/**
* Read the input pattern and parse the range of character class
*
* @return completion value
* Returned value must be freed with ecma_free_completion_value
*/
ecma_completion_value_t
re_parse_char_class (re_parser_ctx_t *parser_ctx_p, /**< number of classes */
re_char_class_callback append_char_class, /**< callback function,
* which adds the char-ranges
* to the bytecode */
void *re_ctx_p, /**< regexp compiler context */
re_token_t *out_token_p) /**< out: output token */
{
re_token_type_t token_type = ((re_compiler_ctx_t *) re_ctx_p)->current_token.type;
out_token_p->qmax = out_token_p->qmin = 1;
uint32_t start = RE_CHAR_UNDEF;
bool is_range = false;
parser_ctx_p->num_of_classes = 0;
if (lit_utf8_peek_prev (parser_ctx_p->input_curr_p) != LIT_CHAR_LEFT_SQUARE)
{
lit_utf8_decr (&parser_ctx_p->input_curr_p);
lit_utf8_decr (&parser_ctx_p->input_curr_p);
}
do
{
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
return ecma_raise_syntax_error ("invalid character class, end of string");
}
uint32_t ch = lit_utf8_read_next (&parser_ctx_p->input_curr_p);
if (ch == LIT_CHAR_RIGHT_SQUARE)
{
if (start != RE_CHAR_UNDEF)
{
append_char_class (re_ctx_p, start, start);
}
break;
}
else if (ch == LIT_CHAR_MINUS)
{
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
return ecma_raise_syntax_error ("invalid character class, end of string after '-'");
}
if (start != RE_CHAR_UNDEF
&& !is_range
&& *parser_ctx_p->input_curr_p != LIT_CHAR_RIGHT_SQUARE)
{
is_range = true;
continue;
}
}
else if (ch == LIT_CHAR_BACKSLASH)
{
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
return ecma_raise_syntax_error ("invalid character class, end of string after '\\'");
}
ch = *parser_ctx_p->input_curr_p++;
if (ch == LIT_CHAR_LOWERCASE_B)
{
ch = LIT_CHAR_BS;
}
else if (ch == LIT_CHAR_LOWERCASE_F)
{
ch = LIT_CHAR_FF;
}
else if (ch == LIT_CHAR_LOWERCASE_N)
{
ch = LIT_CHAR_LF;
}
else if (ch == LIT_CHAR_LOWERCASE_T)
{
ch = LIT_CHAR_TAB;
}
else if (ch == LIT_CHAR_LOWERCASE_R)
{
ch = LIT_CHAR_CR;
}
else if (ch == LIT_CHAR_LOWERCASE_V)
{
ch = LIT_CHAR_VTAB;
}
else if (ch == LIT_CHAR_LOWERCASE_C)
{
if (parser_ctx_p->input_curr_p < parser_ctx_p->input_end_p)
{
ch = *parser_ctx_p->input_curr_p;
if ((ch >= LIT_CHAR_ASCII_UPPERCASE_LETTERS_BEGIN && ch <= LIT_CHAR_ASCII_UPPERCASE_LETTERS_END)
|| (ch >= LIT_CHAR_ASCII_LOWERCASE_LETTERS_BEGIN && ch <= LIT_CHAR_ASCII_LOWERCASE_LETTERS_END)
|| (ch >= LIT_CHAR_0 && ch <= LIT_CHAR_9))
{
/* See ECMA-262 v5, 15.10.2.10 (Point 3) */
ch = (ch % 32);
parser_ctx_p->input_curr_p++;
}
else
{
ch = LIT_CHAR_LOWERCASE_C;
}
}
}
else if (ch == LIT_CHAR_LOWERCASE_X)
{
lit_code_point_t code_point;
if (!lit_read_code_point_from_hex (parser_ctx_p->input_curr_p, 2, &code_point))
{
return ecma_raise_syntax_error ("invalid character class, end of string after '\\x'");
}
parser_ctx_p->input_curr_p += 2;
append_char_class (re_ctx_p, code_point, code_point);
}
else if (ch == LIT_CHAR_LOWERCASE_U)
{
lit_code_point_t code_point;
if (!lit_read_code_point_from_hex (parser_ctx_p->input_curr_p, 4, &code_point))
{
