Rework RegExp engine and add support for proper unicode matching (#3746)

This change includes several bugfixes, general improvements, and support
for additional features.
- Added full support for web compatibility syntax defined in Annex B
- Implemented parsing and matching patterns in unicode mode
- Fixed capture results when iterating with nested capturing groups
- Significantly reduced regexp bytecode size
- Reduced stack usage during regexp execution
- Improved matching performance

JerryScript-DCO-1.0-Signed-off-by: Dániel Bátyai dbatyai@inf.u-szeged.hu
This commit is contained in:
Dániel Bátyai
2020-05-26 15:28:54 +02:00
committed by GitHub
parent 908240ba62
commit 8f76a1f382
30 changed files with 3641 additions and 2647 deletions
@@ -223,13 +223,13 @@ ecma_builtin_global_object_decode_uri_helper (lit_utf8_byte_t *input_start_p, /*
continue;
}
ecma_char_t decoded_byte;
if (!lit_read_code_unit_from_hex (input_char_p + 1, 2, &decoded_byte))
uint32_t hex_value = lit_char_hex_lookup (input_char_p + 1, input_end_p, 2);
if (hex_value == UINT32_MAX)
{
return ecma_raise_uri_error (ECMA_ERR_MSG ("Invalid hexadecimal value."));
}
ecma_char_t decoded_byte = (ecma_char_t) hex_value;
input_char_p += URI_ENCODED_BYTE_SIZE;
if (decoded_byte <= LIT_UTF8_1_BYTE_CODE_POINT_MAX)
@@ -272,20 +272,18 @@ ecma_builtin_global_object_decode_uri_helper (lit_utf8_byte_t *input_start_p, /*
/* Input decode. */
if (*input_char_p != '%')
{
*output_char_p = *input_char_p;
output_char_p++;
input_char_p++;
*output_char_p++ = *input_char_p++;
continue;
}
ecma_char_t decoded_byte;
if (!lit_read_code_unit_from_hex (input_char_p + 1, 2, &decoded_byte))
uint32_t hex_value = lit_char_hex_lookup (input_char_p + 1, input_end_p, 2);
if (hex_value == UINT32_MAX)
{
ret_value = ecma_raise_uri_error (ECMA_ERR_MSG ("Invalid hexadecimal value."));
break;
}
ecma_char_t decoded_byte = (ecma_char_t) hex_value;
input_char_p += URI_ENCODED_BYTE_SIZE;
if (decoded_byte <= LIT_UTF8_1_BYTE_CODE_POINT_MAX)
@@ -337,17 +335,16 @@ ecma_builtin_global_object_decode_uri_helper (lit_utf8_byte_t *input_start_p, /*
}
else
{
ecma_char_t chr;
hex_value = lit_char_hex_lookup (input_char_p + 1, input_end_p, 2);
if (!lit_read_code_unit_from_hex (input_char_p + 1, 2, &chr)
|| ((chr & LIT_UTF8_EXTRA_BYTE_MASK) != LIT_UTF8_EXTRA_BYTE_MARKER))
if (hex_value == UINT32_MAX || (hex_value & LIT_UTF8_EXTRA_BYTE_MASK) != LIT_UTF8_EXTRA_BYTE_MARKER)
{
is_valid = false;
break;
}
octets[i] = (lit_utf8_byte_t) chr;
input_char_p += URI_ENCODED_BYTE_SIZE;
octets[i] = (lit_utf8_byte_t) hex_value;
}
}
@@ -174,18 +174,13 @@ ecma_builtin_json_parse_string (ecma_json_token_t *token_p) /**< token argument
}
case LIT_CHAR_LOWERCASE_U:
{
if ((end_p - current_p <= ECMA_JSON_HEX_ESCAPE_SEQUENCE_LENGTH))
uint32_t hex_value = lit_char_hex_lookup (current_p + 1, end_p, ECMA_JSON_HEX_ESCAPE_SEQUENCE_LENGTH);
if (hex_value == UINT32_MAX)
{
goto invalid_string;
}
ecma_char_t code_unit;
if (!(lit_read_code_unit_from_hex (current_p + 1, ECMA_JSON_HEX_ESCAPE_SEQUENCE_LENGTH, &code_unit)))
{
goto invalid_string;
}
ecma_stringbuilder_append_char (&result_builder, code_unit);
ecma_stringbuilder_append_char (&result_builder, (ecma_char_t) hex_value);
current_p += ECMA_JSON_HEX_ESCAPE_SEQUENCE_LENGTH + 1;
break;
}
@@ -505,12 +505,10 @@ ecma_instantiate_builtin (ecma_builtin_id_t obj_builtin_id) /**< built-in id */
ext_object_p->u.class_prop.class_id = LIT_MAGIC_STRING_REGEXP_UL;
const re_compiled_code_t *bc_p = NULL;
ecma_value_t ret_value = re_compile_bytecode (&bc_p,
ecma_get_magic_string (LIT_MAGIC_STRING_EMPTY_NON_CAPTURE_GROUP),
RE_FLAG_EMPTY);
re_compiled_code_t *bc_p = re_compile_bytecode (ecma_get_magic_string (LIT_MAGIC_STRING_EMPTY_NON_CAPTURE_GROUP),
RE_FLAG_EMPTY);
JERRY_ASSERT (ecma_is_value_empty (ret_value));
JERRY_ASSERT (bc_p != NULL);
ECMA_SET_INTERNAL_VALUE_POINTER (ext_object_p->u.class_prop.u.value, bc_p);