?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
merge the formfield patch from ooo-build
[ooovba.git] / tools / source / inet / inetmime.cxx
blob10b7d390e0781089cc292d723c2c173ed897a1df
1 /*************************************************************************
2 *
3 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
4 *
5 * Copyright 2008 by Sun Microsystems, Inc.
6 *
7 * OpenOffice.org - a multi-platform office productivity suite
8 *
9 * $RCSfile: inetmime.cxx,v $
10 * $Revision: 1.14 $
11 *
12 * This file is part of OpenOffice.org.
13 *
14 * OpenOffice.org is free software: you can redistribute it and/or modify
15 * it under the terms of the GNU Lesser General Public License version 3
16 * only, as published by the Free Software Foundation.
17 *
18 * OpenOffice.org is distributed in the hope that it will be useful,
19 * but WITHOUT ANY WARRANTY; without even the implied warranty of
20 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
21 * GNU Lesser General Public License version 3 for more details
22 * (a copy is included in the LICENSE file that accompanied this code).
23 *
24 * You should have received a copy of the GNU Lesser General Public License
25 * version 3 along with OpenOffice.org. If not, see
26 * <http://www.openoffice.org/license.html>
27 * for a copy of the LGPLv3 License.
28 *
29 ************************************************************************/
30
31 // MARKER(update_precomp.py): autogen include statement, do not remove
32 #include "precompiled_tools.hxx"
33
34 #include <cstddef>
35 #include <limits>
36
37 #include "rtl/tencinfo.h"
38 #include <tools/datetime.hxx>
39 #include <tools/inetmime.hxx>
40
41 namespace unnamed_tools_inetmime {} using namespace unnamed_tools_inetmime;
42 // unnamed namespaces don't work well yet
43
44 //============================================================================
45 namespace unnamed_tools_inetmime {
46
47 class Charset
48 {
49 rtl_TextEncoding m_eEncoding;
50 const sal_uInt32 * m_pRanges;
51
52 public:
53 inline Charset(rtl_TextEncoding eTheEncoding,
54 const sal_uInt32 * pTheRanges);
55
56 rtl_TextEncoding getEncoding() const { return m_eEncoding; }
57
58 bool contains(sal_uInt32 nChar) const;
59 };
60
61 inline Charset::Charset(rtl_TextEncoding eTheEncoding,
62 const sal_uInt32 * pTheRanges):
63 m_eEncoding(eTheEncoding),
64 m_pRanges(pTheRanges)
65 {
66 DBG_ASSERT(m_pRanges, "Charset::Charset(): Bad ranges");
67 }
68
69 //============================================================================
70 void appendISO88591(UniString & rText, sal_Char const * pBegin,
71 sal_Char const * pEnd);
72
73 }
74
75 //============================================================================
76 class INetMIMECharsetList_Impl
77 {
78 struct Node
79 {
80 Charset m_aCharset;
81 bool m_bDisabled;
82 Node * m_pNext;
83
84 inline Node(const Charset & rTheCharset, bool bTheDisabled,
85 Node * pTheNext);
86 };
87
88 Node * m_pFirst;
89
90 public:
91 INetMIMECharsetList_Impl(): m_pFirst(0) {}
92
93 ~INetMIMECharsetList_Impl();
94
95 void prepend(const Charset & rCharset)
96 { m_pFirst = new Node(rCharset, false, m_pFirst); }
97
98 void includes(sal_uInt32 nChar);
99
100 rtl_TextEncoding getPreferredEncoding(rtl_TextEncoding eDefault
101 = RTL_TEXTENCODING_DONTKNOW)
102 const;
103
104 void reset();
105 };
106
107 inline INetMIMECharsetList_Impl::Node::Node(const Charset & rTheCharset,
108 bool bTheDisabled,
109 Node * pTheNext):
110 m_aCharset(rTheCharset),
111 m_bDisabled(bTheDisabled),
112 m_pNext(pTheNext)
113 {}
114
115 //============================================================================
116 namespace unnamed_tools_inetmime {
117
118 struct Parameter
119 {
120 Parameter * m_pNext;
121 ByteString m_aAttribute;
122 ByteString m_aCharset;
123 ByteString m_aLanguage;
124 ByteString m_aValue;
125 sal_uInt32 m_nSection;
126 bool m_bExtended;
127
128 inline Parameter(Parameter * pTheNext, ByteString const & rTheAttribute,
129 ByteString const & rTheCharset,
130 ByteString const & rTheLanguage,
131 ByteString const & rTheValue, sal_uInt32 nTheSection,
132 bool bTheExtended);
133 };
134
135 inline Parameter::Parameter(Parameter * pTheNext,
136 ByteString const & rTheAttribute,
137 ByteString const & rTheCharset,
138 ByteString const & rTheLanguage,
139 ByteString const & rTheValue,
140 sal_uInt32 nTheSection, bool bTheExtended):
141 m_pNext(pTheNext),
142 m_aAttribute(rTheAttribute),
143 m_aCharset(rTheCharset),
144 m_aLanguage(rTheLanguage),
145 m_aValue(rTheValue),
146 m_nSection(nTheSection),
147 m_bExtended(bTheExtended)
148 {}
149
150 //============================================================================
151 struct ParameterList
152 {
153 Parameter * m_pList;
154
155 ParameterList(): m_pList(0) {}
156
157 inline ~ParameterList();
158
159 Parameter ** find(ByteString const & rAttribute, sal_uInt32 nSection,
160 bool & rPresent);
161 };
162
163 inline ParameterList::~ParameterList()
164 {
165 while (m_pList)
166 {
167 Parameter * pNext = m_pList->m_pNext;
168 delete m_pList;
169 m_pList = pNext;
170 }
171 }
172
173 //============================================================================
174 bool parseParameters(ParameterList const & rInput,
175 INetContentTypeParameterList * pOutput);
176
177 }
178
179 //============================================================================
180 //
181 // Charset
182 //
183 //============================================================================
184
185 bool Charset::contains(sal_uInt32 nChar) const
186 {
187 for (const sal_uInt32 * p = m_pRanges;;)
188 {
189 if (nChar < *p++)
190 return false;
191 if (nChar <= *p++)
192 return true;
193 }
194 }
195
196 //============================================================================
197 //
198 // appendISO88591
199 //
200 //============================================================================
201
202 namespace unnamed_tools_inetmime {
203
204 void appendISO88591(UniString & rText, sal_Char const * pBegin,
205 sal_Char const * pEnd)
206 {
207 xub_StrLen nLength = static_cast< xub_StrLen >(pEnd - pBegin);
208 sal_Unicode * pBuffer = new sal_Unicode[nLength];
209 for (sal_Unicode * p = pBuffer; pBegin != pEnd;)
210 *p++ = sal_uChar(*pBegin++);
211 rText.Append(pBuffer, nLength);
212 delete[] pBuffer;
213 }
214
215 }
216
217 //============================================================================
218 //
219 // INetMIMECharsetList_Impl
220 //
221 //============================================================================
222
223 INetMIMECharsetList_Impl::~INetMIMECharsetList_Impl()
224 {
225 while (m_pFirst)
226 {
227 Node * pRemove = m_pFirst;
228 m_pFirst = m_pFirst->m_pNext;
229 delete pRemove;
230 }
231 }
232
233 //============================================================================
234 void INetMIMECharsetList_Impl::includes(sal_uInt32 nChar)
235 {
236 for (Node * p = m_pFirst; p; p = p->m_pNext)
237 if (!(p->m_bDisabled || p->m_aCharset.contains(nChar)))
238 p->m_bDisabled = true;
239 }
240
241 //============================================================================
242 rtl_TextEncoding
243 INetMIMECharsetList_Impl::getPreferredEncoding(rtl_TextEncoding eDefault)
244 const
245 {
246 for (Node * p = m_pFirst; p; p = p->m_pNext)
247 if (!p->m_bDisabled)
248 return p->m_aCharset.getEncoding();
249 return eDefault;
250 }
251
252 //============================================================================
253 void INetMIMECharsetList_Impl::reset()
254 {
255 for (Node * p = m_pFirst; p; p = p->m_pNext)
256 p->m_bDisabled = false;
257 }
258
259 //============================================================================
260 //
261 // ParameterList
262 //
263 //============================================================================
264
265 Parameter ** ParameterList::find(ByteString const & rAttribute,
266 sal_uInt32 nSection, bool & rPresent)
267 {
268 Parameter ** p = &m_pList;
269 for (; *p; p = &(*p)->m_pNext)
270 {
271 StringCompare eCompare = rAttribute.CompareTo((*p)->m_aAttribute);
272 if (eCompare == COMPARE_GREATER)
273 break;
274 else if (eCompare == COMPARE_EQUAL)
275 {
276 if (nSection > (*p)->m_nSection)
277 break;
278 else if (nSection == (*p)->m_nSection)
279 {
280 rPresent = true;
281 return p;
282 }
283 }
284 }
285 rPresent = false;
286 return p;
287 }
288
289 //============================================================================
290 //
291 // parseParameters
292 //
293 //============================================================================
294
295 namespace unnamed_tools_inetmime {
296
297 bool parseParameters(ParameterList const & rInput,
298 INetContentTypeParameterList * pOutput)
299 {
300 if (pOutput)
301 pOutput->Clear();
302
303 Parameter * pPrev = 0;
304 for (Parameter * p = rInput.m_pList; p; p = p->m_pNext)
305 {
306 if (p->m_nSection > 0
307 && (!pPrev
308 || pPrev->m_nSection != p->m_nSection - 1
309 || pPrev->m_aAttribute != p->m_aAttribute))
310 return false;
311 pPrev = p;
312 }
313
314 if (pOutput)
315 for (Parameter * p = rInput.m_pList; p;)
316 {
317 bool bCharset = p->m_aCharset.Len() != 0;
318 rtl_TextEncoding eEncoding = RTL_TEXTENCODING_DONTKNOW;
319 if (bCharset)
320 eEncoding
321 = INetMIME::getCharsetEncoding(p->m_aCharset.GetBuffer(),
322 p->m_aCharset.GetBuffer()
323 + rInput.m_pList->
324 m_aCharset.
325 Len());
326 UniString aValue;
327 bool bBadEncoding = false;
328 Parameter * pNext = p;
329 do
330 {
331 sal_Size nSize;
332 sal_Unicode * pUnicode
333 = INetMIME::convertToUnicode(pNext->m_aValue.GetBuffer(),
334 pNext->m_aValue.GetBuffer()
335 + pNext->m_aValue.Len(),
336 bCharset && p->m_bExtended ?
337 eEncoding :
338 RTL_TEXTENCODING_UTF8,
339 nSize);
340 if (!pUnicode && !(bCharset && p->m_bExtended))
341 pUnicode = INetMIME::convertToUnicode(
342 pNext->m_aValue.GetBuffer(),
343 pNext->m_aValue.GetBuffer()
344 + pNext->m_aValue.Len(),
345 RTL_TEXTENCODING_ISO_8859_1, nSize);
346 if (!pUnicode)
347 {
348 bBadEncoding = true;
349 break;
350 }
351 aValue += UniString(pUnicode, static_cast< xub_StrLen >(nSize));
352 delete[] pUnicode;
353 pNext = pNext->m_pNext;
354 }
355 while (pNext && pNext->m_nSection > 0);
356 if (bBadEncoding)
357 {
358 aValue.Erase();
359 for (pNext = p;;)
360 {
361 if (pNext->m_bExtended)
362 for (xub_StrLen i = 0; i < pNext->m_aValue.Len(); ++i)
363 aValue += sal_Unicode(
364 sal_Unicode(
365 sal_uChar(pNext->m_aValue.GetChar(i)))
366 | 0xF800);
367 else
368 for (xub_StrLen i = 0; i < pNext->m_aValue.Len(); ++i)
369 aValue
370 += sal_Unicode(sal_uChar
371 (pNext->
372 m_aValue.GetChar(i)));
373 pNext = pNext->m_pNext;
374 if (!pNext || pNext->m_nSection == 0)
375 break;
376 };
377 }
378 pOutput->Insert(new INetContentTypeParameter(p->m_aAttribute,
379 p->m_aCharset,
380 p->m_aLanguage,
381 aValue,
382 !bBadEncoding),
383 LIST_APPEND);
384 p = pNext;
385 }
386 return true;
387 }
388
389 }
390
391 //============================================================================
392 //
393 // INetMIME
394 //
395 //============================================================================
396
397 // static
398 bool INetMIME::isAtomChar(sal_uInt32 nChar)
399 {
400 static const bool aMap[128]
401 = { false, false, false, false, false, false, false, false,
402 false, false, false, false, false, false, false, false,
403 false, false, false, false, false, false, false, false,
404 false, false, false, false, false, false, false, false,
405 false, true, false, true, true, true, true, true, // !"#$%&'
406 false, false, true, true, false, true, false, true, //()*+,-./
407 true, true, true, true, true, true, true, true, //01234567
408 true, true, false, false, false, true, false, true, //89:;<=>?
409 false, true, true, true, true, true, true, true, //@ABCDEFG
410 true, true, true, true, true, true, true, true, //HIJKLMNO
411 true, true, true, true, true, true, true, true, //PQRSTUVW
412 true, true, true, false, false, false, true, true, //XYZ[\]^_
413 true, true, true, true, true, true, true, true, //`abcdefg
414 true, true, true, true, true, true, true, true, //hijklmno
415 true, true, true, true, true, true, true, true, //pqrstuvw
416 true, true, true, true, true, true, true, false //xyz{|}~
417 };
418 return isUSASCII(nChar) && aMap[nChar];
419 }
420
421 //============================================================================
422 // static
423 bool INetMIME::isTokenChar(sal_uInt32 nChar)
424 {
425 static const sal_Char aMap[128]
426 = { false, false, false, false, false, false, false, false,
427 false, false, false, false, false, false, false, false,
428 false, false, false, false, false, false, false, false,
429 false, false, false, false, false, false, false, false,
430 false, true, false, true, true, true, true, true, // !"#$%&'
431 false, false, true, true, false, true, true, false, //()*+,-./
432 true, true, true, true, true, true, true, true, //01234567
433 true, true, false, false, false, false, false, false, //89:;<=>?
434 false, true, true, true, true, true, true, true, //@ABCDEFG
435 true, true, true, true, true, true, true, true, //HIJKLMNO
436 true, true, true, true, true, true, true, true, //PQRSTUVW
437 true, true, true, false, false, false, true, true, //XYZ[\]^_
438 true, true, true, true, true, true, true, true, //`abcdefg
439 true, true, true, true, true, true, true, true, //hijklmno
440 true, true, true, true, true, true, true, true, //pqrstuvw
441 true, true, true, true, true, true, true, false //xyz{|}~
442 };
443 return isUSASCII(nChar) && aMap[nChar];
444 }
445
446 //============================================================================
447 // static
448 bool INetMIME::isEncodedWordTokenChar(sal_uInt32 nChar)
449 {
450 static const sal_Char aMap[128]
451 = { false, false, false, false, false, false, false, false,
452 false, false, false, false, false, false, false, false,
453 false, false, false, false, false, false, false, false,
454 false, false, false, false, false, false, false, false,
455 false, true, false, true, true, true, true, true, // !"#$%&'
456 false, false, true, true, false, true, false, false, //()*+,-./
457 true, true, true, true, true, true, true, true, //01234567
458 true, true, false, false, false, false, false, false, //89:;<=>?
