_codecs_cn.c 9.1 KB
Newer Older
1 2 3 4 5 6 7 8 9
/*
 * _codecs_cn.c: Codecs collection for Mainland Chinese encodings
 *
 * Written by Hye-Shik Chang <perky@FreeBSD.org>
 */

#include "cjkcodecs.h"
#include "mappings_cn.h"

10 11 12 13 14 15 16 17
/**
 * hz is predefined as 100 on AIX. So we undefine it to avoid
 * conflict against hz codec's.
 */
#ifdef _AIX
#undef hz
#endif

18 19
/* GBK and GB2312 map differently in few codepoints that are listed below:
 *
20 21 22 23
 *              gb2312                          gbk
 * A1A4         U+30FB KATAKANA MIDDLE DOT      U+00B7 MIDDLE DOT
 * A1AA         U+2015 HORIZONTAL BAR           U+2014 EM DASH
 * A844         undefined                       U+2015 HORIZONTAL BAR
24 25 26
 */

#define GBK_DECODE(dc1, dc2, assi) \
27 28 29 30 31
    if ((dc1) == 0xa1 && (dc2) == 0xaa) (assi) = 0x2014; \
    else if ((dc1) == 0xa8 && (dc2) == 0x44) (assi) = 0x2015; \
    else if ((dc1) == 0xa1 && (dc2) == 0xa4) (assi) = 0x00b7; \
    else TRYMAP_DEC(gb2312, assi, dc1 ^ 0x80, dc2 ^ 0x80); \
    else TRYMAP_DEC(gbkext, assi, dc1, dc2);
32 33

#define GBK_ENCODE(code, assi) \
34 35 36 37
    if ((code) == 0x2014) (assi) = 0xa1aa; \
    else if ((code) == 0x2015) (assi) = 0xa844; \
    else if ((code) == 0x00b7) (assi) = 0xa1a4; \
    else if ((code) != 0x30fb && TRYMAP_ENC_COND(gbcommon, assi, code));
38 39 40 41 42 43 44

/*
 * GB2312 codec
 */

ENCODER(gb2312)
{
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
    while (inleft > 0) {
        Py_UNICODE c = IN1;
        DBCHAR code;

        if (c < 0x80) {
            WRITE1((unsigned char)c)
            NEXT(1, 1)
            continue;
        }
        UCS4INVALID(c)

        REQUIRE_OUTBUF(2)
        TRYMAP_ENC(gbcommon, code, c);
        else return 1;

        if (code & 0x8000) /* MSB set: GBK */
            return 1;

        OUT1((code >> 8) | 0x80)
        OUT2((code & 0xFF) | 0x80)
        NEXT(1, 2)
    }

    return 0;
69 70 71 72
}

DECODER(gb2312)
{
73 74
    while (inleft > 0) {
        unsigned char c = **inbuf;
75

76
        REQUIRE_OUTBUF(1)
77

78 79 80 81 82
        if (c < 0x80) {
            OUT1(c)
            NEXT(1, 1)
            continue;
        }
83

84 85 86 87
        REQUIRE_INBUF(2)
        TRYMAP_DEC(gb2312, **outbuf, c ^ 0x80, IN2 ^ 0x80) {
            NEXT(2, 1)
        }
88
        else return 1;
89
    }
90

91
    return 0;
92 93 94 95 96 97 98 99 100
}


/*
 * GBK codec
 */

ENCODER(gbk)
{
101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125
    while (inleft > 0) {
        Py_UNICODE c = IN1;
        DBCHAR code;

        if (c < 0x80) {
            WRITE1((unsigned char)c)
            NEXT(1, 1)
            continue;
        }
        UCS4INVALID(c)

        REQUIRE_OUTBUF(2)

        GBK_ENCODE(c, code)
        else return 1;

        OUT1((code >> 8) | 0x80)
        if (code & 0x8000)
            OUT2((code & 0xFF)) /* MSB set: GBK */
        else
            OUT2((code & 0xFF) | 0x80) /* MSB unset: GB2312 */
        NEXT(1, 2)
    }

    return 0;
126 127 128 129
}

DECODER(gbk)
{
130 131
    while (inleft > 0) {
        unsigned char c = IN1;
132

