1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ******************************************************************************* |
5 | * |
6 | * Copyright (C) 1999-2010, International Business Machines |
7 | * Corporation and others. All Rights Reserved. |
8 | * |
9 | ******************************************************************************* |
10 | * file name: uinvchar.c |
11 | * encoding: UTF-8 |
12 | * tab size: 8 (not used) |
13 | * indentation:2 |
14 | * |
15 | * created on: 2004sep14 |
16 | * created by: Markus W. Scherer |
17 | * |
18 | * Functions for handling invariant characters, moved here from putil.c |
19 | * for better modularization. |
20 | */ |
21 | |
22 | #include "unicode/utypes.h" |
23 | #include "unicode/ustring.h" |
24 | #include "udataswp.h" |
25 | #include "cstring.h" |
26 | #include "cmemory.h" |
27 | #include "uassert.h" |
28 | #include "uinvchar.h" |
29 | |
30 | /* invariant-character handling --------------------------------------------- */ |
31 | |
32 | /* |
33 | * These maps for ASCII to/from EBCDIC map invariant characters (see utypes.h) |
34 | * appropriately for most EBCDIC codepages. |
35 | * |
36 | * They currently also map most other ASCII graphic characters, |
37 | * appropriately for codepages 37 and 1047. |
38 | * Exceptions: The characters for []^ have different codes in 37 & 1047. |
39 | * Both versions are mapped to ASCII. |
40 | * |
41 | * ASCII 37 1047 |
42 | * [ 5B BA AD |
43 | * ] 5D BB BD |
44 | * ^ 5E B0 5F |
45 | * |
46 | * There are no mappings for variant characters from Unicode to EBCDIC. |
47 | * |
48 | * Currently, C0 control codes are also included in these maps. |
49 | * Exceptions: S/390 Open Edition swaps LF and NEL codes compared with other |
50 | * EBCDIC platforms; both codes (15 and 25) are mapped to ASCII LF (0A), |
51 | * but there is no mapping for ASCII LF back to EBCDIC. |
52 | * |
53 | * ASCII EBCDIC S/390-OE |
54 | * LF 0A 25 15 |
55 | * NEL 85 15 25 |
56 | * |
57 | * The maps below explicitly exclude the variant |
58 | * control and graphical characters that are in ASCII-based |
59 | * codepages at 0x80 and above. |
60 | * "No mapping" is expressed by mapping to a 00 byte. |
61 | * |
62 | * These tables do not establish a converter or a codepage. |
63 | */ |
64 | |
65 | static const uint8_t asciiFromEbcdic[256]={ |
66 | 0x00, 0x01, 0x02, 0x03, 0x00, 0x09, 0x00, 0x7f, 0x00, 0x00, 0x00, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, |
67 | 0x10, 0x11, 0x12, 0x13, 0x00, 0x0a, 0x08, 0x00, 0x18, 0x19, 0x00, 0x00, 0x1c, 0x1d, 0x1e, 0x1f, |
68 | 0x00, 0x00, 0x00, 0x00, 0x00, 0x0a, 0x17, 0x1b, 0x00, 0x00, 0x00, 0x00, 0x00, 0x05, 0x06, 0x07, |
69 | 0x00, 0x00, 0x16, 0x00, 0x00, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00, 0x00, 0x14, 0x15, 0x00, 0x1a, |
70 | |
71 | 0x20, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x2e, 0x3c, 0x28, 0x2b, 0x7c, |
72 | 0x26, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x21, 0x24, 0x2a, 0x29, 0x3b, 0x5e, |
73 | 0x2d, 0x2f, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x2c, 0x25, 0x5f, 0x3e, 0x3f, |
74 | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x60, 0x3a, 0x23, 0x40, 0x27, 0x3d, 0x22, |
75 | |
76 | 0x00, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
77 | 0x00, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, 0x70, 0x71, 0x72, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
78 | 0x00, 0x7e, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x00, 0x00, 0x00, 0x5b, 0x00, 0x00, |
79 | 0x5e, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x5b, 0x5d, 0x00, 0x5d, 0x00, 0x00, |
