source: rtems-tools/rtemstoolkit/ConvertUTF.c @ e0a52a4

5
Last change on this file since e0a52a4 was e0a52a4, checked in by Chris Johns <chrisj@…>, on 04/30/18 at 05:50:03

rtemstoolkit: Remove warnings generated on Windows for fall-thru.

  • Property mode set to 100644
File size: 18.8 KB
Line 
1/*
2 * Copyright 2001-2004 Unicode, Inc.
3 *
4 * Disclaimer
5 *
6 * This source code is provided as is by Unicode, Inc. No claims are
7 * made as to fitness for any particular purpose. No warranties of any
8 * kind are expressed or implied. The recipient agrees to determine
9 * applicability of information provided. If this file has been
10 * purchased on magnetic or optical media from Unicode, Inc., the
11 * sole remedy for any claim will be exchange of defective media
12 * within 90 days of receipt.
13 *
14 * Limitations on Rights to Redistribute This Code
15 *
16 * Unicode, Inc. hereby grants the right to freely use the information
17 * supplied in this file in the creation of products supporting the
18 * Unicode Standard, and to make copies of this file in any form
19 * for internal or external distribution as long as this notice
20 * remains attached.
21 */
22
23/* ---------------------------------------------------------------------
24
25    Conversions between UTF32, UTF-16, and UTF-8. Source code file.
26    Author: Mark E. Davis, 1994.
27    Rev History: Rick McGowan, fixes & updates May 2001.
28    Sept 2001: fixed const & error conditions per
29        mods suggested by S. Parent & A. Lillich.
30    June 2002: Tim Dodd added detection and handling of incomplete
31        source sequences, enhanced error detection, added casts
32        to eliminate compiler warnings.
33    July 2003: slight mods to back out aggressive FFFE detection.
34    Jan 2004: updated switches in from-UTF8 conversions.
35    Oct 2004: updated to use UNI_MAX_LEGAL_UTF32 in UTF-32 conversions.
36
37    See the header file "ConvertUTF.h" for complete documentation.
38
39------------------------------------------------------------------------ */
40
41
42#include "ConvertUTF.h"
43#ifdef CVTUTF_DEBUG
44#include <stdio.h>
45#endif
46
47static const int halfShift  = 10; /* used for shifting by 10 bits */
48
49static const UTF32 halfBase = 0x0010000UL;
50static const UTF32 halfMask = 0x3FFUL;
51
52#define UNI_SUR_HIGH_START  (UTF32)0xD800
53#define UNI_SUR_HIGH_END    (UTF32)0xDBFF
54#define UNI_SUR_LOW_START   (UTF32)0xDC00
55#define UNI_SUR_LOW_END     (UTF32)0xDFFF
56#define false      0
57#define true        1
58
59/* --------------------------------------------------------------------- */
60
61ConversionResult ConvertUTF32toUTF16 (
62        const UTF32** sourceStart, const UTF32* sourceEnd,
63        UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
64    ConversionResult result = conversionOK;
65    const UTF32* source = *sourceStart;
66    UTF16* target = *targetStart;
67    while (source < sourceEnd) {
68        UTF32 ch;
69        if (target >= targetEnd) {
70            result = targetExhausted; break;
71        }
72        ch = *source++;
73        if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
74            /* UTF-16 surrogate values are illegal in UTF-32; 0xffff or 0xfffe are both reserved values */
75            if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
76                if (flags == strictConversion) {
77                    --source; /* return to the illegal value itself */
78                    result = sourceIllegal;
79                    break;
80                } else {
81                    *target++ = UNI_REPLACEMENT_CHAR;
82                }
83            } else {
84                *target++ = (UTF16)ch; /* normal case */
85            }
86        } else if (ch > UNI_MAX_LEGAL_UTF32) {
87            if (flags == strictConversion) {
88                result = sourceIllegal;
89            } else {
90                *target++ = UNI_REPLACEMENT_CHAR;
91            }
92        } else {
93            /* target is a character in range 0xFFFF - 0x10FFFF. */
94            if (target + 1 >= targetEnd) {
95                --source; /* Back up source pointer! */
96                result = targetExhausted; break;
97            }
98            ch -= halfBase;
99            *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
100            *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
101        }
102    }
103    *sourceStart = source;
104    *targetStart = target;
105    return result;
106}
107
108/* --------------------------------------------------------------------- */
109
110ConversionResult ConvertUTF16toUTF32 (