133
        REQUIRE_OUTBUF(1)
134

135 136 137 138 139
        if (c < 0x80) {
            OUT1(c)
            NEXT(1, 1)
            continue;
        }
140

141
        REQUIRE_INBUF(2)
142

143
        GBK_DECODE(c, IN2, **outbuf)
144
        else return 1;
145

146 147
        NEXT(2, 1)
    }
148

149
    return 0;
150 151 152 153 154 155 156 157 158
}


/*
 * GB18030 codec
 */

ENCODER(gb18030)
{
159 160 161 162 163 164 165 166 167 168 169 170
    while (inleft > 0) {
        ucs4_t c = IN1;
        DBCHAR code;

        if (c < 0x80) {
            WRITE1(c)
            NEXT(1, 1)
            continue;
        }

        DECODE_SURROGATE(c)
        if (c > 0x10FFFF)
171
#if Py_UNICODE_SIZE == 2
172
            return 2; /* surrogates pair */
173
#else
174
            return 1;
175
#endif
176 177
        else if (c >= 0x10000) {
            ucs4_t tc = c - 0x10000;
178

179
            REQUIRE_OUTBUF(4)
180

181 182 183 184 185 186 187
            OUT4((unsigned char)(tc % 10) + 0x30)
            tc /= 10;
            OUT3((unsigned char)(tc % 126) + 0x81)
            tc /= 126;
            OUT2((unsigned char)(tc % 10) + 0x30)
            tc /= 10;
            OUT1((unsigned char)(tc + 0x90))
188 189

#if Py_UNICODE_SIZE == 2
190
            NEXT(2, 4) /* surrogates pair */
191
#else
192
            NEXT(1, 4)
193
#endif
194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242
            continue;
        }

        REQUIRE_OUTBUF(2)

        GBK_ENCODE(c, code)
        else TRYMAP_ENC(gb18030ext, code, c);
        else {
            const struct _gb18030_to_unibmp_ranges *utrrange;

            REQUIRE_OUTBUF(4)

            for (utrrange = gb18030_to_unibmp_ranges;
                 utrrange->first != 0;
                 utrrange++)
                if (utrrange->first <= c &&
                    c <= utrrange->last) {
                    Py_UNICODE tc;

                    tc = c - utrrange->first +
                         utrrange->base;

                    OUT4((unsigned char)(tc % 10) + 0x30)
                    tc /= 10;
                    OUT3((unsigned char)(tc % 126) + 0x81)
                    tc /= 126;
                    OUT2((unsigned char)(tc % 10) + 0x30)
                    tc /= 10;
                    OUT1((unsigned char)tc + 0x81)

                    NEXT(1, 4)
                    break;
                }

            if (utrrange->first == 0)
                return 1;
            continue;
        }

        OUT1((code >> 8) | 0x80)
        if (code & 0x8000)
            OUT2((code & 0xFF)) /* MSB set: GBK or GB18030ext */
        else
            OUT2((code & 0xFF) | 0x80) /* MSB unset: GB2312 */

        NEXT(1, 2)
    }

    return 0;
243 244 245 246
}

DECODER(gb18030)
{
247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269
    while (inleft > 0) {
        unsigned char c = IN1, c2;

        REQUIRE_OUTBUF(1)

        if (c < 0x80) {
            OUT1(c)
            NEXT(1, 1)
            continue;
        }

        REQUIRE_INBUF(2)

        c2 = IN2;
        if (c2 >= 0x30 && c2 <= 0x39) { /* 4 bytes seq */
            const struct _gb18030_to_unibmp_ranges *utr;
            unsigned char c3, c4;
            ucs4_t lseq;

