/**
* Parse a comment
*
* @return true if newline was met during parsing
* false - otherwise
*/
static bool
lexer_parse_comment (void)
{
ecma_char_t c = LA (0);
bool multiline;
bool was_newlines = false;
JERRY_ASSERT (LA (0) == LIT_CHAR_SLASH);
JERRY_ASSERT (LA (1) == LIT_CHAR_SLASH || LA (1) == LIT_CHAR_ASTERISK);
multiline = (LA (1) == LIT_CHAR_ASTERISK);
consume_char ();
consume_char ();
while (!lit_utf8_iterator_is_eos (&src_iter))
{
c = LA (0);
if (!multiline)
{
if (lit_char_is_line_terminator (c))
{
return true;
}
}
else
{
if (c == LIT_CHAR_ASTERISK
&& LA (1) == LIT_CHAR_SLASH)
{
consume_char ();
consume_char ();
return was_newlines;
}
else if (lit_char_is_line_terminator (c))
{
was_newlines = true;
}
}
consume_char ();
}
if (multiline)
{
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "Unclosed multiline comment", lit_utf8_iterator_get_pos (&src_iter));
}
return false;
} /* lexer_parse_comment */
/**
* 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 */
/**
* Dump current line
*/
static void
dump_current_line (void)
{
if (!allow_dump_lines)
{
return;
}
printf ("// ");
lit_utf8_iterator_t iter = src_iter;
while (!lit_utf8_iterator_is_eos (&iter))
{
ecma_char_t code_unit = lit_utf8_iterator_read_next (&iter);
if (lit_char_is_line_terminator (code_unit))
{
if (code_unit == LIT_CHAR_CR
&& !lit_utf8_iterator_is_eos (&iter)
&& lit_utf8_iterator_peek_next (&iter) == LIT_CHAR_LF)
{
lit_utf8_iterator_incr (&iter);
}
break;
}
lit_put_ecma_char (code_unit);
}
lit_put_ecma_char (LIT_CHAR_LF);
} /* dump_current_line */
/**
* Dump specified line of the source script
*/
void
lexer_dump_line (size_t line) /**< line number */
{
size_t l = 0;
lit_utf8_iterator_t iter = src_iter;
lit_utf8_iterator_seek_bos (&iter);
while (!lit_utf8_iterator_is_eos (&iter))
{
ecma_char_t code_unit;
if (l == line)
{
while (!lit_utf8_iterator_is_eos (&iter))
{
code_unit = lit_utf8_iterator_read_next (&iter);
if (lit_char_is_line_terminator (code_unit))
{
break;
}
lit_put_ecma_char (code_unit);
}
return;
}
code_unit = lit_utf8_iterator_read_next (&iter);
if (lit_char_is_line_terminator (code_unit))
{
l++;
if (code_unit == LIT_CHAR_CR
&& !lit_utf8_iterator_is_eos (&iter)
&& lit_utf8_iterator_peek_next (&iter) == LIT_CHAR_LF)
{
lit_utf8_iterator_incr (&iter);
}
}
}
} /* lexer_dump_line */
/**
* Convert locus to line and column
*/
void
lexer_locus_to_line_and_column (lit_utf8_iterator_pos_t locus, /**< iterator position in the source script */
size_t *line, /**< @out: line number */
size_t *column) /**< @out: column number */
{
JERRY_ASSERT ((lit_utf8_size_t) (locus.offset + locus.is_non_bmp_middle) <= buffer_size);
lit_utf8_iterator_t iter = lit_utf8_iterator_create (buffer_start, (lit_utf8_size_t) buffer_size);
lit_utf8_iterator_pos_t iter_pos = lit_utf8_iterator_get_pos (&iter);
size_t l = 0, c = 0;
while (!lit_utf8_iterator_is_eos (&iter) && lit_utf8_iterator_pos_cmp (iter_pos, locus) < 0)
{
ecma_char_t code_unit = lit_utf8_iterator_read_next (&iter);
iter_pos = lit_utf8_iterator_get_pos (&iter);
if (lit_char_is_line_terminator (code_unit))
{
if (code_unit == LIT_CHAR_CR
&& !lit_utf8_iterator_is_eos (&iter)
&& lit_utf8_iterator_peek_next (&iter) == LIT_CHAR_LF)
{
lit_utf8_iterator_incr (&iter);
}
c = 0;
l++;
continue;
}
c++;
}
if (line)
{
*line = l;
}
if (column)
{
*column = c;
}
} /* lexer_locus_to_line_and_column */
/**
* Recursive function for RegExp matching. Tests for a regular expression
* match and returns a MatchResult value.