459 false, true, true, true, true, true, true, true, //@ABCDEFG
460 true, true, true, true, true, true, true, true, //HIJKLMNO
461 true, true, true, true, true, true, true, true, //PQRSTUVW
462 true, true, true, false, false, false, true, true, //XYZ[\]^_
463 true, true, true, true, true, true, true, true, //`abcdefg
464 true, true, true, true, true, true, true, true, //hijklmno
465 true, true, true, true, true, true, true, true, //pqrstuvw
466 true, true, true, true, true, true, true, false //xyz{|}~
467 };
468 return isUSASCII(nChar) && aMap[nChar];
469 }
470
471 //============================================================================
472 // static
473 bool INetMIME::isIMAPAtomChar(sal_uInt32 nChar)
474 {
475 static const sal_Char aMap[128]
476 = { false, false, false, false, false, false, false, false,
477 false, false, false, false, false, false, false, false,
478 false, false, false, false, false, false, false, false,
479 false, false, false, false, false, false, false, false,
480 false, true, false, true, true, false, true, true, // !"#$%&'
481 false, false, false, true, true, true, true, true, //()*+,-./
482 true, true, true, true, true, true, true, true, //01234567
483 true, true, true, true, true, true, true, true, //89:;<=>?
484 true, true, true, true, true, true, true, true, //@ABCDEFG
485 true, true, true, true, true, true, true, true, //HIJKLMNO
486 true, true, true, true, true, true, true, true, //PQRSTUVW
487 true, true, true, true, false, true, true, true, //XYZ[\]^_
488 true, true, true, true, true, true, true, true, //`abcdefg
489 true, true, true, true, true, true, true, true, //hijklmno
490 true, true, true, true, true, true, true, true, //pqrstuvw
491 true, true, true, false, true, true, true, false //xyz{|}~
492 };
493 return isUSASCII(nChar) && aMap[nChar];
494 }
495
496 //============================================================================
497 // static
498 sal_uInt32 INetMIME::getDigit(int nWeight)
499 {
500 DBG_ASSERT(nWeight >= 0 && nWeight < 10,
501 "INetMIME::getDigit(): Bad weight");
502
503 static const sal_Char aDigits[16]
504 = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9' };
505 return aDigits[nWeight];
506 }
507
508 //============================================================================
509 // static
510 sal_uInt32 INetMIME::getHexDigit(int nWeight)
511 {
512 DBG_ASSERT(nWeight >= 0 && nWeight < 16,
513 "INetMIME::getHexDigit(): Bad weight");
514
515 static const sal_Char aDigits[16]
516 = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C',
517 'D', 'E', 'F' };
518 return aDigits[nWeight];
519 }
520
521 //============================================================================
522 // static
523 sal_uInt32 INetMIME::getBase64Digit(int nWeight)
524 {
525 DBG_ASSERT(nWeight >= 0 && nWeight < 64,
526 "INetMIME::getBase64Digit(): Bad weight");
527
528 static const sal_Char aDigits[64]
529 = { 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M',
530 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z',
531 'a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l', 'm',
532 'n', 'o', 'p', 'q', 'r', 's', 't', 'u', 'v', 'w', 'x', 'y', 'z',
533 '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '+', '/' };
534 return aDigits[nWeight];
535 }
536
537 //============================================================================
538 // static
539 bool INetMIME::equalIgnoreCase(const sal_Char * pBegin1,
540 const sal_Char * pEnd1,
541 const sal_Char * pBegin2,
542 const sal_Char * pEnd2)
543 {
544 DBG_ASSERT(pBegin1 && pBegin1 <= pEnd1 && pBegin2 && pBegin2 <= pEnd2,
545 "INetMIME::equalIgnoreCase(): Bad sequences");
546
547 if (pEnd1 - pBegin1 != pEnd2 - pBegin2)
548 return false;
549 while (pBegin1 != pEnd1)
550 if (toUpperCase(*pBegin1++) != toUpperCase(*pBegin2++))
551 return false;
552 return true;
553 }
554
555 //============================================================================
556 // static
557 bool INetMIME::equalIgnoreCase(const sal_Char * pBegin1,
558 const sal_Char * pEnd1,
559 const sal_Char * pString2)
560 {
561 DBG_ASSERT(pBegin1 && pBegin1 <= pEnd1 && pString2,
562 "INetMIME::equalIgnoreCase(): Bad sequences");
563
564 while (*pString2 != 0)
565 if (pBegin1 == pEnd1
566 || toUpperCase(*pBegin1++) != toUpperCase(*pString2++))
567 return false;
568 return pBegin1 == pEnd1;
569 }
570
571 //============================================================================
572 // static
573 bool INetMIME::equalIgnoreCase(const sal_Unicode * pBegin1,
574 const sal_Unicode * pEnd1,
575 const sal_Char * pString2)
576 {
577 DBG_ASSERT(pBegin1 && pBegin1 <= pEnd1 && pString2,
578 "INetMIME::equalIgnoreCase(): Bad sequences");
579
580 while (*pString2 != 0)
581 if (pBegin1 == pEnd1
582 || toUpperCase(*pBegin1++) != toUpperCase(*pString2++))
583 return false;
584 return pBegin1 == pEnd1;
585 }
586
587 //============================================================================
588 // static
589 const sal_Char * INetMIME::skipLinearWhiteSpace(const sal_Char * pBegin,
590 const sal_Char * pEnd)
591 {
592 DBG_ASSERT(pBegin && pBegin <= pEnd,
593 "INetMIME::skipLinearWhiteSpace(): Bad sequence");
594
595 while (pBegin != pEnd)
596 switch (*pBegin)
597 {
598 case '\t':
599 case ' ':
600 ++pBegin;
601 break;
602
603 case 0x0D: // CR
604 if (startsWithLineFolding(pBegin, pEnd))
605 pBegin += 3;
606 else
607 return pBegin;
608 break;
609
610 default:
611 return pBegin;
612 }
613 return pBegin;
614 }
615
616 //============================================================================
617 // static
618 const sal_Unicode * INetMIME::skipLinearWhiteSpace(const sal_Unicode * pBegin,
619 const sal_Unicode * pEnd)
620 {
621 DBG_ASSERT(pBegin && pBegin <= pEnd,
622 "INetMIME::skipLinearWhiteSpace(): Bad sequence");
623
624 while (pBegin != pEnd)
625 switch (*pBegin)
626 {
627 case '\t':
628 case ' ':
629 ++pBegin;
630 break;
631
632 case 0x0D: // CR
633 if (startsWithLineFolding(pBegin, pEnd))
634 pBegin += 3;
635 else
636 return pBegin;
637 break;
638
639 default:
640 return pBegin;
641 }
642 return pBegin;
643 }
644
645 //============================================================================
646 // static
647 const sal_Char * INetMIME::skipComment(const sal_Char * pBegin,
648 const sal_Char * pEnd)
649 {
650 DBG_ASSERT(pBegin && pBegin <= pEnd,
651 "INetMIME::skipComment(): Bad sequence");
652
653 if (pBegin != pEnd && *pBegin == '(')
654 {
655 sal_uInt32 nLevel = 0;
656 for (const sal_Char * p = pBegin; p != pEnd;)
657 switch (*p++)
658 {
659 case '(':
660 ++nLevel;
661 break;
662
663 case ')':
664 if (--nLevel == 0)
665 return p;
666 break;
667
668 case '\\':
669 if (p != pEnd)
670 ++p;
671 break;
672 }
673 }
674 return pBegin;
675 }
676
677 //============================================================================
678 // static
679 const sal_Unicode * INetMIME::skipComment(const sal_Unicode * pBegin,
680 const sal_Unicode * pEnd)
681 {
682 DBG_ASSERT(pBegin && pBegin <= pEnd,
683 "INetMIME::skipComment(): Bad sequence");
684
685 if (pBegin != pEnd && *pBegin == '(')
686 {
687 sal_uInt32 nLevel = 0;
688 for (const sal_Unicode * p = pBegin; p != pEnd;)
689 switch (*p++)
690 {
691 case '(':
692 ++nLevel;
693 break;
694
695 case ')':
696 if (--nLevel == 0)
697 return p;
698 break;
699
700 case '\\':
701 if (p != pEnd)
702 ++p;
703 break;
704 }
705 }
706 return pBegin;
707 }
708
709 //============================================================================
710 // static
711 const sal_Char * INetMIME::skipLinearWhiteSpaceComment(const sal_Char *
712 pBegin,
713 const sal_Char * pEnd)
714 {
715 DBG_ASSERT(pBegin && pBegin <= pEnd,
716 "INetMIME::skipLinearWhiteSpaceComment(): Bad sequence");
717
718 while (pBegin != pEnd)
719 switch (*pBegin)
720 {
721 case '\t':
722 case ' ':
723 ++pBegin;
724 break;
725
726 case 0x0D: // CR
727 if (startsWithLineFolding(pBegin, pEnd))
728 pBegin += 3;
729 else
730 return pBegin;
731 break;
732
733 case '(':
734 {
735 const sal_Char * p = skipComment(pBegin, pEnd);
736 if (p == pBegin)
737 return pBegin;
738 pBegin = p;
739 break;
740 }
741
742 default:
743 return pBegin;
744 }
745 return pBegin;
746 }
747
748 //============================================================================
749 // static
750 const sal_Unicode * INetMIME::skipLinearWhiteSpaceComment(const sal_Unicode *
751 pBegin,
752 const sal_Unicode *
753 pEnd)
754 {
755 DBG_ASSERT(pBegin && pBegin <= pEnd,
756 "INetMIME::skipLinearWhiteSpaceComment(): Bad sequence");
757
758 while (pBegin != pEnd)
759 switch (*pBegin)
760 {
761 case '\t':
762 case ' ':
763 ++pBegin;
764 break;
765
766 case 0x0D: // CR
767 if (startsWithLineFolding(pBegin, pEnd))
768 pBegin += 3;
769 else
770 return pBegin;
771 break;
772
773 case '(':
774 {
775 const sal_Unicode * p = skipComment(pBegin, pEnd);
776 if (p == pBegin)
777 return pBegin;
778 pBegin = p;
779 break;
780 }
781
782 default:
783 return pBegin;
784 }
785 return pBegin;
786 }
787
788 //============================================================================
789 // static
790 const sal_Char * INetMIME::skipQuotedString(const sal_Char * pBegin,
791 const sal_Char * pEnd)
792 {
793 DBG_ASSERT(pBegin && pBegin <= pEnd,
794 "INetMIME::skipQuotedString(): Bad sequence");
795
796 if (pBegin != pEnd && *pBegin == '"')
797 for (const sal_Char * p = pBegin + 1; p != pEnd;)
798 switch (*p++)
799 {
800 case 0x0D: // CR
801 if (pEnd - p < 2 || *p++ != 0x0A // LF
802 || !isWhiteSpace(*p++))
803 return pBegin;
804 break;
805
806 case '"':
807 return p;
808
809 case '\\':
810 if (p != pEnd)
811 ++p;
812 break;
813 }
814 return pBegin;
815 }
816
817 //============================================================================
818 // static
819 const sal_Unicode * INetMIME::skipQuotedString(const sal_Unicode * pBegin,
820 const sal_Unicode * pEnd)
821 {
822 DBG_ASSERT(pBegin && pBegin <= pEnd,
823 "INetMIME::skipQuotedString(): Bad sequence");
824
825 if (pBegin != pEnd && *pBegin == '"')
826 for (const sal_Unicode * p = pBegin + 1; p != pEnd;)
827 switch (*p++)
828 {
829 case 0x0D: // CR
830 if (pEnd - p < 2 || *p++ != 0x0A // LF
831 || !isWhiteSpace(*p++))
832 return pBegin;
833 break;
834
835 case '"':
836 return p;
837
838 case '\\':
839 if (p != pEnd)
840 ++p;
841 break;
842 }
843 return pBegin;
844 }
845
846 //============================================================================