111        const UTF16** sourceStart, const UTF16* sourceEnd,
112        UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
113    ConversionResult result = conversionOK;
114    const UTF16* source = *sourceStart;
115    UTF32* target = *targetStart;
116    UTF32 ch, ch2;
117    while (source < sourceEnd) {
118        const UTF16* oldSource = source; /*  In case we have to back up because of target overflow. */
119        ch = *source++;
120        /* If we have a surrogate pair, convert to UTF32 first. */
121        if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
122            /* If the 16 bits following the high surrogate are in the source buffer... */
123            if (source < sourceEnd) {
124                ch2 = *source;
125                /* If it's a low surrogate, convert to UTF32. */
126                if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
127                    ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
128                        + (ch2 - UNI_SUR_LOW_START) + halfBase;
129                    ++source;
130                } else if (flags == strictConversion) { /* it's an unpaired high surrogate */
131                    --source; /* return to the illegal value itself */
132                    result = sourceIllegal;
133                    break;
134                }
135            } else { /* We don't have the 16 bits following the high surrogate. */
136                --source; /* return to the high surrogate */
137                result = sourceExhausted;
138                break;
139            }
140        } else if (flags == strictConversion) {
141            /* UTF-16 surrogate values are illegal in UTF-32 */
142            if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
143                --source; /* return to the illegal value itself */
144                result = sourceIllegal;
145                break;
146            }
147        }
148        if (target >= targetEnd) {
149            source = oldSource; /* Back up source pointer! */
150            result = targetExhausted; break;
151        }
152        *target++ = ch;
153    }
154    *sourceStart = source;
155    *targetStart = target;
156#ifdef CVTUTF_DEBUG
157if (result == sourceIllegal) {
158    fprintf(stderr, "ConvertUTF16toUTF32 illegal seq 0x%04x,%04x\n", ch, ch2);
159    fflush(stderr);
160}
161#endif
162    return result;
163}
164
165/* --------------------------------------------------------------------- */
166
167/*
168 * Index into the table below with the first byte of a UTF-8 sequence to
169 * get the number of trailing bytes that are supposed to follow it.
170 * Note that *legal* UTF-8 values can't have 4 or 5-bytes. The table is
171 * left as-is for anyone who may want to do such conversion, which was
172 * allowed in earlier algorithms.
173 */
174static const char trailingBytesForUTF8[256] = {
175    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,
176    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,
177    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,
178    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,
179    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,
180    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,
181    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,
182    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
183};
184
185/*
186 * Magic values subtracted from a buffer value during UTF8 conversion.
187 * This table contains as many values as there might be trailing bytes
188 * in a UTF-8 sequence.
189 */
190static const UTF32 offsetsFromUTF8[6] = { 0x00000000UL, 0x00003080UL, 0x000E2080UL,
191                     0x03C82080UL, 0xFA082080UL, 0x82082080UL };
192
193/*
194 * Once the bits are split out into bytes of UTF-8, this is a mask OR-ed
195 * into the first byte, depending on how many bytes follow.  There are
196 * as many entries in this table as there are UTF-8 sequence types.
197 * (I.e., one byte sequence, two byte... etc.). Remember that sequencs
198 * for *legal* UTF-8 will be 4 or fewer bytes total.
199 */
200static const UTF8 firstByteMark[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };
201
202/* --------------------------------------------------------------------- */
203
204/* The interface converts a whole buffer to avoid function-call overhead.
205 * Constants have been gathered. Loops & conditionals have been removed as
206 * much as possible for efficiency, in favor of drop-through switches.
207 * (See "Note A" at the bottom of the file for equivalent code.)
208 * If your compiler supports it, the "isLegalUTF8" call can be turned
209 * into an inline function.