*
* See also:
* ECMA-262 v5, 15.10.2.1
*
* @return completion value
* Returned value must be freed with ecma_free_completion_value
*/
static ecma_completion_value_t
re_match_regexp (re_matcher_ctx_t *re_ctx_p, /**< RegExp matcher context */
re_bytecode_t *bc_p, /**< pointer to the current RegExp bytecode */
lit_utf8_iterator_t iter, /**< input string iterator */
lit_utf8_iterator_t *out_iter_p) /**< Output: matching substring iterator */
{
ecma_completion_value_t ret_value = ecma_make_empty_completion_value ();
re_opcode_t op;
while ((op = re_get_opcode (&bc_p)))
{
switch (op)
{
case RE_OP_MATCH:
{
JERRY_DDLOG ("Execute RE_OP_MATCH: match\n");
*out_iter_p = iter;
ret_value = ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_TRUE);
return ret_value; /* match */
}
case RE_OP_CHAR:
{
if (lit_utf8_iterator_is_eos (&iter))
{
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
bool is_ignorecase = re_ctx_p->flags & RE_FLAG_IGNORE_CASE;
ecma_char_t ch1 = (ecma_char_t) re_get_value (&bc_p); /* Already canonicalized. */
ecma_char_t ch2 = re_canonicalize (lit_utf8_iterator_read_next (&iter), is_ignorecase);
JERRY_DDLOG ("Character matching %d to %d: ", ch1, ch2);
if (ch1 != ch2)
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
JERRY_DDLOG ("match\n");
break; /* tail merge */
}
case RE_OP_PERIOD:
{
if (lit_utf8_iterator_is_eos (&iter))
{
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
ecma_char_t ch = lit_utf8_iterator_read_next (&iter);
JERRY_DDLOG ("Period matching '.' to %d: ", (uint32_t) ch);
if (lit_char_is_line_terminator (ch))
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
JERRY_DDLOG ("match\n");
break; /* tail merge */
}
case RE_OP_ASSERT_START:
{
JERRY_DDLOG ("Execute RE_OP_ASSERT_START: ");
if ((iter.buf_p + iter.buf_pos.offset) <= re_ctx_p->input_start_p)
{
JERRY_DDLOG ("match\n");
break;
}
if (!(re_ctx_p->flags & RE_FLAG_MULTILINE))
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
if (lit_char_is_line_terminator (lit_utf8_iterator_peek_prev (&iter)))
{
JERRY_DDLOG ("match\n");
break;
}
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
case RE_OP_ASSERT_END:
{
JERRY_DDLOG ("Execute RE_OP_ASSERT_END: ");
if ((iter.buf_p + iter.buf_pos.offset) >= re_ctx_p->input_end_p)
{
JERRY_DDLOG ("match\n");
break; /* tail merge */
}
if (!(re_ctx_p->flags & RE_FLAG_MULTILINE))
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
if (lit_char_is_line_terminator (lit_utf8_iterator_peek_next (&iter)))
{
JERRY_DDLOG ("match\n");
break; /* tail merge */
}
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
case RE_OP_ASSERT_WORD_BOUNDARY:
case RE_OP_ASSERT_NOT_WORD_BOUNDARY:
{
bool is_wordchar_left, is_wordchar_right;
if ((iter.buf_p + iter.buf_pos.offset) <= re_ctx_p->input_start_p)
{
is_wordchar_left = false; /* not a wordchar */
}
else
{