847 // static
848 const sal_Char * INetMIME::scanAtom(const sal_Char * pBegin,
849 const sal_Char * pEnd)
850 {
851 while (pBegin != pEnd && isAtomChar(*pBegin))
852 ++pBegin;
853 return pBegin;
854 }
855
856 //============================================================================
857 // static
858 const sal_Unicode * INetMIME::scanAtom(const sal_Unicode * pBegin,
859 const sal_Unicode * pEnd)
860 {
861 while (pBegin != pEnd && isAtomChar(*pBegin))
862 ++pBegin;
863 return pBegin;
864 }
865
866 //============================================================================
867 // static
868 bool INetMIME::scanUnsigned(const sal_Char *& rBegin, const sal_Char * pEnd,
869 bool bLeadingZeroes, sal_uInt32 & rValue)
870 {
871 sal_uInt64 nTheValue = 0;
872 const sal_Char * p = rBegin;
873 for ( ; p != pEnd; ++p)
874 {
875 int nWeight = getWeight(*p);
876 if (nWeight < 0)
877 break;
878 nTheValue = 10 * nTheValue + nWeight;
879 if (nTheValue > std::numeric_limits< sal_uInt32 >::max())
880 return false;
881 }
882 if (nTheValue == 0 && (p == rBegin || (!bLeadingZeroes && p - rBegin != 1)))
883 return false;
884 rBegin = p;
885 rValue = sal_uInt32(nTheValue);
886 return true;
887 }
888
889 //============================================================================
890 // static
891 bool INetMIME::scanUnsigned(const sal_Unicode *& rBegin,
892 const sal_Unicode * pEnd, bool bLeadingZeroes,
893 sal_uInt32 & rValue)
894 {
895 sal_uInt64 nTheValue = 0;
896 const sal_Unicode * p = rBegin;
897 for ( ; p != pEnd; ++p)
898 {
899 int nWeight = getWeight(*p);
900 if (nWeight < 0)
901 break;
902 nTheValue = 10 * nTheValue + nWeight;
903 if (nTheValue > std::numeric_limits< sal_uInt32 >::max())
904 return false;
905 }
906 if (nTheValue == 0 && (p == rBegin || (!bLeadingZeroes && p - rBegin != 1)))
907 return false;
908 rBegin = p;
909 rValue = sal_uInt32(nTheValue);
910 return true;
911 }
912
913 //============================================================================
914 // static
915 bool INetMIME::scanUnsignedHex(const sal_Char *& rBegin,
916 const sal_Char * pEnd, bool bLeadingZeroes,
917 sal_uInt32 & rValue)
918 {
919 sal_uInt64 nTheValue = 0;
920 const sal_Char * p = rBegin;
921 for ( p = rBegin; p != pEnd; ++p)
922 {
923 int nWeight = getHexWeight(*p);
924 if (nWeight < 0)
925 break;
926 nTheValue = nTheValue << 4 | nWeight;
927 if (nTheValue > std::numeric_limits< sal_uInt32 >::max())
928 return false;
929 }
930 if (nTheValue == 0 && (p == rBegin || (!bLeadingZeroes && p - rBegin != 1)))
931 return false;
932 rBegin = p;
933 rValue = sal_uInt32(nTheValue);
934 return true;
935 }
936
937 //============================================================================
938 // static
939 bool INetMIME::scanUnsignedHex(const sal_Unicode *& rBegin,
940 const sal_Unicode * pEnd, bool bLeadingZeroes,
941 sal_uInt32 & rValue)
942 {
943 sal_uInt64 nTheValue = 0;
944 const sal_Unicode * p = rBegin;
945 for ( ; p != pEnd; ++p)
946 {
947 int nWeight = getHexWeight(*p);
948 if (nWeight < 0)
949 break;
950 nTheValue = nTheValue << 4 | nWeight;
951 if (nTheValue > std::numeric_limits< sal_uInt32 >::max())
952 return false;
953 }
954 if (nTheValue == 0 && (p == rBegin || (!bLeadingZeroes && p - rBegin != 1)))
955 return false;
956 rBegin = p;
957 rValue = sal_uInt32(nTheValue);
958 return true;
959 }
960
961 //============================================================================
962 // static
963 const sal_Char * INetMIME::scanQuotedBlock(const sal_Char * pBegin,
964 const sal_Char * pEnd,
965 sal_uInt32 nOpening,
966 sal_uInt32 nClosing,
967 sal_Size & rLength,
968 bool & rModify)
969 {
970 DBG_ASSERT(pBegin && pBegin <= pEnd,
971 "INetMIME::scanQuotedBlock(): Bad sequence");
972
973 if (pBegin != pEnd && static_cast< unsigned char >(*pBegin) == nOpening)
974 {
975 ++rLength;
976 ++pBegin;
977 while (pBegin != pEnd)
978 if (static_cast< unsigned char >(*pBegin) == nClosing)
979 {
980 ++rLength;
981 return ++pBegin;
982 }
983 else
984 {
985 sal_uInt32 c = *pBegin++;
986 switch (c)
987 {
988 case 0x0D: // CR
989 if (pBegin != pEnd && *pBegin == 0x0A) // LF
990 if (pEnd - pBegin >= 2 && isWhiteSpace(pBegin[1]))
991 {
992 ++rLength;
993 rModify = true;
994 pBegin += 2;
995 }
996 else
997 {
998 rLength += 3;
999 rModify = true;
1000 ++pBegin;
1001 }
1002 else
1003 ++rLength;
1004 break;
1005
1006 case '\\':
1007 ++rLength;
1008 if (pBegin != pEnd)
1009 {
1010 if (startsWithLineBreak(pBegin, pEnd)
1011 && (pEnd - pBegin < 3
1012 || !isWhiteSpace(pBegin[2])))
1013 {
1014 rLength += 3;
1015 rModify = true;
1016 pBegin += 2;
1017 }
1018 else
1019 ++pBegin;
1020 }
1021 break;
1022
1023 default:
1024 ++rLength;
1025 if (!isUSASCII(c))
1026 rModify = true;
1027 break;
1028 }
1029 }
1030 }
1031 return pBegin;
1032 }
1033
1034 //============================================================================
1035 // static
1036 const sal_Unicode * INetMIME::scanQuotedBlock(const sal_Unicode * pBegin,
1037 const sal_Unicode * pEnd,
1038 sal_uInt32 nOpening,
1039 sal_uInt32 nClosing,
1040 sal_Size & rLength,
1041 bool & rModify)
1042 {
1043 DBG_ASSERT(pBegin && pBegin <= pEnd,
1044 "INetMIME::scanQuotedBlock(): Bad sequence");
1045
1046 if (pBegin != pEnd && *pBegin == nOpening)
1047 {
1048 ++rLength;
1049 ++pBegin;
1050 while (pBegin != pEnd)
1051 if (*pBegin == nClosing)
1052 {
1053 ++rLength;
1054 return ++pBegin;
1055 }
1056 else
1057 {
1058 sal_uInt32 c = *pBegin++;
1059 switch (c)
1060 {
1061 case 0x0D: // CR
1062 if (pBegin != pEnd && *pBegin == 0x0A) // LF
1063 if (pEnd - pBegin >= 2 && isWhiteSpace(pBegin[1]))
1064 {
1065 ++rLength;
1066 rModify = true;
1067 pBegin += 2;
1068 }
1069 else
1070 {
1071 rLength += 3;
1072 rModify = true;
1073 ++pBegin;
1074 }
1075 else
1076 ++rLength;
1077 break;
1078
1079 case '\\':
1080 ++rLength;
1081 if (pBegin != pEnd)
1082 {
1083 if (startsWithLineBreak(pBegin, pEnd)
1084 && (pEnd - pBegin < 3
1085 || !isWhiteSpace(pBegin[2])))
1086 {
1087 rLength += 3;
1088 rModify = true;
1089 pBegin += 2;
1090 }
1091 else
1092 ++pBegin;
1093 }
1094 break;
1095
1096 default:
1097 ++rLength;
1098 if (!isUSASCII(c))
1099 rModify = true;
1100 break;
1101 }
1102 }
1103 }
1104 return pBegin;
1105 }
1106
1107 //============================================================================
1108 // static
1109 sal_Char const * INetMIME::scanParameters(sal_Char const * pBegin,
1110 sal_Char const * pEnd,
1111 INetContentTypeParameterList *
1112 pParameters)
1113 {
1114 ParameterList aList;
1115 sal_Char const * pParameterBegin = pBegin;
1116 for (sal_Char const * p = pParameterBegin;; pParameterBegin = p)
1117 {
1118 pParameterBegin = skipLinearWhiteSpaceComment(p, pEnd);
1119 if (pParameterBegin == pEnd || *pParameterBegin != ';')
1120 break;
1121 p = pParameterBegin + 1;
1122
1123 sal_Char const * pAttributeBegin = skipLinearWhiteSpaceComment(p,
1124 pEnd);
1125 p = pAttributeBegin;
1126 bool bDowncaseAttribute = false;
1127 while (p != pEnd && isTokenChar(*p) && *p != '*')
1128 {
1129 bDowncaseAttribute = bDowncaseAttribute || isUpperCase(*p);
1130 ++p;
1131 }
1132 if (p == pAttributeBegin)
1133 break;
1134 ByteString aAttribute(
1135 pAttributeBegin, static_cast< xub_StrLen >(p - pAttributeBegin));
1136 if (bDowncaseAttribute)
1137 aAttribute.ToLowerAscii();
1138
1139 sal_uInt32 nSection = 0;
1140 if (p != pEnd && *p == '*')
1141 {
1142 ++p;
1143 if (p != pEnd && isDigit(*p)
1144 && !scanUnsigned(p, pEnd, false, nSection))
1145 break;
1146 }
1147
1148 bool bPresent;
1149 Parameter ** pPos = aList.find(aAttribute, nSection, bPresent);
1150 if (bPresent)
1151 break;
1152
1153 bool bExtended = false;
1154 if (p != pEnd && *p == '*')
1155 {
1156 ++p;
1157 bExtended = true;
1158 }
1159
1160 p = skipLinearWhiteSpaceComment(p, pEnd);
1161
1162 if (p == pEnd || *p != '=')
1163 break;
1164
1165 p = skipLinearWhiteSpaceComment(p + 1, pEnd);
1166
1167 ByteString aCharset;
1168 ByteString aLanguage;
1169 ByteString aValue;
1170 if (bExtended)
1171 {
1172 if (nSection == 0)
1173 {
1174 sal_Char const * pCharsetBegin = p;
1175 bool bDowncaseCharset = false;
1176 while (p != pEnd && isTokenChar(*p) && *p != '\'')
1177 {
1178 bDowncaseCharset = bDowncaseCharset || isUpperCase(*p);
1179 ++p;
1180 }
1181 if (p == pCharsetBegin)
1182 break;
1183 if (pParameters)
1184 {
1185 aCharset = ByteString(
1186 pCharsetBegin,
1187 static_cast< xub_StrLen >(p - pCharsetBegin));
1188 if (bDowncaseCharset)
1189 aCharset.ToLowerAscii();
1190 }
1191
1192 if (p == pEnd || *p != '\'')
1193 break;
1194 ++p;
1195
1196 sal_Char const * pLanguageBegin = p;
1197 bool bDowncaseLanguage = false;
1198 int nLetters = 0;
1199 for (; p != pEnd; ++p)
1200 if (isAlpha(*p))
1201 {
1202 if (++nLetters > 8)
1203 break;
1204 bDowncaseLanguage = bDowncaseLanguage
1205 || isUpperCase(*p);
1206 }
1207 else if (*p == '-')
1208 {
1209 if (nLetters == 0)
1210 break;
1211 nLetters = 0;
1212 }
1213 else
1214 break;
1215 if (nLetters == 0 || nLetters > 8)
1216 break;
1217 if (pParameters)
1218 {
1219 aLanguage = ByteString(
1220 pLanguageBegin,
1221 static_cast< xub_StrLen >(p - pLanguageBegin));
1222 if (bDowncaseLanguage)
1223 aLanguage.ToLowerAscii();
1224 }
1225
1226 if (p == pEnd || *p != '\'')
1227 break;
1228 ++p;
1229 }
1230 if (pParameters)
1231 while (p != pEnd && (isTokenChar(*p) || !isUSASCII(*p)))
1232 {
1233 if (*p == '%')
1234 {
1235 if (p + 2 < pEnd)
1236 {
1237 int nWeight1 = getHexWeight(p[1]);
1238 int nWeight2 = getHexWeight(p[2]);
1239 if (nWeight1 >= 0 && nWeight2 >= 0)
1240 {
1241 aValue += sal_Char(nWeight1 << 4 | nWeight2);
1242 p += 3;
1243 continue;
1244 }
1245 }
1246 }
1247 aValue += *p++;
1248 }
1249 else
1250 while (p != pEnd && (isTokenChar(*p) || !isUSASCII(*p)))
1251 ++p;
1252 }
1253 else if (p != pEnd && *p == '"')
1254 if (pParameters)
1255 {
1256 bool bInvalid = false;
1257 for (++p;;)
1258 {
1259 if (p == pEnd)
1260 {
1261 bInvalid = true;
1262 break;
1263 }
1264 else if (*p == '"')
1265 {
1266 ++p;
1267 break;
1268 }
1269 else if (*p == 0x0D) // CR
1270 {
1271 if (pEnd - p < 3 || p[1] != 0x0A // LF
1272 || !isWhiteSpace(p[2]))
1273 {
1274 bInvalid = true;
1275 break;
1276 }
1277 p += 2;
1278 }
1279 else if (*p == '\\' && ++p == pEnd)
1280 {
1281 bInvalid = true;
1282 break;
1283 }
1284 aValue += *p++;
1285 }
1286 if (bInvalid)
1287 break;
1288 }
1289 else
1290 {
1291 sal_Char const * pStringEnd = skipQuotedString(p, pEnd);