return ecma_raise_syntax_error ("invalid character class, end of string after '\\u'");
}
parser_ctx_p->input_curr_p += 4;
append_char_class (re_ctx_p, code_point, code_point);
}
else if (ch == LIT_CHAR_LOWERCASE_D)
{
/* See ECMA-262 v5, 15.10.2.12 */
append_char_class (re_ctx_p, LIT_CHAR_ASCII_DIGITS_BEGIN, LIT_CHAR_ASCII_DIGITS_END);
ch = RE_CHAR_UNDEF;
}
else if (ch == LIT_CHAR_UPPERCASE_D)
{
/* See ECMA-262 v5, 15.10.2.12 */
append_char_class (re_ctx_p, LIT_CHAR_NULL, LIT_CHAR_ASCII_DIGITS_BEGIN - 1);
append_char_class (re_ctx_p, LIT_CHAR_ASCII_DIGITS_END + 1, LIT_UTF16_CODE_UNIT_MAX);
ch = RE_CHAR_UNDEF;
}
else if (ch == LIT_CHAR_LOWERCASE_S)
{
/* See ECMA-262 v5, 15.10.2.12 */
append_char_class (re_ctx_p, LIT_CHAR_TAB, LIT_CHAR_CR);
append_char_class (re_ctx_p, LIT_CHAR_SP, LIT_CHAR_SP);
append_char_class (re_ctx_p, LIT_CHAR_NBSP, LIT_CHAR_NBSP);
append_char_class (re_ctx_p, 0x1680UL, 0x1680UL); /* Ogham Space Mark */
append_char_class (re_ctx_p, 0x180EUL, 0x180EUL); /* Mongolian Vowel Separator */
append_char_class (re_ctx_p, 0x2000UL, 0x200AUL); /* En Quad - Hair Space */
append_char_class (re_ctx_p, LIT_CHAR_LS, LIT_CHAR_PS);
append_char_class (re_ctx_p, 0x202FUL, 0x202FUL); /* Narrow No-Break Space */
append_char_class (re_ctx_p, 0x205FUL, 0x205FUL); /* Medium Mathematical Space */
append_char_class (re_ctx_p, 0x3000UL, 0x3000UL); /* Ideographic Space */
append_char_class (re_ctx_p, LIT_CHAR_BOM, LIT_CHAR_BOM);
ch = RE_CHAR_UNDEF;
}
else if (ch == LIT_CHAR_UPPERCASE_S)
{
/* See ECMA-262 v5, 15.10.2.12 */
append_char_class (re_ctx_p, LIT_CHAR_NULL, LIT_CHAR_TAB - 1);
append_char_class (re_ctx_p, LIT_CHAR_CR + 1, LIT_CHAR_SP - 1);
append_char_class (re_ctx_p, LIT_CHAR_SP + 1, LIT_CHAR_NBSP - 1);
append_char_class (re_ctx_p, LIT_CHAR_NBSP + 1, 0x167FUL);
append_char_class (re_ctx_p, 0x1681UL, 0x180DUL);
append_char_class (re_ctx_p, 0x180FUL, 0x1FFFUL);
append_char_class (re_ctx_p, 0x200BUL, LIT_CHAR_LS - 1);
append_char_class (re_ctx_p, LIT_CHAR_PS + 1, 0x202EUL);
append_char_class (re_ctx_p, 0x2030UL, 0x205EUL);
append_char_class (re_ctx_p, 0x2060UL, 0x2FFFUL);
append_char_class (re_ctx_p, 0x3001UL, LIT_CHAR_BOM - 1);
append_char_class (re_ctx_p, LIT_CHAR_BOM + 1, LIT_UTF16_CODE_UNIT_MAX);
ch = RE_CHAR_UNDEF;
}
else if (ch == LIT_CHAR_LOWERCASE_W)
{
/* See ECMA-262 v5, 15.10.2.12 */
append_char_class (re_ctx_p, LIT_CHAR_0, LIT_CHAR_9);
append_char_class (re_ctx_p, LIT_CHAR_UPPERCASE_A, LIT_CHAR_UPPERCASE_Z);
append_char_class (re_ctx_p, LIT_CHAR_UNDERSCORE, LIT_CHAR_UNDERSCORE);
append_char_class (re_ctx_p, LIT_CHAR_LOWERCASE_A, LIT_CHAR_LOWERCASE_Z);
ch = RE_CHAR_UNDEF;
}
else if (ch == LIT_CHAR_UPPERCASE_W)
{
/* See ECMA-262 v5, 15.10.2.12 */
append_char_class (re_ctx_p, LIT_CHAR_NULL, LIT_CHAR_0 - 1);
append_char_class (re_ctx_p, LIT_CHAR_9 + 1, LIT_CHAR_UPPERCASE_A - 1);
append_char_class (re_ctx_p, LIT_CHAR_UPPERCASE_Z + 1, LIT_CHAR_UNDERSCORE - 1);
append_char_class (re_ctx_p, LIT_CHAR_UNDERSCORE + 1, LIT_CHAR_LOWERCASE_A - 1);
append_char_class (re_ctx_p, LIT_CHAR_LOWERCASE_Z + 1, LIT_UTF16_CODE_UNIT_MAX);