is_wordchar_left = lit_char_is_word_char (lit_utf8_iterator_peek_prev (&iter));
}
if ((iter.buf_p + iter.buf_pos.offset) >= re_ctx_p->input_end_p)
{
is_wordchar_right = false; /* not a wordchar */
}
else
{
is_wordchar_right = lit_char_is_word_char (lit_utf8_iterator_peek_next (&iter));
}
if (op == RE_OP_ASSERT_WORD_BOUNDARY)
{
JERRY_DDLOG ("Execute RE_OP_ASSERT_WORD_BOUNDARY: ");
if (is_wordchar_left == is_wordchar_right)
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
}
else
{
JERRY_ASSERT (op == RE_OP_ASSERT_NOT_WORD_BOUNDARY);
JERRY_DDLOG ("Execute RE_OP_ASSERT_NOT_WORD_BOUNDARY: ");
if (is_wordchar_left != is_wordchar_right)
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
}
JERRY_DDLOG ("match\n");
break; /* tail merge */
}
case RE_OP_LOOKAHEAD_POS:
case RE_OP_LOOKAHEAD_NEG:
{
ecma_completion_value_t match_value = ecma_make_empty_completion_value ();
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
uint32_t array_size = re_ctx_p->num_of_captures + re_ctx_p->num_of_non_captures;
MEM_DEFINE_LOCAL_ARRAY (saved_bck_p, array_size, lit_utf8_iterator_t);
size_t size = (size_t) (array_size) * sizeof (lit_utf8_iterator_t);
memcpy (saved_bck_p, re_ctx_p->saved_p, size);
do
{
uint32_t offset = re_get_value (&bc_p);
if (!sub_iter.buf_p)
{
match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (ecma_is_completion_value_throw (match_value))
{
break;
}
}
bc_p += offset;
}
while (re_get_opcode (&bc_p) == RE_OP_ALTERNATIVE);
if (!ecma_is_completion_value_throw (match_value))
{
JERRY_DDLOG ("Execute RE_OP_LOOKAHEAD_POS/NEG: ");
ecma_free_completion_value (match_value);
if ((op == RE_OP_LOOKAHEAD_POS && sub_iter.buf_p)
|| (op == RE_OP_LOOKAHEAD_NEG && !sub_iter.buf_p))
{
JERRY_DDLOG ("match\n");
match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
}
else
{
JERRY_DDLOG ("fail\n");
match_value = ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
}
if (!ecma_is_completion_value_throw (match_value))
{
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
}
else
{
JERRY_ASSERT (ecma_is_value_boolean (match_value));
/* restore saved */
memcpy (re_ctx_p->saved_p, saved_bck_p, size);
}
}
MEM_FINALIZE_LOCAL_ARRAY (saved_bck_p);
return match_value;
}
case RE_OP_CHAR_CLASS:
case RE_OP_INV_CHAR_CLASS:
{
uint32_t num_of_ranges;
bool is_match;
JERRY_DDLOG ("Execute RE_OP_CHAR_CLASS/RE_OP_INV_CHAR_CLASS, ");
if (lit_utf8_iterator_is_eos (&iter))
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
bool is_ignorecase = re_ctx_p->flags & RE_FLAG_IGNORE_CASE;
ecma_char_t curr_ch = re_canonicalize (lit_utf8_iterator_read_next (&iter), is_ignorecase);
num_of_ranges = re_get_value (&bc_p);
is_match = false;
while (num_of_ranges)
{
ecma_char_t ch1 = re_canonicalize ((ecma_char_t) re_get_value (&bc_p), is_ignorecase);
ecma_char_t ch2 = re_canonicalize ((ecma_char_t) re_get_value (&bc_p), is_ignorecase);