80 | |
81 | 0x7b, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
82 | 0x7d, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, 0x50, 0x51, 0x52, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
83 | 0x5c, 0x00, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
84 | 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 |
85 | }; |
86 | |
87 | static const uint8_t ebcdicFromAscii[256]={ |
88 | 0x00, 0x01, 0x02, 0x03, 0x37, 0x2d, 0x2e, 0x2f, 0x16, 0x05, 0x00, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, |
89 | 0x10, 0x11, 0x12, 0x13, 0x3c, 0x3d, 0x32, 0x26, 0x18, 0x19, 0x3f, 0x27, 0x1c, 0x1d, 0x1e, 0x1f, |
90 | 0x40, 0x00, 0x7f, 0x00, 0x00, 0x6c, 0x50, 0x7d, 0x4d, 0x5d, 0x5c, 0x4e, 0x6b, 0x60, 0x4b, 0x61, |
91 | 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xf9, 0x7a, 0x5e, 0x4c, 0x7e, 0x6e, 0x6f, |
92 | |
93 | 0x00, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, 0xc8, 0xc9, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, |
94 | 0xd7, 0xd8, 0xd9, 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9, 0x00, 0x00, 0x00, 0x00, 0x6d, |
95 | 0x00, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, |
96 | 0x97, 0x98, 0x99, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, 0xa8, 0xa9, 0x00, 0x00, 0x00, 0x00, 0x07, |
97 | |
98 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
99 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
100 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
101 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
102 | |
103 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
104 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
105 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
106 | 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 |
107 | }; |
108 | |
109 | /* Same as asciiFromEbcdic[] except maps all letters to lowercase. */ |
110 | static const uint8_t lowercaseAsciiFromEbcdic[256]={ |
111 | 0x00, 0x01, 0x02, 0x03, 0x00, 0x09, 0x00, 0x7f, 0x00, 0x00, 0x00, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, |
112 | 0x10, 0x11, 0x12, 0x13, 0x00, 0x0a, 0x08, 0x00, 0x18, 0x19, 0x00, 0x00, 0x1c, 0x1d, 0x1e, 0x1f, |
113 | 0x00, 0x00, 0x00, 0x00, 0x00, 0x0a, 0x17, 0x1b, 0x00, 0x00, 0x00, 0x00, 0x00, 0x05, 0x06, 0x07, |
114 | 0x00, 0x00, 0x16, 0x00, 0x00, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00, 0x00, 0x14, 0x15, 0x00, 0x1a, |
115 | |
116 | 0x20, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x2e, 0x3c, 0x28, 0x2b, 0x7c, |
117 | 0x26, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x21, 0x24, 0x2a, 0x29, 0x3b, 0x5e, |
118 | 0x2d, 0x2f, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x2c, 0x25, 0x5f, 0x3e, 0x3f, |
119 | 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x60, 0x3a, 0x23, 0x40, 0x27, 0x3d, 0x22, |
120 | |
121 | 0x00, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
122 | 0x00, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, 0x70, 0x71, 0x72, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
123 | 0x00, 0x7e, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x00, 0x00, 0x00, 0x5b, 0x00, 0x00, |
124 | 0x5e, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x5b, 0x5d, 0x00, 0x5d, 0x00, 0x00, |
125 | |
126 | 0x7b, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
127 | 0x7d, 0x6a, 0x6b, 0x6c, 0x6d, 0x6e, 0x6f, 0x70, 0x71, 0x72, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
128 | 0x7c, 0x00, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, |
129 | 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 |
130 | }; |
