source: libcaca/trunk/cucul/charset.c @ 2306

Last change on this file since 2306 was 2306, checked in by Sam Hocevar, 12 years ago
  • Documentation fixes (due to renamings, some parameters were no longer documented).
  • Property svn:keywords set to Id
File size: 12.3 KB
Line 
1/*
2 *  libcucul      Canvas for ultrafast compositing of Unicode letters
3 *  Copyright (c) 2002-2007 Sam Hocevar <sam@zoy.org>
4 *                2007 Ben Wiley Sittler <bsittler@gmail.com>
5 *                All Rights Reserved
6 *
7 *  $Id: charset.c 2306 2008-04-19 19:40:25Z sam $
8 *
9 *  This library is free software. It comes without any warranty, to
10 *  the extent permitted by applicable law. You can redistribute it
11 *  and/or modify it under the terms of the Do What The Fuck You Want
12 *  To Public License, Version 2, as published by Sam Hocevar. See
13 *  http://sam.zoy.org/wtfpl/COPYING for more details.
14 */
15
16/*
17 *  This file contains functions for converting characters between
18 *  various character sets.
19 */
20
21#include "config.h"
22
23#if !defined(__KERNEL__)
24#   include <string.h>
25#endif
26
27#include "cucul.h"
28#include "cucul_internals.h"
29
30/*
31 * UTF-8 handling
32 */
33
34static uint8_t const trailing[256] =
35{
36    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
37    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
38    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
39    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
40    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
41    0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,
42    1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
43    2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 3,3,3,3,3,3,3,3,4,4,4,4,5,5,5,5
44};
45
46static uint32_t const offsets[6] =
47{
48    0x00000000UL, 0x00003080UL, 0x000E2080UL,
49    0x03C82080UL, 0xFA082080UL, 0x82082080UL
50};
51
52/*
53 * CP437 handling
54 */
55
56static uint32_t const cp437_lookup1[] =
57{
58    /* 0x01 - 0x0f: ☺ ☻ ♥ ♦ ♣ ♠ • ◘ ○ ◙ ♂ ♀ ♪ ♫ ☼ */
59            0x263a, 0x263b, 0x2665, 0x2666, 0x2663, 0x2660, 0x2022,
60    0x25d8, 0x25cb, 0x25d9, 0x2642, 0x2640, 0x266a, 0x266b, 0x263c,
61    /* 0x10 - 0x1f: ► ◄ ↕ ‼ ¶ § ▬ ↨ ↑ ↓ → ← ∟ ↔ ▲ ▼ */
62    0x25ba, 0x25c4, 0x2195, 0x203c, 0xb6, 0xa7, 0x25ac, 0x21a8,
63    0x2191, 0x2193, 0x2192, 0x2190, 0x221f, 0x2194, 0x25b2, 0x25bc
64};
65
66static uint32_t const cp437_lookup2[] =
67{
68    /* 0x7f: ⌂ */
69    0x2302,
70    /* 0x80 - 0x8f: Ç ü é â ä à å ç ê ë è ï î ì Ä Å */
71    0xc7, 0xfc, 0xe9, 0xe2, 0xe4, 0xe0, 0xe5, 0xe7,
72    0xea, 0xeb, 0xe8, 0xef, 0xee, 0xec, 0xc4, 0xc5,
73    /* 0x90 - 0x9f: É æ Æ ô ö ò û ù ÿ Ö Ü ¢ £ ¥ ₧ ƒ */
74    0xc9, 0xe6, 0xc6, 0xf4, 0xf6, 0xf2, 0xfb, 0xf9,
75    0xff, 0xd6, 0xdc, 0xa2, 0xa3, 0xa5, 0x20a7, 0x192,
76    /* 0xa0 - 0xaf: á í ó ú ñ Ñ ª º ¿ ⌐ ¬ ½ ¼ ¡ « » */
77    0xe1, 0xed, 0xf3, 0xfa, 0xf1, 0xd1, 0xaa, 0xba,
78    0xbf, 0x2310, 0xac, 0xbd, 0xbc, 0xa1, 0xab, 0xbb,
79    /* 0xb0 - 0xbf: ░ ▒ ▓ │ ┤ ╡ ╢ ╖ ╕ ╣ ║ ╗ ╝ ╜ ╛ ┐ */
80    0x2591, 0x2592, 0x2593, 0x2502, 0x2524, 0x2561, 0x2562, 0x2556,
81    0x2555, 0x2563, 0x2551, 0x2557, 0x255d, 0x255c, 0x255b, 0x2510,
82    /* 0xc0 - 0xcf: └ ┴ ┬ ├ ─ ┼ ╞ ╟ ╚ ╔ ╩ ╦ ╠ ═ ╬ ╧ */
83    0x2514, 0x2534, 0x252c, 0x251c, 0x2500, 0x253c, 0x255e, 0x255f,
