encoding.c: Support the ISO 8859-15 encoding.

src/include/espeak-ng/espeak_ng.h

@@ -204,6 +204,7 @@ typedef enum
 	// ISO-8859-12 is not a valid encoding.
 	ESPEAKNG_ENCODING_ISO_8859_13,
 	ESPEAKNG_ENCODING_ISO_8859_14,
+	ESPEAKNG_ENCODING_ISO_8859_15,
 } espeak_ng_ENCODING;
 
 ESPEAK_NG_API espeak_ng_ENCODING

src/libespeak-ng/encoding.c

@@ -58,6 +58,7 @@ MNEM_TAB mnem_encoding[] = {
 	{ "ISO_8859-10:1992", ESPEAKNG_ENCODING_ISO_8859_10 },
 	{ "ISO_8859-14",      ESPEAKNG_ENCODING_ISO_8859_14 },
 	{ "ISO_8859-14:1998", ESPEAKNG_ENCODING_ISO_8859_14 },
+	{ "ISO_8859-15",      ESPEAKNG_ENCODING_ISO_8859_15 },
 	{ "ISO646-US",        ESPEAKNG_ENCODING_US_ASCII },
 	{ "ISO-8859-1",       ESPEAKNG_ENCODING_ISO_8859_1 },
 	{ "ISO-8859-2",       ESPEAKNG_ENCODING_ISO_8859_2 },
@@ -72,6 +73,8 @@ MNEM_TAB mnem_encoding[] = {
 	{ "ISO-8859-11",      ESPEAKNG_ENCODING_ISO_8859_11 },
 	{ "ISO-8859-13",      ESPEAKNG_ENCODING_ISO_8859_13 },
 	{ "ISO-8859-14",      ESPEAKNG_ENCODING_ISO_8859_14 },
+	{ "ISO-8859-15",      ESPEAKNG_ENCODING_ISO_8859_15 },
+	{ "Latin-9",          ESPEAKNG_ENCODING_ISO_8859_15 },
 	{ "TIS-620",          ESPEAKNG_ENCODING_ISO_8859_11 },
 	{ "US-ASCII",         ESPEAKNG_ENCODING_US_ASCII },
 	{ "cp367",            ESPEAKNG_ENCODING_US_ASCII },
@@ -79,6 +82,7 @@ MNEM_TAB mnem_encoding[] = {
 	{ "csASCII",          ESPEAKNG_ENCODING_US_ASCII },
 	{ "csISO885913",      ESPEAKNG_ENCODING_ISO_8859_13 },
 	{ "csISO885914",      ESPEAKNG_ENCODING_ISO_8859_14 },
+	{ "csISO885915",      ESPEAKNG_ENCODING_ISO_8859_15 },
 	{ "csISOLatin1",      ESPEAKNG_ENCODING_ISO_8859_1 },
 	{ "csISOLatin2",      ESPEAKNG_ENCODING_ISO_8859_2 },
 	{ "csISOLatin3",      ESPEAKNG_ENCODING_ISO_8859_3 },
@@ -394,6 +398,26 @@ static const uint32_t ISO_8859_14[0x80] = {
 	0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x0177, 0x00ff, // f8
 };
 
+// Reference: http://www.unicode.org/Public/MAPPINGS/ISO8859/8859-15.TXT
+static const uint32_t ISO_8859_15[0x80] = {
+	0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087, // 80
+	0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f, // 88
+	0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097, // 90
+	0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f, // 98
+	0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x20ac, 0x00a5, 0x0160, 0x00a7, // a0
+	0x0161, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af, // a8
+	0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x017d, 0x00b5, 0x00b6, 0x00b7, // b0
+	0x017e, 0x00b9, 0x00ba, 0x00bb, 0x0152, 0x0153, 0x0178, 0x00bf, // b8
+	0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7, // c0
+	0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf, // c8
+	0x00d0, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7, // d0
+	0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df, // d8
+	0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7, // e0
+	0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef, // e8
+	0x00f0, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7, // f0
+	0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x00ff, // f8
+};
+
 static uint32_t
 string_decoder_getc_us_ascii(espeak_ng_TEXT_DECODER *decoder)
 {
@@ -439,6 +463,7 @@ static const encoding_t string_decoders[] = {
 	// ISO-8859-12 is not a valid encoding.
 	{ string_decoder_getc_codepage, ISO_8859_13 },
 	{ string_decoder_getc_codepage, ISO_8859_14 },
+	{ string_decoder_getc_codepage, ISO_8859_15 },
 };
 
 espeak_ng_TEXT_DECODER *
@@ -466,7 +491,7 @@ text_decoder_decode_string(espeak_ng_TEXT_DECODER *decoder,
                            int length,
                            espeak_ng_ENCODING encoding)
 {
-	if (encoding > ESPEAKNG_ENCODING_ISO_8859_14)
+	if (encoding > ESPEAKNG_ENCODING_ISO_8859_15)
 		return ENS_UNKNOWN_TEXT_ENCODING;
 
 	const encoding_t *enc = string_decoders + encoding;

tests/encoding.c

@@ -489,6 +489,34 @@ test_iso_8859_14_encoding()
 	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);
+}
+
 int
 main(int argc, char **argv)
 {
@@ -511,6 +539,7 @@ main(int argc, char **argv)
 	// ISO-8859-12 is not a valid encoding.
 	test_iso_8859_13_encoding();
 	test_iso_8859_14_encoding();
+	test_iso_8859_15_encoding();
 
 	printf("done\n");