131 | |
132 | /* |
133 | * Bit sets indicating which characters of the ASCII repertoire |
134 | * (by ASCII/Unicode code) are "invariant". |
135 | * See utypes.h for more details. |
136 | * |
137 | * As invariant are considered the characters of the ASCII repertoire except |
138 | * for the following: |
139 | * 21 '!' <exclamation mark> |
140 | * 23 '#' <number sign> |
141 | * 24 '$' <dollar sign> |
142 | * |
143 | * 40 '@' <commercial at> |
144 | * |
145 | * 5b '[' <left bracket> |
146 | * 5c '\' <backslash> |
147 | * 5d ']' <right bracket> |
148 | * 5e '^' <circumflex> |
149 | * |
150 | * 60 '`' <grave accent> |
151 | * |
152 | * 7b '{' <left brace> |
153 | * 7c '|' <vertical line> |
154 | * 7d '}' <right brace> |
155 | * 7e '~' <tilde> |
156 | */ |
157 | static const uint32_t invariantChars[4]={ |
158 | 0xfffffbff, /* 00..1f but not 0a */ |
159 | 0xffffffe5, /* 20..3f but not 21 23 24 */ |
160 | 0x87fffffe, /* 40..5f but not 40 5b..5e */ |
161 | 0x87fffffe /* 60..7f but not 60 7b..7e */ |
162 | }; |
163 | |
164 | /* |
165 | * test unsigned types (or values known to be non-negative) for invariant characters, |
166 | * tests ASCII-family character values |
167 | */ |
168 | #define UCHAR_IS_INVARIANT(c) (((c)<=0x7f) && (invariantChars[(c)>>5]&((uint32_t)1<<((c)&0x1f)))!=0) |
169 | |
170 | /* test signed types for invariant characters, adds test for positive values */ |
171 | #define SCHAR_IS_INVARIANT(c) ((0<=(c)) && UCHAR_IS_INVARIANT(c)) |
172 | |
173 | #if U_CHARSET_FAMILY==U_ASCII_FAMILY |
174 | #define CHAR_TO_UCHAR(c) c |
175 | #define UCHAR_TO_CHAR(c) c |
176 | #elif U_CHARSET_FAMILY==U_EBCDIC_FAMILY |
177 | #define CHAR_TO_UCHAR(u) asciiFromEbcdic[u] |
178 | #define UCHAR_TO_CHAR(u) ebcdicFromAscii[u] |
179 | #else |
180 | # error U_CHARSET_FAMILY is not valid |
181 | #endif |
182 | |
183 | |
184 | U_CAPI void U_EXPORT2 |
185 | u_charsToUChars(const char *cs, char16_t *us, int32_t length) { |
186 | char16_t u; |
187 | uint8_t c; |
188 | |
189 | /* |
190 | * Allow the entire ASCII repertoire to be mapped _to_ Unicode. |
191 | * For EBCDIC systems, this works for characters with codes from |
192 | * codepages 37 and 1047 or compatible. |
193 | */ |
194 | while(length>0) { |
195 | c=(uint8_t)(*cs++); |
196 | u=(char16_t)CHAR_TO_UCHAR(c); |
197 | U_ASSERT((u!=0 || c==0)); /* only invariant chars converted? */ |
198 | *us++=u; |
199 | --length; |
200 | } |
201 | } |
202 | |
203 | U_CAPI void U_EXPORT2 |
204 | u_UCharsToChars(const char16_t *us, char *cs, int32_t length) { |
205 | char16_t u; |
206 | |
207 | while(length>0) { |
208 | u=*us++; |
209 | if(!UCHAR_IS_INVARIANT(u)) { |
210 | U_ASSERT(false); /* Variant characters were used. These are not portable in ICU. */ |
211 | u=0; |
212 | } |
213 | *cs++=(char)UCHAR_TO_CHAR(u); |
214 | --length; |
215 | } |
216 | } |
217 | |
218 | U_CAPI UBool U_EXPORT2 |
219 | uprv_isInvariantString(const char *s, int32_t length) { |
220 | uint8_t c; |
221 | |
222 | for(;;) { |
223 | if(length<0) { |
224 | /* NUL-terminated */ |
225 | c=(uint8_t)*s++; |
226 | if(c==0) { |
227 | break; |
228 | } |
229 | } else { |
230 | /* count length */ |
231 | if(length==0) { |
232 | break; |
233 | } |
234 | --length; |
235 | c=(uint8_t)*s++; |
236 | if(c==0) { |
237 | continue; /* NUL is invariant */ |
238 | } |
239 | } |
240 | /* c!=0 now, one branch below checks c==0 for variant characters */ |
241 | |
242 | /* |
243 | * no assertions here because these functions are legitimately called |
244 | * for strings with variant characters |
245 | */ |
246 | #if U_CHARSET_FAMILY==U_ASCII_FAMILY |