1292 if (p == pStringEnd)
1293 break;
1294 p = pStringEnd;
1295 }
1296 else
1297 {
1298 sal_Char const * pTokenBegin = p;
1299 while (p != pEnd && (isTokenChar(*p) || !isUSASCII(*p)))
1300 ++p;
1301 if (p == pTokenBegin)
1302 break;
1303 if (pParameters)
1304 aValue = ByteString(
1305 pTokenBegin, static_cast< xub_StrLen >(p - pTokenBegin));
1306 }
1307
1308 *pPos = new Parameter(*pPos, aAttribute, aCharset, aLanguage, aValue,
1309 nSection, bExtended);
1310 }
1311 return parseParameters(aList, pParameters) ? pParameterBegin : pBegin;
1312 }
1313
1314 //============================================================================
1315 // static
1316 sal_Unicode const * INetMIME::scanParameters(sal_Unicode const * pBegin,
1317 sal_Unicode const * pEnd,
1318 INetContentTypeParameterList *
1319 pParameters)
1320 {
1321 ParameterList aList;
1322 sal_Unicode const * pParameterBegin = pBegin;
1323 for (sal_Unicode const * p = pParameterBegin;; pParameterBegin = p)
1324 {
1325 pParameterBegin = skipLinearWhiteSpaceComment(p, pEnd);
1326 if (pParameterBegin == pEnd || *pParameterBegin != ';')
1327 break;
1328 p = pParameterBegin + 1;
1329
1330 sal_Unicode const * pAttributeBegin
1331 = skipLinearWhiteSpaceComment(p, pEnd);
1332 p = pAttributeBegin;
1333 bool bDowncaseAttribute = false;
1334 while (p != pEnd && isTokenChar(*p) && *p != '*')
1335 {
1336 bDowncaseAttribute = bDowncaseAttribute || isUpperCase(*p);
1337 ++p;
1338 }
1339 if (p == pAttributeBegin)
1340 break;
1341 ByteString aAttribute = ByteString(
1342 pAttributeBegin, static_cast< xub_StrLen >(p - pAttributeBegin),
1343 RTL_TEXTENCODING_ASCII_US);
1344 if (bDowncaseAttribute)
1345 aAttribute.ToLowerAscii();
1346
1347 sal_uInt32 nSection = 0;
1348 if (p != pEnd && *p == '*')
1349 {
1350 ++p;
1351 if (p != pEnd && isDigit(*p)
1352 && !scanUnsigned(p, pEnd, false, nSection))
1353 break;
1354 }
1355
1356 bool bPresent;
1357 Parameter ** pPos = aList.find(aAttribute, nSection, bPresent);
1358 if (bPresent)
1359 break;
1360
1361 bool bExtended = false;
1362 if (p != pEnd && *p == '*')
1363 {
1364 ++p;
1365 bExtended = true;
1366 }
1367
1368 p = skipLinearWhiteSpaceComment(p, pEnd);
1369
1370 if (p == pEnd || *p != '=')
1371 break;
1372
1373 p = skipLinearWhiteSpaceComment(p + 1, pEnd);
1374
1375 ByteString aCharset;
1376 ByteString aLanguage;
1377 ByteString aValue;
1378 if (bExtended)
1379 {
1380 if (nSection == 0)
1381 {
1382 sal_Unicode const * pCharsetBegin = p;
1383 bool bDowncaseCharset = false;
1384 while (p != pEnd && isTokenChar(*p) && *p != '\'')
1385 {
1386 bDowncaseCharset = bDowncaseCharset || isUpperCase(*p);
1387 ++p;
1388 }
1389 if (p == pCharsetBegin)
1390 break;
1391 if (pParameters)
1392 {
1393 aCharset = ByteString(
1394 pCharsetBegin,
1395 static_cast< xub_StrLen >(p - pCharsetBegin),
1396 RTL_TEXTENCODING_ASCII_US);
1397 if (bDowncaseCharset)
1398 aCharset.ToLowerAscii();
1399 }
1400
1401 if (p == pEnd || *p != '\'')
1402 break;
1403 ++p;
1404
1405 sal_Unicode const * pLanguageBegin = p;
1406 bool bDowncaseLanguage = false;
1407 int nLetters = 0;
1408 for (; p != pEnd; ++p)
1409 if (isAlpha(*p))
1410 {
1411 if (++nLetters > 8)
1412 break;
1413 bDowncaseLanguage = bDowncaseLanguage
1414 || isUpperCase(*p);
1415 }
1416 else if (*p == '-')
1417 {
1418 if (nLetters == 0)
1419 break;
1420 nLetters = 0;
1421 }
1422 else
1423 break;
1424 if (nLetters == 0 || nLetters > 8)
1425 break;
1426 if (pParameters)
1427 {
1428 aLanguage = ByteString(
1429 pLanguageBegin,
1430 static_cast< xub_StrLen >(p - pLanguageBegin),
1431 RTL_TEXTENCODING_ASCII_US);
1432 if (bDowncaseLanguage)
1433 aLanguage.ToLowerAscii();
1434 }
1435
1436 if (p == pEnd || *p != '\'')
1437 break;
1438 ++p;
1439 }
1440 if (pParameters)
1441 {
1442 INetMIMEStringOutputSink
1443 aSink(0, INetMIMEOutputSink::NO_LINE_LENGTH_LIMIT);
1444 while (p != pEnd)
1445 {
1446 sal_uInt32 nChar = INetMIME::getUTF32Character(p, pEnd);
1447 if (isUSASCII(nChar) && !isTokenChar(nChar))
1448 break;
1449 if (nChar == '%' && p + 1 < pEnd)
1450 {
1451 int nWeight1 = getHexWeight(p[0]);
1452 int nWeight2 = getHexWeight(p[1]);
1453 if (nWeight1 >= 0 && nWeight2 >= 0)
1454 {
1455 aSink << sal_Char(nWeight1 << 4 | nWeight2);
1456 p += 2;
1457 continue;
1458 }
1459 }
1460 INetMIME::writeUTF8(aSink, nChar);
1461 }
1462 aValue = aSink.takeBuffer();
1463 }
1464 else
1465 while (p != pEnd && (isTokenChar(*p) || !isUSASCII(*p)))
1466 ++p;
1467 }
1468 else if (p != pEnd && *p == '"')
1469 if (pParameters)
1470 {
1471 INetMIMEStringOutputSink
1472 aSink(0, INetMIMEOutputSink::NO_LINE_LENGTH_LIMIT);
1473 bool bInvalid = false;
1474 for (++p;;)
1475 {
1476 if (p == pEnd)
1477 {
1478 bInvalid = true;
1479 break;
1480 }
1481 sal_uInt32 nChar = INetMIME::getUTF32Character(p, pEnd);
1482 if (nChar == '"')
1483 break;
1484 else if (nChar == 0x0D) // CR
1485 {
1486 if (pEnd - p < 2 || *p++ != 0x0A // LF
1487 || !isWhiteSpace(*p))
1488 {
1489 bInvalid = true;
1490 break;
1491 }
1492 nChar = sal_uChar(*p++);
1493 }
1494 else if (nChar == '\\')
1495 {
1496 if (p == pEnd)
1497 {
1498 bInvalid = true;
1499 break;
1500 }
1501 nChar = INetMIME::getUTF32Character(p, pEnd);
1502 }
1503 INetMIME::writeUTF8(aSink, nChar);
1504 }
1505 if (bInvalid)
1506 break;
1507 aValue = aSink.takeBuffer();
1508 }
1509 else
1510 {
1511 sal_Unicode const * pStringEnd = skipQuotedString(p, pEnd);
1512 if (p == pStringEnd)
1513 break;
1514 p = pStringEnd;
1515 }
1516 else
1517 {
1518 sal_Unicode const * pTokenBegin = p;
1519 while (p != pEnd && (isTokenChar(*p) || !isUSASCII(*p)))
1520 ++p;
1521 if (p == pTokenBegin)
1522 break;
1523 if (pParameters)
1524 aValue = ByteString(
1525 pTokenBegin, static_cast< xub_StrLen >(p - pTokenBegin),
1526 RTL_TEXTENCODING_UTF8);
1527 }
1528
1529 *pPos = new Parameter(*pPos, aAttribute, aCharset, aLanguage, aValue,
1530 nSection, bExtended);
1531 }
1532 return parseParameters(aList, pParameters) ? pParameterBegin : pBegin;
1533 }
1534
1535 //============================================================================
1536 // static
1537 const sal_Char * INetMIME::getCharsetName(rtl_TextEncoding eEncoding)
1538 {
1539 if (rtl_isOctetTextEncoding(eEncoding))
1540 {
1541 char const * p = rtl_getMimeCharsetFromTextEncoding(eEncoding);
1542 DBG_ASSERT(p, "INetMIME::getCharsetName(): Unsupported encoding");
1543 return p;
1544 }
1545 else
1546 switch (eEncoding)
1547 {
1548 case RTL_TEXTENCODING_UCS4:
1549 return "ISO-10646-UCS-4";
1550
1551 case RTL_TEXTENCODING_UCS2:
1552 return "ISO-10646-UCS-2";
1553
1554 default:
1555 DBG_ERROR("INetMIME::getCharsetName(): Unsupported encoding");
1556 return 0;
1557 }
1558 }
1559
1560 //============================================================================
1561 namespace unnamed_tools_inetmime {
1562
1563 struct EncodingEntry
1564 {
1565 sal_Char const * m_aName;
1566 rtl_TextEncoding m_eEncoding;
1567 };
1568
1569 //============================================================================
1570 // The source for the following table is <ftp://ftp.iana.org/in-notes/iana/
1571 // assignments/character-sets> as of Jan, 21 2000 12:46:00, unless otherwise
1572 // noted:
1573 EncodingEntry const aEncodingMap[]
1574 = { { "US-ASCII", RTL_TEXTENCODING_ASCII_US },
1575 { "ANSI_X3.4-1968", RTL_TEXTENCODING_ASCII_US },
1576 { "ISO-IR-6", RTL_TEXTENCODING_ASCII_US },
1577 { "ANSI_X3.4-1986", RTL_TEXTENCODING_ASCII_US },
1578 { "ISO_646.IRV:1991", RTL_TEXTENCODING_ASCII_US },
1579 { "ASCII", RTL_TEXTENCODING_ASCII_US },
1580 { "ISO646-US", RTL_TEXTENCODING_ASCII_US },
1581 { "US", RTL_TEXTENCODING_ASCII_US },
1582 { "IBM367", RTL_TEXTENCODING_ASCII_US },
1583 { "CP367", RTL_TEXTENCODING_ASCII_US },
1584 { "CSASCII", RTL_TEXTENCODING_ASCII_US },
1585 { "ISO-8859-1", RTL_TEXTENCODING_ISO_8859_1 },
1586 { "ISO_8859-1:1987", RTL_TEXTENCODING_ISO_8859_1 },
1587 { "ISO-IR-100", RTL_TEXTENCODING_ISO_8859_1 },
1588 { "ISO_8859-1", RTL_TEXTENCODING_ISO_8859_1 },
1589 { "LATIN1", RTL_TEXTENCODING_ISO_8859_1 },
1590 { "L1", RTL_TEXTENCODING_ISO_8859_1 },
1591 { "IBM819", RTL_TEXTENCODING_ISO_8859_1 },
1592 { "CP819", RTL_TEXTENCODING_ISO_8859_1 },
1593 { "CSISOLATIN1", RTL_TEXTENCODING_ISO_8859_1 },
1594 { "ISO-8859-2", RTL_TEXTENCODING_ISO_8859_2 },
1595 { "ISO_8859-2:1987", RTL_TEXTENCODING_ISO_8859_2 },
1596 { "ISO-IR-101", RTL_TEXTENCODING_ISO_8859_2 },
1597 { "ISO_8859-2", RTL_TEXTENCODING_ISO_8859_2 },
1598 { "LATIN2", RTL_TEXTENCODING_ISO_8859_2 },
1599 { "L2", RTL_TEXTENCODING_ISO_8859_2 },
1600 { "CSISOLATIN2", RTL_TEXTENCODING_ISO_8859_2 },
1601 { "ISO-8859-3", RTL_TEXTENCODING_ISO_8859_3 },
1602 { "ISO_8859-3:1988", RTL_TEXTENCODING_ISO_8859_3 },
1603 { "ISO-IR-109", RTL_TEXTENCODING_ISO_8859_3 },
1604 { "ISO_8859-3", RTL_TEXTENCODING_ISO_8859_3 },
1605 { "LATIN3", RTL_TEXTENCODING_ISO_8859_3 },
1606 { "L3", RTL_TEXTENCODING_ISO_8859_3 },
1607 { "CSISOLATIN3", RTL_TEXTENCODING_ISO_8859_3 },
1608 { "ISO-8859-4", RTL_TEXTENCODING_ISO_8859_4 },
1609 { "ISO_8859-4:1988", RTL_TEXTENCODING_ISO_8859_4 },
1610 { "ISO-IR-110", RTL_TEXTENCODING_ISO_8859_4 },
1611 { "ISO_8859-4", RTL_TEXTENCODING_ISO_8859_4 },
1612 { "LATIN4", RTL_TEXTENCODING_ISO_8859_4 },
1613 { "L4", RTL_TEXTENCODING_ISO_8859_4 },
1614 { "CSISOLATIN4", RTL_TEXTENCODING_ISO_8859_4 },
1615 { "ISO-8859-5", RTL_TEXTENCODING_ISO_8859_5 },
1616 { "ISO_8859-5:1988", RTL_TEXTENCODING_ISO_8859_5 },
1617 { "ISO-IR-144", RTL_TEXTENCODING_ISO_8859_5 },
1618 { "ISO_8859-5", RTL_TEXTENCODING_ISO_8859_5 },
1619 { "CYRILLIC", RTL_TEXTENCODING_ISO_8859_5 },
1620 { "CSISOLATINCYRILLIC", RTL_TEXTENCODING_ISO_8859_5 },
1621 { "ISO-8859-6", RTL_TEXTENCODING_ISO_8859_6 },
1622 { "ISO_8859-6:1987", RTL_TEXTENCODING_ISO_8859_6 },
1623 { "ISO-IR-127", RTL_TEXTENCODING_ISO_8859_6 },
1624 { "ISO_8859-6", RTL_TEXTENCODING_ISO_8859_6 },
1625 { "ECMA-114", RTL_TEXTENCODING_ISO_8859_6 },
1626 { "ASMO-708", RTL_TEXTENCODING_ISO_8859_6 },
1627 { "ARABIC", RTL_TEXTENCODING_ISO_8859_6 },
1628 { "CSISOLATINARABIC", RTL_TEXTENCODING_ISO_8859_6 },
1629 { "ISO-8859-7", RTL_TEXTENCODING_ISO_8859_7 },
1630 { "ISO_8859-7:1987", RTL_TEXTENCODING_ISO_8859_7 },
1631 { "ISO-IR-126", RTL_TEXTENCODING_ISO_8859_7 },
1632 { "ISO_8859-7", RTL_TEXTENCODING_ISO_8859_7 },
1633 { "ELOT_928", RTL_TEXTENCODING_ISO_8859_7 },
1634 { "ECMA-118", RTL_TEXTENCODING_ISO_8859_7 },
1635 { "GREEK", RTL_TEXTENCODING_ISO_8859_7 },
1636 { "GREEK8", RTL_TEXTENCODING_ISO_8859_7 },