ch = RE_CHAR_UNDEF;
}
else if (ch <= LIT_UTF16_CODE_UNIT_MAX
&& lit_char_is_octal_digit ((ecma_char_t) ch)
&& ch != LIT_CHAR_0)
{
parser_ctx_p->input_curr_p--;
ch = re_parse_octal (parser_ctx_p);
}
} /* ch == LIT_CHAR_BACKSLASH */
if (ch == RE_CHAR_UNDEF)
{
if (start != RE_CHAR_UNDEF)
{
if (is_range)
{
return ecma_raise_syntax_error ("invalid character class, invalid range");
}
else
{
append_char_class (re_ctx_p, start, start);
start = RE_CHAR_UNDEF;
}
}
}
else
{
if (start != RE_CHAR_UNDEF)
{
if (is_range)
{
if (start > ch)
{
return ecma_raise_syntax_error ("invalid character class, wrong order");
}
else
{
append_char_class (re_ctx_p, start, ch);
start = RE_CHAR_UNDEF;
is_range = false;
}
}
else
{
append_char_class (re_ctx_p, start, start);
start = ch;
}
}
else
{
start = ch;
}
}
}
while (token_type == RE_TOK_START_CHAR_CLASS || token_type == RE_TOK_START_INV_CHAR_CLASS);
return re_parse_iterator (parser_ctx_p, out_token_p);
} /* re_parse_char_class */
/**
* Read the input pattern and parse the next token for the RegExp compiler
*
* @return completion value
* Returned value must be freed with ecma_free_completion_value
*/
ecma_completion_value_t
re_parse_next_token (re_parser_ctx_t *parser_ctx_p, /**< RegExp parser context */
re_token_t *out_token_p) /**< out: output token */
{
ecma_completion_value_t ret_value = ecma_make_empty_completion_value ();
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
out_token_p->type = RE_TOK_EOF;
return ret_value;
}
ecma_char_t ch = lit_utf8_read_next (&parser_ctx_p->input_curr_p);
switch (ch)
{
case LIT_CHAR_VLINE:
{
out_token_p->type = RE_TOK_ALTERNATIVE;
break;
}
case LIT_CHAR_CIRCUMFLEX:
{
out_token_p->type = RE_TOK_ASSERT_START;
break;
}
case LIT_CHAR_DOLLAR_SIGN:
{
out_token_p->type = RE_TOK_ASSERT_END;
break;
}
case LIT_CHAR_DOT:
{
out_token_p->type = RE_TOK_PERIOD;
ret_value = re_parse_iterator (parser_ctx_p, out_token_p);
break;
}
case LIT_CHAR_BACKSLASH:
{
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
return ecma_raise_syntax_error ("invalid regular experssion");
}
out_token_p->type = RE_TOK_CHAR;
ch = lit_utf8_read_next (&parser_ctx_p->input_curr_p);
if (ch == LIT_CHAR_LOWERCASE_B)
{
out_token_p->type = RE_TOK_ASSERT_WORD_BOUNDARY;
}
else if (ch == LIT_CHAR_UPPERCASE_B)
{
out_token_p->type = RE_TOK_ASSERT_NOT_WORD_BOUNDARY;
}
else if (ch == LIT_CHAR_LOWERCASE_F)
{
out_token_p->value = LIT_CHAR_FF;
}
else if (ch == LIT_CHAR_LOWERCASE_N)
{
out_token_p->value = LIT_CHAR_LF;
}
else if (ch == LIT_CHAR_LOWERCASE_T)
{
out_token_p->value = LIT_CHAR_TAB;
}
else if (ch == LIT_CHAR_LOWERCASE_R)
{
out_token_p->value = LIT_CHAR_CR;
}
else if (ch == LIT_CHAR_LOWERCASE_V)
{
out_token_p->value = LIT_CHAR_VTAB;
}
else if (ch == LIT_CHAR_LOWERCASE_C)
{
if (parser_ctx_p->input_curr_p < parser_ctx_p->input_end_p)
{
ch = *parser_ctx_p->input_curr_p;
if ((ch >= LIT_CHAR_ASCII_UPPERCASE_LETTERS_BEGIN && ch <= LIT_CHAR_ASCII_UPPERCASE_LETTERS_END)
|| (ch >= LIT_CHAR_ASCII_LOWERCASE_LETTERS_BEGIN && ch <= LIT_CHAR_ASCII_LOWERCASE_LETTERS_END))
{
out_token_p->value = (ch % 32);
