/**
* Check if a literal which holds the passed string exists.
* If it doesn't exist, create a new one.
*
* @return pointer to existing or newly created record
*/
literal_t
lit_find_or_create_literal_from_utf8_string (const lit_utf8_byte_t *str_p, /**< string, could be non-zero-terminated */
lit_utf8_size_t str_size) /**< length of the string */
{
literal_t lit = lit_find_literal_by_utf8_string (str_p, str_size);
if (lit == NULL)
{
lit = lit_create_literal_from_utf8_string (str_p, str_size);
}
return lit;
} /* lit_find_or_create_literal_from_utf8_string */
/**
* Convert string to token of specified type
*
* @return token descriptor
*/
static token
convert_string_to_token (token_type tt, /**< token type */
const lit_utf8_byte_t *str_p, /**< characters buffer */
lit_utf8_size_t length) /**< string's length */
{
JERRY_ASSERT (str_p != NULL);
literal_t lit = lit_find_literal_by_utf8_string (str_p, length);
if (lit != NULL)
{
return create_token_from_lit (tt, lit);
}
lit = lit_create_literal_from_utf8_string (str_p, length);
JERRY_ASSERT (lit->get_type () == LIT_STR_T
|| lit->get_type () == LIT_MAGIC_STR_T
|| lit->get_type () == LIT_MAGIC_STR_EX_T);
return create_token_from_lit (tt, lit);
}
/**
* Create token of specified type from charset
*
* @return token descriptor
*/
static token
lexer_create_token_for_charset (token_type tt, /**< token type */
const lit_utf8_byte_t *charset_p, /**< charset buffer */
lit_utf8_size_t size) /**< size of the charset */
{
JERRY_ASSERT (charset_p != NULL);
literal_t lit = lit_find_literal_by_utf8_string (charset_p, size);
if (lit != NULL)
{
return create_token_from_lit (tt, lit);
}
lit = lit_create_literal_from_utf8_string (charset_p, size);
JERRY_ASSERT (lit->get_type () == LIT_STR_T
|| lit->get_type () == LIT_MAGIC_STR_T
|| lit->get_type () == LIT_MAGIC_STR_EX_T);
return create_token_from_lit (tt, lit);
} /* lexer_create_token_for_charset */
int
main (int __attr_unused___ argc,
char __attr_unused___ **argv)
{
TEST_INIT ();
const lit_utf8_byte_t *ptrs[test_sub_iters];
ecma_number_t numbers[test_sub_iters];
lit_utf8_byte_t strings[test_sub_iters][max_characters_in_string + 1];
lit_utf8_size_t lengths[test_sub_iters];
jmem_init ();
lit_init ();
for (uint32_t i = 0; i < test_iters; i++)
{
memset (numbers, 0, sizeof (ecma_number_t) * test_sub_iters);
memset (lengths, 0, sizeof (lit_utf8_size_t) * test_sub_iters);
memset (ptrs, 0, sizeof (lit_utf8_byte_t *) * test_sub_iters);
for (uint32_t j = 0; j < test_sub_iters; j++)
{
int type = rand () % 3;
if (type == 0)
{
lengths[j] = (lit_utf8_size_t) (rand () % max_characters_in_string + 1);
generate_string (strings[j], lengths[j]);
lit_create_literal_from_utf8_string (strings[j], lengths[j]);
strings[j][lengths[j]] = '\0';
ptrs[j] = strings[j];
JERRY_ASSERT (ptrs[j]);
}
else if (type == 1)
{
lit_magic_string_id_t msi = (lit_magic_string_id_t) (rand () % LIT_MAGIC_STRING__COUNT);
ptrs[j] = lit_get_magic_string_utf8 (msi);
JERRY_ASSERT (ptrs[j]);