84    0x255a, 0x2554, 0x2569, 0x2566, 0x2560, 0x2550, 0x256c, 0x2567,
85    /* 0xd0 - 0xdf: ╨ ╤ ╥ ╙ ╘ ╒ ╓ ╫ ╪ ┘ ┌ █ ▄ ▌ ▐ ▀ */
86    0x2568, 0x2564, 0x2565, 0x2559, 0x2558, 0x2552, 0x2553, 0x256b,
87    0x256a, 0x2518, 0x250c, 0x2588, 0x2584, 0x258c, 0x2590, 0x2580,
88    /* 0xe0 - 0xef: α ß Γ π Σ σ µ τ Φ Θ Ω δ ∞ φ ε ∩ */
89    0x3b1, 0xdf, 0x393, 0x3c0, 0x3a3, 0x3c3, 0xb5, 0x3c4,
90    0x3a6, 0x398, 0x3a9, 0x3b4, 0x221e, 0x3c6, 0x3b5, 0x2229,
91    /* 0xf0 - 0xff: ≡ ± ≥ ≤ ⌠ ⌡ ÷ ≈ ° ∙ · √ ⁿ ² ■ <nbsp> */
92    0x2261, 0xb1, 0x2265, 0x2264, 0x2320, 0x2321, 0xf7, 0x2248,
93    0xb0, 0x2219, 0xb7, 0x221a, 0x207f, 0xb2, 0x25a0, 0xa0
94};
95
96/** \brief Convert a UTF-8 character to UTF-32.
97 *
98 *  Convert a UTF-8 character read from a string and return its value in
99 *  the UTF-32 character set. If the second argument is not null, the total
100 *  number of read bytes is written in it.
101 *
102 *  If a null byte was reached before the expected end of the UTF-8 sequence,
103 *  this function returns zero and the number of read bytes is set to zero.
104 *
105 *  This function never fails, but its behaviour with illegal UTF-8 sequences
106 *  is undefined.
107 *
108 *  \param s A string containing the UTF-8 character.
109 *  \param bytes A pointer to a size_t to store the number of bytes in the
110 *         character, or NULL.
111 *  \return The corresponding UTF-32 character, or zero if the character
112 *  is incomplete.
113 */
114uint32_t cucul_utf8_to_utf32(char const *s, size_t *bytes)
115{
116    int todo = trailing[(int)(unsigned char)*s];
117    int i = 0;
118    uint32_t ret = 0;
119
120    for(;;)
121    {
122        if(!*s)
123        {
124            if(bytes)
125                *bytes = 0;
126            return 0;
127        }
128
129        ret += ((uint32_t)(unsigned char)*s++) << (6 * (todo - i));
130
131        if(todo == i++)
132        {
133            if(bytes)
134                *bytes = i;
135            return ret - offsets[todo];
136        }
137    }
138}
139
140/** \brief Convert a UTF-32 character to UTF-8.
141 *
142 *  Convert a UTF-32 character read from a string and write its value in
143 *  the UTF-8 character set into the given buffer.
144 *
145 *  This function never fails, but its behaviour with illegal UTF-32 characters
146 *  is undefined.
147 *
148 *  \param buf A pointer to a character buffer where the UTF-8 sequence will
149 *  be written.
150 *  \param ch The UTF-32 character.
151 *  \return The number of bytes written.
152 */
153size_t cucul_utf32_to_utf8(char *buf, uint32_t ch)
154{
155    static const uint8_t mark[7] =
156    {
157        0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC
158    };
159
160    char *parser = buf;
161    size_t bytes;
162
163    if(ch < 0x80)
164    {
165        *parser++ = ch;
166        return 1;
167    }
168
169    bytes = (ch < 0x800) ? 2 : (ch < 0x10000) ? 3 : 4;
170    parser += bytes;
171
172    switch(bytes)
173    {
174        case 4: *--parser = (ch | 0x80) & 0xbf; ch >>= 6;
175        case 3: *--parser = (ch | 0x80) & 0xbf; ch >>= 6;
176        case 2: *--parser = (ch | 0x80) & 0xbf; ch >>= 6;
177    }
178    *--parser = ch | mark[bytes];
179
180    return bytes;
181}
182
183/** \brief Convert a UTF-32 character to CP437.