            REQUIRE_INBUF(4)
            c3 = IN3;
            c4 = IN4;
            if (c < 0x81 || c3 < 0x81 || c4 < 0x30 || c4 > 0x39)
270
                return 1;
271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294
            c -= 0x81;  c2 -= 0x30;
            c3 -= 0x81; c4 -= 0x30;

            if (c < 4) { /* U+0080 - U+FFFF */
                lseq = ((ucs4_t)c * 10 + c2) * 1260 +
                    (ucs4_t)c3 * 10 + c4;
                if (lseq < 39420) {
                    for (utr = gb18030_to_unibmp_ranges;
                         lseq >= (utr + 1)->base;
                         utr++) ;
                    OUT1(utr->first - utr->base + lseq)
                    NEXT(4, 1)
                    continue;
                }
            }
            else if (c >= 15) { /* U+10000 - U+10FFFF */
                lseq = 0x10000 + (((ucs4_t)c-15) * 10 + c2)
                    * 1260 + (ucs4_t)c3 * 10 + c4;
                if (lseq <= 0x10FFFF) {
                    WRITEUCS4(lseq);
                    NEXT_IN(4)
                    continue;
                }
            }
295
            return 1;
296 297 298 299
        }

        GBK_DECODE(c, c2, **outbuf)
        else TRYMAP_DEC(gb18030ext, **outbuf, c, c2);
300
        else return 1;
301 302 303 304 305

        NEXT(2, 1)
    }

    return 0;
306 307 308 309 310 311 312 313 314
}


/*
 * HZ codec
 */

ENCODER_INIT(hz)
{
315 316
    state->i = 0;
    return 0;
317 318 319 320
}

ENCODER_RESET(hz)
{
321 322 323 324 325 326
    if (state->i != 0) {
        WRITE2('~', '}')
        state->i = 0;
        NEXT_OUT(2)
    }
    return 0;
327 328 329 330
}

ENCODER(hz)
{
331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367
    while (inleft > 0) {
        Py_UNICODE c = IN1;
        DBCHAR code;

        if (c < 0x80) {
            if (state->i == 0) {
                WRITE1((unsigned char)c)
                NEXT(1, 1)
            }
            else {
                WRITE3('~', '}', (unsigned char)c)
                NEXT(1, 3)
                state->i = 0;
            }
            continue;
        }

        UCS4INVALID(c)

        TRYMAP_ENC(gbcommon, code, c);
        else return 1;

        if (code & 0x8000) /* MSB set: GBK */
            return 1;

        if (state->i == 0) {
            WRITE4('~', '{', code >> 8, code & 0xff)
            NEXT(1, 4)
            state->i = 1;
        }
        else {
            WRITE2(code >> 8, code & 0xff)
            NEXT(1, 2)
        }
    }

    return 0;
368 369 370 371
}

DECODER_INIT(hz)
{
372 373
    state->i = 0;
    return 0;
374 375 376 377
}

DECODER_RESET(hz)
{
378 379
    state->i = 0;
    return 0;
380 381 382 383
}

DECODER(hz)
{
384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
    while (inleft > 0) {
        unsigned char c = IN1;

        if (c == '~') {
            unsigned char c2 = IN2;

            REQUIRE_INBUF(2)
            if (c2 == '~') {
                WRITE1('~')
                NEXT(2, 1)
                continue;
            }
            else if (c2 == '{' && state->i == 0)
                state->i = 1; /* set GB */
            else if (c2 == '}' && state->i == 1)
                state->i = 0; /* set ASCII */
            else if (c2 == '\n')
                ; /* line-continuation */
            else
403
                return 1;
404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421
            NEXT(2, 0);
            continue;
        }

        if (c & 0x80)
            return 1;

        if (state->i == 0) { /* ASCII mode */
            WRITE1(c)
            NEXT(1, 1)
        }
        else { /* GB mode */
            REQUIRE_INBUF(2)
            REQUIRE_OUTBUF(1)
            TRYMAP_DEC(gb2312, **outbuf, c, IN2) {
                NEXT(2, 1)
            }
            else
422
                return 1;
423 424 425 426
        }
    }

    return 0;
427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444
}


BEGIN_MAPPINGS_LIST
  MAPPING_DECONLY(gb2312)
  MAPPING_DECONLY(gbkext)
  MAPPING_ENCONLY(gbcommon)
  MAPPING_ENCDEC(gb18030ext)
END_MAPPINGS_LIST

BEGIN_CODECS_LIST
  CODEC_STATELESS(gb2312)
  CODEC_STATELESS(gbk)
  CODEC_STATELESS(gb18030)
  CODEC_STATEFUL(hz)
END_CODECS_LIST

I_AM_A_MODULE_FOR(cn)