/**
* Verifies that the default encoding (UTF-16) works for wstrings in giop 1.1, which uses the
* length indicator to specify the number of characters rather than bytes and require a two-byte
* null terminator. Wide characters in 1.1 do not take width bytes
*/
@Test
public void testDefaultEncodingWCharGiop1_1() throws Exception {
byte[] codedText = {
0,
0,
0,
5, // string length in bytes, not chars
0x30,
(byte) (0xDF & 0xff), // Mitsubishi, in Katakana
0x30,
(byte) (0xC4 & 0xff),
0x30,
(byte) (0xFA & 0xff),
0x30,
(byte) (0xB7 & 0xff),
0,
0, // two-byte null terminator
0x5,
(byte) (0xD1 & 0xff), // Hebrew letter beis
};
CDRInputStream stream = new CDRInputStream(orb, codedText);
stream.setGIOPMinor(1);
assertEquals("wstring value", "\u30DF\u30C4\u30FA\u30B7", stream.read_wstring());
assertEquals("wchar 1", '\u05D1', stream.read_wchar());
stream.close();
}
/**
* Verifies that the UTF-8 works for wchar, wchar arrays, and wstrings. Reading the wstring forces
* alignment of the 4-byte length. Note that byte-ordering is fixed by the encoding.
*/
@Test
public void testUTF8EncodingWChar() throws Exception {
byte[] codedText = {
1,
'x', // Latin-l lowercase x
2,
(byte) (0xD7 & 0xff),
(byte) (0x90 & 0xff), // Hebrew letter aleph
3,
(byte) (0xE3 & 0xff),
(byte) (0x81 & 0xff),
(byte) (0x91 & 0xff), // Hiragana syllable ha
3,
(byte) (0xE3 & 0xff),
(byte) (0x81 & 0xff),
(byte) (0xB4 & 0xff), // Hiragana syllable pi
2,
(byte) (0xD7 & 0xff),
(byte) (0x91 & 0xff), // Hebrew letter beis
2,
(byte) (0xD7 & 0xff),
(byte) (0x92 & 0xff), // Hebrew letter gimmel
2,
(byte) (0xD7 & 0xff),
(byte) (0x93 & 0xff), // Hebrew letter dalet
45,
73, // bytes ignored by 'long' alignment
0,
0,
0,
12, // string length in bytes, not chars
(byte) (0xE3 & 0xff),
(byte) (0x83 & 0xff),
(byte) (0x9F & 0xff), // Mitsubishi, in Katakana
(byte) (0xE3 & 0xff),
(byte) (0x83 & 0xff),
(byte) (0x84 & 0xff),
(byte) (0xE3 & 0xff),
(byte) (0x83 & 0xff),
(byte) (0xBA & 0xff),
(byte) (0xE3 & 0xff),
(byte) (0x82 & 0xff),
(byte) (0xB7 & 0xff),
};
CDRInputStream stream = new CDRInputStream(orb, codedText);
selectCodeSets(stream, "ISO8859_1", "UTF8");
assertEquals("wchar 1", 'x', stream.read_wchar());
assertEquals("wchar 2", '\u05D0', stream.read_wchar());
assertEquals("wchar 3", '\u3051', stream.read_wchar());
assertEquals("wchar 4", '\u3074', stream.read_wchar());
char[] buffer = new char[4];
buffer[0] = '\u05D0';
stream.read_wchar_array(buffer, 1, 3);
assertEquals("wchar array", "\u05D0\u05D1\u05D2\u05D3", new String(buffer));
assertEquals("wstring value", "\u30DF\u30C4\u30FA\u30B7", stream.read_wstring());
stream.close();
}
/**
* Verifies that the default encoding (UTF-16) works for wchar, wchar arrays, and wstrings with no
* byte-order-marker. Reading the wstring forces alignment of the 4-byte length.
*/
@Test
public void testDefaultEncodingWChar() throws Exception {
byte[] codedText = {
2,
0x5,
(byte) (0xD0 & 0xff), // Hebrew letter aleph
2,
0x30,
0x51, // Hiragana syllable ha
2,
0x30,
0x74, // Hiragana syllable pi
2, // Hebrew letter beis: length byte
(byte) (0xFE & 0xff), // Big-endian indicator
(byte) (0xFF & 0xff),
0x5,
(byte) (0xD1 & 0xff), // UTF16 encoding (big-endian)
2, // Hebrew letter gimmel: length byte
(byte) (0xFF & 0xff), // Little-endian indicator
(byte) (0xFE & 0xff),
(byte) (0xD2 & 0xff),
0x5, // UTF16 encoding (little-endian)
2,
0x5,
(byte) (0xD3 & 0xff), // Hebrew letter dalet
45,
23, // bytes ignored by 'long' alignment
0,
0,
0,
8, // string length in bytes, not chars
0x30,
(byte) (0xDF & 0xff), // Mitsubishi, in Katakana
0x30,
(byte) (0xC4 & 0xff),
0x30,
(byte) (0xFA & 0xff),
0x30,
(byte) (0xB7 & 0xff),
};
CDRInputStream stream = new CDRInputStream(orb, codedText);
assertEquals("wchar 1", '\u05D0', stream.read_wchar());
assertEquals("wchar 2", '\u3051', stream.read_wchar());
assertEquals("wchar 3", '\u3074', stream.read_wchar());
char[] buffer = new char[4];
buffer[0] = '\u05D0';
stream.read_wchar_array(buffer, 1, 3);
assertEquals("wchar array", "\u05D0\u05D1\u05D2\u05D3", new String(buffer));
assertEquals("wstring value", "\u30DF\u30C4\u30FA\u30B7", stream.read_wstring());
stream.close();
}