1637 { "CSISOLATINGREEK", RTL_TEXTENCODING_ISO_8859_7 },
1638 { "ISO-8859-8", RTL_TEXTENCODING_ISO_8859_8 },
1639 { "ISO_8859-8:1988", RTL_TEXTENCODING_ISO_8859_8 },
1640 { "ISO-IR-138", RTL_TEXTENCODING_ISO_8859_8 },
1641 { "ISO_8859-8", RTL_TEXTENCODING_ISO_8859_8 },
1642 { "HEBREW", RTL_TEXTENCODING_ISO_8859_8 },
1643 { "CSISOLATINHEBREW", RTL_TEXTENCODING_ISO_8859_8 },
1644 { "ISO-8859-9", RTL_TEXTENCODING_ISO_8859_9 },
1645 { "ISO_8859-9:1989", RTL_TEXTENCODING_ISO_8859_9 },
1646 { "ISO-IR-148", RTL_TEXTENCODING_ISO_8859_9 },
1647 { "ISO_8859-9", RTL_TEXTENCODING_ISO_8859_9 },
1648 { "LATIN5", RTL_TEXTENCODING_ISO_8859_9 },
1649 { "L5", RTL_TEXTENCODING_ISO_8859_9 },
1650 { "CSISOLATIN5", RTL_TEXTENCODING_ISO_8859_9 },
1651 { "ISO-8859-14", RTL_TEXTENCODING_ISO_8859_14 }, // RFC 2047
1652 { "ISO_8859-15", RTL_TEXTENCODING_ISO_8859_15 },
1653 { "ISO-8859-15", RTL_TEXTENCODING_ISO_8859_15 }, // RFC 2047
1654 { "MACINTOSH", RTL_TEXTENCODING_APPLE_ROMAN },
1655 { "MAC", RTL_TEXTENCODING_APPLE_ROMAN },
1656 { "CSMACINTOSH", RTL_TEXTENCODING_APPLE_ROMAN },
1657 { "IBM437", RTL_TEXTENCODING_IBM_437 },
1658 { "CP437", RTL_TEXTENCODING_IBM_437 },
1659 { "437", RTL_TEXTENCODING_IBM_437 },
1660 { "CSPC8CODEPAGE437", RTL_TEXTENCODING_IBM_437 },
1661 { "IBM850", RTL_TEXTENCODING_IBM_850 },
1662 { "CP850", RTL_TEXTENCODING_IBM_850 },
1663 { "850", RTL_TEXTENCODING_IBM_850 },
1664 { "CSPC850MULTILINGUAL", RTL_TEXTENCODING_IBM_850 },
1665 { "IBM860", RTL_TEXTENCODING_IBM_860 },
1666 { "CP860", RTL_TEXTENCODING_IBM_860 },
1667 { "860", RTL_TEXTENCODING_IBM_860 },
1668 { "CSIBM860", RTL_TEXTENCODING_IBM_860 },
1669 { "IBM861", RTL_TEXTENCODING_IBM_861 },
1670 { "CP861", RTL_TEXTENCODING_IBM_861 },
1671 { "861", RTL_TEXTENCODING_IBM_861 },
1672 { "CP-IS", RTL_TEXTENCODING_IBM_861 },
1673 { "CSIBM861", RTL_TEXTENCODING_IBM_861 },
1674 { "IBM863", RTL_TEXTENCODING_IBM_863 },
1675 { "CP863", RTL_TEXTENCODING_IBM_863 },
1676 { "863", RTL_TEXTENCODING_IBM_863 },
1677 { "CSIBM863", RTL_TEXTENCODING_IBM_863 },
1678 { "IBM865", RTL_TEXTENCODING_IBM_865 },
1679 { "CP865", RTL_TEXTENCODING_IBM_865 },
1680 { "865", RTL_TEXTENCODING_IBM_865 },
1681 { "CSIBM865", RTL_TEXTENCODING_IBM_865 },
1682 { "IBM775", RTL_TEXTENCODING_IBM_775 },
1683 { "CP775", RTL_TEXTENCODING_IBM_775 },
1684 { "CSPC775BALTIC", RTL_TEXTENCODING_IBM_775 },
1685 { "IBM852", RTL_TEXTENCODING_IBM_852 },
1686 { "CP852", RTL_TEXTENCODING_IBM_852 },
1687 { "852", RTL_TEXTENCODING_IBM_852 },
1688 { "CSPCP852", RTL_TEXTENCODING_IBM_852 },
1689 { "IBM855", RTL_TEXTENCODING_IBM_855 },
1690 { "CP855", RTL_TEXTENCODING_IBM_855 },
1691 { "855", RTL_TEXTENCODING_IBM_855 },
1692 { "CSIBM855", RTL_TEXTENCODING_IBM_855 },
1693 { "IBM857", RTL_TEXTENCODING_IBM_857 },
1694 { "CP857", RTL_TEXTENCODING_IBM_857 },
1695 { "857", RTL_TEXTENCODING_IBM_857 },
1696 { "CSIBM857", RTL_TEXTENCODING_IBM_857 },
1697 { "IBM862", RTL_TEXTENCODING_IBM_862 },
1698 { "CP862", RTL_TEXTENCODING_IBM_862 },
1699 { "862", RTL_TEXTENCODING_IBM_862 },
1700 { "CSPC862LATINHEBREW", RTL_TEXTENCODING_IBM_862 },
1701 { "IBM864", RTL_TEXTENCODING_IBM_864 },
1702 { "CP864", RTL_TEXTENCODING_IBM_864 },
1703 { "CSIBM864", RTL_TEXTENCODING_IBM_864 },
1704 { "IBM866", RTL_TEXTENCODING_IBM_866 },
1705 { "CP866", RTL_TEXTENCODING_IBM_866 },
1706 { "866", RTL_TEXTENCODING_IBM_866 },
1707 { "CSIBM866", RTL_TEXTENCODING_IBM_866 },
1708 { "IBM869", RTL_TEXTENCODING_IBM_869 },
1709 { "CP869", RTL_TEXTENCODING_IBM_869 },
1710 { "869", RTL_TEXTENCODING_IBM_869 },
1711 { "CP-GR", RTL_TEXTENCODING_IBM_869 },
1712 { "CSIBM869", RTL_TEXTENCODING_IBM_869 },
1713 { "WINDOWS-1250", RTL_TEXTENCODING_MS_1250 },
1714 { "WINDOWS-1251", RTL_TEXTENCODING_MS_1251 },
1715 { "WINDOWS-1253", RTL_TEXTENCODING_MS_1253 },
1716 { "WINDOWS-1254", RTL_TEXTENCODING_MS_1254 },
1717 { "WINDOWS-1255", RTL_TEXTENCODING_MS_1255 },
1718 { "WINDOWS-1256", RTL_TEXTENCODING_MS_1256 },
1719 { "WINDOWS-1257", RTL_TEXTENCODING_MS_1257 },
1720 { "WINDOWS-1258", RTL_TEXTENCODING_MS_1258 },
1721 { "SHIFT_JIS", RTL_TEXTENCODING_SHIFT_JIS },
1722 { "MS_KANJI", RTL_TEXTENCODING_SHIFT_JIS },
1723 { "CSSHIFTJIS", RTL_TEXTENCODING_SHIFT_JIS },
1724 { "GB2312", RTL_TEXTENCODING_GB_2312 },
1725 { "CSGB2312", RTL_TEXTENCODING_GB_2312 },
1726 { "BIG5", RTL_TEXTENCODING_BIG5 },
1727 { "CSBIG5", RTL_TEXTENCODING_BIG5 },
1728 { "EUC-JP", RTL_TEXTENCODING_EUC_JP },
1729 { "EXTENDED_UNIX_CODE_PACKED_FORMAT_FOR_JAPANESE",
1730 RTL_TEXTENCODING_EUC_JP },
1731 { "CSEUCPKDFMTJAPANESE", RTL_TEXTENCODING_EUC_JP },
1732 { "ISO-2022-JP", RTL_TEXTENCODING_ISO_2022_JP },
1733 { "CSISO2022JP", RTL_TEXTENCODING_ISO_2022_JP },
1734 { "ISO-2022-CN", RTL_TEXTENCODING_ISO_2022_CN },
1735 { "KOI8-R", RTL_TEXTENCODING_KOI8_R },
1736 { "CSKOI8R", RTL_TEXTENCODING_KOI8_R },
1737 { "UTF-7", RTL_TEXTENCODING_UTF7 },
1738 { "UTF-8", RTL_TEXTENCODING_UTF8 },
1739 { "ISO-8859-10", RTL_TEXTENCODING_ISO_8859_10 }, // RFC 2047
1740 { "ISO-8859-13", RTL_TEXTENCODING_ISO_8859_13 }, // RFC 2047
1741 { "EUC-KR", RTL_TEXTENCODING_EUC_KR },
1742 { "CSEUCKR", RTL_TEXTENCODING_EUC_KR },
1743 { "ISO-2022-KR", RTL_TEXTENCODING_ISO_2022_KR },
1744 { "CSISO2022KR", RTL_TEXTENCODING_ISO_2022_KR },
1745 { "ISO-10646-UCS-4", RTL_TEXTENCODING_UCS4 },
1746 { "CSUCS4", RTL_TEXTENCODING_UCS4 },
1747 { "ISO-10646-UCS-2", RTL_TEXTENCODING_UCS2 },
1748 { "CSUNICODE", RTL_TEXTENCODING_UCS2 } };
1749
1750 //============================================================================
1751 template< typename T >
1752 inline rtl_TextEncoding getCharsetEncoding_Impl(T const * pBegin,
1753 T const * pEnd)
1754 {
1755 for (sal_Size i = 0; i < sizeof aEncodingMap / sizeof (EncodingEntry);
1756 ++i)
1757 if (INetMIME::equalIgnoreCase(pBegin, pEnd, aEncodingMap[i].m_aName))
1758 return aEncodingMap[i].m_eEncoding;
1759 return RTL_TEXTENCODING_DONTKNOW;
1760 }
1761
1762 }
1763
1764 //============================================================================
1765 // static
1766 rtl_TextEncoding INetMIME::getCharsetEncoding(sal_Char const * pBegin,
1767 sal_Char const * pEnd)
1768 {
1769 return getCharsetEncoding_Impl(pBegin, pEnd);
1770 }
1771
1772 //============================================================================
1773 // static
1774 rtl_TextEncoding INetMIME::getCharsetEncoding(sal_Unicode const * pBegin,
1775 sal_Unicode const * pEnd)
1776 {
1777 return getCharsetEncoding_Impl(pBegin, pEnd);
1778 }
1779
1780 //============================================================================
1781 // static
1782 INetMIMECharsetList_Impl *
1783 INetMIME::createPreferredCharsetList(rtl_TextEncoding eEncoding)
1784 {
1785 static const sal_uInt32 aUSASCIIRanges[] = { 0, 0x7F, sal_uInt32(-1) };
1786
1787 static const sal_uInt32 aISO88591Ranges[] = { 0, 0xFF, sal_uInt32(-1) };
1788 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-1.TXT> version
1789 // 1.0 of 1999 July 27
1790
1791 static const sal_uInt32 aISO88592Ranges[]
1792 = { 0, 0xA0, 0xA4, 0xA4, 0xA7, 0xA8, 0xAD, 0xAD, 0xB0, 0xB0,
1793 0xB4, 0xB4, 0xB8, 0xB8, 0xC1, 0xC2, 0xC4, 0xC4, 0xC7, 0xC7,
1794 0xC9, 0xC9, 0xCB, 0xCB, 0xCD, 0xCE, 0xD3, 0xD4, 0xD6, 0xD7,
1795 0xDA, 0xDA, 0xDC, 0xDD, 0xDF, 0xDF, 0xE1, 0xE2, 0xE4, 0xE4,
1796 0xE7, 0xE7, 0xE9, 0xE9, 0xEB, 0xEB, 0xED, 0xEE, 0xF3, 0xF4,
1797 0xF6, 0xF7, 0xFA, 0xFA, 0xFC, 0xFD, 0x102, 0x107, 0x10C, 0x111,
1798 0x118, 0x11B, 0x139, 0x13A, 0x13D, 0x13E, 0x141, 0x144,
1799 0x147, 0x148, 0x150, 0x151, 0x154, 0x155, 0x158, 0x15B,
1800 0x15E, 0x165, 0x16E, 0x171, 0x179, 0x17E, 0x2C7, 0x2C7,
1801 0x2D8, 0x2D9, 0x2DB, 0x2DB, 0x2DD, 0x2DD, sal_uInt32(-1) };
1802 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-2.TXT> version
1803 // 1.0 of 1999 July 27
1804
1805 static const sal_uInt32 aISO88593Ranges[]
1806 = { 0, 0xA0, 0xA3, 0xA4, 0xA7, 0xA8, 0xAD, 0xAD, 0xB0, 0xB0,
1807 0xB2, 0xB5, 0xB7, 0xB8, 0xBD, 0xBD, 0xC0, 0xC2, 0xC4, 0xC4,
1808 0xC7, 0xCF, 0xD1, 0xD4, 0xD6, 0xD7, 0xD9, 0xDC, 0xDF, 0xE2,
1809 0xE4, 0xE4, 0xE7, 0xEF, 0xF1, 0xF4, 0xF6, 0xF7, 0xF9, 0xFC,
1810 0x108, 0x10B, 0x11C, 0x121, 0x124, 0x127, 0x130, 0x131,
1811 0x134, 0x135, 0x15C, 0x15F, 0x16C, 0x16D, 0x17B, 0x17C,
1812 0x2D8, 0x2D9, sal_uInt32(-1) };
1813 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-3.TXT> version
1814 // 1.0 of 1999 July 27
1815
1816 static const sal_uInt32 aISO88594Ranges[]
1817 = { 0, 0xA0, 0xA4, 0xA4, 0xA7, 0xA8, 0xAD, 0xAD, 0xAF, 0xB0,
1818 0xB4, 0xB4, 0xB8, 0xB8, 0xC1, 0xC6, 0xC9, 0xC9, 0xCB, 0xCB,
1819 0xCD, 0xCE, 0xD4, 0xD8, 0xDA, 0xDC, 0xDF, 0xDF, 0xE1, 0xE6,
1820 0xE9, 0xE9, 0xEB, 0xEB, 0xED, 0xEE, 0xF4, 0xF8, 0xFA, 0xFC,
1821 0x100, 0x101, 0x104, 0x105, 0x10C, 0x10D, 0x110, 0x113,
1822 0x116, 0x119, 0x122, 0x123, 0x128, 0x12B, 0x12E, 0x12F,
1823 0x136, 0x138, 0x13B, 0x13C, 0x145, 0x146, 0x14A, 0x14D,
1824 0x156, 0x157, 0x160, 0x161, 0x166, 0x16B, 0x172, 0x173,
1825 0x17D, 0x17E, 0x2C7, 0x2C7, 0x2D9, 0x2D9, 0x2DB, 0x2DB,
1826 sal_uInt32(-1) };
1827 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-4.TXT> version
1828 // 1.0 of 1999 July 27
1829
1830 static const sal_uInt32 aISO88595Ranges[]
1831 = { 0, 0xA0, 0xA7, 0xA7, 0xAD, 0xAD, 0x401, 0x40C, 0x40E, 0x44F,
1832 0x451, 0x45C, 0x45E, 0x45F, 0x2116, 0x2116, sal_uInt32(-1) };
1833 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-5.TXT> version
1834 // 1.0 of 1999 July 27
1835
1836 static const sal_uInt32 aISO88596Ranges[]
1837 = { 0, 0xA0, 0xA4, 0xA4, 0xAD, 0xAD, 0x60C, 0x60C, 0x61B, 0x61B,
1838 0x61F, 0x61F, 0x621, 0x63A, 0x640, 0x652, sal_uInt32(-1) };
1839 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-6.TXT> version
1840 // 1.0 of 1999 July 27
1841
1842 static const sal_uInt32 aISO88597Ranges[]
1843 = { 0, 0xA0, 0xA3, 0xA3, 0xA6, 0xA9, 0xAB, 0xAD, 0xB0, 0xB3,
1844 0xB7, 0xB7, 0xBB, 0xBB, 0xBD, 0xBD, 0x384, 0x386, 0x388, 0x38A,
1845 0x38C, 0x38C, 0x38E, 0x3A1, 0x3A3, 0x3CE, 0x2015, 0x2015,
1846 0x2018, 0x2019, sal_uInt32(-1) };
1847 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-7.TXT> version
1848 // 1.0 of 1999 July 27
1849
1850 static const sal_uInt32 aISO88598Ranges[]
1851 = { 0, 0xA0, 0xA2, 0xA9, 0xAB, 0xB9, 0xBB, 0xBE, 0xD7, 0xD7,
1852 0xF7, 0xF7, 0x5D0, 0x5EA, 0x200E, 0x200F, 0x2017, 0x2017,
1853 sal_uInt32(-1) };
1854 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-8.TXT> version
1855 // 1.1 of 2000-Jan-03
1856
1857 static const sal_uInt32 aISO88599Ranges[]
1858 = { 0, 0xCF, 0xD1, 0xDC, 0xDF, 0xEF, 0xF1, 0xFC, 0xFF, 0xFF,