parser_ctx_p->input_curr_p++;
}
else
{
out_token_p->value = LIT_CHAR_BACKSLASH;
parser_ctx_p->input_curr_p--;
}
}
else
{
out_token_p->value = LIT_CHAR_BACKSLASH;
parser_ctx_p->input_curr_p--;
}
}
else if (ch == LIT_CHAR_LOWERCASE_X
&& re_hex_lookup (parser_ctx_p, 2))
{
lit_code_point_t code_point;
if (!lit_read_code_point_from_hex (parser_ctx_p->input_curr_p, 2, &code_point))
{
return ecma_raise_syntax_error ("decode error");
}
parser_ctx_p->input_curr_p += 2;
out_token_p->value = code_point;
}
else if (ch == LIT_CHAR_LOWERCASE_U
&& re_hex_lookup (parser_ctx_p, 4))
{
lit_code_point_t code_point;
if (!lit_read_code_point_from_hex (parser_ctx_p->input_curr_p, 4, &code_point))
{
return ecma_raise_syntax_error ("decode error");
}
parser_ctx_p->input_curr_p += 4;
out_token_p->value = code_point;
}
else if (ch == LIT_CHAR_LOWERCASE_D)
{
out_token_p->type = RE_TOK_DIGIT;
break;
}
else if (ch == LIT_CHAR_UPPERCASE_D)
{
out_token_p->type = RE_TOK_NOT_DIGIT;
break;
}
else if (ch == LIT_CHAR_LOWERCASE_S)
{
out_token_p->type = RE_TOK_WHITE;
break;
}
else if (ch == LIT_CHAR_UPPERCASE_S)
{
out_token_p->type = RE_TOK_NOT_WHITE;
break;
}
else if (ch == LIT_CHAR_LOWERCASE_W)
{
out_token_p->type = RE_TOK_WORD_CHAR;
break;
}
else if (ch == LIT_CHAR_UPPERCASE_W)
{
out_token_p->type = RE_TOK_NOT_WORD_CHAR;
break;
}
else if (lit_char_is_decimal_digit (ch))
{
if (ch == LIT_CHAR_0)
{
if (parser_ctx_p->input_curr_p < parser_ctx_p->input_end_p
&& lit_char_is_decimal_digit (*parser_ctx_p->input_curr_p))
{
return ecma_raise_syntax_error ("RegExp escape pattern error.");
}
out_token_p->value = LIT_UNICODE_CODE_POINT_NULL;
}
else
{
if (parser_ctx_p->num_of_groups == -1)
{
re_count_num_of_groups (parser_ctx_p);
}
if (parser_ctx_p->num_of_groups)
{
parser_ctx_p->input_curr_p--;
uint32_t number = 0;
int index = 0;
do
{
if (index >= RE_MAX_RE_DECESC_DIGITS)
{
ret_value = ecma_raise_syntax_error ("RegExp escape pattern error: decimal escape too long.");
return ret_value;
}
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
break;
}
ecma_char_t digit = *parser_ctx_p->input_curr_p++;
if (!lit_char_is_decimal_digit (digit))
{
parser_ctx_p->input_curr_p--;
break;
}
number = number * 10 + lit_char_hex_to_int (digit);
index++;
}
while (true);
if ((int) number <= parser_ctx_p->num_of_groups)
{
out_token_p->type = RE_TOK_BACKREFERENCE;
}
else
/* Invalid backreference, fallback to octal */
{
/* Rewind to start of number. */
parser_ctx_p->input_curr_p -= index;
/* Try to reparse as octal. */
ecma_char_t digit = *parser_ctx_p->input_curr_p;
if (!lit_char_is_octal_digit (digit))
{
/* Not octal, keep digit character value. */
number = digit;
parser_ctx_p->input_curr_p++;
}
else
{
number = re_parse_octal (parser_ctx_p);
}
}
out_token_p->value = number;
}
else
/* Invalid backreference, fallback to octal if possible */
{
if (!lit_char_is_octal_digit (ch))
{
/* Not octal, keep character value. */
out_token_p->value = ch;
}
else
{
parser_ctx_p->input_curr_p--;
out_token_p->value = re_parse_octal (parser_ctx_p);
}
}
}
}
else
{
out_token_p->value = ch;