210 */
211
212/* --------------------------------------------------------------------- */
213
214ConversionResult ConvertUTF16toUTF8 (
215        const UTF16** sourceStart, const UTF16* sourceEnd,
216        UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
217    ConversionResult result = conversionOK;
218    const UTF16* source = *sourceStart;
219    UTF8* target = *targetStart;
220    while (source < sourceEnd) {
221        UTF32 ch;
222        unsigned short bytesToWrite = 0;
223        const UTF32 byteMask = 0xBF;
224        const UTF32 byteMark = 0x80;
225        const UTF16* oldSource = source; /* In case we have to back up because of target overflow. */
226        ch = *source++;
227        /* If we have a surrogate pair, convert to UTF32 first. */
228        if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_HIGH_END) {
229            /* If the 16 bits following the high surrogate are in the source buffer... */
230            if (source < sourceEnd) {
231                UTF32 ch2 = *source;
232                /* If it's a low surrogate, convert to UTF32. */
233                if (ch2 >= UNI_SUR_LOW_START && ch2 <= UNI_SUR_LOW_END) {
234                    ch = ((ch - UNI_SUR_HIGH_START) << halfShift)
235                        + (ch2 - UNI_SUR_LOW_START) + halfBase;
236                    ++source;
237                } else if (flags == strictConversion) { /* it's an unpaired high surrogate */
238                    --source; /* return to the illegal value itself */
239                    result = sourceIllegal;
240                    break;
241                }
242            } else { /* We don't have the 16 bits following the high surrogate. */
243                --source; /* return to the high surrogate */
244                result = sourceExhausted;
245                break;
246            }
247        } else if (flags == strictConversion) {
248            /* UTF-16 surrogate values are illegal in UTF-32 */
249            if (ch >= UNI_SUR_LOW_START && ch <= UNI_SUR_LOW_END) {
250                --source; /* return to the illegal value itself */
251                result = sourceIllegal;
252                break;
253            }
254        }
255        /* Figure out how many bytes the result will require */
256        if (ch < (UTF32)0x80) {      bytesToWrite = 1;
257        } else if (ch < (UTF32)0x800) {     bytesToWrite = 2;
258        } else if (ch < (UTF32)0x10000) {   bytesToWrite = 3;
259        } else if (ch < (UTF32)0x110000) {  bytesToWrite = 4;
260        } else {                            bytesToWrite = 3;
261                                            ch = UNI_REPLACEMENT_CHAR;
262        }
263
264        target += bytesToWrite;
265        if (target > targetEnd) {
266            source = oldSource; /* Back up source pointer! */
267            target -= bytesToWrite; result = targetExhausted; break;
268        }
269        switch (bytesToWrite) { /* note: everything falls through. */
270            case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
271            case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
272            case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
273            case 1: *--target =  (UTF8)(ch | firstByteMark[bytesToWrite]);
274        }
275        target += bytesToWrite;
276    }
277    *sourceStart = source;
278    *targetStart = target;
279    return result;
280}
281
282/* --------------------------------------------------------------------- */
283
284/*
285 * Utility routine to tell whether a sequence of bytes is legal UTF-8.
286 * This must be called with the length pre-determined by the first byte.
287 * If not calling this from ConvertUTF8to*, then the length can be set by:
288 *  length = trailingBytesForUTF8[*source]+1;
289 * and the sequence is illegal right away if there aren't that many bytes
290 * available.
291 * If presented with a length > 4, this returns false.  The Unicode
292 * definition of UTF-8 goes up to 4-byte sequences.
293 */
294
295static Boolean isLegalUTF8(const UTF8 *source, int length) {
296    UTF8 a;
297    const UTF8 *srcptr = source+length;
298    switch (length) {
299    default: return false;
300        /* Everything else falls through when "true"... */
301    case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
302    case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
303    case 2: if ((a = (*--srcptr)) > 0xBF) return false;
304
305        switch (*source) {
306            /* no fall-through in this inner switch */
307            case 0xE0: if (a < 0xA0) return false; break;
308            case 0xED: if (a > 0x9F) return false; break;
309            case 0xF0: if (a < 0x90) return false; break;
310            case 0xF4: if (a > 0x8F) return false; break;
311            default:   if (a < 0x80) return false;
312        }
313
314    case 1: if (*source >= 0x80 && *source < 0xC2) return false;
315    }
316    if (*source > 0xF4) return false;
317    return true;
318}
319
320/* --------------------------------------------------------------------- */
321
322/*
323 * Exported function to return whether a UTF-8 sequence is legal or not.