1859 0x11E, 0x11F, 0x130, 0x131, 0x15E, 0x15F, sal_uInt32(-1) };
1860 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-9.TXT> version
1861 // 1.0 of 1999 July 27
1862
1863 static const sal_uInt32 aISO885910Ranges[]
1864 = { 0, 0xA0, 0xA7, 0xA7, 0xAD, 0xAD, 0xB0, 0xB0, 0xB7, 0xB7,
1865 0xC1, 0xC6, 0xC9, 0xC9, 0xCB, 0xCB, 0xCD, 0xD0, 0xD3, 0xD6,
1866 0xD8, 0xD8, 0xDA, 0xDF, 0xE1, 0xE6, 0xE9, 0xE9, 0xEB, 0xEB,
1867 0xED, 0xF0, 0xF3, 0xF6, 0xF8, 0xF8, 0xFA, 0xFE, 0x100, 0x101,
1868 0x104, 0x105, 0x10C, 0x10D, 0x110, 0x113, 0x116, 0x119,
1869 0x122, 0x123, 0x128, 0x12B, 0x12E, 0x12F, 0x136, 0x138,
1870 0x13B, 0x13C, 0x145, 0x146, 0x14A, 0x14D, 0x160, 0x161,
1871 0x166, 0x16B, 0x172, 0x173, 0x17D, 0x17E, 0x2015, 0x2015,
1872 sal_uInt32(-1) };
1873 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-10.TXT> version
1874 // 1.1 of 1999 October 11
1875
1876 static const sal_uInt32 aISO885913Ranges[]
1877 = { 0, 0xA0, 0xA2, 0xA4, 0xA6, 0xA7, 0xA9, 0xA9, 0xAB, 0xAE,
1878 0xB0, 0xB3, 0xB5, 0xB7, 0xB9, 0xB9, 0xBB, 0xBE, 0xC4, 0xC6,
1879 0xC9, 0xC9, 0xD3, 0xD3, 0xD5, 0xD8, 0xDC, 0xDC, 0xDF, 0xDF,
1880 0xE4, 0xE6, 0xE9, 0xE9, 0xF3, 0xF3, 0xF5, 0xF8, 0xFC, 0xFC,
1881 0x100, 0x101, 0x104, 0x107, 0x10C, 0x10D, 0x112, 0x113,
1882 0x116, 0x119, 0x122, 0x123, 0x12A, 0x12B, 0x12E, 0x12F,
1883 0x136, 0x137, 0x13B, 0x13C, 0x141, 0x146, 0x14C, 0x14D,
1884 0x156, 0x157, 0x15A, 0x15B, 0x160, 0x161, 0x16A, 0x16B,
1885 0x172, 0x173, 0x179, 0x17E, 0x2019, 0x2019, 0x201C, 0x201E,
1886 sal_uInt32(-1) };
1887 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-13.TXT> version
1888 // 1.0 of 1999 July 27
1889
1890 static const sal_uInt32 aISO885914Ranges[]
1891 = { 0, 0xA0, 0xA3, 0xA3, 0xA7, 0xA7, 0xA9, 0xA9, 0xAD, 0xAE,
1892 0xB6, 0xB6, 0xC0, 0xCF, 0xD1, 0xD6, 0xD8, 0xDD, 0xDF, 0xEF,
1893 0xF1, 0xF6, 0xF8, 0xFD, 0xFF, 0xFF, 0x10A, 0x10B, 0x120, 0x121,
1894 0x174, 0x178, 0x1E02, 0x1E03, 0x1E0A, 0x1E0B, 0x1E1E, 0x1E1F,
1895 0x1E40, 0x1E41, 0x1E56, 0x1E57, 0x1E60, 0x1E61, 0x1E6A, 0x1E6B,
1896 0x1E80, 0x1E85, 0x1EF2, 0x1EF3, sal_uInt32(-1) };
1897 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-14.TXT> version
1898 // 1.0 of 1999 July 27
1899
1900 static const sal_uInt32 aISO885915Ranges[]
1901 = { 0, 0xA3, 0xA5, 0xA5, 0xA7, 0xA7, 0xA9, 0xB3, 0xB5, 0xB7,
1902 0xB9, 0xBB, 0xBF, 0xFF, 0x152, 0x153, 0x160, 0x161, 0x178, 0x178,
1903 0x17D, 0x17E, 0x20AC, 0x20AC, sal_uInt32(-1) };
1904 // <ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-15.TXT> version
1905 // 1.0 of 1999 July 27
1906
1907 static const sal_uInt32 aKOI8RRanges[]
1908 = { 0, 0x7F, 0xA0, 0xA0, 0xA9, 0xA9, 0xB0, 0xB0, 0xB2, 0xB2,
1909 0xB7, 0xB7, 0xF7, 0xF7, 0x401, 0x401, 0x410, 0x44F, 0x451, 0x451,
1910 0x2219, 0x221A, 0x2248, 0x2248, 0x2264, 0x2265, 0x2320, 0x2321,
1911 0x2500, 0x2500, 0x2502, 0x2502, 0x250C, 0x250C, 0x2510, 0x2510,
1912 0x2514, 0x2514, 0x2518, 0x2518, 0x251C, 0x251C, 0x2524, 0x2524,
1913 0x252C, 0x252C, 0x2534, 0x2534, 0x253C, 0x253C, 0x2550, 0x256C,
1914 0x2580, 0x2580, 0x2584, 0x2584, 0x2588, 0x2588, 0x258C, 0x258C,
1915 0x2590, 0x2593, 0x25A0, 0x25A0, sal_uInt32(-1) };
1916 // <ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MISC/KOI8-R.TXT>
1917 // version 1.0 of 18 August 1999
1918
1919 #if defined WNT
1920 static const sal_uInt32 aWindows1252Ranges[]
1921 = { 0, 0x7F, 0xA0, 0xFF, 0x152, 0x153, 0x160, 0x161, 0x178, 0x178,
1922 0x17D, 0x17E, 0x192, 0x192, 0x2C6, 0x2C6, 0x2DC, 0x2DC,
1923 0x2013, 0x2014, 0x2018, 0x201A, 0x201C, 0x201E, 0x2020, 0x2022,
1924 0x2026, 0x2026, 0x2030, 0x2030, 0x2039, 0x203A, 0x20AC, 0x20AC,
1925 0x2122, 0x2122, sal_uInt32(-1) };
1926 // <ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/
1927 // CP1252.TXT> version 2.01 of 04/15/98
1928 #endif // WNT
1929
1930 INetMIMECharsetList_Impl * pList = new INetMIMECharsetList_Impl;
1931 switch (eEncoding)
1932 {
1933 case RTL_TEXTENCODING_MS_1252:
1934 #if defined WNT
1935 pList->prepend(Charset(RTL_TEXTENCODING_MS_1252,
1936 aWindows1252Ranges));
1937 #endif // WNT
1938 case RTL_TEXTENCODING_ISO_8859_1:
1939 case RTL_TEXTENCODING_UTF7:
1940 case RTL_TEXTENCODING_UTF8:
1941 break;
1942
1943 case RTL_TEXTENCODING_ISO_8859_2:
1944 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_2,
1945 aISO88592Ranges));
1946 break;
1947
1948 case RTL_TEXTENCODING_ISO_8859_3:
1949 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_3,
1950 aISO88593Ranges));
1951 break;
1952
1953 case RTL_TEXTENCODING_ISO_8859_4:
1954 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_4,
1955 aISO88594Ranges));
1956 break;
1957
1958 case RTL_TEXTENCODING_ISO_8859_5:
1959 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_5,
1960 aISO88595Ranges));
1961 break;
1962
1963 case RTL_TEXTENCODING_ISO_8859_6:
1964 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_6,
1965 aISO88596Ranges));
1966 break;
1967
1968 case RTL_TEXTENCODING_ISO_8859_7:
1969 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_7,
1970 aISO88597Ranges));
1971 break;
1972
1973 case RTL_TEXTENCODING_ISO_8859_8:
1974 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_8,
1975 aISO88598Ranges));
1976 break;
1977
1978 case RTL_TEXTENCODING_ISO_8859_9:
1979 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_9,
1980 aISO88599Ranges));
1981 break;
1982
1983 case RTL_TEXTENCODING_ISO_8859_10:
1984 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_10,
1985 aISO885910Ranges));
1986 break;
1987
1988 case RTL_TEXTENCODING_ISO_8859_13:
1989 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_13,
1990 aISO885913Ranges));
1991 break;
1992
1993 case RTL_TEXTENCODING_ISO_8859_14:
1994 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_14,
1995 aISO885914Ranges));
1996 break;
1997
1998 case RTL_TEXTENCODING_ISO_8859_15:
1999 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_15,
2000 aISO885915Ranges));
2001 break;
2002
2003 case RTL_TEXTENCODING_MS_1250:
2004 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_2,
2005 aISO88592Ranges));
2006 break;
2007
2008 case RTL_TEXTENCODING_MS_1251:
2009 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_5,
2010 aISO88595Ranges));
2011 break;
2012
2013 case RTL_TEXTENCODING_MS_1253:
2014 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_7,
2015 aISO88597Ranges));
2016 break;
2017
2018 case RTL_TEXTENCODING_MS_1254:
2019 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_9,
2020 aISO88599Ranges));
2021 break;
2022
2023 case RTL_TEXTENCODING_MS_1255:
2024 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_8,
2025 aISO88598Ranges));
2026 break;
2027
2028 case RTL_TEXTENCODING_MS_1256:
2029 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_6,
2030 aISO88596Ranges));
2031 break;
2032
2033 case RTL_TEXTENCODING_MS_1257:
2034 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_4,
2035 aISO88594Ranges));
2036 break;
2037
2038 case RTL_TEXTENCODING_KOI8_R:
2039 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_5,
2040 aISO88595Ranges));
2041 pList->prepend(Charset(RTL_TEXTENCODING_KOI8_R, aKOI8RRanges));
2042 break;
2043
2044 default: //@@@ more cases are missing!
2045 DBG_ERROR("INetMIME::createPreferredCharsetList():"
2046 " Unsupported encoding");
2047 break;
2048 }
2049 pList->prepend(Charset(RTL_TEXTENCODING_ISO_8859_1, aISO88591Ranges));
2050 pList->prepend(Charset(RTL_TEXTENCODING_ASCII_US, aUSASCIIRanges));
2051 return pList;
2052 }
2053
2054 //============================================================================
2055 // static
2056 sal_Unicode * INetMIME::convertToUnicode(const sal_Char * pBegin,
2057 const sal_Char * pEnd,
2058 rtl_TextEncoding eEncoding,
2059 sal_Size & rSize)
2060 {
2061 if (eEncoding == RTL_TEXTENCODING_DONTKNOW)
2062 return 0;
2063 rtl_TextToUnicodeConverter hConverter
2064 = rtl_createTextToUnicodeConverter(eEncoding);
2065 rtl_TextToUnicodeContext hContext
2066 = rtl_createTextToUnicodeContext(hConverter);
2067 sal_Unicode * pBuffer;
2068 sal_uInt32 nInfo;
2069 for (sal_Size nBufferSize = pEnd - pBegin;;
2070 nBufferSize += nBufferSize / 3 + 1)
2071 {
2072 pBuffer = new sal_Unicode[nBufferSize];
2073 sal_Size nSrcCvtBytes;
2074 rSize = rtl_convertTextToUnicode(
2075 hConverter, hContext, pBegin, pEnd - pBegin, pBuffer,
2076 nBufferSize,
2077 RTL_TEXTTOUNICODE_FLAGS_UNDEFINED_ERROR
2078 | RTL_TEXTTOUNICODE_FLAGS_MBUNDEFINED_ERROR
2079 | RTL_TEXTTOUNICODE_FLAGS_INVALID_ERROR,
2080 &nInfo, &nSrcCvtBytes);
2081 if (nInfo != RTL_TEXTTOUNICODE_INFO_DESTBUFFERTOSMALL)
2082 break;
2083 delete[] pBuffer;
2084 rtl_resetTextToUnicodeContext(hConverter, hContext);
2085 }
2086 rtl_destroyTextToUnicodeContext(hConverter, hContext);
2087 rtl_destroyTextToUnicodeConverter(hConverter);
2088 if (nInfo != 0)
2089 {
2090 delete[] pBuffer;
2091 pBuffer = 0;
2092 }
2093 return pBuffer;
2094 }
2095
2096 //============================================================================
2097 // static
2098 sal_Char * INetMIME::convertFromUnicode(const sal_Unicode * pBegin,
2099 const sal_Unicode * pEnd,
2100 rtl_TextEncoding eEncoding,
2101 sal_Size & rSize)
2102 {
2103 if (eEncoding == RTL_TEXTENCODING_DONTKNOW)
2104 return 0;
2105 rtl_UnicodeToTextConverter hConverter
2106 = rtl_createUnicodeToTextConverter(eEncoding);
2107 rtl_UnicodeToTextContext hContext
2108 = rtl_createUnicodeToTextContext(hConverter);
2109 sal_Char * pBuffer;
2110 sal_uInt32 nInfo;
2111 for (sal_Size nBufferSize = pEnd - pBegin;;
2112 nBufferSize += nBufferSize / 3 + 1)
2113 {
2114 pBuffer = new sal_Char[nBufferSize];
2115 sal_Size nSrcCvtBytes;
2116 rSize = rtl_convertUnicodeToText(
2117 hConverter, hContext, pBegin, pEnd - pBegin, pBuffer,
2118 nBufferSize,
2119 RTL_UNICODETOTEXT_FLAGS_UNDEFINED_ERROR
2120 | RTL_UNICODETOTEXT_FLAGS_INVALID_ERROR
2121 | RTL_UNICODETOTEXT_FLAGS_UNDEFINED_REPLACE
2122 | RTL_UNICODETOTEXT_FLAGS_UNDEFINED_REPLACESTR,
2123 &nInfo, &nSrcCvtBytes);
2124 if (nInfo != RTL_UNICODETOTEXT_INFO_DESTBUFFERTOSMALL)
2125 break;
2126 delete[] pBuffer;
2127 rtl_resetUnicodeToTextContext(hConverter, hContext);
2128 }
2129 rtl_destroyUnicodeToTextContext(hConverter, hContext);
2130 rtl_destroyUnicodeToTextConverter(hConverter);
2131 if (nInfo != 0)
2132 {
2133 delete[] pBuffer;
2134 pBuffer = 0;
2135 }
2136 return pBuffer;
2137 }
2138
2139 //============================================================================
2140 // static
2141 void INetMIME::writeUTF8(INetMIMEOutputSink & rSink, sal_uInt32 nChar)
2142 {
2143 // See RFC 2279 for a discussion of UTF-8.
