From 39f3ea54cfe3fa977aed46c5d1d87ed1549d4c82 Mon Sep 17 00:00:00 2001 From: "Reece H. Dunn" Date: Wed, 29 Mar 2017 07:49:48 +0100 Subject: [PATCH] encoding.c: Support the KOI8-R encoding. --- src/include/espeak-ng/espeak_ng.h | 1 + src/libespeak-ng/encoding.c | 26 +++++++++++++++++++++++++- tests/encoding.c | 27 +++++++++++++++++++++++++++ 3 files changed, 53 insertions(+), 1 deletion(-) diff --git a/src/include/espeak-ng/espeak_ng.h b/src/include/espeak-ng/espeak_ng.h index dbcd4412..7cb5db09 100644 --- a/src/include/espeak-ng/espeak_ng.h +++ b/src/include/espeak-ng/espeak_ng.h @@ -206,6 +206,7 @@ typedef enum ESPEAKNG_ENCODING_ISO_8859_14, ESPEAKNG_ENCODING_ISO_8859_15, ESPEAKNG_ENCODING_ISO_8859_16, + ESPEAKNG_ENCODING_KOI8_R, } espeak_ng_ENCODING; ESPEAK_NG_API espeak_ng_ENCODING diff --git a/src/libespeak-ng/encoding.c b/src/libespeak-ng/encoding.c index b543efda..8d18d1b7 100644 --- a/src/libespeak-ng/encoding.c +++ b/src/libespeak-ng/encoding.c @@ -77,6 +77,7 @@ MNEM_TAB mnem_encoding[] = { { "ISO-8859-14", ESPEAKNG_ENCODING_ISO_8859_14 }, { "ISO-8859-15", ESPEAKNG_ENCODING_ISO_8859_15 }, { "ISO-8859-16", ESPEAKNG_ENCODING_ISO_8859_16 }, + { "KOI8-R", ESPEAKNG_ENCODING_KOI8_R }, { "Latin-9", ESPEAKNG_ENCODING_ISO_8859_15 }, { "TIS-620", ESPEAKNG_ENCODING_ISO_8859_11 }, { "US-ASCII", ESPEAKNG_ENCODING_US_ASCII }, @@ -97,6 +98,7 @@ MNEM_TAB mnem_encoding[] = { { "csISOLatinCyrillic",ESPEAKNG_ENCODING_ISO_8859_5 }, { "csISOLatinGreek", ESPEAKNG_ENCODING_ISO_8859_7 }, { "csISOLatinHebrew", ESPEAKNG_ENCODING_ISO_8859_8 }, + { "csKOI8R", ESPEAKNG_ENCODING_KOI8_R }, { "csTIS620", ESPEAKNG_ENCODING_ISO_8859_11 }, { "arabic", ESPEAKNG_ENCODING_ISO_8859_6 }, { "cyrillic", ESPEAKNG_ENCODING_ISO_8859_5 }, @@ -445,6 +447,27 @@ static const uint32_t ISO_8859_16[0x80] = { 0x0171, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0119, 0x021b, 0x00ff, // f8 }; +// Reference: http://www.iana.org/go/rfc1489 +// Reference: http://www.unicode.org/Public/MAPPINGS/VENDORS/MISC/KOI8-R.TXT +static const uint32_t KOI8_R[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, 0x0104, 0x0105, 0x0141, 0x20ac, 0x201e, 0x0160, 0x00a7, // a0 + 0x0161, 0x00a9, 0x0218, 0x00ab, 0x0179, 0x00ad, 0x017a, 0x017b, // a8 + 0x00b0, 0x00b1, 0x010c, 0x0142, 0x017d, 0x201d, 0x00b6, 0x00b7, // b0 + 0x017e, 0x010d, 0x0219, 0x00bb, 0x0152, 0x0153, 0x0178, 0x017c, // b8 + 0x00c0, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0106, 0x00c6, 0x00c7, // c0 + 0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf, // c8 + 0x0110, 0x0143, 0x00d2, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x015a, // d0 + 0x0170, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x0118, 0x021a, 0x00df, // d8 + 0x00e0, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x0107, 0x00e6, 0x00e7, // e0 + 0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef, // e8 + 0x0111, 0x0144, 0x00f2, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x015b, // f0 + 0x0171, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x0119, 0x021b, 0x00ff, // f8 +}; + static uint32_t string_decoder_getc_us_ascii(espeak_ng_TEXT_DECODER *decoder) { @@ -492,6 +515,7 @@ static const encoding_t string_decoders[] = { { string_decoder_getc_codepage, ISO_8859_14 }, { string_decoder_getc_codepage, ISO_8859_15 }, { string_decoder_getc_codepage, ISO_8859_16 }, + { string_decoder_getc_codepage, KOI8_R }, }; espeak_ng_TEXT_DECODER * @@ -519,7 +543,7 @@ text_decoder_decode_string(espeak_ng_TEXT_DECODER *decoder, int length, espeak_ng_ENCODING encoding) { - if (encoding > ESPEAKNG_ENCODING_ISO_8859_16) + if (encoding > ESPEAKNG_ENCODING_KOI8_R) return ENS_UNKNOWN_TEXT_ENCODING; const encoding_t *enc = string_decoders + encoding; diff --git a/tests/encoding.c b/tests/encoding.c index ddbab0e6..a949551b 100644 --- a/tests/encoding.c +++ b/tests/encoding.c @@ -91,6 +91,32 @@ test_us_ascii_encoding() 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_iso_8859_1_encoding() { @@ -555,6 +581,7 @@ main(int argc, char **argv) test_unknown_encoding(); test_us_ascii_encoding(); + test_koi8_r_encoding(); test_iso_8859_1_encoding(); test_iso_8859_2_encoding();