void
test_char_decoder()
{
printf("testing char decoder\n");
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
// null string
assert(text_decoder_decode_string(decoder, NULL, -1, ESPEAKNG_ENCODING_ISO_8859_1) == ENS_OK);
assert(text_decoder_eof(decoder) == 1);
assert(text_decoder_getc(decoder) == 0);
assert(text_decoder_eof(decoder) == 1);
// string length
assert(text_decoder_decode_string(decoder, "aG", -1, ESPEAKNG_ENCODING_ISO_8859_1) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_10_encoding()
{
printf("testing ISO-8859-10 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-10") == ESPEAKNG_ENCODING_ISO_8859_10);
assert(espeak_ng_EncodingFromName("ISO_8859-10") == ESPEAKNG_ENCODING_ISO_8859_10);
assert(espeak_ng_EncodingFromName("ISO_8859-10:1992") == ESPEAKNG_ENCODING_ISO_8859_10);
assert(espeak_ng_EncodingFromName("iso-ir-157") == ESPEAKNG_ENCODING_ISO_8859_10);
assert(espeak_ng_EncodingFromName("latin6") == ESPEAKNG_ENCODING_ISO_8859_10);
assert(espeak_ng_EncodingFromName("l6") == ESPEAKNG_ENCODING_ISO_8859_10);
assert(espeak_ng_EncodingFromName("csISOLatin6") == ESPEAKNG_ENCODING_ISO_8859_10);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_ISO_8859_10) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x00de);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_8_encoding()
{
printf("testing ISO-8859-8 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-8") == ESPEAKNG_ENCODING_ISO_8859_8);
assert(espeak_ng_EncodingFromName("ISO_8859-8") == ESPEAKNG_ENCODING_ISO_8859_8);
assert(espeak_ng_EncodingFromName("ISO_8859-8:1988") == ESPEAKNG_ENCODING_ISO_8859_8);
assert(espeak_ng_EncodingFromName("iso-ir-138") == ESPEAKNG_ENCODING_ISO_8859_8);
assert(espeak_ng_EncodingFromName("hebrew") == ESPEAKNG_ENCODING_ISO_8859_8);
assert(espeak_ng_EncodingFromName("csISOLatinHebrew") == ESPEAKNG_ENCODING_ISO_8859_8);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xEE", 5, ESPEAKNG_ENCODING_ISO_8859_8) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x05de);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_7_encoding()
{
printf("testing ISO-8859-7 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-7") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("ISO_8859-7") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("ISO_8859-7:1987") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("iso-ir-126") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("ECMA-118") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("ELOT_928") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("greek") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("greek8") == ESPEAKNG_ENCODING_ISO_8859_7);
assert(espeak_ng_EncodingFromName("csISOLatinGreek") == ESPEAKNG_ENCODING_ISO_8859_7);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_ISO_8859_7) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x03AE);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_6_encoding()
{
printf("testing ISO-8859-6 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-6") == ESPEAKNG_ENCODING_ISO_8859_6);
assert(espeak_ng_EncodingFromName("ISO_8859-6") == ESPEAKNG_ENCODING_ISO_8859_6);
assert(espeak_ng_EncodingFromName("ISO_8859-6:1987") == ESPEAKNG_ENCODING_ISO_8859_6);
assert(espeak_ng_EncodingFromName("iso-ir-127") == ESPEAKNG_ENCODING_ISO_8859_6);
assert(espeak_ng_EncodingFromName("ECMA-114") == ESPEAKNG_ENCODING_ISO_8859_6);
assert(espeak_ng_EncodingFromName("ASMO-708") == ESPEAKNG_ENCODING_ISO_8859_6);
assert(espeak_ng_EncodingFromName("arabic") == ESPEAKNG_ENCODING_ISO_8859_6);
assert(espeak_ng_EncodingFromName("csISOLatinArabic") == ESPEAKNG_ENCODING_ISO_8859_6);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDA", 5, ESPEAKNG_ENCODING_ISO_8859_6) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x063A);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_5_encoding()
{
printf("testing ISO-8859-5 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-5") == ESPEAKNG_ENCODING_ISO_8859_5);
assert(espeak_ng_EncodingFromName("ISO_8859-5") == ESPEAKNG_ENCODING_ISO_8859_5);
assert(espeak_ng_EncodingFromName("ISO_8859-5:1988") == ESPEAKNG_ENCODING_ISO_8859_5);
assert(espeak_ng_EncodingFromName("iso-ir-144") == ESPEAKNG_ENCODING_ISO_8859_5);
assert(espeak_ng_EncodingFromName("cyrillic") == ESPEAKNG_ENCODING_ISO_8859_5);
assert(espeak_ng_EncodingFromName("csISOLatinCyrillic") == ESPEAKNG_ENCODING_ISO_8859_5);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_ISO_8859_5) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x043E);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_11_encoding()
{
printf("testing ISO-8859-11 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-11") == ESPEAKNG_ENCODING_ISO_8859_11);
assert(espeak_ng_EncodingFromName("TIS-620") == ESPEAKNG_ENCODING_ISO_8859_11);
