Global object unescape routine

JerryScript-DCO-1.0-Signed-off-by: Szilard Ledan szledan.u-szeged@partner.samsung.com

Author: szledan

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

@@ -1329,6 +1329,123 @@ ecma_builtin_global_object_escape (ecma_value_t this_arg __attr_unused___, /**<
   return ret_value;
 } /* ecma_builtin_global_object_escape */
 
+/**
+ * The Global object's 'unescape' routine
+ *
+ * See also:
+ *          ECMA-262 v5, B.2.2
+ *
+ * @return completion value
+ *         Returned value must be freed with ecma_free_completion_value.
+ */
+static ecma_completion_value_t
+ecma_builtin_global_object_unescape (ecma_value_t this_arg __attr_unused___, /**< this argument */
+                                     ecma_value_t arg) /**< routine's first argument */
+{
+  ecma_completion_value_t ret_value = ecma_make_empty_completion_value ();
+
+  /* 1. */
+  ECMA_TRY_CATCH (string, ecma_op_to_string (arg), ret_value);
+  ecma_string_t *input_string_p = ecma_get_string_from_value (string);
+  /* 2. */
+  lit_utf8_size_t input_size = ecma_string_get_size (input_string_p);
+
+  /* 3. */
+  MEM_DEFINE_LOCAL_ARRAY (input_start_p, input_size, lit_utf8_byte_t);
+  ssize_t sz = ecma_string_to_utf8_string (input_string_p, input_start_p, (ssize_t) (input_size));
+  JERRY_ASSERT (sz >= 0);
+
+  lit_utf8_iterator_t iterator = lit_utf8_iterator_create (input_start_p, input_size);
+  /* 4. */
+  /* The length of input string is always greater than output string
+   * so we re-use the input string buffer.
+   * The %xx is three byte long, and the maximum encoded value is 0xff,
+   * which maximum encoded length is two byte. Similar to this, the maximum
+   * encoded length of %uxxxx is four byte. */
+  lit_utf8_byte_t *output_char_p = input_start_p;
+
+  /* The state of parsing that tells us where we are in an escape pattern.
+   * 0    we are outside of pattern,
+   * 1    found '%', start of pattern,
+   * 2    found first hex digit of '%xy' pattern
+   * 3    found valid '%xy' pattern
+   * 4    found 'u', start of '%uwxyz' pattern
+   * 5-7  found hex digits of '%uwxyz' pattern
+   * 8    found valid '%uwxyz' pattern
+   */
+  uint8_t status = 0;
+  lit_code_point_t high_surrogate = 0;
+  lit_code_point_t hex_digits = 0;
+  /* 5. */
+  while (!lit_utf8_iterator_is_eos (&iterator))
+  {
+    /* 6. */
+    lit_code_point_t code_point = lit_utf8_iterator_read_next (&iterator);
+
+    /* 7-8. */
+    if (status == 0 && code_point == LIT_CHAR_PERCENT)
+    {
+      /* Found '%' char, start of escape sequence. */
+      status = 1;
+    }
+    /* 9-10. */
+    else if (status == 1 && code_point == LIT_CHAR_LOWERCASE_U)
+    {
+      /* Found 'u' char after '%'. */
+      status = 4;
+    }
+    else if (status > 0 && lit_char_is_hex_digit ((ecma_char_t) code_point))
+    {
+      /* Found hexadecimal digit in escape sequence. */
+      hex_digits = hex_digits * 16 + lit_char_hex_to_int ((ecma_char_t) code_point);
+      status++;
+    }
+
+    /* 11-17. Found valid '%uwxyz' or '%xy' escape. */
+    if (status == 8 || status == 3)
+    {
+      output_char_p -= (status == 3) ? 2 : 5;
+      status = 0;
+      code_point = (ecma_char_t) hex_digits;
+      hex_digits = 0;
+    }
+
+    /* Handle surrogate pairs. */
+    bool is_non_bmp_middle = iterator.buf_pos.is_non_bmp_middle;
+    if (!high_surrogate && lit_is_code_unit_high_surrogate ((ecma_char_t) code_point))
+    {
+      high_surrogate = code_point;
+
+      if (is_non_bmp_middle)
+      {
+        code_point = lit_utf8_iterator_read_next (&iterator);
+      }
+    }
+
+    if (high_surrogate && lit_is_code_unit_low_surrogate ((ecma_char_t) code_point))
+    {
+      output_char_p -= is_non_bmp_middle ? 0 : 3;
+      code_point = lit_convert_surrogate_pair_to_code_point ((ecma_char_t) high_surrogate,
+                                                             (ecma_char_t) code_point);
+      high_surrogate = 0;
+    }
+
+    /* Copying character. */
+    lit_utf8_size_t lit_size = lit_code_point_to_utf8 (code_point, output_char_p);
+    output_char_p += lit_size;
+    JERRY_ASSERT (output_char_p - input_start_p <= iterator.buf_pos.offset);
+  }
+
+  lit_utf8_size_t output_length = (lit_utf8_size_t) (output_char_p - input_start_p);
+  ecma_string_t *output_string_p = ecma_new_ecma_string_from_utf8 (input_start_p, output_length);
+  ret_value = ecma_make_normal_completion_value (ecma_make_string_value (output_string_p));
+
+  MEM_FINALIZE_LOCAL_ARRAY (input_start_p);
+
+  ECMA_FINALIZE (string);
+  return ret_value;
+} /* ecma_builtin_global_object_unescape */
+
 #endif /* CONFIG_ECMA_COMPACT_PROFILE_DISABLE_ANNEXB_BUILTIN */
 