2144 DBG_ASSERT(nChar < 0x80000000, "INetMIME::writeUTF8(): Bad char");
2145
2146 if (nChar < 0x80)
2147 rSink << sal_Char(nChar);
2148 else if (nChar < 0x800)
2149 rSink << sal_Char(nChar >> 6 | 0xC0)
2150 << sal_Char((nChar & 0x3F) | 0x80);
2151 else if (nChar < 0x10000)
2152 rSink << sal_Char(nChar >> 12 | 0xE0)
2153 << sal_Char((nChar >> 6 & 0x3F) | 0x80)
2154 << sal_Char((nChar & 0x3F) | 0x80);
2155 else if (nChar < 0x200000)
2156 rSink << sal_Char(nChar >> 18 | 0xF0)
2157 << sal_Char((nChar >> 12 & 0x3F) | 0x80)
2158 << sal_Char((nChar >> 6 & 0x3F) | 0x80)
2159 << sal_Char((nChar & 0x3F) | 0x80);
2160 else if (nChar < 0x4000000)
2161 rSink << sal_Char(nChar >> 24 | 0xF8)
2162 << sal_Char((nChar >> 18 & 0x3F) | 0x80)
2163 << sal_Char((nChar >> 12 & 0x3F) | 0x80)
2164 << sal_Char((nChar >> 6 & 0x3F) | 0x80)
2165 << sal_Char((nChar & 0x3F) | 0x80);
2166 else
2167 rSink << sal_Char(nChar >> 30 | 0xFC)
2168 << sal_Char((nChar >> 24 & 0x3F) | 0x80)
2169 << sal_Char((nChar >> 18 & 0x3F) | 0x80)
2170 << sal_Char((nChar >> 12 & 0x3F) | 0x80)
2171 << sal_Char((nChar >> 6 & 0x3F) | 0x80)
2172 << sal_Char((nChar & 0x3F) | 0x80);
2173 }
2174
2175 //============================================================================
2176 // static
2177 void INetMIME::writeUnsigned(INetMIMEOutputSink & rSink, sal_uInt32 nValue,
2178 int nMinDigits)
2179 {
2180 sal_Char aBuffer[10];
2181 // max unsigned 32 bit value (4294967295) has 10 places
2182 sal_Char * p = aBuffer;
2183 for (; nValue > 0; nValue /= 10)
2184 *p++ = sal_Char(getDigit(nValue % 10));
2185 nMinDigits -= p - aBuffer;
2186 while (nMinDigits-- > 0)
2187 rSink << '0';
2188 while (p != aBuffer)
2189 rSink << *--p;
2190 }
2191
2192 //============================================================================
2193 // static
2194 void INetMIME::writeDateTime(INetMIMEOutputSink & rSink,
2195 const DateTime & rUTC)
2196 {
2197 static const sal_Char aDay[7][3]
2198 = { { 'M', 'o', 'n' },
2199 { 'T', 'u', 'e' },
2200 { 'W', 'e', 'd' },
2201 { 'T', 'h', 'u' },
2202 { 'F', 'r', 'i' },
2203 { 'S', 'a', 't' },
2204 { 'S', 'u', 'n' } };
2205 const sal_Char * pTheDay = aDay[rUTC.GetDayOfWeek()];
2206 rSink.write(pTheDay, pTheDay + 3);
2207 rSink << ", ";
2208 writeUnsigned(rSink, rUTC.GetDay());
2209 rSink << ' ';
2210 static const sal_Char aMonth[12][3]
2211 = { { 'J', 'a', 'n' },
2212 { 'F', 'e', 'b' },
2213 { 'M', 'a', 'r' },
2214 { 'A', 'p', 'r' },
2215 { 'M', 'a', 'y' },
2216 { 'J', 'u', 'n' },
2217 { 'J', 'u', 'l' },
2218 { 'A', 'u', 'g' },
2219 { 'S', 'e', 'p' },
2220 { 'O', 'c', 't' },
2221 { 'N', 'o', 'v' },
2222 { 'D', 'e', 'c' } };
2223 const sal_Char * pTheMonth = aMonth[rUTC.GetMonth() - 1];
2224 rSink.write(pTheMonth, pTheMonth + 3);
2225 rSink << ' ';
2226 writeUnsigned(rSink, rUTC.GetYear());
2227 rSink << ' ';
2228 writeUnsigned(rSink, rUTC.GetHour(), 2);
2229 rSink << ':';
2230 writeUnsigned(rSink, rUTC.GetMin(), 2);
2231 rSink << ':';
2232 writeUnsigned(rSink, rUTC.GetSec(), 2);
2233 rSink << " +0000";
2234 }
2235
2236 //============================================================================
2237 // static
2238 void INetMIME::writeHeaderFieldBody(INetMIMEOutputSink & rSink,
2239 HeaderFieldType eType,
2240 const ByteString & rBody,
2241 rtl_TextEncoding ePreferredEncoding,
2242 bool bInitialSpace)
2243 {
2244 writeHeaderFieldBody(rSink, eType,
2245 UniString(rBody, RTL_TEXTENCODING_UTF8),
2246 ePreferredEncoding, bInitialSpace);
2247 }
2248
2249 //============================================================================
2250 // static
2251 void INetMIME::writeHeaderFieldBody(INetMIMEOutputSink & rSink,
2252 HeaderFieldType eType,
2253 const UniString & rBody,
2254 rtl_TextEncoding ePreferredEncoding,
2255 bool bInitialSpace)
2256 {
2257 if (eType == HEADER_FIELD_TEXT)
2258 {
2259 INetMIMEEncodedWordOutputSink
2260 aOutput(rSink, INetMIMEEncodedWordOutputSink::CONTEXT_TEXT,
2261 bInitialSpace ?
2262 INetMIMEEncodedWordOutputSink::SPACE_ALWAYS :
2263 INetMIMEEncodedWordOutputSink::SPACE_NO,
2264 ePreferredEncoding);
2265 aOutput.write(rBody.GetBuffer(), rBody.GetBuffer() + rBody.Len());
2266 aOutput.flush();
2267 }
2268 else
2269 {
2270 enum Brackets { BRACKETS_OUTSIDE, BRACKETS_OPENING, BRACKETS_INSIDE };
2271 Brackets eBrackets = BRACKETS_OUTSIDE;
2272
2273 const sal_Unicode * pBodyPtr = rBody.GetBuffer();
2274 const sal_Unicode * pBodyEnd = pBodyPtr + rBody.Len();
2275 while (pBodyPtr != pBodyEnd)
2276 switch (*pBodyPtr)
2277 {
2278 case '\t':
2279 case ' ':
2280 // A WSP adds to accumulated space:
2281 bInitialSpace = true;
2282 ++pBodyPtr;
2283 break;
2284
2285 case '(':
2286 {
2287 // Write a pending '<' if necessary:
2288 if (eBrackets == BRACKETS_OPENING)
2289 {
2290 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2291 >= rSink.getLineLengthLimit())
2292 rSink << INetMIMEOutputSink::endl << ' ';
2293 else if (bInitialSpace)
2294 rSink << ' ';
2295 rSink << '<';
2296 bInitialSpace = false;
2297 eBrackets = BRACKETS_INSIDE;
2298 }
2299
2300 // Write the comment, introducing encoded-words where
2301 // necessary:
2302 int nLevel = 0;
2303 INetMIMEEncodedWordOutputSink
2304 aOutput(
2305 rSink,
2306 INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT,
2307 INetMIMEEncodedWordOutputSink::SPACE_NO,
2308 ePreferredEncoding);
2309 while (pBodyPtr != pBodyEnd)
2310 switch (*pBodyPtr)
2311 {
2312 case '(':
2313 aOutput.flush();
2314 if (rSink.getColumn()
2315 + (bInitialSpace ? 1 : 0)
2316 >= rSink.getLineLengthLimit())
2317 rSink << INetMIMEOutputSink::endl << ' ';
2318 else if (bInitialSpace)
2319 rSink << ' ';
2320 rSink << '(';
2321 bInitialSpace = false;
2322 ++nLevel;
2323 ++pBodyPtr;
2324 break;
2325
2326 case ')':
2327 aOutput.flush();
2328 if (rSink.getColumn()
2329 >= rSink.getLineLengthLimit())
2330 rSink << INetMIMEOutputSink::endl << ' ';
2331 rSink << ')';
2332 ++pBodyPtr;
2333 if (--nLevel == 0)
2334 goto comment_done;
2335 break;
2336
2337 case '\\':
2338 if (++pBodyPtr == pBodyEnd)
2339 break;
2340 default:
2341 aOutput << *pBodyPtr++;
2342 break;
2343 }
2344 comment_done:
2345 break;
2346 }
2347
2348 case '<':
2349 // Write an already pending '<' if necessary:
2350 if (eBrackets == BRACKETS_OPENING)
2351 {
2352 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2353 >= rSink.getLineLengthLimit())
2354 rSink << INetMIMEOutputSink::endl << ' ';
2355 else if (bInitialSpace)
2356 rSink << ' ';
2357 rSink << '<';
2358 bInitialSpace = false;
2359 }
2360
2361 // Remember this '<' as pending, and open a bracketed
2362 // block:
2363 eBrackets = BRACKETS_OPENING;
2364 ++pBodyPtr;
2365 break;
2366
2367 case '>':
2368 // Write a pending '<' if necessary:
2369 if (eBrackets == BRACKETS_OPENING)
2370 {
2371 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2372 >= rSink.getLineLengthLimit())
2373 rSink << INetMIMEOutputSink::endl << ' ';
2374 else if (bInitialSpace)
2375 rSink << ' ';
2376 rSink << '<';
2377 bInitialSpace = false;
2378 }
2379
2380 // Write this '>', and close any bracketed block:
2381 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2382 >= rSink.getLineLengthLimit())
2383 rSink << INetMIMEOutputSink::endl << ' ';
2384 else if (bInitialSpace)
2385 rSink << ' ';
2386 rSink << '>';
2387 bInitialSpace = false;
2388 eBrackets = BRACKETS_OUTSIDE;
2389 ++pBodyPtr;
2390 break;
2391
2392 case ',':
2393 case ':':
2394 case ';':
2395 case '\\':
2396 case ']':
2397 // Write a pending '<' if necessary:
2398 if (eBrackets == BRACKETS_OPENING)
2399 {
2400 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2401 >= rSink.getLineLengthLimit())
2402 rSink << INetMIMEOutputSink::endl << ' ';
2403 else if (bInitialSpace)
2404 rSink << ' ';
2405 rSink << '<';
2406 bInitialSpace = false;
2407 eBrackets = BRACKETS_INSIDE;
2408 }
2409
2410 // Write this specials:
2411 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2412 >= rSink.getLineLengthLimit())
2413 rSink << INetMIMEOutputSink::endl << ' ';
2414 else if (bInitialSpace)
2415 rSink << ' ';
2416 rSink << sal_Char(*pBodyPtr++);
2417 bInitialSpace = false;
2418 break;
2419
2420 case '\x0D': // CR
2421 // A <CRLF WSP> adds to accumulated space, a <CR> not
2422 // followed by <LF WSP> starts 'junk':
2423 if (startsWithLineFolding(pBodyPtr, pBodyEnd))
2424 {
2425 bInitialSpace = true;
2426 pBodyPtr += 3;
2427 break;
2428 }
2429 default:
2430 {
2431 // The next token is either one of <"." / "@" / atom /
2432 // quoted-string / domain-literal>, or it's 'junk'; if it
2433 // is not 'junk', it is either a 'phrase' (i.e., it may
2434 // contain encoded-words) or a 'non-phrase' (i.e., it may
2435 // not contain encoded-words):
2436 enum Entity { ENTITY_JUNK, ENTITY_NON_PHRASE,
2437 ENTITY_PHRASE };
2438 Entity eEntity = ENTITY_JUNK;
2439 switch (*pBodyPtr)
2440 {
2441 case '.':
2442 case '@':
2443 case '[':
2444 // A token of <"." / "@" / domain-literal> always
2445 // starts a 'non-phrase':
2446 eEntity = ENTITY_NON_PHRASE;
2447 break;
2448
2449 default:
2450 if (isUSASCII(*pBodyPtr)
2451 && !isAtomChar(*pBodyPtr))
2452 {
2453 eEntity = ENTITY_JUNK;
2454 break;
2455 }
2456 case '"':
2457 // A token of <atom / quoted-string> can either be
2458 // a 'phrase' or a 'non-phrase':
2459 switch (eType)
2460 {
2461 case HEADER_FIELD_STRUCTURED:
2462 eEntity = ENTITY_NON_PHRASE;
2463 break;
2464
2465 case HEADER_FIELD_PHRASE:
2466 eEntity = ENTITY_PHRASE;
2467 break;
2468
2469 case HEADER_FIELD_MESSAGE_ID:
2470 // A 'phrase' if and only if outside any
2471 // bracketed block:
2472 eEntity
2473 = eBrackets == BRACKETS_OUTSIDE ?
2474 ENTITY_PHRASE :
2475 ENTITY_NON_PHRASE;
2476 break;
2477
2478 case HEADER_FIELD_ADDRESS:
2479 {
2480 // A 'non-phrase' if and only if, after
2481 // skipping this token and any following
2482 // <linear-white-space> and <comment>s,
2483 // there is no token left, or the next
2484 // token is any of <"." / "@" / ">" / ","
2485 // / ";">, or the next token is <":"> and
2486 // is within a bracketed block:
2487 const sal_Unicode * pLookAhead = pBodyPtr;
2488 if (*pLookAhead == '"')
2489 {
2490 pLookAhead
2491 = skipQuotedString(pLookAhead,
2492 pBodyEnd);
2493 if (pLookAhead == pBodyPtr)
2494 pLookAhead = pBodyEnd;
2495 }
2496 else
2497 while (pLookAhead != pBodyEnd
2498 && (isAtomChar(*pLookAhead)
2499 || !isUSASCII(
2500 *pLookAhead)))
2501 ++pLookAhead;
2502 while (pLookAhead != pBodyEnd)
2503 switch (*pLookAhead)
2504 {
2505 case '\t':
2506 case ' ':
2507 ++pLookAhead;
2508 break;
2509
2510 case '(':
2511 {
2512 const sal_Unicode * pPast
2513 = skipComment(pLookAhead,
2514 pBodyEnd);
2515 pLookAhead
2516 = pPast == pLookAhead ?
2517 pBodyEnd : pPast;
2518 break;
2519 }
2520
2521 case ',':
2522 case '.':
2523 case ';':
2524 case '>':
2525 case '@':
2526 eEntity = ENTITY_NON_PHRASE;
2527 goto entity_determined;
2528
2529 case ':':
2530 eEntity
2531 = eBrackets
2532 == BRACKETS_OUTSIDE ?