assert(espeak_ng_EncodingFromName("csTIS620") == ESPEAKNG_ENCODING_ISO_8859_11);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xEE", 5, ESPEAKNG_ENCODING_ISO_8859_11) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0e4e);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_14_encoding()
{
printf("testing ISO-8859-14 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-14") == ESPEAKNG_ENCODING_ISO_8859_14);
assert(espeak_ng_EncodingFromName("ISO_8859-14") == ESPEAKNG_ENCODING_ISO_8859_14);
assert(espeak_ng_EncodingFromName("ISO_8859-14:1998") == ESPEAKNG_ENCODING_ISO_8859_14);
assert(espeak_ng_EncodingFromName("iso-ir-199") == ESPEAKNG_ENCODING_ISO_8859_14);
assert(espeak_ng_EncodingFromName("iso-celtic") == ESPEAKNG_ENCODING_ISO_8859_14);
assert(espeak_ng_EncodingFromName("latin8") == ESPEAKNG_ENCODING_ISO_8859_14);
assert(espeak_ng_EncodingFromName("l8") == ESPEAKNG_ENCODING_ISO_8859_14);
assert(espeak_ng_EncodingFromName("csISO885914") == ESPEAKNG_ENCODING_ISO_8859_14);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_ISO_8859_14) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0176);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_koi8_r_encoding()
{
printf("testing KOI8-R encoding\n");
assert(espeak_ng_EncodingFromName("KOI8-R") == ESPEAKNG_ENCODING_KOI8_R);
assert(espeak_ng_EncodingFromName("csKOI8R") == ESPEAKNG_ENCODING_KOI8_R);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_KOI8_R) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x021a);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_peekc()
{
printf("testing peekc\n");
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aGd", 3, ESPEAKNG_ENCODING_US_ASCII) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_peekc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_peekc(decoder) == 'd');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'd');
assert(text_decoder_eof(decoder) == 1);
// Calling peekc past the end of the buffer.
assert(text_decoder_peekc(decoder) == '\0');
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_10646_ucs_2_encoding()
{
printf("testing ISO-10646-UCS-2 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-10646-UCS-2") == ESPEAKNG_ENCODING_ISO_10646_UCS_2);
assert(espeak_ng_EncodingFromName("csUnicode") == ESPEAKNG_ENCODING_ISO_10646_UCS_2);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "a\00G\00\xA0\00\x22\x21\x23\x21", 9, ESPEAKNG_ENCODING_ISO_10646_UCS_2) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x2122);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_us_ascii_encoding()
{
printf("testing US-ASCII encoding\n");
assert(espeak_ng_EncodingFromName("US-ASCII") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("iso-ir-6") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("ANSI_X3.4-1968") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("ANSI_X3.4-1986") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("ISO_646.irv:1991") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("ISO646-US") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("us") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("IBM367") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("cp367") == ESPEAKNG_ENCODING_US_ASCII);
assert(espeak_ng_EncodingFromName("csASCII") == ESPEAKNG_ENCODING_US_ASCII);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_US_ASCII) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_16_encoding()
{
printf("testing ISO-8859-16 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-16") == ESPEAKNG_ENCODING_ISO_8859_16);
assert(espeak_ng_EncodingFromName("ISO_8859-16") == ESPEAKNG_ENCODING_ISO_8859_16);
assert(espeak_ng_EncodingFromName("ISO_8859-16:2001") == ESPEAKNG_ENCODING_ISO_8859_16);
assert(espeak_ng_EncodingFromName("iso-ir-226") == ESPEAKNG_ENCODING_ISO_8859_16);
assert(espeak_ng_EncodingFromName("latin10") == ESPEAKNG_ENCODING_ISO_8859_16);
assert(espeak_ng_EncodingFromName("l10") == ESPEAKNG_ENCODING_ISO_8859_16);
assert(espeak_ng_EncodingFromName("csISO885916") == ESPEAKNG_ENCODING_ISO_8859_16);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_ISO_8859_16) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x021a);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iso_8859_15_encoding()
{
printf("testing ISO-8859-15 encoding\n");
assert(espeak_ng_EncodingFromName("ISO-8859-15") == ESPEAKNG_ENCODING_ISO_8859_15);
assert(espeak_ng_EncodingFromName("ISO_8859-15") == ESPEAKNG_ENCODING_ISO_8859_15);
assert(espeak_ng_EncodingFromName("Latin-9") == ESPEAKNG_ENCODING_ISO_8859_15);
assert(espeak_ng_EncodingFromName("csISO885915") == ESPEAKNG_ENCODING_ISO_8859_15);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xBE", 5, ESPEAKNG_ENCODING_ISO_8859_15) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x92);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xA0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0178);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
static espeak_ng_STATUS
set_text(const char *text, const char *voicename)
{
espeak_ng_STATUS status = espeak_ng_SetVoiceByName(voicename);
