/src/logging-log4cxx/src/main/cpp/charsetencoder.cpp
Line | Count | Source |
1 | | /* |
2 | | * Licensed to the Apache Software Foundation (ASF) under one or more |
3 | | * contributor license agreements. See the NOTICE file distributed with |
4 | | * this work for additional information regarding copyright ownership. |
5 | | * The ASF licenses this file to You under the Apache License, Version 2.0 |
6 | | * (the "License"); you may not use this file except in compliance with |
7 | | * the License. You may obtain a copy of the License at |
8 | | * |
9 | | * http://www.apache.org/licenses/LICENSE-2.0 |
10 | | * |
11 | | * Unless required by applicable law or agreed to in writing, software |
12 | | * distributed under the License is distributed on an "AS IS" BASIS, |
13 | | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
14 | | * See the License for the specific language governing permissions and |
15 | | * limitations under the License. |
16 | | */ |
17 | | #include <log4cxx/logstring.h> |
18 | | #include <log4cxx/helpers/charsetencoder.h> |
19 | | #include <log4cxx/helpers/bytebuffer.h> |
20 | | #include <log4cxx/helpers/exception.h> |
21 | | #include <apr_xlate.h> |
22 | | #include <log4cxx/helpers/stringhelper.h> |
23 | | #include <log4cxx/helpers/transcoder.h> |
24 | | #include <algorithm> |
25 | | |
26 | | #if !defined(LOG4CXX) |
27 | | #define LOG4CXX 1 |
28 | | #endif |
29 | | |
30 | | #include <log4cxx/private/log4cxx_private.h> |
31 | | #include <apr_portable.h> |
32 | | #include <mutex> |
33 | | |
34 | | #ifdef LOG4CXX_HAS_WCSTOMBS |
35 | | #include <stdlib.h> |
36 | | #endif |
37 | | |
38 | | #if 15 < LOG4CXX_ABI_VERSION |
39 | | #define LOG4CXX_16_VIRTUAL_SPECIFIER override |
40 | | #else |
41 | | #define LOG4CXX_16_VIRTUAL_SPECIFIER |
42 | | #endif |
43 | | |
44 | | using namespace LOG4CXX_NS; |
45 | | using namespace LOG4CXX_NS::helpers; |
46 | | |
47 | | IMPLEMENT_LOG4CXX_OBJECT(CharsetEncoder) |
48 | | |
49 | | namespace LOG4CXX_NS |
50 | | { |
51 | | |
52 | | namespace helpers |
53 | | { |
54 | | |
55 | | #if APR_HAS_XLATE |
56 | | /** |
57 | | * A character encoder implemented using apr_xlate. |
58 | | */ |
59 | | class APRCharsetEncoder : public CharsetEncoder |
60 | | { |
61 | | public: |
62 | 0 | APRCharsetEncoder(const LogString& topage) : pool() |
63 | 0 | { |
64 | 0 | #if LOG4CXX_LOGCHAR_IS_WCHAR |
65 | 0 | const char* frompage = "WCHAR_T"; |
66 | 0 | #endif |
67 | | #if LOG4CXX_LOGCHAR_IS_UTF8 |
68 | | const char* frompage = "UTF-8"; |
69 | | #endif |
70 | | #if LOG4CXX_LOGCHAR_IS_UNICHAR |
71 | | const char* frompage = "UTF-16"; |
72 | | #endif |
73 | 0 | std::string tpage(Transcoder::encodeCharsetName(topage)); |
74 | 0 | apr_status_t stat = apr_xlate_open(&convset, |
75 | 0 | tpage.c_str(), |
76 | 0 | frompage, |
77 | 0 | pool.getAPRPool()); |
78 | |
|
79 | 0 | if (stat != APR_SUCCESS) |
80 | 0 | { |
81 | 0 | throw IllegalArgumentException(topage); |
82 | 0 | } |
83 | 0 | } |
84 | | |
85 | | virtual ~APRCharsetEncoder() |
86 | 0 | { |
87 | 0 | } |
88 | | |
89 | | virtual log4cxx_status_t encode(const LogString& in, |
90 | | LogString::const_iterator& iter, |
91 | | ByteBuffer& out) override |
92 | 0 | { |
93 | 0 | apr_status_t stat; |
94 | 0 | size_t outbytes_left = out.remaining(); |
95 | 0 | size_t initial_outbytes_left = outbytes_left; |
96 | 0 | size_t position = out.position(); |
97 | |
