encoding.c: Support the ISO 8859-5 encoding.

espeak-ng · Mar 28, 2017 · 51295d9 · 51295d9
1 parent b5589fc
commit 51295d9
Show file tree

Hide file tree

Showing 3 changed files with 63 additions and 0 deletions.
diff --git a/src/include/espeak-ng/espeak_ng.h b/src/include/espeak-ng/espeak_ng.h
@@ -194,6 +194,7 @@ typedef enum
 	ESPEAKNG_ENCODING_ISO_8859_2,
 	ESPEAKNG_ENCODING_ISO_8859_3,
 	ESPEAKNG_ENCODING_ISO_8859_4,
+	ESPEAKNG_ENCODING_ISO_8859_5,
 } espeak_ng_ENCODING;
 
 ESPEAK_NG_API espeak_ng_ENCODING

diff --git a/src/libespeak-ng/encoding.c b/src/libespeak-ng/encoding.c
@@ -36,6 +36,7 @@ MNEM_TAB mnem_encoding[] = {
 	{ "ISO_8859-2",       ESPEAKNG_ENCODING_ISO_8859_2 },
 	{ "ISO_8859-3",       ESPEAKNG_ENCODING_ISO_8859_3 },
 	{ "ISO_8859-4",       ESPEAKNG_ENCODING_ISO_8859_4 },
+	{ "ISO_8859-5",       ESPEAKNG_ENCODING_ISO_8859_5 },
 	{ "ISO646-US",        ESPEAKNG_ENCODING_US_ASCII },
 	{ "ISO-8859-1",       ESPEAKNG_ENCODING_ISO_8859_1 },
 	{ "ISO-8859-1:1987",  ESPEAKNG_ENCODING_ISO_8859_1 },
@@ -45,6 +46,8 @@ MNEM_TAB mnem_encoding[] = {
 	{ "ISO-8859-3:1988",  ESPEAKNG_ENCODING_ISO_8859_3 },
 	{ "ISO-8859-4",       ESPEAKNG_ENCODING_ISO_8859_4 },
 	{ "ISO-8859-4:1988",  ESPEAKNG_ENCODING_ISO_8859_4 },
+	{ "ISO-8859-5",       ESPEAKNG_ENCODING_ISO_8859_5 },
+	{ "ISO-8859-5:1988",  ESPEAKNG_ENCODING_ISO_8859_5 },
 	{ "US-ASCII",         ESPEAKNG_ENCODING_US_ASCII },
 	{ "cp367",            ESPEAKNG_ENCODING_US_ASCII },
 	{ "cp819",            ESPEAKNG_ENCODING_ISO_8859_1 },
@@ -53,11 +56,14 @@ MNEM_TAB mnem_encoding[] = {
 	{ "csISOLatin2",      ESPEAKNG_ENCODING_ISO_8859_2 },
 	{ "csISOLatin3",      ESPEAKNG_ENCODING_ISO_8859_3 },
 	{ "csISOLatin4",      ESPEAKNG_ENCODING_ISO_8859_4 },
+	{ "csISOLatinCyrillic",ESPEAKNG_ENCODING_ISO_8859_5 },
+	{ "cyrillic",         ESPEAKNG_ENCODING_ISO_8859_5 },
 	{ "iso-ir-6",         ESPEAKNG_ENCODING_US_ASCII },
 	{ "iso-ir-100",       ESPEAKNG_ENCODING_ISO_8859_1 },
 	{ "iso-ir-101",       ESPEAKNG_ENCODING_ISO_8859_2 },
 	{ "iso-ir-109",       ESPEAKNG_ENCODING_ISO_8859_3 },
 	{ "iso-ir-110",       ESPEAKNG_ENCODING_ISO_8859_4 },
+	{ "iso-ir-144",       ESPEAKNG_ENCODING_ISO_8859_5 },
 	{ "latin1",           ESPEAKNG_ENCODING_ISO_8859_1 },
 	{ "latin2",           ESPEAKNG_ENCODING_ISO_8859_2 },
 	{ "latin3",           ESPEAKNG_ENCODING_ISO_8859_3 },
@@ -152,6 +158,27 @@ static const uint32_t ISO_8859_4[0x80] = {
 	0x00f8, 0x0173, 0x00fa, 0x00fb, 0x00fc, 0x0169, 0x016b, 0x02d9, // f8
 };
 
+// Reference: http://www.iana.org/go/rfc1345
+// Reference: http://www.unicode.org/Public/MAPPINGS/ISO8859/8859-5.TXT
+static const uint32_t ISO_8859_5[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, 0x0401, 0x0402, 0x0403, 0x0404, 0x0405, 0x0406, 0x0407, // a0
+	0x0408, 0x0409, 0x040a, 0x040b, 0x040c, 0x00ad, 0x040e, 0x040f, // a8
+	0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417, // b0
+	0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f, // b8
+	0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427, // c0
+	0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f, // c8
+	0x0430, 0x0431, 0x0432, 0x0433, 0x0434, 0x0435, 0x0436, 0x0437, // d0
+	0x0438, 0x0439, 0x043a, 0x043b, 0x043c, 0x043d, 0x043e, 0x043f, // d8
+	0x0440, 0x0441, 0x0442, 0x0443, 0x0444, 0x0445, 0x0446, 0x0447, // e0
+	0x0448, 0x0449, 0x044a, 0x044b, 0x044c, 0x044d, 0x044e, 0x044f, // e8
+	0x2116, 0x0451, 0x0452, 0x0453, 0x0454, 0x0455, 0x0456, 0x0457, // f0
+	0x0458, 0x0459, 0x045a, 0x045b, 0x045c, 0x00a7, 0x045e, 0x045f, // f8
+};
+
 static uint32_t
 text_decoder_getc_us_ascii(espeak_ng_TEXT_DECODER *decoder)
 {
@@ -219,6 +246,10 @@ initialize_encoding(espeak_ng_TEXT_DECODER *decoder,
 		decoder->get = text_decoder_getc_codepage;
 		decoder->codepage = ISO_8859_4;
 		break;
+	case ESPEAKNG_ENCODING_ISO_8859_5:
+		decoder->get = text_decoder_getc_codepage;
+		decoder->codepage = ISO_8859_5;
+		break;
 	default:
 		return 0;
 	}

diff --git a/tests/encoding.c b/tests/encoding.c
@@ -217,6 +217,36 @@ test_iso_8859_4_encoding()
 	destroy_text_decoder(decoder);
 }
 
+void
+test_iso_8859_5_encoding()
+{
+	printf("testing ISO-8859-5 encoding\n");
+
+	assert(espeak_ng_EncodingFromName("ISO-8859-5:1988") == ESPEAKNG_ENCODING_ISO_8859_5);
+	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-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);
+}
+
 int
 main(int argc, char **argv)
 {
@@ -227,6 +257,7 @@ main(int argc, char **argv)
 	test_iso_8859_2_encoding();
 	test_iso_8859_3_encoding();
 	test_iso_8859_4_encoding();
+	test_iso_8859_5_encoding();
 	printf("done\n");
 
 	return EXIT_SUCCESS;