247 | if(!UCHAR_IS_INVARIANT(c)) { |
248 | return false; /* found a variant char */ |
249 | } |
250 | #elif U_CHARSET_FAMILY==U_EBCDIC_FAMILY |
251 | c=CHAR_TO_UCHAR(c); |
252 | if(c==0 || !UCHAR_IS_INVARIANT(c)) { |
253 | return false; /* found a variant char */ |
254 | } |
255 | #else |
256 | # error U_CHARSET_FAMILY is not valid |
257 | #endif |
258 | } |
259 | return true; |
260 | } |
261 | |
262 | U_CAPI UBool U_EXPORT2 |
263 | uprv_isInvariantUString(const char16_t *s, int32_t length) { |
264 | char16_t c; |
265 | |
266 | for(;;) { |
267 | if(length<0) { |
268 | /* NUL-terminated */ |
269 | c=*s++; |
270 | if(c==0) { |
271 | break; |
272 | } |
273 | } else { |
274 | /* count length */ |
275 | if(length==0) { |
276 | break; |
277 | } |
278 | --length; |
279 | c=*s++; |
280 | } |
281 | |
282 | /* |
283 | * no assertions here because these functions are legitimately called |
284 | * for strings with variant characters |
285 | */ |
286 | if(!UCHAR_IS_INVARIANT(c)) { |
287 | return false; /* found a variant char */ |
288 | } |
289 | } |
290 | return true; |
291 | } |
292 | |
293 | /* UDataSwapFn implementations used in udataswp.c ------- */ |
294 | |
295 | /* convert ASCII to EBCDIC and verify that all characters are invariant */ |
296 | U_CAPI int32_t U_EXPORT2 |
297 | uprv_ebcdicFromAscii(const UDataSwapper *ds, |
298 | const void *inData, int32_t length, void *outData, |
299 | UErrorCode *pErrorCode) { |
300 | const uint8_t *s; |
301 | uint8_t *t; |
302 | uint8_t c; |
303 | |
304 | int32_t count; |
305 | |
306 | if(pErrorCode==nullptr || U_FAILURE(*pErrorCode)) { |
307 | return 0; |
308 | } |
309 | if(ds==nullptr || inData==nullptr || length<0 || (length>0 && outData==nullptr)) { |
310 | *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR; |
311 | return 0; |
312 | } |
313 | |
314 | /* setup and swapping */ |
315 | s=(const uint8_t *)inData; |
316 | t=(uint8_t *)outData; |
317 | count=length; |
318 | while(count>0) { |
319 | c=*s++; |
320 | if(!UCHAR_IS_INVARIANT(c)) { |
321 | udata_printError(ds, "uprv_ebcdicFromAscii() string[%d] contains a variant character in position %d\n" , |
322 | length, length-count); |
323 | *pErrorCode=U_INVALID_CHAR_FOUND; |
324 | return 0; |
325 | } |
326 | *t++=ebcdicFromAscii[c]; |
327 | --count; |
328 | } |
329 | |
330 | return length; |
331 | } |
332 | |
333 | /* this function only checks and copies ASCII strings without conversion */ |
334 | U_CFUNC int32_t |
335 | uprv_copyAscii(const UDataSwapper *ds, |
336 | const void *inData, int32_t length, void *outData, |
337 | UErrorCode *pErrorCode) { |
338 | const uint8_t *s; |
339 | uint8_t c; |
340 | |
341 | int32_t count; |
342 | |
343 | if(pErrorCode==nullptr || U_FAILURE(*pErrorCode)) { |
344 | return 0; |
345 | } |
346 | if(ds==nullptr || inData==nullptr || length<0 || (length>0 && outData==nullptr)) { |
347 | *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR; |
348 | return 0; |
349 | } |
350 | |
351 | /* setup and checking */ |
352 | s=(const uint8_t *)inData; |
353 | count=length; |
354 | while(count>0) { |
355 | c=*s++; |
356 | if(!UCHAR_IS_INVARIANT(c)) { |
357 | udata_printError(ds, "uprv_copyFromAscii() string[%d] contains a variant character in position %d\n" , |
358 | length, length-count); |
359 | *pErrorCode=U_INVALID_CHAR_FOUND; |
360 | return 0; |
361 | } |
362 | --count; |
363 | } |
364 | |
365 | if(length>0 && inData!=outData) { |
366 | uprv_memcpy(outData, inData, length); |
367 | } |
368 | |
369 | return length; |
370 | } |
371 | |
372 | /* convert EBCDIC to ASCII and verify that all characters are invariant */ |
