Fix ISO-2022-JP encoder, and incorporate encoder attack changes from …

…spec
inexorabletash · Feb 12, 2016 · aaa9e9f · aaa9e9f
1 parent 5716244
commit aaa9e9f
Show file tree

Hide file tree

Showing 2 changed files with 38 additions and 13 deletions.
diff --git a/lib/encoding.js b/lib/encoding.js
@@ -2601,6 +2601,7 @@ if (typeof module !== "undefined" && module.exports) {
       if (code_point === end_of_stream &&
           iso2022jp_state !== states.ASCII) {
         stream.prepend(code_point);
+        iso2022jp_state = states.ASCII;
         return [0x1B, 0x28, 0x42];
       }
 
@@ -2609,19 +2610,28 @@ if (typeof module !== "undefined" && module.exports) {
       if (code_point === end_of_stream && iso2022jp_state === states.ASCII)
         return finished;
 
-      // 3. If iso-2022-jp encoder state is ASCII and code point is an
+      // 3. If ISO-2022-JP encoder state is ASCII or Roman, and code
+      // point is U+000E, U+000F, or U+001B, return error with U+FFFD.
+      if ((iso2022jp_state === states.ASCII ||
+           iso2022jp_state === states.Roman) &&
+          (code_point === 0x000E || code_point === 0x000F ||
+           code_point === 0x001B)) {
+        return encoderError(0xFFFD);
+      }
+
+      // 4. If iso-2022-jp encoder state is ASCII and code point is an
       // ASCII code point, return a byte whose value is code point.
       if (iso2022jp_state === states.ASCII &&
           isASCIICodePoint(code_point))
         return code_point;
 
-      // 4. If iso-2022-jp encoder state is Roman and code point is an
+      // 5. If iso-2022-jp encoder state is Roman and code point is an
       // ASCII code point, excluding U+005C and U+007E, or is U+00A5
       // or U+203E, run these substeps:
       if (iso2022jp_state === states.Roman &&
-          (isASCIICodePoint(code_point) &&
+          ((isASCIICodePoint(code_point) &&
            code_point !== 0x005C && code_point !== 0x007E) ||
-          (code_point == 0x00A5 || code_point == 0x203E)) {
+          (code_point == 0x00A5 || code_point == 0x203E))) {
 
         // 1. If code point is an ASCII code point, return a byte
         // whose value is code point.
@@ -2637,7 +2647,7 @@ if (typeof module !== "undefined" && module.exports) {
           return 0x7E;
       }
 
-      // 5. If code point is an ASCII code point, and iso-2022-jp
+      // 6. If code point is an ASCII code point, and iso-2022-jp
       // encoder state is not ASCII, prepend code point to stream, set
       // iso-2022-jp encoder state to ASCII, and return three bytes
       // 0x1B 0x28 0x42.
@@ -2648,7 +2658,7 @@ if (typeof module !== "undefined" && module.exports) {
         return [0x1B, 0x28, 0x42];
       }
 
-      // 6. If code point is either U+00A5 or U+203E, and iso-2022-jp
+      // 7. If code point is either U+00A5 or U+203E, and iso-2022-jp
       // encoder state is not Roman, prepend code point to stream, set
       // iso-2022-jp encoder state to Roman, and return three bytes
       // 0x1B 0x28 0x4A.
@@ -2659,19 +2669,19 @@ if (typeof module !== "undefined" && module.exports) {
         return [0x1B, 0x28, 0x4A];
       }
 
-      // 7. If code point is U+2212, set it to U+FF0D.
+      // 8. If code point is U+2212, set it to U+FF0D.
       if (code_point === 0x2212)
         code_point = 0xFF0D;
 
-      // 8. Let pointer be the index pointer for code point in index
+      // 9. Let pointer be the index pointer for code point in index
       // jis0208.
       var pointer = indexPointerFor(code_point, index('jis0208'));
 
-      // 9. If pointer is null, return error with code point.
+      // 10. If pointer is null, return error with code point.
       if (pointer === null)
         return encoderError(code_point);
 
-      // 10. If iso-2022-jp encoder state is not jis0208, prepend code
+      // 11. If iso-2022-jp encoder state is not jis0208, prepend code
       // point to stream, set iso-2022-jp encoder state to jis0208,
       // and return three bytes 0x1B 0x24 0x42.
       if (iso2022jp_state !== states.jis0208) {
@@ -2680,13 +2690,13 @@ if (typeof module !== "undefined" && module.exports) {
         return [0x1B, 0x24, 0x42];
       }
 
-      // 11. Let lead be floor(pointer / 94) + 0x21.
+      // 12. Let lead be floor(pointer / 94) + 0x21.
       var lead = floor(pointer / 94) + 0x21;
 
-      // 12. Let trail be pointer % 94 + 0x21.
+      // 13. Let trail be pointer % 94 + 0x21.
       var trail = pointer % 94 + 0x21;
 
-      // 13. Return two bytes whose values are lead and trail.
+      // 14. Return two bytes whose values are lead and trail.
       return [lead, trail];
     };
   }

diff --git a/test/test-misc.js b/test/test-misc.js
@@ -342,3 +342,18 @@ test(function() {
         .encode('\uE5E5');
     });
 }, 'NONSTANDARD - gb18030: U+E5E5 (encoding)');
+
+
+test(function() {
+  // Regression test for https://github.com/whatwg/encoding/issues/15
+  var encoder =
+      new TextEncoder('iso-2022-jp', {NONSTANDARD_allowLegacyEncoding: true});
+
+  [
+    //'\u000E', '\u000F', '\u001B',
+    '\u00A5\u000E', //'\u00A5\u000F',  '\u00A5\u001B'
+  ].forEach(function(s) {
+    assert_throws(new TypeError, function() { encoder.encode(s); });
+  });
+
+}, 'NONSTANDARD - iso-2022-jp encoding attack (encoding)');