184 *
185 *  Convert a UTF-32 character read from a string and return its value in
186 *  the CP437 character set, or "?" if the character has no equivalent.
187 *
188 *  This function never fails.
189 *
190 *  \param ch The UTF-32 character.
191 *  \return The corresponding CP437 character, or "?" if not representable.
192 */
193uint8_t cucul_utf32_to_cp437(uint32_t ch)
194{
195    unsigned int i;
196
197    if(ch < 0x00000020)
198        return '?';
199
200    if(ch < 0x00000080)
201        return ch;
202
203    for(i = 0; i < sizeof(cp437_lookup1) / sizeof(*cp437_lookup1); i++)
204        if(cp437_lookup1[i] == ch)
205            return 0x01 + i;
206
207    for(i = 0; i < sizeof(cp437_lookup2) / sizeof(*cp437_lookup2); i++)
208        if(cp437_lookup2[i] == ch)
209            return 0x7f + i;
210
211    return '?';
212}
213
214/** \brief Convert a CP437 character to UTF-32.
215 *
216 *  Convert a CP437 character read from a string and return its value in
217 *  the UTF-32 character set, or zero if the character is a CP437 control
218 *  character.
219 *
220 *  This function never fails.
221 *
222 *  \param ch The CP437 character.
223 *  \return The corresponding UTF-32 character, or zero if not representable.
224 */
225uint32_t cucul_cp437_to_utf32(uint8_t ch)
226{
227    if(ch > 0x7f)
228        return cp437_lookup2[ch - 0x7f];
229
230    if(ch >= 0x20)
231        return (uint32_t)ch;
232
233    if(ch > 0)
234        return cp437_lookup1[ch - 0x01];
235
236    return 0x00000000;
237}
238
239/** \brief Convert a UTF-32 character to ASCII.
240 *
241 *  Convert a UTF-32 character into an ASCII character. When no equivalent
242 *  exists, a graphically close equivalent is sought.
243 *
244 *  This function never fails, but its behaviour with illegal UTF-32 characters
245 *  is undefined.
246 *
247 *  \param ch The UTF-32 character.
248 *  \return The corresponding ASCII character, or a graphically close
249 *  equivalent if found, or "?" if not representable.
250 */
251char cucul_utf32_to_ascii(uint32_t ch)
252{
253    /* Standard ASCII */
254    if(ch < 0x80)
255        return ch;
256
257    /* Fullwidth Forms */
258    if(ch > 0x0000ff00 && ch < 0x0000ff5f)
259        return ' ' + (ch - 0x0000ff00);
260
261    switch (ch)
262    {
263    case 0x000000a0: /*   (nbsp) */
264    case 0x00003000: /*   (ideographic space) */
265        return ' ';
266    case 0x000000a3: /* £ */
267        return 'f';
268    case 0x000000b0: /* ° */
269        return '\'';
270    case 0x000000b1: /* ± */
271        return '#';
272    case 0x000000b7: /* · */
273    case 0x00002219: /* ∙ */
274    case 0x000030fb: /* ・ */
275        return '.';
276    case 0x000003c0: /* π */
277        return '*';
278    case 0x00002018: /* ‘ */
279    case 0x00002019: /* ’ */
280        return '\'';
281    case 0x0000201c: /* “ */
282    case 0x0000201d: /* ” */
283        return '"';
284    case 0x00002190: /* ← */
285        return '<';
286    case 0x00002191: /* ↑ */
287        return '^';
288    case 0x00002192: /* → */
289        return '>';
290    case 0x00002193: /* ↓ */
291        return 'v';
292    case 0x00002260: /* ≠ */
293        return '!';
294    case 0x00002261: /* ≡ */
295        return '=';
296    case 0x00002264: /* ≤ */
297        return '<';
298    case 0x00002265: /* ≥ */
299        return '>';
300    case 0x000023ba: /* ⎺ */
301    case 0x000023bb: /* ⎻ */
302    case 0x000023bc: /* ⎼ */
303    case 0x000023bd: /* ⎽ */
304    case 0x00002500: /* ─ */
305    case 0x00002550: /* ═ */
306        return '-';
307    case 0x00002502: /* │ */