JERRY_DDLOG ("num_of_ranges=%d, ch1=%d, ch2=%d, curr_ch=%d; ",
num_of_ranges, ch1, ch2, curr_ch);
if (curr_ch >= ch1 && curr_ch <= ch2)
{
/* We must read all the ranges from bytecode. */
is_match = true;
}
num_of_ranges--;
}
if (op == RE_OP_CHAR_CLASS)
{
if (!is_match)
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
}
else
{
JERRY_ASSERT (op == RE_OP_INV_CHAR_CLASS);
if (is_match)
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
}
JERRY_DDLOG ("match\n");
break; /* tail merge */
}
case RE_OP_BACKREFERENCE:
{
uint32_t backref_idx;
backref_idx = re_get_value (&bc_p);
JERRY_DDLOG ("Execute RE_OP_BACKREFERENCE (idx: %d): ", backref_idx);
backref_idx *= 2; /* backref n -> saved indices [n*2, n*2+1] */
JERRY_ASSERT (backref_idx >= 2 && backref_idx + 1 < re_ctx_p->num_of_captures);
if (!re_ctx_p->saved_p[backref_idx].buf_p || !re_ctx_p->saved_p[backref_idx + 1].buf_p)
{
JERRY_DDLOG ("match\n");
break; /* capture is 'undefined', always matches! */
}
lit_utf8_iterator_t sub_iter = re_ctx_p->saved_p[backref_idx];
while (sub_iter.buf_pos.offset < re_ctx_p->saved_p[backref_idx + 1].buf_pos.offset)
{
ecma_char_t ch1, ch2;
if ((iter.buf_p + iter.buf_pos.offset) >= re_ctx_p->input_end_p)
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
ch1 = lit_utf8_iterator_read_next (&sub_iter);
ch2 = lit_utf8_iterator_read_next (&iter);
if (ch1 != ch2)
{
JERRY_DDLOG ("fail\n");
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
}
JERRY_DDLOG ("match\n");
break; /* tail merge */
}
case RE_OP_SAVE_AT_START:
{
re_bytecode_t *old_bc_p;
JERRY_DDLOG ("Execute RE_OP_SAVE_AT_START\n");
lit_utf8_iterator_t old_start_p = re_ctx_p->saved_p[RE_GLOBAL_START_IDX];
re_ctx_p->saved_p[RE_GLOBAL_START_IDX] = iter;
do
{
uint32_t offset = re_get_value (&bc_p);
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
bc_p += offset;
old_bc_p = bc_p;
}
while (re_get_opcode (&bc_p) == RE_OP_ALTERNATIVE);
bc_p = old_bc_p;
re_ctx_p->saved_p[RE_GLOBAL_START_IDX] = old_start_p;
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
case RE_OP_SAVE_AND_MATCH:
{
JERRY_DDLOG ("End of pattern is reached: match\n");
re_ctx_p->saved_p[RE_GLOBAL_END_IDX] = iter;
*out_iter_p = iter;
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_TRUE); /* match */
}
case RE_OP_ALTERNATIVE:
{
/*
* Alternatives should be jump over, when alternative opcode appears.
*/
uint32_t offset = re_get_value (&bc_p);
JERRY_DDLOG ("Execute RE_OP_ALTERNATIVE");
bc_p += offset;
while (*bc_p == RE_OP_ALTERNATIVE)
{
JERRY_DDLOG (", jump: %d");
bc_p++;
offset = re_get_value (&bc_p);
bc_p += offset;
}
JERRY_DDLOG ("\n");
break; /* tail merge */
}
case RE_OP_CAPTURE_NON_GREEDY_ZERO_GROUP_START:
case RE_OP_NON_CAPTURE_NON_GREEDY_ZERO_GROUP_START:
{
/*
* On non-greedy iterations we have to execute the bytecode
* after the group first, if zero iteration is allowed.
