Fix ecma_builtin_global_object_decode_uri_helper.

JerryScript-DCO-1.0-Signed-off-by: Ruben Ayrapetyan r.ayrapetyan@samsung.com

jerry-core/ecma/builtin-objects/ecma-builtin-global.cpp

@@ -857,7 +857,12 @@ ecma_builtin_global_object_decode_uri_helper (ecma_value_t uri __attr_unused___,
 
       lit_code_point_t decoded_byte;
 
-      lit_read_code_point_from_hex (input_char_p + 1, 2, &decoded_byte);
+      if (!lit_read_code_point_from_hex (input_char_p + 1, 2, &decoded_byte))
+      {
+        ret_value = ecma_make_throw_obj_completion_value (ecma_new_standard_error (ECMA_ERROR_URI));
+        break;
+      }
+
       input_char_p += URI_ENCODED_BYTE_SIZE;
 
       if (decoded_byte <= LIT_UTF8_1_BYTE_CODE_POINT_MAX)
@@ -910,9 +915,9 @@ ecma_builtin_global_object_decode_uri_helper (ecma_value_t uri __attr_unused___,
           else
           {
             lit_code_point_t cp;
-            lit_read_code_point_from_hex (input_char_p + 1, 2, &cp);
 
-            if ((cp & LIT_UTF8_EXTRA_BYTE_MASK) != LIT_UTF8_EXTRA_BYTE_MARKER)
+            if (!lit_read_code_point_from_hex (input_char_p + 1, 2, &cp)
+                || ((cp & LIT_UTF8_EXTRA_BYTE_MASK) != LIT_UTF8_EXTRA_BYTE_MARKER))
             {
               is_valid = false;
               break;
@@ -923,7 +928,8 @@ ecma_builtin_global_object_decode_uri_helper (ecma_value_t uri __attr_unused___,
           }
         }
 
-        if (!is_valid)
+        if (!is_valid
+            || !lit_is_utf8_string_valid (octets, bytes_count))
         {
           ret_value = ecma_make_throw_obj_completion_value (ecma_new_standard_error (ECMA_ERROR_URI));
           break;
@@ -932,12 +938,8 @@ ecma_builtin_global_object_decode_uri_helper (ecma_value_t uri __attr_unused___,
         lit_code_point_t cp;
         lit_read_code_point_from_utf8 (octets, bytes_count, &cp);
 
-        if ((bytes_count == 2 && cp <= LIT_UTF8_1_BYTE_CODE_POINT_MAX)
+        if (lit_is_code_point_utf16_high_surrogate (cp)
-            || (bytes_count == 3 && cp <= LIT_UTF8_2_BYTE_CODE_POINT_MAX)
+            || lit_is_code_point_utf16_low_surrogate (cp))
-            || (bytes_count == 4 && cp <= LIT_UTF8_3_BYTE_CODE_POINT_MAX)
-            || lit_is_code_unit_high_surrogate ((ecma_char_t) cp)
-            || lit_is_code_unit_low_surrogate ((ecma_char_t) cp)
-            || cp > LIT_UNICODE_CODE_POINT_MAX)
         {
           ret_value = ecma_make_throw_obj_completion_value (ecma_new_standard_error (ECMA_ERROR_URI));
           break;
@@ -1067,7 +1069,7 @@ ecma_builtin_global_object_encode_uri_helper (ecma_value_t uri, /**< uri argumen
     /* Input validation, we need to reject stray surrogates. */
     input_char_p += lit_read_code_unit_from_utf8 (input_char_p, &ch);
 
-    if (lit_is_code_unit_low_surrogate (ch))
+    if (lit_is_code_point_utf16_low_surrogate (ch))
     {
       ret_value = ecma_make_throw_obj_completion_value (ecma_new_standard_error (ECMA_ERROR_URI));
       break;
@@ -1075,7 +1077,7 @@ ecma_builtin_global_object_encode_uri_helper (ecma_value_t uri, /**< uri argumen
 
     cp = ch;
 
-    if (lit_is_code_unit_high_surrogate (ch))
+    if (lit_is_code_point_utf16_high_surrogate (ch))
     {
       if (input_char_p == input_end_p)
       {
@@ -1086,7 +1088,7 @@ ecma_builtin_global_object_encode_uri_helper (ecma_value_t uri, /**< uri argumen
       ecma_char_t next_ch;
       lit_utf8_size_t read_size = lit_read_code_unit_from_utf8 (input_char_p, &next_ch);
 
-      if (lit_is_code_unit_low_surrogate (next_ch))
+      if (lit_is_code_point_utf16_low_surrogate (next_ch))
       {
         cp = lit_convert_surrogate_pair_to_code_point (ch, next_ch);
         input_char_p += read_size;
@@ -1132,12 +1134,12 @@ ecma_builtin_global_object_encode_uri_helper (ecma_value_t uri, /**< uri argumen
       input_char_p += lit_read_code_unit_from_utf8 (input_char_p, &ch);
       cp = ch;
 