|
98 | 0 | if (iter == in.end()) |
99 | 0 | { |
100 | 0 | std::lock_guard<std::mutex> lock(mutex); |
101 | 0 | stat = apr_xlate_conv_buffer(convset, NULL, NULL, |
102 | 0 | out.data() + position, &outbytes_left); |
103 | 0 | } |
104 | 0 | else |
105 | 0 | { |
106 | 0 | LogString::size_type inOffset = (iter - in.begin()); |
107 | 0 | apr_size_t inbytes_left = |
108 | 0 | (in.size() - inOffset) * sizeof(LogString::value_type); |
109 | 0 | apr_size_t initial_inbytes_left = inbytes_left; |
110 | 0 | { |
111 | 0 | std::lock_guard<std::mutex> lock(mutex); |
112 | 0 | stat = apr_xlate_conv_buffer(convset, |
113 | 0 | (const char*) (in.data() + inOffset), |
114 | 0 | &inbytes_left, |
115 | 0 | out.data() + position, |
116 | 0 | &outbytes_left); |
117 | 0 | } |
118 | 0 | iter += ((initial_inbytes_left - inbytes_left) / sizeof(LogString::value_type)); |
119 | 0 | } |
120 | |
|
121 | 0 | out.increment_position((initial_outbytes_left - outbytes_left)); |
122 | 0 | return stat; |
123 | 0 | } |
124 | | |
125 | | /** |
126 | | * Add onto \c out an encoded equivalent of \c codePoint. |
127 | | */ |
128 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
129 | 0 | { |
130 | 0 | apr_status_t result = APR_SUCCESS; |
131 | 0 | if (codePoint <= 0x10FFFF) |
132 | 0 | Transcoder::encodeUTF8(codePoint, out); |
133 | 0 | else |
134 | 0 | result = APR_BADARG; |
135 | 0 | return result; |
136 | 0 | } |
137 | | |
138 | | private: |
139 | | APRCharsetEncoder(const APRCharsetEncoder&); |
140 | | APRCharsetEncoder& operator=(const APRCharsetEncoder&); |
141 | | Pool pool; |
142 | | std::mutex mutex; |
143 | | apr_xlate_t* convset; |
144 | | }; |
145 | | #endif |
146 | | |
147 | | #if LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS |
148 | | /** |
149 | | * A character encoder implemented using wcstombs. |
150 | | */ |
151 | | class WcstombsCharsetEncoder : public CharsetEncoder |
152 | | { |
153 | | public: |
154 | | WcstombsCharsetEncoder() |
155 | 0 | { |
156 | 0 | } |
157 | | |
158 | | /** |
159 | | * Converts a wchar_t to the default external multibyte encoding. |
160 | | */ |
161 | | log4cxx_status_t encode(const LogString& in, |
162 | | LogString::const_iterator& iter, |
163 | | ByteBuffer& out) override |
164 | 0 | { |
165 | 0 | log4cxx_status_t stat = APR_SUCCESS; |
166 | 0 |
|
167 | 0 | if (iter != in.end()) |
168 | 0 | { |
169 | 0 | size_t outbytes_left = out.remaining(); |
170 | 0 | size_t position = out.position(); |
171 | 0 | std::wstring::size_type inOffset = (iter - in.begin()); |
172 | 0 | enum { BUFSIZE = 256 }; |
173 | 0 | wchar_t buf[BUFSIZE]; |
174 | 0 | size_t chunkSize = BUFSIZE - 1; |
175 | 0 |
|
176 | 0 | if (chunkSize * MB_LEN_MAX > outbytes_left) |
177 | 0 | { |
178 | 0 | chunkSize = outbytes_left / MB_LEN_MAX; |
179 | 0 | } |
180 | 0 |
|
181 | 0 | if (chunkSize > in.length() - inOffset) |
182 | 0 | { |
183 | 0 | chunkSize = in.length() - inOffset; |
184 | 0 | } |
185 | 0 |
|
186 | 0 | memset(buf, 0, BUFSIZE * sizeof(wchar_t)); |
187 | 0 | memcpy(buf, |
188 | 0 | in.data() + inOffset, |
189 | 0 | chunkSize * sizeof(wchar_t)); |
190 | 0 | size_t converted = wcstombs(out.data() + position, buf, outbytes_left); |
191 | 0 |
|
192 | 0 | if (converted == (size_t) -1) |
193 | 0 | { |
194 | 0 | stat = APR_BADARG; |
195 | 0 |
|
196 | 0 | // |
197 | 0 | // if unconvertable character was encountered |