*/
uint32_t start_idx, iter_idx, offset;
lit_utf8_iterator_t old_start = lit_utf8_iterator_create (NULL, 0);
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
re_bytecode_t *old_bc_p;
old_bc_p = bc_p; /* save the bytecode start position of the group start */
start_idx = re_get_value (&bc_p);
offset = re_get_value (&bc_p);
if (RE_IS_CAPTURE_GROUP (op))
{
JERRY_ASSERT (start_idx <= re_ctx_p->num_of_captures / 2);
iter_idx = start_idx - 1;
start_idx *= 2;
old_start = re_ctx_p->saved_p[start_idx];
re_ctx_p->saved_p[start_idx] = iter;
}
else
{
JERRY_ASSERT (start_idx < re_ctx_p->num_of_non_captures);
iter_idx = start_idx + (re_ctx_p->num_of_captures / 2) - 1;
start_idx += re_ctx_p->num_of_captures;
}
re_ctx_p->num_of_iterations_p[iter_idx] = 0;
/* Jump all over to the end of the END opcode. */
bc_p += offset;
/* Try to match after the close paren if zero is allowed */
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
if (RE_IS_CAPTURE_GROUP (op))
{
re_ctx_p->saved_p[start_idx] = old_start;
}
bc_p = old_bc_p;
/* FALLTHRU */
}
case RE_OP_CAPTURE_GROUP_START:
case RE_OP_CAPTURE_GREEDY_ZERO_GROUP_START:
case RE_OP_NON_CAPTURE_GROUP_START:
case RE_OP_NON_CAPTURE_GREEDY_ZERO_GROUP_START:
{
uint32_t start_idx, iter_idx, old_iteration_cnt, offset;
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
re_bytecode_t *old_bc_p;
re_bytecode_t *end_bc_p = NULL;
start_idx = re_get_value (&bc_p);
if (op != RE_OP_CAPTURE_GROUP_START
&& op != RE_OP_NON_CAPTURE_GROUP_START)
{
offset = re_get_value (&bc_p);
end_bc_p = bc_p + offset;
}
if (RE_IS_CAPTURE_GROUP (op))
{
JERRY_ASSERT (start_idx <= re_ctx_p->num_of_captures / 2);
iter_idx = start_idx - 1;
start_idx *= 2;
}
else
{
JERRY_ASSERT (start_idx < re_ctx_p->num_of_non_captures);
iter_idx = start_idx + (re_ctx_p->num_of_captures / 2) - 1;
start_idx += re_ctx_p->num_of_captures;
}
lit_utf8_iterator_t old_start = re_ctx_p->saved_p[start_idx];
old_iteration_cnt = re_ctx_p->num_of_iterations_p[iter_idx];
re_ctx_p->saved_p[start_idx] = iter;
re_ctx_p->num_of_iterations_p[iter_idx] = 0;
do
{
offset = re_get_value (&bc_p);
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
bc_p += offset;
old_bc_p = bc_p;
}
while (re_get_opcode (&bc_p) == RE_OP_ALTERNATIVE);
bc_p = old_bc_p;
re_ctx_p->num_of_iterations_p[iter_idx] = old_iteration_cnt;
/* Try to match after the close paren if zero is allowed. */
if (op == RE_OP_CAPTURE_GREEDY_ZERO_GROUP_START
|| op == RE_OP_NON_CAPTURE_GREEDY_ZERO_GROUP_START)
{
JERRY_ASSERT (end_bc_p);
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, end_bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
}
re_ctx_p->saved_p[start_idx] = old_start;
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
case RE_OP_CAPTURE_NON_GREEDY_GROUP_END:
case RE_OP_NON_CAPTURE_NON_GREEDY_GROUP_END:
{
uint32_t end_idx, iter_idx, min, max;
re_bytecode_t *old_bc_p;
/*
* On non-greedy iterations we have to execute the bytecode
* after the group first. Try to iterate only if it fails.
*/
old_bc_p = bc_p; /* save the bytecode start position of the group end */
end_idx = re_get_value (&bc_p);
min = re_get_value (&bc_p);
max = re_get_value (&bc_p);
re_get_value (&bc_p); /* start offset */
if (RE_IS_CAPTURE_GROUP (op))
{
JERRY_ASSERT (end_idx <= re_ctx_p->num_of_captures / 2);
iter_idx = end_idx - 1;
end_idx = (end_idx * 2) + 1;
}
else
{
JERRY_ASSERT (end_idx <= re_ctx_p->num_of_non_captures);
iter_idx = end_idx + (re_ctx_p->num_of_captures / 2) - 1;
end_idx += re_ctx_p->num_of_captures;
}
re_ctx_p->num_of_iterations_p[iter_idx]++;
if (re_ctx_p->num_of_iterations_p[iter_idx] >= min
&& re_ctx_p->num_of_iterations_p[iter_idx] <= max)
{
lit_utf8_iterator_t old_end = re_ctx_p->saved_p[end_idx];
re_ctx_p->saved_p[end_idx] = iter;
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
re_ctx_p->saved_p[end_idx] = old_end;
}
re_ctx_p->num_of_iterations_p[iter_idx]--;
bc_p = old_bc_p;
/* If non-greedy fails and try to iterate... */
/* FALLTHRU */
}
case RE_OP_CAPTURE_GREEDY_GROUP_END:
case RE_OP_NON_CAPTURE_GREEDY_GROUP_END:
{
uint32_t start_idx, end_idx, iter_idx, min, max, offset;
lit_utf8_iterator_t old_start = lit_utf8_iterator_create (NULL, 0);
lit_utf8_iterator_t old_end = lit_utf8_iterator_create (NULL, 0);
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
re_bytecode_t *old_bc_p;
end_idx = re_get_value (&bc_p);
min = re_get_value (&bc_p);
max = re_get_value (&bc_p);
