/**
* Skip any whitespace and comment tokens
*
* @return true - if a newline token was skipped,
* false - otherwise
*/
static bool
lexer_skip_whitespace_and_comments (void)
{
bool new_lines_occurred = false;
while (true)
{
ecma_char_t c = LA (0);
if (lit_char_is_white_space (c))
{
do
{
consume_char ();
c = LA (0);
}
while (lit_char_is_white_space (c));
}
else if (lit_char_is_line_terminator (c))
{
dump_current_line ();
new_lines_occurred = true;
do
{
consume_char ();
c = LA (0);
}
while (lit_char_is_line_terminator (c));
}
else if (c == LIT_CHAR_SLASH
&& (LA (1) == LIT_CHAR_SLASH
|| LA (1) == LIT_CHAR_ASTERISK))
{
/* ECMA-262 v5, 7.4, SingleLineComment or MultiLineComment */
if (lexer_parse_comment ())
{
new_lines_occurred = true;
}
}
else
{
break;
}
}
return new_lines_occurred;
} /* lexer_skip_whitespace_and_comments */
/**
* The String.prototype object's 'trim' routine
*
* See also:
* ECMA-262 v5, 15.5.4.20
*
* @return completion value
* Returned value must be freed with ecma_free_completion_value.
*/
static ecma_completion_value_t
ecma_builtin_string_prototype_object_trim (ecma_value_t this_arg) /**< this argument */
{
ecma_completion_value_t ret_value = ecma_make_empty_completion_value ();
/* 1 */
ECMA_TRY_CATCH (check_coercible_val,
ecma_op_check_object_coercible (this_arg),
ret_value);
/* 2 */
ECMA_TRY_CATCH (to_string_val,
ecma_op_to_string (this_arg),
ret_value);
ecma_string_t *original_string_p = ecma_get_string_from_value (to_string_val);
/* 3 */
const lit_utf8_size_t size = ecma_string_get_size (original_string_p);
const ecma_length_t length = ecma_string_get_size (original_string_p);
/* Workaround: avoid repeated call of ecma_string_get_char_at_pos() because its overhead */
lit_utf8_byte_t *original_utf8_str_p = (lit_utf8_byte_t *) mem_heap_alloc_block (size + 1,
MEM_HEAP_ALLOC_SHORT_TERM);
ecma_string_to_utf8_string (original_string_p, original_utf8_str_p, (ssize_t) size);
uint32_t prefix = 0, postfix = 0;
uint32_t new_len = 0;
while (prefix < length)
{
ecma_char_t next_char = lit_utf8_string_code_unit_at (original_utf8_str_p,
size,
prefix);
if (lit_char_is_white_space (next_char)
|| lit_char_is_line_terminator (next_char))
{
prefix++;
}
else
{
break;
}
}
while (postfix < length - prefix)
{
ecma_char_t next_char = lit_utf8_string_code_unit_at (original_utf8_str_p,
size,
length - postfix - 1);
if (lit_char_is_white_space (next_char)
|| lit_char_is_line_terminator (next_char))
{
postfix++;
}
else
{
break;
}
}
new_len = prefix < size ? size - prefix - postfix : 0;
ecma_string_t *new_str_p = ecma_string_substr (original_string_p, prefix, prefix + new_len);
/* 4 */
ret_value = ecma_make_normal_completion_value (ecma_make_string_value (new_str_p));
mem_heap_free_block (original_utf8_str_p);
ECMA_FINALIZE (to_string_val);
ECMA_FINALIZE (check_coercible_val);
return ret_value;
} /* ecma_builtin_string_prototype_object_trim */
/**
* Parse and construct lexer token
*
* Note:
* Currently, lexer token doesn't fully correspond to Token, defined in ECMA-262, v5, 7.5.
* For example, there is no new-line token type in the token definition of ECMA-262 v5.