324 * This is not used here; it's just exported.
325 */
326Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd) {
327    int length = trailingBytesForUTF8[*source]+1;
328    if (source+length > sourceEnd) {
329        return false;
330    }
331    return isLegalUTF8(source, length);
332}
333
334/* --------------------------------------------------------------------- */
335
336ConversionResult ConvertUTF8toUTF16 (
337        const UTF8** sourceStart, const UTF8* sourceEnd,
338        UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
339    ConversionResult result = conversionOK;
340    const UTF8* source = *sourceStart;
341    UTF16* target = *targetStart;
342    while (source < sourceEnd) {
343        UTF32 ch = 0;
344        unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
345        if (source + extraBytesToRead >= sourceEnd) {
346            result = sourceExhausted; break;
347        }
348        /* Do this check whether lenient or strict */
349        if (! isLegalUTF8(source, extraBytesToRead+1)) {
350            result = sourceIllegal;
351            break;
352        }
353        /*
354         * The cases all fall through. See "Note A" below.
355         */
356        switch (extraBytesToRead) {
357            case 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
358            case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
359            case 3: ch += *source++; ch <<= 6;
360            case 2: ch += *source++; ch <<= 6;
361            case 1: ch += *source++; ch <<= 6;
362            case 0: ch += *source++;
363        }
364        ch -= offsetsFromUTF8[extraBytesToRead];
365
366        if (target >= targetEnd) {
367            source -= (extraBytesToRead+1); /* Back up source pointer! */
368            result = targetExhausted; break;
369        }
370        if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
371            /* UTF-16 surrogate values are illegal in UTF-32 */
372            if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
373                if (flags == strictConversion) {
374                    source -= (extraBytesToRead+1); /* return to the illegal value itself */
375                    result = sourceIllegal;
376                    break;
377                } else {
378                    *target++ = UNI_REPLACEMENT_CHAR;
379                }
380            } else {
381                *target++ = (UTF16)ch; /* normal case */
382            }
383        } else if (ch > UNI_MAX_UTF16) {
384            if (flags == strictConversion) {
385                result = sourceIllegal;
386                source -= (extraBytesToRead+1); /* return to the start */
387                break; /* Bail out; shouldn't continue */
388            } else {
389                *target++ = UNI_REPLACEMENT_CHAR;
390            }
391        } else {
392            /* target is a character in range 0xFFFF - 0x10FFFF. */
393            if (target + 1 >= targetEnd) {
394                source -= (extraBytesToRead+1); /* Back up source pointer! */
395                result = targetExhausted; break;
396            }
397            ch -= halfBase;
398            *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
399            *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
400        }
401    }
402    *sourceStart = source;
403    *targetStart = target;
404    return result;
405}
406
407/* --------------------------------------------------------------------- */
408
409ConversionResult ConvertUTF32toUTF8 (
410        const UTF32** sourceStart, const UTF32* sourceEnd,
411        UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
412    ConversionResult result = conversionOK;
413    const UTF32* source = *sourceStart;
414    UTF8* target = *targetStart;
415    while (source < sourceEnd) {
416        UTF32 ch;
417        unsigned short bytesToWrite = 0;
418        const UTF32 byteMask = 0xBF;
419        const UTF32 byteMark = 0x80;
420        ch = *source++;
421        if (flags == strictConversion ) {
422            /* UTF-16 surrogate values are illegal in UTF-32 */
423            if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
424                --source; /* return to the illegal value itself */
425                result = sourceIllegal;
426                break;
427            }
428        }
429        /*
430         * Figure out how many bytes the result will require. Turn any
431         * illegally large UTF32 things (> Plane 17) into replacement chars.