offset = re_get_value (&bc_p);
if (RE_IS_CAPTURE_GROUP (op))
{
JERRY_ASSERT (end_idx <= re_ctx_p->num_of_captures / 2);
iter_idx = end_idx - 1;
start_idx = end_idx * 2;
end_idx = start_idx + 1;
}
else
{
JERRY_ASSERT (end_idx <= re_ctx_p->num_of_non_captures);
iter_idx = end_idx + (re_ctx_p->num_of_captures / 2) - 1;
end_idx += re_ctx_p->num_of_captures;
start_idx = end_idx;
}
/* Check the empty iteration if the minimum number of iterations is reached. */
if (re_ctx_p->num_of_iterations_p[iter_idx] >= min
&& iter.buf_p == re_ctx_p->saved_p[start_idx].buf_p
&& iter.buf_pos.offset == re_ctx_p->saved_p[start_idx].buf_pos.offset)
{
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
re_ctx_p->num_of_iterations_p[iter_idx]++;
old_bc_p = bc_p; /* Save the bytecode end position of the END opcodes for matching after it. */
old_end = re_ctx_p->saved_p[end_idx];
re_ctx_p->saved_p[end_idx] = iter;
if (re_ctx_p->num_of_iterations_p[iter_idx] < max)
{
bc_p -= offset;
offset = re_get_value (&bc_p);
old_start = re_ctx_p->saved_p[start_idx];
re_ctx_p->saved_p[start_idx] = iter;
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
re_ctx_p->saved_p[start_idx] = old_start;
/* Try to match alternatives if any. */
bc_p += offset;
while (*bc_p == RE_OP_ALTERNATIVE)
{
bc_p++; /* RE_OP_ALTERNATIVE */
offset = re_get_value (&bc_p);
old_start = re_ctx_p->saved_p[start_idx];
re_ctx_p->saved_p[start_idx] = iter;
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
re_ctx_p->saved_p[start_idx] = old_start;
bc_p += offset;
}
}
if (re_ctx_p->num_of_iterations_p[iter_idx] >= min
&& re_ctx_p->num_of_iterations_p[iter_idx] <= max)
{
/* Try to match the rest of the bytecode. */
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, old_bc_p, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
}
/* restore if fails */
re_ctx_p->saved_p[end_idx] = old_end;
re_ctx_p->num_of_iterations_p[iter_idx]--;
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
case RE_OP_NON_GREEDY_ITERATOR:
{
uint32_t min, max, offset, num_of_iter;
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
min = re_get_value (&bc_p);
max = re_get_value (&bc_p);
offset = re_get_value (&bc_p);
JERRY_DDLOG ("Non-greedy iterator, min=%lu, max=%lu, offset=%ld\n",
(unsigned long) min, (unsigned long) max, (long) offset);
num_of_iter = 0;
while (num_of_iter <= max)
{
if (num_of_iter >= min)
{
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p + offset, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
}
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (!ecma_is_value_true (match_value))
{
if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
break;
}
iter = sub_iter;
num_of_iter++;
}
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
case RE_OP_GREEDY_ITERATOR:
{
uint32_t min, max, offset, num_of_iter;
lit_utf8_iterator_t sub_iter = lit_utf8_iterator_create (NULL, 0);
min = re_get_value (&bc_p);
max = re_get_value (&bc_p);
offset = re_get_value (&bc_p);
JERRY_DDLOG ("Greedy iterator, min=%lu, max=%lu, offset=%ld\n",
(unsigned long) min, (unsigned long) max, (long) offset);
num_of_iter = 0;
while (num_of_iter < max)
{
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p, iter, &sub_iter);
if (!ecma_is_value_true (match_value))
{
if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
break;
}
iter = sub_iter;
num_of_iter++;
}
while (num_of_iter >= min)
{
ecma_completion_value_t match_value = re_match_regexp (re_ctx_p, bc_p + offset, iter, &sub_iter);
if (ecma_is_value_true (match_value))
{
*out_iter_p = sub_iter;
return match_value; /* match */
}
else if (ecma_is_completion_value_throw (match_value))
{
return match_value;
}
if (num_of_iter == min)
{
break;
}
lit_utf8_iterator_read_prev (&iter);
num_of_iter--;
}
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
}
default:
{
JERRY_DDLOG ("UNKNOWN opcode (%d)!\n", (uint32_t) op);
return ecma_make_throw_obj_completion_value (ecma_new_standard_error (ECMA_ERROR_COMMON));
}
}
}
JERRY_UNREACHABLE ();
return ecma_make_simple_completion_value (ECMA_SIMPLE_VALUE_FALSE); /* fail */
} /* regexp_match */
/**
* 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 */
/**
* Transforming escape sequences in the charset, outputting converted string to specified buffer
*
* Note:
* Size of string with transformed escape sequences is always
* less or equal to size of corresponding source string.