308    case 0x00002551: /* ║ */
309        return '|';
310    case 0x0000250c: /* ┌ */
311    case 0x00002552: /* ╒ */
312    case 0x00002553: /* ╓ */
313    case 0x00002554: /* ╔ */
314    case 0x00002514: /* └ */
315    case 0x00002558: /* ╘ */
316    case 0x00002559: /* ╙ */
317    case 0x0000255a: /* ╚ */
318    case 0x0000251c: /* ├ */
319    case 0x0000255e: /* ╞ */
320    case 0x0000255f: /* ╟ */
321    case 0x00002560: /* ╠ */
322    case 0x0000252c: /* ┬ */
323    case 0x00002564: /* ╤ */
324    case 0x00002565: /* ╥ */
325    case 0x00002566: /* ╦ */
326    case 0x00002534: /* ┴ */
327    case 0x00002567: /* ╧ */
328    case 0x00002568: /* ╨ */
329    case 0x00002569: /* ╩ */
330    case 0x0000253c: /* ┼ */
331    case 0x0000256a: /* ╪ */
332    case 0x0000256b: /* ╫ */
333    case 0x0000256c: /* ╬ */
334        return '+';
335    case 0x00002510: /* ┐ */
336    case 0x00002555: /* ╕ */
337    case 0x00002556: /* ╖ */
338    case 0x00002557: /* ╗ */
339    case 0x00002518: /* ┘ */
340    case 0x0000255b: /* ╛ */
341    case 0x0000255c: /* ╜ */
342    case 0x0000255d: /* ╝ */
343    case 0x00002524: /* ┤ */
344    case 0x00002561: /* ╡ */
345    case 0x00002562: /* ╢ */
346    case 0x00002563: /* ╣ */
347        return '+';
348    case 0x00002591: /* ░ */
349    case 0x00002592: /* ▒ */
350    case 0x00002593: /* ▓ */
351    case 0x00002588: /* █ */
352    case 0x0000258c: /* ▌ */
353    case 0x00002590: /* ▐ */
354    case 0x000025a0: /* ■ */
355    case 0x000025ac: /* ▬ */
356    case 0x000025ae: /* ▮ */
357        return '#';
358    case 0x00002580: /* ▀ */
359        return '"';
360    case 0x00002584: /* ▄ */
361        return ',';
362    case 0x000025c6: /* ◆ */
363    case 0x00002666: /* ♦ */
364        return '+';
365    case 0x00002022: /* • */
366    case 0x000025cb: /* ○ */
367    case 0x000025cf: /* ● */
368    case 0x00002603: /* ☃ */
369    case 0x0000263c: /* ☼ */
370        return 'o';
371    case 0x0000301c: /* 〜 */
372        return '~';
373    }
374
375    return '?';
376}
377
378/** \brief Tell whether a UTF-32 character is fullwidth.
379 *
380 *  Check whether the given UTF-32 character should be printed at twice
381 *  the normal width (fullwidth characters). If the character is unknown
382 *  or if its status cannot be decided, it is treated as a standard-width
383 *  character.
384 *
385 *  This function never fails.
386 *
387 *  \param ch The UTF-32 character.
388 *  \return 1 if the character is fullwidth, 0 otherwise.
389 */
390int cucul_utf32_is_fullwidth(uint32_t ch)
391{
392    if(ch < 0x2e80) /* Standard stuff */
393        return 0;
394    if(ch < 0xa700) /* Japanese, Korean, CJK, Yi... */
395        return 1;
396    if(ch < 0xac00) /* Modified Tone Letters, Syloti Nagri */
397        return 0;
398    if(ch < 0xd800) /* Hangul Syllables */
399        return 1;
400    if(ch < 0xf900) /* Misc crap */
401        return 0;
402    if(ch < 0xfb00) /* More CJK */
403        return 1;
404    if(ch < 0xfe20) /* Misc crap */
405        return 0;
406    if(ch < 0xfe70) /* More CJK */
407        return 1;
408    if(ch < 0xff00) /* Misc crap */
409        return 0;
410    if(ch < 0xff61) /* Fullwidth forms */
411        return 1;
412    if(ch < 0xffe0) /* Halfwidth forms */
413        return 0;
414    if(ch < 0xffe8) /* More fullwidth forms */
415        return 1;
416    if(ch < 0x20000) /* Misc crap */
417        return 0;
418    if(ch < 0xe0000) /* More CJK */
419        return 1;
420    return 0;
421}
422
Note: See TracBrowser for help on using the repository browser.