if (status != ENS_OK)
return status;
if (p_decoder == NULL)
p_decoder = create_text_decoder();
count_characters = 0;
return text_decoder_decode_string(p_decoder, text, -1, ESPEAKNG_ENCODING_UTF_8);
}
void
test_unknown_encoding()
{
printf("testing unknown encodings\n");
assert(espeak_ng_EncodingFromName(NULL) == ESPEAKNG_ENCODING_UNKNOWN);
assert(espeak_ng_EncodingFromName("") == ESPEAKNG_ENCODING_UNKNOWN);
assert(espeak_ng_EncodingFromName("abcxyz") == ESPEAKNG_ENCODING_UNKNOWN);
assert(espeak_ng_EncodingFromName("US") == ESPEAKNG_ENCODING_UNKNOWN); // wrong case
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_UNKNOWN) == ENS_UNKNOWN_TEXT_ENCODING);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_iscii_encoding()
{
printf("testing ISCII encoding\n");
assert(espeak_ng_EncodingFromName("ISCII") == ESPEAKNG_ENCODING_ISCII);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xE6", 5, ESPEAKNG_ENCODING_ISCII) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'a');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 'G');
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xfffd);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xfffd);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x094c);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
void
test_utf_8_encoding()
{
printf("testing UTF-8 encoding\n");
assert(espeak_ng_EncodingFromName("UTF-8") == ESPEAKNG_ENCODING_UTF_8);
assert(espeak_ng_EncodingFromName("csUTF8") == ESPEAKNG_ENCODING_UTF_8);
espeak_ng_TEXT_DECODER *decoder = create_text_decoder();
// 1-byte UTF-8 sequences
assert(text_decoder_decode_string(decoder, "\x0D\x1E\x20\x35\x42\x57\x65\x77", 8, ESPEAKNG_ENCODING_UTF_8) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x000D);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x001E);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0020);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0035);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0042);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0057);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0065);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0077);
assert(text_decoder_eof(decoder) == 1);
// UTF-8 tail bytes without an initial length indicator character
assert(text_decoder_decode_string(decoder, "\x84\x92\xA8\xB5", 4, ESPEAKNG_ENCODING_UTF_8) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 1);
// 2-byte UTF-8 sequences
assert(text_decoder_decode_string(decoder, "\xC2\xA0\xD0\xB0\xC5\x65\xC2\xA0", 7, ESPEAKNG_ENCODING_UTF_8) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x00A0);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0430);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // \x65 is not a continuation byte, so \xC5 is invalid
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0065);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // incomplete: \xA0 is past the end of the string
assert(text_decoder_eof(decoder) == 1);
// 3-byte UTF-8 sequences
assert(text_decoder_decode_string(decoder, "\xE4\xBA\x8C\xE8\x42\xE2\x93\x44\xE4\xA0\x80", 9, ESPEAKNG_ENCODING_UTF_8) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x4E8C);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // \x42 is not a continuation byte, so \xE8 is invalid
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0042);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // \x44 is not a continuation byte, so \xE2\x93 is invalid
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0044);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // incomplete: \xA0 is past the end of the string
assert(text_decoder_eof(decoder) == 1);
// 4-byte UTF-8 sequences
assert(text_decoder_decode_string(decoder, "\xF0\x90\x8C\x82\xF4\x8F\xBF\xBF\xF3\x61\xF3\xA5\x32\xF3\x87\xB2\x36\xF1\xA0\x80\x80", 18, ESPEAKNG_ENCODING_UTF_8) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x10302);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x10FFFF);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // \x61 is not a continuation byte, so \xF3 is invalid
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0061);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // \x32 is not a continuation byte, so \xF3\xA5 is invalid
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0032);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // \x36 is not a continuation byte, so \xF3\x87\xB2 is invalid
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0x0036);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD); // incomplete: \xA0 is past the end of the string
assert(text_decoder_eof(decoder) == 1);
// out of range (> 0x10FFFF)
assert(text_decoder_decode_string(decoder, "\xF4\x90\x80\x80", 4, ESPEAKNG_ENCODING_UTF_8) == ENS_OK);
assert(text_decoder_eof(decoder) == 0);
assert(text_decoder_getc(decoder) == 0xFFFD);
assert(text_decoder_eof(decoder) == 1);
destroy_text_decoder(decoder);
}