*
* @return size of converted string
*/
static lit_utf8_size_t
lexer_transform_escape_sequences (const jerry_api_char_t *source_str_p, /**< string to convert,
* located in source buffer */
lit_utf8_size_t source_str_size, /**< size of the string and of the output buffer */
jerry_api_char_t *output_str_buf_p) /**< output buffer for converted string */
{
if (source_str_size == 0)
{
return 0;
}
else
{
JERRY_ASSERT (source_str_p != NULL);
}
lit_utf8_byte_t *output_str_buf_iter_p = output_str_buf_p;
const size_t output_str_buf_size = source_str_size;
bool is_correct_sequence = true;
lit_utf8_iterator_t source_str_iter = lit_utf8_iterator_create (source_str_p,
source_str_size);
ecma_char_t prev_converted_char = LIT_CHAR_NULL;
while (!lit_utf8_iterator_is_eos (&source_str_iter))
{
ecma_char_t converted_char;
const ecma_char_t next_char = lit_utf8_iterator_read_next (&source_str_iter);
if (next_char == LIT_CHAR_BACKSLASH)
{
if (lit_utf8_iterator_is_eos (&source_str_iter))
{
is_correct_sequence = false;
break;
}
const ecma_char_t char_after_next = lit_utf8_iterator_read_next (&source_str_iter);
if (lit_char_is_decimal_digit (char_after_next))
{
if (lit_char_is_octal_digit (char_after_next))
{
if (char_after_next == LIT_CHAR_0
&& (lit_utf8_iterator_is_eos (&source_str_iter)
|| !lit_char_is_octal_digit (lit_utf8_iterator_peek_next (&source_str_iter))))
{
converted_char = LIT_CHAR_NULL;
}
else
{
/* Implementation-defined (ECMA-262 v5, B.1.2): octal escape sequences are not implemented */
is_correct_sequence = false;
break;
}
}
else
{
converted_char = char_after_next;
}
}
else if (char_after_next == LIT_CHAR_LOWERCASE_U
|| char_after_next == LIT_CHAR_LOWERCASE_X)
{
if (!lexer_convert_escape_sequence_digits_to_char (&source_str_iter,
char_after_next == LIT_CHAR_LOWERCASE_U,
&converted_char))
{
is_correct_sequence = false;
break;
}
}
else if (lit_char_is_line_terminator (char_after_next))
{
/* Skip \, followed by a LineTerminatorSequence (ECMA-262, v5, 7.3) */
if (char_after_next == LIT_CHAR_CR
&& !lit_utf8_iterator_is_eos (&source_str_iter)
&& lit_utf8_iterator_peek_next (&source_str_iter) == LIT_CHAR_LF)
{
lit_utf8_iterator_incr (&source_str_iter);
}
continue;
}
else
{
lexer_convert_single_escape_character (char_after_next, &converted_char);
}
}
else
{
converted_char = next_char;
}
if (lit_is_code_unit_high_surrogate (prev_converted_char)
&& lit_is_code_unit_low_surrogate (converted_char))
{
output_str_buf_iter_p -= LIT_UTF8_MAX_BYTES_IN_CODE_UNIT;
lit_code_point_t code_point = lit_convert_surrogate_pair_to_code_point (prev_converted_char,
converted_char);
output_str_buf_iter_p += lit_code_point_to_utf8 (code_point, output_str_buf_iter_p);
}
else
{
output_str_buf_iter_p += lit_code_unit_to_utf8 (converted_char, output_str_buf_iter_p);
JERRY_ASSERT (output_str_buf_iter_p <= output_str_buf_p + output_str_buf_size);
}
prev_converted_char = converted_char;
}
if (is_correct_sequence)
{
return (lit_utf8_size_t) (output_str_buf_iter_p - output_str_buf_p);
}
else
{