*
* @return constructed token
*/
static token
lexer_parse_token (void)
{
ecma_char_t c = LA (0);
if (lit_char_is_white_space (c))
{
while (lit_char_is_white_space (c))
{
consume_char ();
c = LA (0);
}
}
if (lit_char_is_line_terminator (c))
{
while (lit_char_is_line_terminator (c))
{
consume_char ();
c = LA (0);
}
return create_token (TOK_NEWLINE, 0);
}
JERRY_ASSERT (is_token_parse_in_progress == false);
/* ECMA-262 v5, 7.6, Identifier */
if (lexer_is_char_can_be_identifier_start (c))
{
return lexer_parse_identifier_or_keyword ();
}
/* ECMA-262 v5, 7.8.3, Numeric literal */
if (lit_char_is_decimal_digit (c)
|| (c == LIT_CHAR_DOT
&& lit_char_is_decimal_digit (LA (1))))
{
return lexer_parse_number ();
}
if (c == LIT_CHAR_LF)
{
consume_char ();
return create_token (TOK_NEWLINE, 0);
}
if (c == LIT_CHAR_NULL)
{
return create_token (TOK_EOF, 0);
}
if (c == LIT_CHAR_SINGLE_QUOTE
|| c == LIT_CHAR_DOUBLE_QUOTE)
{
return lexer_parse_string ();
}
/* ECMA-262 v5, 7.4, SingleLineComment or MultiLineComment */
if (c == LIT_CHAR_SLASH
&& (LA (1) == LIT_CHAR_SLASH
|| LA (1) == LIT_CHAR_ASTERISK))
{
if (lexer_parse_comment ())
{
return create_token (TOK_NEWLINE, 0);
}
else
{
return lexer_parse_token ();
}
}
if (c == LIT_CHAR_SLASH
&& !(prev_non_lf_token.type == TOK_NAME
|| prev_non_lf_token.type == TOK_NULL
|| prev_non_lf_token.type == TOK_BOOL
|| prev_non_lf_token.type == TOK_CLOSE_BRACE
|| prev_non_lf_token.type == TOK_CLOSE_SQUARE
|| prev_non_lf_token.type == TOK_CLOSE_PAREN
|| prev_non_lf_token.type == TOK_SMALL_INT
|| prev_non_lf_token.type == TOK_NUMBER
|| prev_non_lf_token.type == TOK_STRING
|| prev_non_lf_token.type == TOK_REGEXP))
{
return lexer_parse_regexp ();
}
/* ECMA-262 v5, 7.7, Punctuator */
switch (c)
{
case LIT_CHAR_LEFT_BRACE:
{
RETURN_PUNC (TOK_OPEN_BRACE);
break;
}
case LIT_CHAR_RIGHT_BRACE:
{
RETURN_PUNC (TOK_CLOSE_BRACE);
break;
}
case LIT_CHAR_LEFT_PAREN:
{
RETURN_PUNC (TOK_OPEN_PAREN);
break;
}
case LIT_CHAR_RIGHT_PAREN:
{
RETURN_PUNC (TOK_CLOSE_PAREN);
break;
}
case LIT_CHAR_LEFT_SQUARE:
{
RETURN_PUNC (TOK_OPEN_SQUARE);
break;
}
case LIT_CHAR_RIGHT_SQUARE:
{
RETURN_PUNC (TOK_CLOSE_SQUARE);
break;
}
case LIT_CHAR_DOT:
{
RETURN_PUNC (TOK_DOT);
break;
}
case LIT_CHAR_SEMICOLON:
{
RETURN_PUNC (TOK_SEMICOLON);
break;
}
case LIT_CHAR_COMMA:
{
RETURN_PUNC (TOK_COMMA);
break;
}
case LIT_CHAR_TILDE:
{
RETURN_PUNC (TOK_COMPL);
break;
}
case LIT_CHAR_COLON:
{
RETURN_PUNC (TOK_COLON);
break;
}
case LIT_CHAR_QUESTION:
{
RETURN_PUNC (TOK_QUERY);
break;
}
case LIT_CHAR_ASTERISK:
{
IF_LA_IS (LIT_CHAR_EQUALS, TOK_MULT_EQ, TOK_MULT);
break;
}
case LIT_CHAR_SLASH:
{
IF_LA_IS (LIT_CHAR_EQUALS, TOK_DIV_EQ, TOK_DIV);
break;
}
case LIT_CHAR_CIRCUMFLEX:
{
IF_LA_IS (LIT_CHAR_EQUALS, TOK_XOR_EQ, TOK_XOR);
break;
}
case LIT_CHAR_PERCENT:
{
IF_LA_IS (LIT_CHAR_EQUALS, TOK_MOD_EQ, TOK_MOD);
break;
}
case LIT_CHAR_PLUS:
{
IF_LA_IS_OR (LIT_CHAR_PLUS, TOK_DOUBLE_PLUS, LIT_CHAR_EQUALS, TOK_PLUS_EQ, TOK_PLUS);