lengths[j] = (lit_utf8_size_t) lit_zt_utf8_string_size (ptrs[j]);
lit_create_literal_from_utf8_string (ptrs[j], lengths[j]);
}
else
{
ecma_number_t num = generate_number ();
lengths[j] = ecma_number_to_utf8_string (num, strings[j], max_characters_in_string);
lit_create_literal_from_num (num);
}
}
// Add empty string
lit_create_literal_from_utf8_string (NULL, 0);
for (uint32_t j = 0; j < test_sub_iters; j++)
{
lit_literal_t lit1;
lit_literal_t lit2;
if (ptrs[j])
{
lit1 = lit_find_or_create_literal_from_utf8_string (ptrs[j], lengths[j]);
lit2 = lit_find_literal_by_utf8_string (ptrs[j], lengths[j]);
JERRY_ASSERT (compare_utf8_string_and_string_literal (ptrs[j], lengths[j], lit1));
JERRY_ASSERT (compare_utf8_string_and_string_literal (ptrs[j], lengths[j], lit2));
}
else
{
lit1 = lit_find_or_create_literal_from_num (numbers[j]);
lit2 = lit_find_literal_by_num (numbers[j]);
JERRY_ASSERT (numbers[j] == lit_number_literal_get_number (lit1));
JERRY_ASSERT (numbers[j] == lit_number_literal_get_number (lit2));
}
JERRY_ASSERT (lit1);
JERRY_ASSERT (lit2);
JERRY_ASSERT (lit1 == lit2);
}
// Check empty string exists
JERRY_ASSERT (lit_find_literal_by_utf8_string (NULL, 0));
}
lit_finalize ();
jmem_finalize (true);
return 0;
} /* main */
/**
* Create token of specified type from charset
*
* @return token descriptor
*/
static token
lexer_create_token_for_charset (jsp_token_type_t tt, /**< token type */
const lit_utf8_byte_t *charset_p, /**< charset buffer */
lit_utf8_size_t size) /**< size of the charset */
{
JERRY_ASSERT (charset_p != NULL);
lit_utf8_iterator_t iter = lit_utf8_iterator_create (charset_p, (lit_utf8_size_t) size);
lit_utf8_size_t new_size = 0;
lit_utf8_size_t new_length = 0;
bool should_convert = false;
while (!lit_utf8_iterator_is_eos (&iter))
{
if (iter.buf_pos.is_non_bmp_middle)
{
should_convert = true;
}
lit_utf8_iterator_incr (&iter);
new_size += LIT_CESU8_MAX_BYTES_IN_CODE_UNIT;
}
lit_utf8_byte_t *converted_str_p;
if (unlikely (should_convert))
{
lit_utf8_iterator_seek_bos (&iter);
converted_str_p = (lit_utf8_byte_t *) jsp_mm_alloc (new_size);
while (!lit_utf8_iterator_is_eos (&iter))
{
ecma_char_t ch = lit_utf8_iterator_read_next (&iter);
new_length += lit_code_unit_to_utf8 (ch, converted_str_p + new_length);
}
}
else
{
converted_str_p = (lit_utf8_byte_t *) charset_p;
new_length = size;
JERRY_ASSERT (lit_is_cesu8_string_valid (converted_str_p, new_length));
}
lit_literal_t lit = lit_find_literal_by_utf8_string (converted_str_p, new_length);
if (lit != NULL)
{
if (unlikely (should_convert))
{
jsp_mm_free (converted_str_p);
}
return create_token_from_lit (tt, lit);
}
lit = lit_create_literal_from_utf8_string (converted_str_p, new_length);
rcs_record_type_t type = rcs_record_get_type (lit);
JERRY_ASSERT (RCS_RECORD_TYPE_IS_CHARSET (type)
|| RCS_RECORD_TYPE_IS_MAGIC_STR (type)
|| RCS_RECORD_TYPE_IS_MAGIC_STR_EX (type));
if (unlikely (should_convert))
{
jsp_mm_free (converted_str_p);
}
return create_token_from_lit (tt, lit);
} /* lexer_create_token_for_charset */