432         */
433        if (ch < (UTF32)0x80) {      bytesToWrite = 1;
434        } else if (ch < (UTF32)0x800) {     bytesToWrite = 2;
435        } else if (ch < (UTF32)0x10000) {   bytesToWrite = 3;
436        } else if (ch <= UNI_MAX_LEGAL_UTF32) {  bytesToWrite = 4;
437        } else {                            bytesToWrite = 3;
438                                            ch = UNI_REPLACEMENT_CHAR;
439                                            result = sourceIllegal;
440        }
441       
442        target += bytesToWrite;
443        if (target > targetEnd) {
444            --source; /* Back up source pointer! */
445            target -= bytesToWrite; result = targetExhausted; break;
446        }
447        switch (bytesToWrite) { /* note: everything falls through. */
448            case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
449            /* fall-thru */
450            case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; /* fall-thru */
451            case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6; /* fall-thru */
452            case 1: *--target = (UTF8) (ch | firstByteMark[bytesToWrite]);    /* fall-thru */
453        }
454        target += bytesToWrite;
455    }
456    *sourceStart = source;
457    *targetStart = target;
458    return result;
459}
460
461/* --------------------------------------------------------------------- */
462
463ConversionResult ConvertUTF8toUTF32 (
464        const UTF8** sourceStart, const UTF8* sourceEnd,
465        UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
466    ConversionResult result = conversionOK;
467    const UTF8* source = *sourceStart;
468    UTF32* target = *targetStart;
469    while (source < sourceEnd) {
470        UTF32 ch = 0;
471        unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
472        if (source + extraBytesToRead >= sourceEnd) {
473            result = sourceExhausted; break;
474        }
475        /* Do this check whether lenient or strict */
476        if (! isLegalUTF8(source, extraBytesToRead+1)) {
477            result = sourceIllegal;
478            break;
479        }
480        /*
481         * The cases all fall through. See "Note A" below.
482         */
483        switch (extraBytesToRead) {
484            case 5: ch += *source++; ch <<= 6; /* fall-thru */
485            case 4: ch += *source++; ch <<= 6; /* fall-thru */
486            case 3: ch += *source++; ch <<= 6; /* fall-thru */
487            case 2: ch += *source++; ch <<= 6; /* fall-thru */
488            case 1: ch += *source++; ch <<= 6; /* fall-thru */
489            case 0: ch += *source++;           /* fall-thru */
490        }
491        ch -= offsetsFromUTF8[extraBytesToRead];
492
493        if (target >= targetEnd) {
494            source -= (extraBytesToRead+1); /* Back up the source pointer! */
495            result = targetExhausted; break;
496        }
497        if (ch <= UNI_MAX_LEGAL_UTF32) {
498            /*
499             * UTF-16 surrogate values are illegal in UTF-32, and anything
500             * over Plane 17 (> 0x10FFFF) is illegal.
501             */
502            if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
503                if (flags == strictConversion) {
504                    source -= (extraBytesToRead+1); /* return to the illegal value itself */
505                    result = sourceIllegal;
506                    break;
507                } else {
508                    *target++ = UNI_REPLACEMENT_CHAR;
509                }
510            } else {
511                *target++ = ch;
512            }
513        } else { /* i.e., ch > UNI_MAX_LEGAL_UTF32 */
514            result = sourceIllegal;
515            *target++ = UNI_REPLACEMENT_CHAR;
516        }
517    }
518    *sourceStart = source;
519    *targetStart = target;
520    return result;
521}
522
523/* ---------------------------------------------------------------------
524
525    Note A.
526    The fall-through switches in UTF-8 reading code save a
527    temp variable, some decrements & conditionals.  The switches
528    are equivalent to the following loop:
529        {
530            int tmpBytesToRead = extraBytesToRead+1;
531            do {
532                ch += *source++;
533                --tmpBytesToRead;
534                if (tmpBytesToRead) ch <<= 6;
535            } while (tmpBytesToRead > 0);
536        }
537    In UTF-8 writing code, the switches on "bytesToWrite" are
538    similarly unrolled loops.
539
540   --------------------------------------------------------------------- */
Note: See TracBrowser for help on using the repository browser.