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "Illegal escape sequence", token_start_pos);
}
} /* lexer_transform_escape_sequences */
/**
* 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 */
/**
* Parse string literal (ECMA-262 v5, 7.8.5)
*/
static token
lexer_parse_regexp (void)
{
token result;
bool is_char_class = false;
JERRY_ASSERT (LA (0) == LIT_CHAR_SLASH);
new_token ();
/* Eat up '/' */
consume_char ();
while (true)
{
if (lit_utf8_iterator_is_eos (&src_iter))
{
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "Unterminated RegExp literal", token_start_pos);
}
ecma_char_t c = (ecma_char_t) LA (0);
if (lit_char_is_line_terminator (c))
{
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "RegExp literal should not contain newline character", token_start_pos);
}
else if (c == LIT_CHAR_BACKSLASH)
{
consume_char ();
if (lit_char_is_line_terminator (LA (0)))
{
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX,
"RegExp literal backslash sequence should not contain newline character",
token_start_pos);
}
}
else if (c == LIT_CHAR_LEFT_SQUARE)
{
is_char_class = true;
}
else if (c == LIT_CHAR_RIGHT_SQUARE)
{
is_char_class = false;
}
else if (c == LIT_CHAR_SLASH && !is_char_class)
{
/* Eat up '/' */
consume_char ();
break;
}
consume_char ();
}
/* Try to parse RegExp flags */
while (true)
{
ecma_char_t c = (ecma_char_t) LA (0);
if (!lit_char_is_word_char (c) || lit_char_is_line_terminator (c))
{
break;
}
consume_char ();
}
result = lexer_create_token_for_charset (TOK_REGEXP, TOK_START () + 1, TOK_SIZE () - 1);
is_token_parse_in_progress = false;
return result;
} /* lexer_parse_regexp */
/**
* Parse string literal (ECMA-262 v5, 7.8.4)
*/
static token
lexer_parse_string (void)
{
ecma_char_t c = (ecma_char_t) LA (0);
JERRY_ASSERT (c == LIT_CHAR_SINGLE_QUOTE
|| c == LIT_CHAR_DOUBLE_QUOTE);
new_token ();
/* Consume quote character */
consume_char ();
const ecma_char_t end_char = c;
bool is_escape_sequence_occured = false;
do
{
c = LA (0);
consume_char ();
if (lit_char_is_line_terminator (c))
{
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "String literal shall not contain newline character", token_start_pos);
}
else if (c == LIT_CHAR_BACKSLASH)
{
is_escape_sequence_occured = true;
ecma_char_t nc = (ecma_char_t) LA (0);
consume_char ();
if (nc == LIT_CHAR_CR)
{
nc = (ecma_char_t) LA (0);
if (nc == LIT_CHAR_LF)
{
consume_char ();
}
}
}
}
while (c != end_char && !lit_utf8_iterator_is_eos (&src_iter));
if (c == LIT_CHAR_NULL)
{
PARSE_ERROR (JSP_EARLY_ERROR_SYNTAX, "Unclosed string", token_start_pos);
}
const lit_utf8_size_t charset_size = TOK_SIZE () - 2;
token ret;
if (!is_escape_sequence_occured)
{
ret = lexer_create_token_for_charset (TOK_STRING,
TOK_START () + 1,
charset_size);
}
else
{
ret = lexer_create_token_for_charset_transform_escape_sequences (TOK_STRING,
TOK_START () + 1,
charset_size);
}
is_token_parse_in_progress = false;
return ret;
} /* lexer_parse_string */
/**
* 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 */