373 | U_CFUNC int32_t |
374 | uprv_asciiFromEbcdic(const UDataSwapper *ds, |
375 | const void *inData, int32_t length, void *outData, |
376 | UErrorCode *pErrorCode) { |
377 | const uint8_t *s; |
378 | uint8_t *t; |
379 | uint8_t c; |
380 | |
381 | int32_t count; |
382 | |
383 | if(pErrorCode==nullptr || U_FAILURE(*pErrorCode)) { |
384 | return 0; |
385 | } |
386 | if(ds==nullptr || inData==nullptr || length<0 || (length>0 && outData==nullptr)) { |
387 | *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR; |
388 | return 0; |
389 | } |
390 | |
391 | /* setup and swapping */ |
392 | s=(const uint8_t *)inData; |
393 | t=(uint8_t *)outData; |
394 | count=length; |
395 | while(count>0) { |
396 | c=*s++; |
397 | if(c!=0 && ((c=asciiFromEbcdic[c])==0 || !UCHAR_IS_INVARIANT(c))) { |
398 | udata_printError(ds, "uprv_asciiFromEbcdic() string[%d] contains a variant character in position %d\n" , |
399 | length, length-count); |
400 | *pErrorCode=U_INVALID_CHAR_FOUND; |
401 | return 0; |
402 | } |
403 | *t++=c; |
404 | --count; |
405 | } |
406 | |
407 | return length; |
408 | } |
409 | |
410 | /* this function only checks and copies EBCDIC strings without conversion */ |
411 | U_CFUNC int32_t |
412 | uprv_copyEbcdic(const UDataSwapper *ds, |
413 | const void *inData, int32_t length, void *outData, |
414 | UErrorCode *pErrorCode) { |
415 | const uint8_t *s; |
416 | uint8_t c; |
417 | |
418 | int32_t count; |
419 | |
420 | if(pErrorCode==nullptr || U_FAILURE(*pErrorCode)) { |
421 | return 0; |
422 | } |
423 | if(ds==nullptr || inData==nullptr || length<0 || (length>0 && outData==nullptr)) { |
424 | *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR; |
425 | return 0; |
426 | } |
427 | |
428 | /* setup and checking */ |
429 | s=(const uint8_t *)inData; |
430 | count=length; |
431 | while(count>0) { |
432 | c=*s++; |
433 | if(c!=0 && ((c=asciiFromEbcdic[c])==0 || !UCHAR_IS_INVARIANT(c))) { |
434 | udata_printError(ds, "uprv_copyEbcdic() string[%] contains a variant character in position %d\n" , |
435 | length, length-count); |
436 | *pErrorCode=U_INVALID_CHAR_FOUND; |
437 | return 0; |
438 | } |
439 | --count; |
440 | } |
441 | |
442 | if(length>0 && inData!=outData) { |
443 | uprv_memcpy(outData, inData, length); |
444 | } |
445 | |
446 | return length; |
447 | } |
448 | |
449 | U_CFUNC UBool |
450 | uprv_isEbcdicAtSign(char c) { |
451 | static const uint8_t ebcdicAtSigns[] = { |
452 | 0x7C, 0x44, 0x66, 0x80, 0xAC, 0xAE, 0xAF, 0xB5, 0xEC, 0xEF, 0x00 }; |
453 | return c != 0 && uprv_strchr((const char *)ebcdicAtSigns, c) != nullptr; |
454 | } |
455 | |
456 | /* compare invariant strings; variant characters compare less than others and unlike each other */ |
457 | U_CFUNC int32_t |
458 | uprv_compareInvAscii(const UDataSwapper *ds, |
459 | const char *outString, int32_t outLength, |
460 | const char16_t *localString, int32_t localLength) { |
461 | (void)ds; |
462 | int32_t minLength; |
463 | UChar32 c1, c2; |
464 | uint8_t c; |
465 | |
466 | if(outString==nullptr || outLength<-1 || localString==nullptr || localLength<-1) { |
467 | return 0; |
468 | } |
469 | |
470 | if(outLength<0) { |
471 | outLength=(int32_t)uprv_strlen(outString); |
472 | } |
473 | if(localLength<0) { |
474 | localLength=u_strlen(localString); |
475 | } |
476 | |
477 | minLength= outLength<localLength ? outLength : localLength; |
478 | |
479 | while(minLength>0) { |
480 | c=(uint8_t)*outString++; |
481 | if(UCHAR_IS_INVARIANT(c)) { |
482 | c1=c; |
483 | } else { |
484 | c1=-1; |
485 | } |
486 | |
