 {
-				b[i] = byte(c)
-				b = append(b[:i+1], b[i+3:]...)
-			}
-		} else if b[i] == '+' {
-			b[i] = ' '
-		}
-	}
-	return b
-}
-
 var URLEncodingTable = [256]bool{
 	// ASCII
 	true, true, true, true, true, true, true, true,
 	true, true, true, true, true, true, true, true,
 	true, true, true, true, true, true, true, true,
 	true, true, true, true, true, true, true, true,
 
-	false, false, true, true, true, true, true, false, // space, !, '
+	true, false, true, true, true, true, true, false, // !, '
 	false, false, false, true, true, false, false, true, // (, ), *, -, .
 	false, false, false, false, false, false, false, false, // 0, 1, 2, 3, 4, 5, 6, 7
 	false, false, true, true, true, true, true, true, // 8, 9
@@ -408,17 +383,90 @@ var URLEncodingTable = [256]bool{
 	true, true, true, true, true, true, true, true,
 }
 
+// URL encoding for Data URIs, escape only non-printable characters, unicode and %, #, &
+// IE11 additionally requires encoding of \, [, ], ", <, >, `, {, }, |, ^ which is not required by Chrome, Firefox, Opera, Edge, Safari, Yandex
+var DataURIEncodingTable = [256]bool{
+	// ASCII
+	true, true, true, true, true, true, true, true,
+	true, false, false, true, true, false, true, true, // all except \t, \n, \r
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+
+	false, false, true, true, false, true, true, false, // ", #, %, &
+	false, false, false, false, false, false, false, false,
+	false, false, false, false, false, false, false, false,
+	false, false, false, false, true, false, true, false, // <, >
+
+	false, false, false, false, false, false, false, false,
+	false, false, false, false, false, false, false, false,
+	false, false, false, false, false, false, false, false,
+	false, false, false, true, true, true, true, false, // [, \, ], ^
+
+	true, false, false, false, false, false, false, false, // `
+	false, false, false, false, false, false, false, false,
+	false, false, false, false, false, false, false, false,
+	false, false, false, true, true, true, false, true, // {, |, }, DEL
+
+	// non-ASCII
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+	true, true, true, true, true, true, true, true,
+}
+
 func EncodeURL(b []byte, table [256]bool) []byte {
 	for i := 0; i < len(b); i++ {
 		c := b[i]
 		if table[c] {
-			b = append(b, 0, 0)
-			copy(b[i+3:], b[i+1:])
-			b[i+0] = '%'
-			b[i+1] = "0123456789ABCDEF"[c>>4]
-			b[i+2] = "0123456789ABCDEF"[c&15]
-		} else if c == ' ' {
-			b[i] = '+'
+			if c == ' ' {
+				b[i] = '+'
+			} else {
+				b = append(b, 0, 0)
+				copy(b[i+3:], b[i+1:])
+				b[i+0] = '%'
+				b[i+1] = "0123456789ABCDEF"[c>>4]
+				b[i+2] = "0123456789ABCDEF"[c&15]
+			}
+		}
+	}
+	return b
+}
+
+func DecodeURL(b []byte) []byte {
+	for i := 0; i < len(b); i++ {
+		if b[i] == '%' && i+2 < len(b) {
+			j := i + 1
+			c := 0
+			for ; j < i+3 && (b[j] >= '0' && b[j] <= '9' || b[j] >= 'a' && b[j] <= 'z' || b[j] >= 'A' && b[j] <= 'Z'); j++ {
+				if b[j] <= '9' {
+					c = c<<4 + int(b[j]-'0')
+				} else if b[j] <= 'F' {
+					c = c<<4 + int(b[j]-'A') + 10
+				} else if b[j] <= 'f' {
+					c = c<<4 + int(b[j]-'a') + 10
+				}
+			}
+			if j == i+3 && c < 128 {
+				b[i] = byte(c)
+				b = append(b[:i+1], b[i+3:]...)
+			}
+		} else if b[i] == '+' {
+			b[i] = ' '
 		}
 	}
 	return b