break;
}
case LIT_CHAR_MINUS:
{
IF_LA_IS_OR (LIT_CHAR_MINUS, TOK_DOUBLE_MINUS, LIT_CHAR_EQUALS, TOK_MINUS_EQ, TOK_MINUS);
break;
}
case LIT_CHAR_AMPERSAND:
{
IF_LA_IS_OR (LIT_CHAR_AMPERSAND, TOK_DOUBLE_AND, LIT_CHAR_EQUALS, TOK_AND_EQ, TOK_AND);
break;
}
case LIT_CHAR_VLINE:
{
IF_LA_IS_OR (LIT_CHAR_VLINE, TOK_DOUBLE_OR, LIT_CHAR_EQUALS, TOK_OR_EQ, TOK_OR);
break;
}
case LIT_CHAR_LESS_THAN:
{
switch (LA (1))
{
case LIT_CHAR_LESS_THAN: IF_LA_N_IS (LIT_CHAR_EQUALS, TOK_LSHIFT_EQ, TOK_LSHIFT, 2); break;
case LIT_CHAR_EQUALS: RETURN_PUNC_EX (TOK_LESS_EQ, 2); break;
default: RETURN_PUNC (TOK_LESS);
}
break;
}
case LIT_CHAR_GREATER_THAN:
{
switch (LA (1))
{
case LIT_CHAR_GREATER_THAN:
{
switch (LA (2))
{
case LIT_CHAR_GREATER_THAN: IF_LA_N_IS (LIT_CHAR_EQUALS, TOK_RSHIFT_EX_EQ, TOK_RSHIFT_EX, 3); break;
case LIT_CHAR_EQUALS: RETURN_PUNC_EX (TOK_RSHIFT_EQ, 3); break;
default: RETURN_PUNC_EX (TOK_RSHIFT, 2);
}
break;
}
case LIT_CHAR_EQUALS: RETURN_PUNC_EX (TOK_GREATER_EQ, 2); break;
default: RETURN_PUNC (TOK_GREATER);
}
break;
}
case LIT_CHAR_EQUALS:
{
if (LA (1) == LIT_CHAR_EQUALS)
{
IF_LA_N_IS (LIT_CHAR_EQUALS, TOK_TRIPLE_EQ, TOK_DOUBLE_EQ, 2);
}
else
{
RETURN_PUNC (TOK_EQ);
}
break;
}
case LIT_CHAR_EXCLAMATION:
{
if (LA (1) == LIT_CHAR_EQUALS)
{
IF_LA_N_IS (LIT_CHAR_EQUALS, TOK_NOT_DOUBLE_EQ, TOK_NOT_EQ, 2);
}
else
{
RETURN_PUNC (TOK_NOT);
}
break;
}
}
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "Illegal character", lit_utf8_iterator_get_pos (&src_iter));
} /* lexer_parse_token */
/**
* Trim leading and trailing whitespace characters from string.
*
* @return trimmed ecma string
*/
ecma_string_t *
ecma_string_trim (const ecma_string_t *string_p) /**< pointer to an ecma string */
{
ecma_string_t *ret_string_p;
ECMA_STRING_TO_UTF8_STRING (string_p, utf8_str_p, utf8_str_size);
if (utf8_str_size > 0)
{
ecma_char_t ch;
lit_utf8_size_t read_size;
const lit_utf8_byte_t *nonws_start_p = utf8_str_p + utf8_str_size;
const lit_utf8_byte_t *current_p = utf8_str_p;
/* Trim front. */
while (current_p < nonws_start_p)
{
read_size = lit_read_code_unit_from_utf8 (current_p, &ch);
if (!lit_char_is_white_space (ch)
&& !lit_char_is_line_terminator (ch))
{
nonws_start_p = current_p;
break;
}
current_p += read_size;
}
current_p = utf8_str_p + utf8_str_size;
/* Trim back. */
while (current_p > utf8_str_p)
{
read_size = lit_read_prev_code_unit_from_utf8 (current_p, &ch);
if (!lit_char_is_white_space (ch)
&& !lit_char_is_line_terminator (ch))
{
break;
}
current_p -= read_size;
}
/* Construct new string. */
if (current_p > nonws_start_p)
{
ret_string_p = ecma_new_ecma_string_from_utf8 (nonws_start_p,
(lit_utf8_size_t) (current_p - nonws_start_p));
}
else
{
ret_string_p = ecma_get_magic_string (LIT_MAGIC_STRING__EMPTY);
}
}
else
{
ret_string_p = ecma_get_magic_string (LIT_MAGIC_STRING__EMPTY);
}
ECMA_FINALIZE_UTF8_STRING (utf8_str_p, utf8_str_size);
return ret_string_p;
} /* ecma_string_trim */
