From a93b0f3d64fb4cb9d39db31e0b3ce7c9e837ff19 Mon Sep 17 00:00:00 2001 From: "Reece H. Dunn" Date: Mon, 27 Mar 2017 23:55:38 +0100 Subject: [PATCH] encoding.c: Support the ISO 8859-3 encoding. --- src/include/espeak-ng/espeak_ng.h | 1 + src/libespeak-ng/encoding.c | 42 +++++++++++++++++++++++++++---- tests/encoding.c | 32 +++++++++++++++++++++++ 3 files changed, 70 insertions(+), 5 deletions(-) diff --git a/src/include/espeak-ng/espeak_ng.h b/src/include/espeak-ng/espeak_ng.h index 86dbf1f9..316258a7 100644 --- a/src/include/espeak-ng/espeak_ng.h +++ b/src/include/espeak-ng/espeak_ng.h @@ -192,6 +192,7 @@ typedef enum ESPEAKNG_ENCODING_US_ASCII, ESPEAKNG_ENCODING_ISO_8859_1, ESPEAKNG_ENCODING_ISO_8859_2, + ESPEAKNG_ENCODING_ISO_8859_3, } espeak_ng_ENCODING; ESPEAK_NG_API espeak_ng_ENCODING diff --git a/src/libespeak-ng/encoding.c b/src/libespeak-ng/encoding.c index a18ccb8b..480f9462 100644 --- a/src/libespeak-ng/encoding.c +++ b/src/libespeak-ng/encoding.c @@ -34,24 +34,31 @@ MNEM_TAB mnem_encoding[] = { { "ISO_646.irv:1991", ESPEAKNG_ENCODING_US_ASCII }, { "ISO_8859-1", ESPEAKNG_ENCODING_ISO_8859_1 }, { "ISO_8859-2", ESPEAKNG_ENCODING_ISO_8859_2 }, + { "ISO_8859-3", ESPEAKNG_ENCODING_ISO_8859_3 }, { "ISO646-US", ESPEAKNG_ENCODING_US_ASCII }, { "ISO-8859-1", ESPEAKNG_ENCODING_ISO_8859_1 }, { "ISO-8859-1:1987", ESPEAKNG_ENCODING_ISO_8859_1 }, { "ISO-8859-2", ESPEAKNG_ENCODING_ISO_8859_2 }, { "ISO-8859-2:1987", ESPEAKNG_ENCODING_ISO_8859_2 }, + { "ISO-8859-3", ESPEAKNG_ENCODING_ISO_8859_3 }, + { "ISO-8859-3:1988", ESPEAKNG_ENCODING_ISO_8859_3 }, { "US-ASCII", ESPEAKNG_ENCODING_US_ASCII }, { "cp367", ESPEAKNG_ENCODING_US_ASCII }, { "cp819", ESPEAKNG_ENCODING_ISO_8859_1 }, { "csASCII", ESPEAKNG_ENCODING_US_ASCII }, { "csISOLatin1", ESPEAKNG_ENCODING_ISO_8859_1 }, { "csISOLatin2", ESPEAKNG_ENCODING_ISO_8859_2 }, + { "csISOLatin3", ESPEAKNG_ENCODING_ISO_8859_3 }, { "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 }, { "latin1", ESPEAKNG_ENCODING_ISO_8859_1 }, { "latin2", ESPEAKNG_ENCODING_ISO_8859_2 }, + { "latin3", ESPEAKNG_ENCODING_ISO_8859_3 }, { "l1", ESPEAKNG_ENCODING_ISO_8859_1 }, { "l2", ESPEAKNG_ENCODING_ISO_8859_2 }, + { "l3", ESPEAKNG_ENCODING_ISO_8859_3 }, { "us", ESPEAKNG_ENCODING_US_ASCII }, { NULL, ESPEAKNG_ENCODING_UNKNOWN } }; @@ -72,16 +79,16 @@ struct espeak_ng_TEXT_DECODER_ const char *end; uint32_t (*get)(espeak_ng_TEXT_DECODER *decoder); - uint32_t *codepage; + const uint32_t *codepage; }; // Reference: http://www.iana.org/go/rfc1345 // Reference: http://www.unicode.org/Public/MAPPINGS/ISO8859/8859-2.TXT static const uint32_t ISO_8859_2[0x80] = { - 0x0080, 0x0081, 0x0082, 0x0083, 0x0084, 0x0085, 0x0086, 0x0087, // a0 - 0x0088, 0x0089, 0x008a, 0x008b, 0x008c, 0x008d, 0x008e, 0x008f, // a8 - 0x0090, 0x0091, 0x0092, 0x0093, 0x0094, 0x0095, 0x0096, 0x0097, // b0 - 0x0098, 0x0099, 0x009a, 0x009b, 0x009c, 0x009d, 0x009e, 0x009f, // b8 + 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, 0x0104, 0x02d8, 0x0141, 0x00a4, 0x013d, 0x015a, 0x00a7, // a0 0x00a8, 0x0160, 0x015e, 0x0164, 0x0179, 0x00ad, 0x017d, 0x017b, // a8 0x00b0, 0x0105, 0x02db, 0x0142, 