198 | 0 | // repeatedly halve source to get fragment that |
199 | 0 | // can be converted |
200 | 0 | for (chunkSize /= 2; |
201 | 0 | chunkSize > 0; |
202 | 0 | chunkSize /= 2) |
203 | 0 | { |
204 | 0 | buf[chunkSize] = 0; |
205 | 0 | converted = wcstombs(out.data() + position, buf, outbytes_left); |
206 | 0 |
|
207 | 0 | if (converted != (size_t) -1) |
208 | 0 | { |
209 | 0 | iter += chunkSize; |
210 | 0 | out.increment_position(converted); |
211 | 0 | break; |
212 | 0 | } |
213 | 0 | } |
214 | 0 | } |
215 | 0 | else |
216 | 0 | { |
217 | 0 | iter += chunkSize; |
218 | 0 | out.increment_position(converted); |
219 | 0 | } |
220 | 0 | } |
221 | 0 |
|
222 | 0 | return stat; |
223 | 0 | } |
224 | | |
225 | | /** |
226 | | * Add onto \c out an encoded equivalent of \c codePoint. |
227 | | */ |
228 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
229 | 0 | { |
230 | 0 | apr_status_t result = APR_SUCCESS; |
231 | 0 | if (codePoint <= 0x10FFFF && MB_LEN_MAX <= out.remaining()) |
232 | 0 | { |
233 | 0 | auto ch = static_cast<wchar_t>(codePoint); |
234 | 0 | auto converted = wcstombs(out.current(), &ch, 1); |
235 | 0 | if (static_cast<std::size_t>(-1) == converted) |
236 | 0 | result = APR_BADARG; |
237 | 0 | else |
238 | 0 | out.increment_position(converted); |
239 | 0 | } |
240 | 0 | else |
241 | 0 | result = APR_BADARG; |
242 | 0 | return result; |
243 | 0 | } |
244 | | |
245 | | private: |
246 | | WcstombsCharsetEncoder(const WcstombsCharsetEncoder&); |
247 | | WcstombsCharsetEncoder& operator=(const WcstombsCharsetEncoder&); |
248 | | }; |
249 | | #endif |
250 | | |
251 | | |
252 | | /** |
253 | | * Encodes a LogString to US-ASCII. |
254 | | */ |
255 | | class USASCIICharsetEncoder : public CharsetEncoder |
256 | | { |
257 | | public: |
258 | | USASCIICharsetEncoder() |
259 | 0 | { |
260 | 0 | } |
261 | | |
262 | | virtual log4cxx_status_t encode(const LogString& in, |
263 | | LogString::const_iterator& iter, |
264 | | ByteBuffer& out) override |
265 | 0 | { |
266 | 0 | log4cxx_status_t stat = APR_SUCCESS; |
267 | |
|
268 | 0 | if (iter != in.end()) |
269 | 0 | { |
270 | 0 | while (out.remaining() > 0 && iter != in.end()) |
271 | 0 | { |
272 | 0 | LogString::const_iterator prev(iter); |
273 | 0 | unsigned int sv = Transcoder::decode(in, iter); |
274 | |
|
275 | 0 | if (sv <= 0x7F) |
276 | 0 | { |
277 | 0 | out.put((char) sv); |
278 | 0 | } |
279 | 0 | else |
280 | 0 | { |
281 | 0 | iter = prev; |
282 | 0 | stat = APR_BADARG; |
283 | 0 | break; |
284 | 0 | } |
285 | 0 | } |
286 | 0 | } |
287 | |
|
288 | 0 | return stat; |
289 | 0 | } |
290 | | |
291 | | /** |
292 | | * Add onto \c out an encoded equivalent of \c codePoint. |
293 | | */ |
294 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
295 | 0 | { |
296 | 0 | apr_status_t result = APR_SUCCESS; |
297 | 0 | if (out.remaining() < 1) |
298 | 0 | result = APR_BADARG; |
299 | 0 | else if (codePoint <= 0x7F) |
300 | 0 | out.put(static_cast<char>(codePoint)); |
301 | 0 | else if (Transcoder::LOSSCHAR == codePoint) |
302 | 0 | out.put('?'); |
303 | 0 | else |
304 | 0 | result = APR_BADARG; |
305 | 0 | return result; |
306 | 0 | } |
307 | | |
308 | | private: |
309 | | USASCIICharsetEncoder(const USASCIICharsetEncoder&); |
310 | | USASCIICharsetEncoder& operator=(const USASCIICharsetEncoder&); |
311 | | }; |
312 | | |
313 | | /** |
314 | | * Converts a LogString to ISO-8859-1. |
315 | | */ |