-      if (lit_is_code_unit_high_surrogate (ch))
+      if (lit_is_code_point_utf16_high_surrogate (ch))
       {
         ecma_char_t next_ch;
         lit_utf8_size_t read_size = lit_read_code_unit_from_utf8 (input_char_p, &next_ch);
 
-        if (lit_is_code_unit_low_surrogate (next_ch))
+        if (lit_is_code_point_utf16_low_surrogate (next_ch))
         {
           cp = lit_convert_surrogate_pair_to_code_point (ch, next_ch);
           input_char_p += read_size;

jerry-core/lit/lit-strings.cpp

@@ -223,26 +223,26 @@ lit_is_cesu8_string_valid (const lit_utf8_byte_t *utf8_buf_p, /**< utf-8 string
 } /* lit_is_cesu8_string_valid */
 
 /**
- * Check if the code unit type is low surrogate
+ * Check if the code point is UTF-16 low surrogate
  *
  * @return true / false
  */
 bool
-lit_is_code_unit_low_surrogate (ecma_char_t code_unit) /**< code unit */
+lit_is_code_point_utf16_low_surrogate (lit_code_point_t code_point) /**< code point */
 {
-  return LIT_UTF16_LOW_SURROGATE_MIN <= code_unit && code_unit <= LIT_UTF16_LOW_SURROGATE_MAX;
+  return LIT_UTF16_LOW_SURROGATE_MIN <= code_point && code_point <= LIT_UTF16_LOW_SURROGATE_MAX;
-} /* lit_is_code_unit_low_surrogate */
+} /* lit_is_code_point_utf16_low_surrogate */
 
 /**
- * Check if the code unit type is high surrogate
+ * Check if the code point is UTF-16 high surrogate
  *
  * @return true / false
  */
 bool
-lit_is_code_unit_high_surrogate (ecma_char_t code_unit) /**< code unit */
+lit_is_code_point_utf16_high_surrogate (lit_code_point_t code_point) /**< code point */
 {
-  return LIT_UTF16_HIGH_SURROGATE_MIN <= code_unit && code_unit <= LIT_UTF16_HIGH_SURROGATE_MAX;
+  return LIT_UTF16_HIGH_SURROGATE_MIN <= code_point && code_point <= LIT_UTF16_HIGH_SURROGATE_MAX;
-} /* lit_is_code_unit_high_surrogate */
+} /* lit_is_code_point_utf16_high_surrogate */
 
 /**
  * Initialize iterator for traversing utf-8 string as a string of code units
@@ -925,8 +925,8 @@ lit_code_point_t
 lit_convert_surrogate_pair_to_code_point (ecma_char_t high_surrogate, /**< high surrogate code point */
                                           ecma_char_t low_surrogate) /**< low surrogate code point */
 {
-  JERRY_ASSERT (lit_is_code_unit_high_surrogate (high_surrogate));
+  JERRY_ASSERT (lit_is_code_point_utf16_high_surrogate (high_surrogate));
-  JERRY_ASSERT (lit_is_code_unit_low_surrogate (low_surrogate));
+  JERRY_ASSERT (lit_is_code_point_utf16_low_surrogate (low_surrogate));
 
   lit_code_point_t code_point;
   code_point = (uint16_t) (high_surrogate - LIT_UTF16_HIGH_SURROGATE_MIN);

jerry-core/lit/lit-strings.h

@@ -126,8 +126,8 @@ bool lit_is_utf8_string_valid (const lit_utf8_byte_t *, lit_utf8_size_t);
 bool lit_is_cesu8_string_valid (const lit_utf8_byte_t *, lit_utf8_size_t);
 
 /* checks */
-bool lit_is_code_unit_low_surrogate (ecma_char_t);
+bool lit_is_code_point_utf16_low_surrogate (lit_code_point_t);
-bool lit_is_code_unit_high_surrogate (ecma_char_t);
+bool lit_is_code_point_utf16_high_surrogate (lit_code_point_t);
 
 /* iteration */
 lit_utf8_iterator_t lit_utf8_iterator_create (const lit_utf8_byte_t *, lit_utf8_size_t);

jerry-core/parser/js/lexer.cpp

@@ -475,8 +475,8 @@ lexer_transform_escape_sequences (const jerry_api_char_t *source_str_p, /**< str
       converted_char = next_char;
     }
 
-    if (lit_is_code_unit_high_surrogate (prev_converted_char)
+    if (lit_is_code_point_utf16_high_surrogate (prev_converted_char)
-        && lit_is_code_unit_low_surrogate (converted_char))
+        && lit_is_code_point_utf16_low_surrogate (converted_char))
     {
       output_str_buf_iter_p -= LIT_UTF8_MAX_BYTES_IN_CODE_UNIT;