487 | c2=*localString++; |
488 | if(!UCHAR_IS_INVARIANT(c2)) { |
489 | c2=-2; |
490 | } |
491 | |
492 | if((c1-=c2)!=0) { |
493 | return c1; |
494 | } |
495 | |
496 | --minLength; |
497 | } |
498 | |
499 | /* strings start with same prefix, compare lengths */ |
500 | return outLength-localLength; |
501 | } |
502 | |
503 | U_CFUNC int32_t |
504 | uprv_compareInvEbcdic(const UDataSwapper *ds, |
505 | const char *outString, int32_t outLength, |
506 | const char16_t *localString, int32_t localLength) { |
507 | (void)ds; |
508 | int32_t minLength; |
509 | UChar32 c1, c2; |
510 | uint8_t c; |
511 | |
512 | if(outString==nullptr || outLength<-1 || localString==nullptr || localLength<-1) { |
513 | return 0; |
514 | } |
515 | |
516 | if(outLength<0) { |
517 | outLength=(int32_t)uprv_strlen(outString); |
518 | } |
519 | if(localLength<0) { |
520 | localLength=u_strlen(localString); |
521 | } |
522 | |
523 | minLength= outLength<localLength ? outLength : localLength; |
524 | |
525 | while(minLength>0) { |
526 | c=(uint8_t)*outString++; |
527 | if(c==0) { |
528 | c1=0; |
529 | } else if((c1=asciiFromEbcdic[c])!=0 && UCHAR_IS_INVARIANT(c1)) { |
530 | /* c1 is set */ |
531 | } else { |
532 | c1=-1; |
533 | } |
534 | |
535 | c2=*localString++; |
536 | if(!UCHAR_IS_INVARIANT(c2)) { |
537 | c2=-2; |
538 | } |
539 | |
540 | if((c1-=c2)!=0) { |
541 | return c1; |
542 | } |
543 | |
544 | --minLength; |
545 | } |
546 | |
547 | /* strings start with same prefix, compare lengths */ |
548 | return outLength-localLength; |
549 | } |
550 | |
551 | U_CAPI int32_t U_EXPORT2 |
552 | uprv_compareInvEbcdicAsAscii(const char *s1, const char *s2) { |
553 | int32_t c1, c2; |
554 | |
555 | for(;; ++s1, ++s2) { |
556 | c1=(uint8_t)*s1; |
557 | c2=(uint8_t)*s2; |
558 | if(c1!=c2) { |
559 | if(c1!=0 && ((c1=asciiFromEbcdic[c1])==0 || !UCHAR_IS_INVARIANT(c1))) { |
560 | c1=-(int32_t)(uint8_t)*s1; |
561 | } |
562 | if(c2!=0 && ((c2=asciiFromEbcdic[c2])==0 || !UCHAR_IS_INVARIANT(c2))) { |
563 | c2=-(int32_t)(uint8_t)*s2; |
564 | } |
565 | return c1-c2; |
566 | } else if(c1==0) { |
567 | return 0; |
568 | } |
569 | } |
570 | } |
571 | |
572 | U_CAPI char U_EXPORT2 |
573 | uprv_ebcdicToAscii(char c) { |
574 | return (char)asciiFromEbcdic[(uint8_t)c]; |
575 | } |
576 | |
577 | U_CAPI char U_EXPORT2 |
578 | uprv_ebcdicToLowercaseAscii(char c) { |
579 | return (char)lowercaseAsciiFromEbcdic[(uint8_t)c]; |
580 | } |
581 | |
582 | U_CAPI uint8_t* U_EXPORT2 |
583 | uprv_aestrncpy(uint8_t *dst, const uint8_t *src, int32_t n) |
584 | { |
585 | uint8_t *orig_dst = dst; |
586 | |
587 | if(n==-1) { |
588 | n = static_cast<int32_t>(uprv_strlen((const char*)src)+1); /* copy NUL */ |
589 | } |
590 | /* copy non-null */ |
591 | while(*src && n>0) { |
592 | *(dst++) = asciiFromEbcdic[*(src++)]; |
593 | n--; |
594 | } |
595 | /* pad */ |
596 | while(n>0) { |
597 | *(dst++) = 0; |
598 | n--; |
599 | } |
600 | return orig_dst; |
601 | } |
602 | |
603 | U_CAPI uint8_t* U_EXPORT2 |
604 | uprv_eastrncpy(uint8_t *dst, const uint8_t *src, int32_t n) |
605 | { |
606 | uint8_t *orig_dst = dst; |
607 | |
608 | if(n==-1) { |
609 | n = static_cast<int32_t>(uprv_strlen((const char*)src)+1); /* copy NUL */ |
610 | } |
611 | /* copy non-null */ |
612 | while(*src && n>0) { |
613 | char ch = ebcdicFromAscii[*(src++)]; |
614 | if(ch == 0) { |
615 | ch = ebcdicFromAscii[0x3f]; /* questionmark (subchar) */ |
616 | } |
617 | *(dst++) = ch; |
618 | n--; |
619 | } |
620 | /* pad */ |
621 | while(n>0) { |
622 | *(dst++) = 0; |
623 | n--; |
624 | } |
625 | return orig_dst; |
626 | } |
627 | |
628 | |