316 | | class ISOLatinCharsetEncoder : public CharsetEncoder |
317 | | { |
318 | | public: |
319 | | ISOLatinCharsetEncoder() |
320 | 0 | { |
321 | 0 | } |
322 | | |
323 | | virtual log4cxx_status_t encode(const LogString& in, |
324 | | LogString::const_iterator& iter, |
325 | | ByteBuffer& out) override |
326 | 0 | { |
327 | 0 | log4cxx_status_t stat = APR_SUCCESS; |
328 | |
|
329 | 0 | while (out.remaining() > 0 && iter != in.end()) |
330 | 0 | { |
331 | 0 | LogString::const_iterator prev(iter); |
332 | 0 | unsigned int sv = Transcoder::decode(in, iter); |
333 | 0 | if (sv <= 0xFF) |
334 | 0 | out.put(static_cast<char>(sv)); |
335 | 0 | else |
336 | 0 | { |
337 | 0 | iter = prev; |
338 | 0 | stat = APR_BADARG; |
339 | 0 | break; |
340 | 0 | } |
341 | 0 | } |
342 | 0 | return stat; |
343 | 0 | } |
344 | | |
345 | | /** |
346 | | * Add onto \c out an encoded equivalent of \c codePoint. |
347 | | */ |
348 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
349 | 0 | { |
350 | 0 | apr_status_t result = APR_SUCCESS; |
351 | 0 | if (out.remaining() < 1) |
352 | 0 | result = APR_BADARG; |
353 | 0 | else if (codePoint <= 0xFF) |
354 | 0 | out.put(static_cast<char>(codePoint)); |
355 | 0 | else if (Transcoder::LOSSCHAR == codePoint) |
356 | 0 | out.put('?'); |
357 | 0 | else |
358 | 0 | result = APR_BADARG; |
359 | 0 | return result; |
360 | 0 | } |
361 | | |
362 | | private: |
363 | | ISOLatinCharsetEncoder(const ISOLatinCharsetEncoder&); |
364 | | ISOLatinCharsetEncoder& operator=(const ISOLatinCharsetEncoder&); |
365 | | }; |
366 | | |
367 | | /** |
368 | | * Encodes a LogString to a byte array when the encodings are identical. |
369 | | */ |
370 | | class TrivialCharsetEncoder : public CharsetEncoder |
371 | | { |
372 | | public: |
373 | | TrivialCharsetEncoder() |
374 | 0 | { |
375 | 0 | } |
376 | | |
377 | | |
378 | | virtual log4cxx_status_t encode(const LogString& in, |
379 | | LogString::const_iterator& iter, |
380 | | ByteBuffer& out) override |
381 | 0 | { |
382 | 0 | if (iter != in.end()) |
383 | 0 | { |
384 | 0 | size_t requested = in.length() - (iter - in.begin()); |
385 | 0 |
|
386 | 0 | if (requested > out.remaining() / sizeof(logchar)) |
387 | 0 | { |
388 | 0 | requested = out.remaining() / sizeof(logchar); |
389 | 0 | } |
390 | 0 |
|
391 | 0 | memcpy(out.current(), |
392 | 0 | (const char*) in.data() + (iter - in.begin()), |
393 | 0 | requested * sizeof(logchar)); |
394 | 0 | iter += requested; |
395 | 0 | out.increment_position(requested * sizeof(logchar)); |
396 | 0 | } |
397 | 0 |
|
398 | 0 | return APR_SUCCESS; |
399 | 0 | } |
400 | | |
401 | | /** |
402 | | * Add onto \c out an encoded equivalent of \c codePoint. |
403 | | */ |
404 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
405 | 0 | { |
406 | 0 | apr_status_t result = APR_SUCCESS; |
407 | 0 | if (out.remaining() < 1) |
408 | 0 | result = APR_BADARG; |
409 | 0 | else if (codePoint <= 0xFF) |
410 | 0 | out.put(static_cast<char>(codePoint)); |
411 | 0 | else if (Transcoder::LOSSCHAR == codePoint) |
412 | 0 | out.put('?'); |
413 | 0 | else |
414 | 0 | result = APR_BADARG; |
415 | 0 | return result; |
416 | 0 | } |
417 | | |
418 | | private: |
419 | | TrivialCharsetEncoder(const TrivialCharsetEncoder&); |
420 | | TrivialCharsetEncoder& operator=(const TrivialCharsetEncoder&); |
421 | | }; |
422 | | |
423 | | /** |
424 | | * Converts a LogString to UTF-8. |