 /**

jerry-core/ecma/builtin-objects/ecma-builtin-global.inc.h

@@ -237,6 +237,7 @@ ROUTINE (LIT_MAGIC_STRING_PARSE_INT, ecma_builtin_global_object_parse_int, 2, 2)
 
 #ifndef CONFIG_ECMA_COMPACT_PROFILE_DISABLE_ANNEXB_BUILTIN
 ROUTINE (LIT_MAGIC_STRING_ESCAPE, ecma_builtin_global_object_escape, 1, 1)
+ROUTINE (LIT_MAGIC_STRING_UNESCAPE, ecma_builtin_global_object_unescape, 1, 1)
 #endif /* CONFIG_ECMA_COMPACT_PROFILE_DISABLE_ANNEXB_BUILTIN */
 
 #undef OBJECT_ID

jerry-core/lit/lit-magic-strings.inc.h

@@ -73,6 +73,7 @@ LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_DECODE_URI_COMPONENT, "decodeURIComponent
 LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_ENCODE_URI, "encodeURI")
 LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_ENCODE_URI_COMPONENT, "encodeURIComponent")
 LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_ESCAPE, "escape")
+LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_UNESCAPE, "unescape")
 LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_GET_PROTOTYPE_OF_UL, "getPrototypeOf")
 LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_GET_OWN_PROPERTY_DESCRIPTOR_UL, "getOwnPropertyDescriptor")
 LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_GET_OWN_PROPERTY_NAMES_UL, "getOwnPropertyNames")

tests/jerry/global-escaping.js

@@ -28,4 +28,33 @@ assert (escape("\x80\x95\xaf\xfe\xff") === "%80%95%AF%FE%FF");
 assert (escape("\u0100\ud800\udc00") === "%u0100%uD800%uDC00");
 
 assert (escape({}) === "%5Bobject%20Object%5D");
-assert (escape(true) === "true")
+assert (escape(true) === "true");
+
+// Unescaping
+
+assert (unescape ("%00%01%02%03%04%05%06%07%08%09%0A%0B%0C%0D%0E%0F") ===
+        "\x00\x01\x02\x03\x04\x05\x06\x07\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f");
+assert (unescape("%10%11%12%13%14%15%16%17%18%19%1A%1B%1C%1D%1E%1F") ===
+        "\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1a\x1b\x1c\x1d\x1e\x1f");
+assert (unescape("%20%21%22%23%24%25%26%27%28%29*+%2C-./0123456789%3A%3B%3C%3D%3E%3F@ABCDEFGHIJKLMN") ===
+        " !\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMN");
+assert (unescape("OPQRSTUVWXYZ%5B%5C%5D%5E_%60abcdefghijklmnopqrstuvwxyz%7B%7C%7D%7F") ===
+        "OPQRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{|}\x7F");
+assert (unescape("%80%95%AF%FE%FF") === "\x80\x95\xaf\xfe\xff");
+assert (unescape("%ud800") === "\ud800");
+assert (unescape("\ud800") === "\ud800");
+assert (unescape("\ud800\udc00") === "\ud800\udc00");
+assert (unescape("%ud800%udc00") === "\ud800\udc00");
+assert (unescape("\ud800%udc00") === "\ud800\udc00");
+assert (unescape("%ud800\udc00") === "\ud800\udc00");
+
+assert (unescape({}) === "[object Object]");
+assert (unescape(true) === "true")
+assert (unescape() === "undefined");
+assert (unescape(1985) === "1985");
+assert (unescape("%5#%uu") === "%5#%uu");
+
+// Inversion
+
+var str = "\u0001\u0000\uFFFF";
+assert (unescape(escape(str)) === str);