0x00b4, 0x013e, 0x015b, 0x02c7, // b0 @@ -96,6 +103,27 @@ static const uint32_t ISO_8859_2[0x80] = { 0x0159, 0x016f, 0x00fa, 0x0171, 0x00fc, 0x00fd, 0x0163, 0x02d9, // f8 }; +// Reference: http://www.iana.org/go/rfc1345 +// Reference: http://www.unicode.org/Public/MAPPINGS/ISO8859/8859-3.TXT +static const uint32_t ISO_8859_3[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, 0x0126, 0x02d8, 0x00a3, 0x00a4, 0x0000, 0x0124, 0x00a7, // a0 + 0x00a8, 0x0130, 0x015e, 0x011e, 0x0134, 0x00ad, 0x0000, 0x017b, // a8 + 0x00b0, 0x0127, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x0125, 0x00b7, // b0 + 0x00b8, 0x0131, 0x015f, 0x011f, 0x0135, 0x00bd, 0x0000, 0x017c, // b8 + 0x00c0, 0x00c1, 0x00c2, 0x0000, 0x00c4, 0x010a, 0x0108, 0x00c7, // c0 + 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf, // c8 + 0x0000, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x0120, 0x00d6, 0x00d7, // d0 + 0x011c, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x016c, 0x015c, 0x00df, // d8 + 0x00e0, 0x00e1, 0x00e2, 0x0000, 0x00e4, 0x010b, 0x0109, 0x00e7, // e0 + 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef, // e8 + 0x0000, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x0121, 0x00f6, 0x00f7, // f0 + 0x011d, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x016d, 0x015d, 0x02d9, // f8 +}; + static uint32_t text_decoder_getc_us_ascii(espeak_ng_TEXT_DECODER *decoder) { @@ -155,6 +183,10 @@ initialize_encoding(espeak_ng_TEXT_DECODER *decoder, decoder->get = text_decoder_getc_codepage; decoder->codepage = ISO_8859_2; break; + case ESPEAKNG_ENCODING_ISO_8859_3: + decoder->get = text_decoder_getc_codepage; + decoder->codepage = ISO_8859_3; + break; default: return 0; } diff --git a/tests/encoding.c b/tests/encoding.c index 6280207d..232fc88e 100644 --- a/tests/encoding.c +++ b/tests/encoding.c @@ -155,6 +155,37 @@ test_iso_8859_2_encoding() destroy_text_decoder(decoder); } +void +test_iso_8859_3_encoding() +{ + printf("testing ISO-8859-3 encoding\n"); + + assert(espeak_ng_EncodingFromName("ISO-8859-3:1988") == ESPEAKNG_ENCODING_ISO_8859_3); + assert(espeak_ng_EncodingFromName("ISO-8859-3") == ESPEAKNG_ENCODING_ISO_8859_3); + assert(espeak_ng_EncodingFromName("ISO_8859-3") == ESPEAKNG_ENCODING_ISO_8859_3); + assert(espeak_ng_EncodingFromName("iso-ir-109") == ESPEAKNG_ENCODING_ISO_8859_3); + assert(espeak_ng_EncodingFromName("latin3") == ESPEAKNG_ENCODING_ISO_8859_3); + assert(espeak_ng_EncodingFromName("l3") == ESPEAKNG_ENCODING_ISO_8859_3); + assert(espeak_ng_EncodingFromName("csISOLatin3") == ESPEAKNG_ENCODING_ISO_8859_3); + + espeak_ng_TEXT_DECODER *decoder = create_text_decoder(); + + assert(text_decoder_decode_string(decoder, "aG\x92\xA0\xDE", 5, ESPEAKNG_ENCODING_ISO_8859_3) == 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) == 0x015C); + assert(text_decoder_eof(decoder) == 1); + + destroy_text_decoder(decoder); +} + int main(int argc, char **argv) { @@ -163,6 +194,7 @@ main(int argc, char **argv) test_us_ascii_encoding(); test_iso_8859_1_encoding(); test_iso_8859_2_encoding(); + test_iso_8859_3_encoding(); printf("done\n"); return EXIT_SUCCESS;