425 | | */ |
426 | | class UTF8CharsetEncoder : public CharsetEncoder |
427 | | { |
428 | | public: |
429 | | UTF8CharsetEncoder() |
430 | 0 | { |
431 | 0 | } |
432 | | |
433 | | virtual log4cxx_status_t encode(const LogString& in, |
434 | | LogString::const_iterator& iter, |
435 | | ByteBuffer& out) override |
436 | 0 | { |
437 | 0 | while (iter != in.end() && out.remaining() >= 8) |
438 | 0 | { |
439 | 0 | auto sv = Transcoder::getCodePoint(in, iter); |
440 | 0 | Transcoder::encodeUTF8(sv, out); |
441 | 0 | } |
442 | |
|
443 | 0 | return APR_SUCCESS; |
444 | 0 | } |
445 | | |
446 | | /** |
447 | | * Add onto \c out an encoded equivalent of \c codePoint. |
448 | | */ |
449 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
450 | 0 | { |
451 | 0 | apr_status_t result = APR_SUCCESS; |
452 | 0 | if (codePoint <= 0x10FFFF) |
453 | 0 | Transcoder::encodeUTF8(codePoint, out); |
454 | 0 | else |
455 | 0 | result = APR_BADARG; |
456 | 0 | return result; |
457 | 0 | } |
458 | | |
459 | | private: |
460 | | UTF8CharsetEncoder(const UTF8CharsetEncoder&); |
461 | | UTF8CharsetEncoder& operator=(const UTF8CharsetEncoder&); |
462 | | }; |
463 | | |
464 | | /** |
465 | | * Encodes a LogString to UTF16-BE. |
466 | | */ |
467 | | class UTF16BECharsetEncoder : public CharsetEncoder |
468 | | { |
469 | | public: |
470 | | UTF16BECharsetEncoder() |
471 | 0 | { |
472 | 0 | } |
473 | | |
474 | | virtual log4cxx_status_t encode(const LogString& in, |
475 | | LogString::const_iterator& iter, |
476 | | ByteBuffer& out) override |
477 | 0 | { |
478 | 0 | while (iter != in.end() && out.remaining() >= 4) |
479 | 0 | { |
480 | 0 | auto sv = Transcoder::getCodePoint(in, iter); |
481 | 0 | Transcoder::encodeUTF16BE(sv, out); |
482 | 0 | } |
483 | |
|
484 | 0 | return APR_SUCCESS; |
485 | 0 | } |
486 | | |
487 | | /** |
488 | | * Add onto \c out an encoded equivalent of \c codePoint. |
489 | | */ |
490 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
491 | 0 | { |
492 | 0 | apr_status_t result = APR_SUCCESS; |
493 | 0 | if (codePoint <= 0x10FFFF) |
494 | 0 | Transcoder::encodeUTF16BE(codePoint, out); |
495 | 0 | else |
496 | 0 | result = APR_BADARG; |
497 | 0 | return result; |
498 | 0 | } |
499 | | |
500 | | private: |
501 | | UTF16BECharsetEncoder(const UTF16BECharsetEncoder&); |
502 | | UTF16BECharsetEncoder& operator=(const UTF16BECharsetEncoder&); |
503 | | }; |
504 | | |
505 | | /** |
506 | | * Encodes a LogString to UTF16-LE. |
507 | | */ |
508 | | class UTF16LECharsetEncoder : public CharsetEncoder |
509 | | { |
510 | | public: |
511 | | UTF16LECharsetEncoder() |
512 | 0 | { |
513 | 0 | } |
514 | | |
515 | | |
516 | | virtual log4cxx_status_t encode(const LogString& in, |
517 | | LogString::const_iterator& iter, |
518 | | ByteBuffer& out) override |
519 | 0 | { |
520 | 0 | while (iter != in.end() && out.remaining() >= 4) |
521 | 0 | { |
522 | 0 | auto sv = Transcoder::getCodePoint(in, iter); |
523 | 0 | Transcoder::encodeUTF16LE(sv, out); |
524 | 0 | } |
525 | |
|
526 | 0 | return APR_SUCCESS; |
527 | 0 | } |
528 | | |
529 | | /** |
530 | | * Add onto \c out an encoded equivalent of \c codePoint. |
531 | | */ |
532 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
533 | 0 | { |
534 | 0 | apr_status_t result = APR_SUCCESS; |
535 | 0 | if (codePoint <= 0x10FFFF) |
536 | 0 | Transcoder::encodeUTF16LE(codePoint, out); |