}
ret_value = re_parse_iterator (parser_ctx_p, out_token_p);
break;
}
case LIT_CHAR_LEFT_PAREN:
{
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
return ecma_raise_syntax_error ("Unterminated group");
}
if (*parser_ctx_p->input_curr_p == LIT_CHAR_QUESTION)
{
parser_ctx_p->input_curr_p++;
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
return ecma_raise_syntax_error ("Invalid group");
}
ch = *parser_ctx_p->input_curr_p++;
if (ch == LIT_CHAR_EQUALS)
{
/* (?= */
out_token_p->type = RE_TOK_ASSERT_START_POS_LOOKAHEAD;
}
else if (ch == LIT_CHAR_EXCLAMATION)
{
/* (?! */
out_token_p->type = RE_TOK_ASSERT_START_NEG_LOOKAHEAD;
}
else if (ch == LIT_CHAR_COLON)
{
/* (?: */
out_token_p->type = RE_TOK_START_NON_CAPTURE_GROUP;
}
else
{
return ecma_raise_syntax_error ("Invalid group");
}
}
else
{
/* ( */
out_token_p->type = RE_TOK_START_CAPTURE_GROUP;
}
break;
}
case LIT_CHAR_RIGHT_PAREN:
{
out_token_p->type = RE_TOK_END_GROUP;
ret_value = re_parse_iterator (parser_ctx_p, out_token_p);
break;
}
case LIT_CHAR_LEFT_SQUARE:
{
out_token_p->type = RE_TOK_START_CHAR_CLASS;
if (parser_ctx_p->input_curr_p >= parser_ctx_p->input_end_p)
{
return ecma_raise_syntax_error ("invalid character class");
}
if (*parser_ctx_p->input_curr_p == LIT_CHAR_CIRCUMFLEX)
{
out_token_p->type = RE_TOK_START_INV_CHAR_CLASS;
parser_ctx_p->input_curr_p++;
}
break;
}
case LIT_CHAR_QUESTION:
case LIT_CHAR_ASTERISK:
case LIT_CHAR_PLUS:
case LIT_CHAR_LEFT_BRACE:
{
return ecma_raise_syntax_error ("Invalid RegExp token.");
}
case LIT_CHAR_NULL:
{
out_token_p->type = RE_TOK_EOF;
break;
}
default:
{
out_token_p->type = RE_TOK_CHAR;
out_token_p->value = ch;
ret_value = re_parse_iterator (parser_ctx_p, out_token_p);
break;
}
}
return ret_value;
} /* re_parse_next_token */
/**
* Parse and extract string token.
*/
static void
ecma_builtin_json_parse_string (ecma_json_token_t *token_p) /**< token argument */
{
lit_utf8_byte_t *current_p = token_p->current_p;
lit_utf8_byte_t *write_p = current_p;
token_p->u.string.start_p = current_p;
while (*current_p != LIT_CHAR_DOUBLE_QUOTE)
{
if (*current_p <= 0x1f)
{
return;
}
if (*current_p == LIT_CHAR_BACKSLASH)
{
current_p++;
switch (*current_p)
{
case LIT_CHAR_DOUBLE_QUOTE:
case LIT_CHAR_SLASH:
case LIT_CHAR_BACKSLASH:
{
break;
}
case LIT_CHAR_LOWERCASE_B:
{
*current_p = LIT_CHAR_BS;
break;
}
case LIT_CHAR_LOWERCASE_F:
{
*current_p = LIT_CHAR_FF;
break;
}
case LIT_CHAR_LOWERCASE_N:
{
*current_p = LIT_CHAR_LF;
break;
}
case LIT_CHAR_LOWERCASE_R:
{
*current_p = LIT_CHAR_CR;
break;
}
case LIT_CHAR_LOWERCASE_T:
{
*current_p = LIT_CHAR_TAB;
break;
}
case LIT_CHAR_LOWERCASE_U:
{
lit_code_point_t code_point;
if (!(lit_read_code_point_from_hex (current_p + 1, 4, &code_point)))
{
return;
}
current_p += 5;
write_p += lit_code_point_to_cesu8 (code_point, write_p);
continue;
}
default:
{
return;
}
}
}
*write_p++ = *current_p++;
}
token_p->u.string.size = (lit_utf8_size_t) (write_p - token_p->u.string.start_p);
token_p->current_p = current_p + 1;
token_p->type = string_token;
} /* ecma_builtin_json_parse_string */
