static tb_object_ref_t tb_object_xplist_reader_func_string(tb_object_xplist_reader_t* reader, tb_size_t event)
{
// check
tb_assert_and_check_return_val(reader && reader->reader && event, tb_null);
// empty?
if (event == TB_XML_READER_EVENT_ELEMENT_EMPTY)
return tb_object_string_init_from_cstr(tb_null);
// done
tb_bool_t leave = tb_false;
tb_object_ref_t string = tb_null;
while (!leave && (event = tb_xml_reader_next(reader->reader)))
{
switch (event)
{
case TB_XML_READER_EVENT_ELEMENT_END:
{
// name
tb_char_t const* name = tb_xml_reader_element(reader->reader);
tb_assert_and_check_break_state(name, leave, tb_true);
// is end?
if (!tb_stricmp(name, "string"))
{
// empty?
if (!string) string = tb_object_string_init_from_cstr(tb_null);
// leave it
leave = tb_true;
}
}
break;
case TB_XML_READER_EVENT_TEXT:
{
// text
tb_char_t const* text = tb_xml_reader_text(reader->reader);
tb_assert_and_check_break_state(text, leave, tb_true);
tb_trace_d("string: %s", text);
// string
string = tb_object_string_init_from_cstr(text);
tb_assert_and_check_break_state(string, leave, tb_true);
}
break;
default:
break;
}
}
// ok?
return string;
}
static tb_object_ref_t tb_object_bin_reader_func_string(tb_object_bin_reader_t* reader, tb_size_t type, tb_uint64_t size)
{
// check
tb_assert_and_check_return_val(reader && reader->stream && reader->list, tb_null);
// empty?
if (!size) return tb_object_string_init_from_cstr(tb_null);
// make data
tb_char_t* data = tb_malloc0_cstr((tb_size_t)size + 1);
tb_assert_and_check_return_val(data, tb_null);
// read data
if (!tb_stream_bread(reader->stream, (tb_byte_t*)data, (tb_size_t)size))
{
tb_free(data);
return tb_null;
}
// decode string
{
tb_byte_t* pb = (tb_byte_t*)data;
tb_byte_t* pe = (tb_byte_t*)data + size;
tb_byte_t xb = (tb_byte_t)(((size >> 8) & 0xff) | (size & 0xff));
for (; pb < pe; pb++, xb++) *pb ^= xb;
}
// make string
tb_object_ref_t string = tb_object_string_init_from_cstr(data);
// exit data
tb_free(data);
// ok?
return string;
}
static tb_object_ref_t tb_object_json_reader_func_string(tb_object_json_reader_t* reader, tb_char_t type)
{
// check
tb_assert_and_check_return_val(reader && reader->stream && (type == '\"' || type == '\''), tb_null);
// init data
tb_string_t data;
if (!tb_string_init(&data)) return tb_null;
// walk
tb_char_t ch;
while (tb_stream_left(reader->stream))
{
// read one character
ch = tb_stream_bread_s8(reader->stream);
// end?
if (ch == '\"' || ch == '\'') break;
// the escaped character?
else if (ch == '\\')
{
// read one character
ch = tb_stream_bread_s8(reader->stream);
// unicode?
if (ch == 'u')
{
#ifdef TB_CONFIG_MODULE_HAVE_CHARSET
// the unicode string
tb_char_t unicode_str[5];
unicode_str[0] = tb_stream_bread_s8(reader->stream);
unicode_str[1] = tb_stream_bread_s8(reader->stream);
unicode_str[2] = tb_stream_bread_s8(reader->stream);
unicode_str[3] = tb_stream_bread_s8(reader->stream);
unicode_str[4] = '\0';
// the unicode value
tb_uint16_t unicode_val = tb_s16toi32(unicode_str);
// the utf8 stream
tb_char_t utf8_data[16] = {0};
tb_static_stream_t utf8_stream;
tb_static_stream_init(&utf8_stream, (tb_byte_t*)utf8_data, sizeof(utf8_data));
// the unicode stream
tb_static_stream_t unicode_stream = {0};
tb_static_stream_init(&unicode_stream, (tb_byte_t*)&unicode_val, 2);
// unicode to utf8
tb_long_t utf8_size = tb_charset_conv_bst(TB_CHARSET_TYPE_UCS2 | TB_CHARSET_TYPE_NE, TB_CHARSET_TYPE_UTF8, &unicode_stream, &utf8_stream);
if (utf8_size > 0) tb_string_cstrncat(&data, utf8_data, utf8_size);
#else
// trace
tb_trace1_e("unicode type is not supported, please enable charset module config if you want to use it!");
// only append it
tb_string_chrcat(&data, ch);
#endif
}
// append escaped character
else tb_string_chrcat(&data, ch);
}
// append character
else tb_string_chrcat(&data, ch);
}
// init string
tb_object_ref_t string = tb_object_string_init_from_cstr(tb_string_cstr(&data));
// trace
tb_trace_d("string: %s", tb_string_cstr(&data));
// exit data
tb_string_exit(&data);
// ok?
return string;
}
static tb_object_ref_t tb_object_bplist_reader_func_string(tb_object_bplist_reader_t* reader, tb_size_t type, tb_size_t size, tb_size_t item_size)
{
// check
tb_assert_and_check_return_val(reader && reader->stream, tb_null);
// init
tb_char_t* utf8 = tb_null;
tb_char_t* utf16 = tb_null;
tb_object_ref_t object = tb_null;
// read
switch (type)
{
case TB_OBJECT_BPLIST_TYPE_STRING:
{
// size is too large?
if (size == 0x0f)
{
// read size
tb_long_t val = tb_object_bplist_reader_func_size(reader, item_size);
tb_assert_and_check_return_val(val >= 0, tb_null);
size = (tb_size_t)val;
}
// read string
if (size)
{
// init utf8
utf8 = tb_malloc_cstr(size + 1);
tb_assert_and_check_break(utf8);
// read utf8
if (!tb_stream_bread(reader->stream, (tb_byte_t*)utf8, size)) break;
utf8[size] = '\0';
}
// init object
object = tb_object_string_init_from_cstr(utf8);
}
break;
case TB_OBJECT_BPLIST_TYPE_UNICODE:
{
#ifdef TB_CONFIG_MODULE_HAVE_CHARSET
// size is too large?
if (size == 0x0f)
{
// read size
tb_long_t val = tb_object_bplist_reader_func_size(reader, item_size);
tb_assert_and_check_return_val(val >= 0, tb_null);
size = (tb_size_t)val;
}
// read string
if (size)
{
// init utf8 & utf16 data
utf8 = tb_malloc_cstr((size + 1) << 2);
utf16 = tb_malloc_cstr(size << 1);
tb_assert_and_check_break(utf8 && utf16);
// read utf16
if (!tb_stream_bread(reader->stream, (tb_byte_t*)utf16, size << 1)) break;
// utf16 to utf8
tb_long_t osize = tb_charset_conv_data(TB_CHARSET_TYPE_UTF16, TB_CHARSET_TYPE_UTF8, (tb_byte_t*)utf16, size << 1, (tb_byte_t*)utf8, (size + 1) << 2);
tb_assert_and_check_break(osize > 0 && osize < (tb_long_t)((size + 1) << 2));
utf8[osize] = '\0';
// init object
object = tb_object_string_init_from_cstr(utf8);
}
#else
// trace
tb_trace1_e("unicode type is not supported, please enable charset module config if you want to use it!");
#endif
}
break;
default:
break;
}
// exit
if (utf8) tb_free(utf8);
if (utf16) tb_free(utf16);
// ok?
return object;
}