537 | 0 | else |
538 | 0 | result = APR_BADARG; |
539 | 0 | return result; |
540 | 0 | } |
541 | | private: |
542 | | UTF16LECharsetEncoder(const UTF16LECharsetEncoder&); |
543 | | UTF16LECharsetEncoder& operator=(const UTF16LECharsetEncoder&); |
544 | | }; |
545 | | |
546 | | /** |
547 | | * Charset encoder that uses current locale settings. |
548 | | */ |
549 | | class LocaleCharsetEncoder : public CharsetEncoder |
550 | | { |
551 | | public: |
552 | 0 | LocaleCharsetEncoder() : state() |
553 | 0 | { |
554 | 0 | } |
555 | | log4cxx_status_t encode |
556 | | ( const LogString& in |
557 | | , LogString::const_iterator& nextCodePoint |
558 | | , ByteBuffer& out |
559 | | ) override |
560 | 0 | { |
561 | 0 | log4cxx_status_t result = APR_SUCCESS; |
562 | 0 | #if !LOG4CXX_CHARSET_EBCDIC |
563 | 0 | char* current = out.current(); |
564 | 0 | size_t availableByteCount = out.remaining(); |
565 | 0 | size_t byteCount = 0; |
566 | 0 | if (std::mbsinit(&this->state)) // ByteBuffer not partially encoded? |
567 | 0 | { |
568 | | // Copy single byte characters |
569 | 0 | for (; |
570 | 0 | nextCodePoint != in.end() && byteCount < availableByteCount && static_cast<unsigned int>(*nextCodePoint) < 0x80; |
571 | 0 | ++nextCodePoint, ++byteCount, ++current) |
572 | 0 | { |
573 | 0 | *current = static_cast<char>(*nextCodePoint); |
574 | 0 | } |
575 | 0 | } |
576 | 0 | #endif |
577 | | // Encode characters that may require multiple bytes |
578 | 0 | while (nextCodePoint != in.end() && byteCount < availableByteCount && MB_CUR_MAX <= (availableByteCount - byteCount)) |
579 | 0 | { |
580 | 0 | auto ch = Transcoder::getCodePoint(in, nextCodePoint); |
581 | 0 | auto n = std::wcrtomb(current, ch, &this->state); |
582 | 0 | if (static_cast<std::size_t>(-1) == n) // not a valid wide character? |
583 | 0 | { |
584 | 0 | result = APR_BADARG; |
585 | 0 | break; |
586 | 0 | } |
587 | 0 | byteCount += n; |
588 | 0 | current += n; |
589 | 0 | } |
590 | 0 | out.increment_position(byteCount); |
591 | 0 | return result; |
592 | 0 | } |
593 | | |
594 | | /** |
595 | | * Add onto \c out an encoded equivalent of \c codePoint. |
596 | | */ |
597 | | log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER |
598 | 0 | { |
599 | 0 | apr_status_t result = APR_SUCCESS; |
600 | 0 | if (MB_CUR_MAX <= out.remaining()) |
601 | 0 | { |
602 | 0 | auto n = std::wcrtomb(out.current(), codePoint, &this->state); |
603 | 0 | if (static_cast<std::size_t>(-1) == n) // not a valid wide character? |
604 | 0 | result = APR_BADARG; |
605 | 0 | else |
606 | 0 | out.increment_position(n); |
607 | 0 | } |
608 | 0 | else |
609 | 0 | result = APR_BADARG; |
610 | 0 | return result; |
611 | 0 | } |
612 | | |
613 | | private: |
614 | | std::mbstate_t state; |
615 | | }; |
616 | | |
617 | | |
618 | | } // namespace helpers |
619 | | |
620 | | } //namespace log4cxx |
621 | | |
622 | | |
623 | | |
624 | | CharsetEncoder::CharsetEncoder() |
625 | 0 | { |
626 | 0 | } |
627 | | |
628 | | CharsetEncoder::~CharsetEncoder() |
629 | 0 | { |
630 | 0 | } |
631 | | |
632 | | CharsetEncoderPtr CharsetEncoder::getDefaultEncoder() |
633 | 0 | { |
634 | 0 | static WideLife<CharsetEncoderPtr> encoder(createDefaultEncoder()); |
635 | | |
636 | | // |
637 | | // if invoked after static variable destruction |
638 | | // (if logging is called in the destructor of a static object) |
639 | | // then create a new decoder. |
640 | | // |
641 | 0 | if (encoder.value() == 0) |
642 | 0 | { |
643 | 0 | return CharsetEncoderPtr( createDefaultEncoder() ); |
644 | 0 | } |
645 | | |
646 | 0 | return encoder; |
647 | 0 | } |
648 | | |
649 | | CharsetEncoder* CharsetEncoder::createDefaultEncoder() |
650 | 0 | { |
651 | 0 | #if LOG4CXX_CHARSET_UTF8 |
652 | | #if LOG4CXX_LOGCHAR_IS_UTF8 |
653 | | return new TrivialCharsetEncoder(); |
654 | | #else |
655 | 0 | return new UTF8CharsetEncoder(); |
656 | 0 | #endif |
657 | | #elif LOG4CXX_CHARSET_ISO88591 |
658 | | return new ISOLatinCharsetEncoder(); |
659 | | #elif LOG4CXX_CHARSET_USASCII |
660 | | return new USASCIICharsetEncoder(); |
661 | | #elif LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS |
662 | | return new WcstombsCharsetEncoder(); |
663 | | #else |
664 | | return new LocaleCharsetEncoder(); |
665 | | #endif |
666 | 0 | } |
667 | | |
668 | | |
669 | | CharsetEncoderPtr CharsetEncoder::getUTF8Encoder() |
670 | 0 | { |
671 | 0 | return std::make_shared<UTF8CharsetEncoder>(); |
672 | 0 | } |
673 | | |
674 | | |
675 | | |
676 | | CharsetEncoderPtr CharsetEncoder::getEncoder(const LogString& charset) |
677 | 0 | { |
678 | 0 | if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-8"), LOG4CXX_STR("utf-8")) |
679 | 0 | || StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP65001"), LOG4CXX_STR("cp65001"))) |
680 | 0 | { |
681 | 0 | return std::make_shared<UTF8CharsetEncoder>(); |
682 | 0 | } |
683 | 0 | else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("C"), LOG4CXX_STR("c")) || |
684 | 0 | charset == LOG4CXX_STR("646") || |
685 | 0 | StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("US-ASCII"), LOG4CXX_STR("us-ascii")) || |
686 | 0 | StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ISO646-US"), LOG4CXX_STR("iso646-US")) || |
687 | 0 | StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ANSI_X3.4-1968"), LOG4CXX_STR("ansi_x3.4-1968")) || |
688 | 0 | StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP20127"), LOG4CXX_STR("cp20127"))) |
689 | 0 | { |
690 | 0 | return std::make_shared<USASCIICharsetEncoder>(); |
691 | 0 | } |
692 | 0 | else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ISO-8859-1"), LOG4CXX_STR("iso-8859-1")) || |
693 | 0 | StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ISO-LATIN-1"), LOG4CXX_STR("iso-latin-1")) || |
694 | 0 | StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP1252"), LOG4CXX_STR("cp1252"))) |
695 | 0 | { |
696 | 0 | return std::make_shared<ISOLatinCharsetEncoder>(); |
697 | 0 | } |
698 | 0 | else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-16BE"), LOG4CXX_STR("utf-16be")) |
699 | 0 | || StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-16"), LOG4CXX_STR("utf-16")) |
700 | 0 | || StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP1200"), LOG4CXX_STR("cp1200"))) |
701 | 0 | { |
702 | 0 | return std::make_shared<UTF16BECharsetEncoder>(); |
703 | 0 | } |
704 | 0 | else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-16LE"), LOG4CXX_STR("utf-16le"))) |
705 | 0 | { |
706 | 0 | return std::make_shared<UTF16LECharsetEncoder>(); |
707 | 0 | } |
708 | 0 | else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("LOCALE"), LOG4CXX_STR("locale"))) |
709 | 0 | { |
710 | 0 | return std::make_shared<LocaleCharsetEncoder>(); |
711 | 0 | } |
712 | | |
713 | 0 | #if APR_HAS_XLATE |
714 | 0 | return std::make_shared<APRCharsetEncoder>(charset); |
715 | | #else |
716 | | throw IllegalArgumentException(charset); |
717 | | #endif |
718 | 0 | } |
719 | | |
720 | | |
721 | | void CharsetEncoder::reset() |
722 | 0 | { |
723 | 0 | } |
724 | | |
725 | | #if LOG4CXX_ABI_VERSION <= 15 |
726 | | void CharsetEncoder::flush(ByteBuffer& /* out */ ) |
727 | 0 | { |
728 | 0 | } |
729 | | #endif |
730 | | |
731 | | void CharsetEncoder::encode(CharsetEncoderPtr& enc, |
732 | | const LogString& src, |
733 | | LogString::const_iterator& iter, |
734 | | ByteBuffer& dst) |
735 | 0 | { |
736 | 0 | log4cxx_status_t stat = enc->encode(src, iter, dst); |
737 | |
|
738 | 0 | if (stat != APR_SUCCESS && iter != src.end()) |
739 | 0 | { |
740 | 0 | #if LOG4CXX_LOGCHAR_IS_WCHAR || LOG4CXX_LOGCHAR_IS_UNICHAR |
741 | 0 | iter++; |
742 | | #elif LOG4CXX_LOGCHAR_IS_UTF8 |
743 | | |
744 | | // advance past this character and all continuation characters |
745 | | do |
746 | | { |
747 | | ++iter; |
748 | | } |
749 | | while (iter != src.end() && |
750 | | (*iter & 0xC0) == 0x80); |
751 | | |
752 | | #else |
753 | | #error logchar is unrecognized |
754 | | #endif |
755 | | #if 15 < LOG4CXX_ABI_VERSION |
756 | | enc->encode(Transcoder::LOSSCHAR, dst); |
757 | | #else // LOG4CXX_ABI_VERSION <= 15 |
758 | 0 | if (auto p = dynamic_cast<TrivialCharsetEncoder*>(enc.get())) |
759 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
760 | 0 | else if (auto p = dynamic_cast<UTF8CharsetEncoder*>(enc.get())) |
761 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
762 | 0 | else if (auto p = dynamic_cast<LocaleCharsetEncoder*>(enc.get())) |
763 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
764 | 0 | else if (auto p = dynamic_cast<USASCIICharsetEncoder*>(enc.get())) |
765 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
766 | 0 | else if (auto p = dynamic_cast<ISOLatinCharsetEncoder*>(enc.get())) |
767 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
768 | 0 | else if (auto p = dynamic_cast<UTF16BECharsetEncoder*>(enc.get())) |
769 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
770 | 0 | else if (auto p = dynamic_cast<UTF16LECharsetEncoder*>(enc.get())) |
771 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
772 | 0 | #if LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS |
773 | 0 | else if (auto p = dynamic_cast<WcstombsCharsetEncoder*>(enc.get())) |
774 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
775 | 0 | #endif // LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS |
776 | 0 | #if APR_HAS_XLATE |
777 | 0 | else if (auto p = dynamic_cast<APRCharsetEncoder*>(enc.get())) |
778 | 0 | p->encode(Transcoder::LOSSCHAR, dst); |
779 | 0 | #endif // APR_HAS_XLATE |
780 | 0 | #endif // LOG4CXX_ABI_VERSION <= 15 |
781 | 0 | } |
782 | 0 | } |
783 | | |
784 | | bool CharsetEncoder::isTriviallyCopyable(const LogString& src, const CharsetEncoderPtr& enc) |
785 | 0 | { |
786 | 0 | bool result; |
787 | 0 | #if !LOG4CXX_CHARSET_EBCDIC |
788 | 0 | if (dynamic_cast<LocaleCharsetEncoder*>(enc.get())) |
789 | 0 | { |
790 | 0 | result = src.end() == std::find_if(src.begin(), src.end() |
791 | 0 | , [](const logchar& ch) -> bool { return 0x80 <= (unsigned int)ch; }); |
792 | 0 | } |
793 | 0 | else |
794 | 0 | #endif |
795 | 0 | result = !!dynamic_cast<TrivialCharsetEncoder*>(enc.get()); |
796 | 0 | return result; |
797 | 0 | } |