2533 ENTITY_PHRASE :
2534 ENTITY_NON_PHRASE;
2535 goto entity_determined;
2536
2537 case '\x0D': // CR
2538 if (startsWithLineFolding(
2539 pLookAhead, pBodyEnd))
2540 {
2541 pLookAhead += 3;
2542 break;
2543 }
2544 default:
2545 eEntity = ENTITY_PHRASE;
2546 goto entity_determined;
2547 }
2548 eEntity = ENTITY_NON_PHRASE;
2549 entity_determined:
2550 break;
2551 }
2552
2553 case HEADER_FIELD_TEXT:
2554 OSL_ASSERT(false);
2555 break;
2556 }
2557
2558 // In a 'non-phrase', a non-US-ASCII character
2559 // cannot be part of an <atom>, but instead the
2560 // whole entity is 'junk' rather than 'non-
2561 // phrase':
2562 if (eEntity == ENTITY_NON_PHRASE
2563 && !isUSASCII(*pBodyPtr))
2564 eEntity = ENTITY_JUNK;
2565 break;
2566 }
2567
2568 switch (eEntity)
2569 {
2570 case ENTITY_JUNK:
2571 {
2572 // Write a pending '<' if necessary:
2573 if (eBrackets == BRACKETS_OPENING)
2574 {
2575 if (rSink.getColumn()
2576 + (bInitialSpace ? 1 : 0)
2577 >= rSink.getLineLengthLimit())
2578 rSink << INetMIMEOutputSink::endl << ' ';
2579 else if (bInitialSpace)
2580 rSink << ' ';
2581 rSink << '<';
2582 bInitialSpace = false;
2583 eBrackets = BRACKETS_INSIDE;
2584 }
2585
2586 // Calculate the length of in- and output:
2587 const sal_Unicode * pStart = pBodyPtr;
2588 sal_Size nLength = 0;
2589 bool bModify = false;
2590 bool bEnd = false;
2591 while (pBodyPtr != pBodyEnd && !bEnd)
2592 switch (*pBodyPtr)
2593 {
2594 case '\x0D': // CR
2595 if (startsWithLineFolding(pBodyPtr,
2596 pBodyEnd))
2597 bEnd = true;
2598 else if (startsWithLineBreak(
2599 pBodyPtr, pBodyEnd))
2600 {
2601 nLength += 3;
2602 bModify = true;
2603 pBodyPtr += 2;
2604 }
2605 else
2606 {
2607 ++nLength;
2608 ++pBodyPtr;
2609 }
2610 break;
2611
2612 case '\t':
2613 case ' ':
2614 bEnd = true;
2615 break;
2616
2617 default:
2618 if (isVisible(*pBodyPtr))
2619 bEnd = true;
2620 else if (isUSASCII(*pBodyPtr))
2621 {
2622 ++nLength;
2623 ++pBodyPtr;
2624 }
2625 else
2626 {
2627 nLength += getUTF8OctetCount(
2628 *pBodyPtr++);
2629 bModify = true;
2630 }
2631 break;
2632 }
2633
2634 // Write the output:
2635 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2636 + nLength
2637 > rSink.getLineLengthLimit())
2638 rSink << INetMIMEOutputSink::endl << ' ';
2639 else if (bInitialSpace)
2640 rSink << ' ';
2641 bInitialSpace = false;
2642 if (bModify)
2643 while (pStart != pBodyPtr)
2644 if (startsWithLineBreak(pStart, pBodyPtr))
2645 {
2646 rSink << "\x0D\\\x0A"; // CR, '\', LF
2647 pStart += 2;
2648 }
2649 else
2650 writeUTF8(rSink, *pStart++);
2651 else
2652 rSink.write(pStart, pBodyPtr);
2653 break;
2654 }
2655
2656 case ENTITY_NON_PHRASE:
2657 {
2658 // Calculate the length of in- and output:
2659 const sal_Unicode * pStart = pBodyPtr;
2660 sal_Size nLength = 0;
2661 bool bBracketedBlock = false;
2662 bool bSymbol = *pStart != '.' && *pStart != '@';
2663 bool bModify = false;
2664 bool bEnd = false;
2665 while (pBodyPtr != pBodyEnd && !bEnd)
2666 switch (*pBodyPtr)
2667 {
2668 case '\t':
2669 case ' ':
2670 case '\x0D': // CR
2671 {
2672 const sal_Unicode * pLookAhead
2673 = skipLinearWhiteSpace(pBodyPtr,
2674 pBodyEnd);
2675 if (pLookAhead < pBodyEnd
2676 && (bSymbol ?
2677 isAtomChar(*pLookAhead)
2678 || *pLookAhead == '"'
2679 || *pLookAhead == '[' :
2680 *pLookAhead == '.'
2681 || *pLookAhead == '@'
2682 || (*pLookAhead == '>'
2683 && eType
2684 >= HEADER_FIELD_MESSAGE_ID
2685 && eBrackets
2686 == BRACKETS_OPENING)))
2687 {
2688 bModify = true;
2689 pBodyPtr = pLookAhead;
2690 }
2691 else
2692 bEnd = true;
2693 break;
2694 }
2695
2696 case '"':
2697 if (bSymbol)
2698 {
2699 pBodyPtr
2700 = scanQuotedBlock(pBodyPtr,
2701 pBodyEnd,
2702 '"', '"',
2703 nLength,
2704 bModify);
2705 bSymbol = false;
2706 }
2707 else
2708 bEnd = true;
2709 break;
2710
2711 case '[':
2712 if (bSymbol)
2713 {
2714 pBodyPtr
2715 = scanQuotedBlock(pBodyPtr,
2716 pBodyEnd,
2717 '[', ']',
2718 nLength,
2719 bModify);
2720 bSymbol = false;
2721 }
2722 else
2723 bEnd = true;
2724 break;
2725
2726 case '.':
2727 case '@':
2728 if (bSymbol)
2729 bEnd = true;
2730 else
2731 {
2732 ++nLength;
2733 bSymbol = true;
2734 ++pBodyPtr;
2735 }
2736 break;
2737
2738 case '>':
2739 if (eBrackets == BRACKETS_OPENING
2740 && eType
2741 >= HEADER_FIELD_MESSAGE_ID)
2742 {
2743 ++nLength;
2744 bBracketedBlock = true;
2745 ++pBodyPtr;
2746 }
2747 bEnd = true;
2748 break;
2749
2750 default:
2751 if (isAtomChar(*pBodyPtr) && bSymbol)
2752 {
2753 while (pBodyPtr != pBodyEnd
2754 && isAtomChar(*pBodyPtr))
2755 {
2756 ++nLength;
2757 ++pBodyPtr;
2758 }
2759 bSymbol = false;
2760 }
2761 else
2762 {
2763 if (!isUSASCII(*pBodyPtr))
2764 bModify = true;
2765 bEnd = true;
2766 }
2767 break;
2768 }
2769
2770 // Write a pending '<' if necessary:
2771 if (eBrackets == BRACKETS_OPENING
2772 && !bBracketedBlock)
2773 {
2774 if (rSink.getColumn()
2775 + (bInitialSpace ? 1 : 0)
2776 >= rSink.getLineLengthLimit())
2777 rSink << INetMIMEOutputSink::endl << ' ';
2778 else if (bInitialSpace)
2779 rSink << ' ';
2780 rSink << '<';
2781 bInitialSpace = false;
2782 eBrackets = BRACKETS_INSIDE;
2783 }
2784
2785 // Write the output:
2786 if (rSink.getColumn() + (bInitialSpace ? 1 : 0)
2787 + nLength
2788 > rSink.getLineLengthLimit())
2789 rSink << INetMIMEOutputSink::endl << ' ';
2790 else if (bInitialSpace)
2791 rSink << ' ';
2792 bInitialSpace = false;
2793 if (bBracketedBlock)
2794 {
2795 rSink << '<';
2796 eBrackets = BRACKETS_OUTSIDE;
2797 }
2798 if (bModify)
2799 {
2800 enum Mode { MODE_PLAIN, MODE_QUOTED_STRING,
2801 MODE_DOMAIN_LITERAL };
2802 Mode eMode = MODE_PLAIN;
2803 while (pStart != pBodyPtr)
2804 switch (*pStart)
2805 {
2806 case '\x0D': // CR
2807 if (startsWithLineFolding(
2808 pStart, pBodyPtr))
2809 {
2810 if (eMode != MODE_PLAIN)
2811 rSink << sal_Char(
2812 pStart[2]);
2813 pStart += 3;
2814 }
2815 else if (startsWithLineBreak(
2816 pStart, pBodyPtr))
2817 {
2818 rSink << "\x0D\\\x0A";
2819 // CR, '\', LF
2820 pStart += 2;
2821 }
2822 else
2823 {
2824 rSink << '\x0D'; // CR
2825 ++pStart;
2826 }
2827 break;
2828
2829 case '\t':
2830 case ' ':
2831 if (eMode != MODE_PLAIN)
2832 rSink << sal_Char(*pStart);
2833 ++pStart;
2834 break;
2835
2836 case '"':
2837 if (eMode == MODE_PLAIN)
2838 eMode = MODE_QUOTED_STRING;
2839 else if (eMode
2840 == MODE_QUOTED_STRING)
2841 eMode = MODE_PLAIN;
2842 rSink << '"';
2843 ++pStart;
2844 break;
2845
2846 case '[':
2847 if (eMode == MODE_PLAIN)
2848 eMode = MODE_DOMAIN_LITERAL;
2849 rSink << '[';
2850 ++pStart;
2851 break;
2852
2853 case ']':
2854 if (eMode == MODE_DOMAIN_LITERAL)
2855 eMode = MODE_PLAIN;
2856 rSink << ']';
2857 ++pStart;
2858 break;
2859
2860 case '\\':
2861 rSink << '\\';
2862 if (++pStart < pBodyPtr)
2863 writeUTF8(rSink, *pStart++);
2864 break;
2865
2866 default:
2867 writeUTF8(rSink, *pStart++);
2868 break;
2869 }
2870 }
2871 else
2872 rSink.write(pStart, pBodyPtr);
2873 break;
2874 }
2875
2876 case ENTITY_PHRASE:
2877 {
2878 // Write a pending '<' if necessary:
2879 if (eBrackets == BRACKETS_OPENING)
2880 {
2881 if (rSink.getColumn()
2882 + (bInitialSpace ? 1 : 0)
2883 >= rSink.getLineLengthLimit())
2884 rSink << INetMIMEOutputSink::endl << ' ';
2885 else if (bInitialSpace)
2886 rSink << ' ';
2887 rSink << '<';
2888 bInitialSpace = false;
2889 eBrackets = BRACKETS_INSIDE;
2890 }
2891
2892 // Calculate the length of in- and output:
2893 const sal_Unicode * pStart = pBodyPtr;
2894 bool bQuotedString = false;
2895 bool bEnd = false;
2896 while (pBodyPtr != pBodyEnd && !bEnd)
2897 switch (*pBodyPtr)
2898 {
2899 case '\t':
2900 case ' ':
2901 case '\x0D': // CR
2902 if (bQuotedString)
2903 ++pBodyPtr;
2904 else
2905 {
2906 const sal_Unicode * pLookAhead
2907 = skipLinearWhiteSpace(
2908 pBodyPtr, pBodyEnd);
2909 if (pLookAhead != pBodyEnd
2910 && (isAtomChar(*pLookAhead)
2911 || !isUSASCII(*pLookAhead)
2912 || *pLookAhead == '"'))
2913 pBodyPtr = pLookAhead;
2914 else
2915 bEnd = true;
2916 }
2917 break;
2918
2919 case '"':
2920 bQuotedString = !bQuotedString;
2921 ++pBodyPtr;
2922 break;
2923
2924 case '\\':
2925 if (bQuotedString)
2926 {
2927 if (++pBodyPtr != pBodyEnd)
2928 ++pBodyPtr;
2929 }
2930 else
2931 bEnd = true;
2932 break;
2933
2934 default:
2935 if (bQuotedString
2936 || isAtomChar(*pBodyPtr)
2937 || !isUSASCII(*pBodyPtr))
2938 ++pBodyPtr;
2939 else
2940 bEnd = true;
2941 break;
2942 }
2943
2944 // Write the phrase, introducing encoded-words
2945 // where necessary:
2946 INetMIMEEncodedWordOutputSink
2947 aOutput(
2948 rSink,
2949 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE,
2950 bInitialSpace ?
2951 INetMIMEEncodedWordOutputSink::SPACE_ALWAYS :
2952 INetMIMEEncodedWordOutputSink::SPACE_ENCODED,
2953 ePreferredEncoding);
2954 while (pStart != pBodyPtr)
2955 switch (*pStart)
2956 {
2957 case '"':
2958 ++pStart;
2959 break;
2960
2961 case '\\':
2962 if (++pStart != pBodyPtr)
2963 aOutput << *pStart++;
2964 break;
2965
2966 case '\x0D': // CR
2967 pStart += 2;
2968 aOutput << *pStart++;
2969 break;
2970
2971 default:
2972 aOutput << *pStart++;
2973 break;
2974 }
2975 bInitialSpace = aOutput.flush();
2976 break;
2977 }
2978 }
2979 break;
2980 }
2981 }
2982 }
2983 }
2984
2985 //============================================================================
2986 // static
2987 bool INetMIME::translateUTF8Char(const sal_Char *& rBegin,
2988 const sal_Char * pEnd,
2989 rtl_TextEncoding eEncoding,
2990 sal_uInt32 & rCharacter)
2991 {
2992 if (rBegin == pEnd || static_cast< unsigned char >(*rBegin) < 0x80
2993 || static_cast< unsigned char >(*rBegin) >= 0xFE)
2994 return false;
2995
2996 int nCount;
2997 sal_uInt32 nMin;
2998 sal_uInt32 nUCS4;
2999 const sal_Char * p = rBegin;
3000 if (static_cast< unsigned char >(*p) < 0xE0)
3001 {
3002 nCount = 1;
3003 nMin = 0x80;
3004 nUCS4 = static_cast< unsigned char >(*p) & 0x1F;
3005 }
3006 else if (static_cast< unsigned char >(*p) < 0xF0)
3007 {
3008 nCount = 2;
3009 nMin = 0x800;
3010 nUCS4 = static_cast< unsigned char >(*p) & 0xF;
3011 }
3012 else if (static_cast< unsigned char >(*p) < 0xF8)
3013 {
3014 nCount = 3;
3015 nMin = 0x10000;
3016 nUCS4 = static_cast< unsigned char >(*p) & 7;
3017 }
3018 else if (static_cast< unsigned char >(*p) < 0xFC)
3019 {
3020 nCount = 4;
3021 nMin = 0x200000;
3022 nUCS4 = static_cast< unsigned char >(*p) & 3;
3023 }
3024 else
3025 {
3026 nCount = 5;
3027 nMin = 0x4000000;
3028 nUCS4 = static_cast< unsigned char >(*p) & 1;
3029 }
3030 ++p;
3031
3032 for (; nCount-- > 0; ++p)
3033 if ((static_cast< unsigned char >(*p) & 0xC0) == 0x80)
3034 nUCS4 = (nUCS4 << 6) | (static_cast< unsigned char >(*p) & 0x3F);
3035 else
3036 return false;
3037
3038 if (nUCS4 < nMin || nUCS4 > 0x10FFFF)
3039 return false;
3040
3041 if (eEncoding >= RTL_TEXTENCODING_UCS4)
3042 rCharacter = nUCS4;
3043 else
3044 {
3045 sal_Unicode aUTF16[2];
3046 const sal_Unicode * pUTF16End = putUTF32Character(aUTF16, nUCS4);
3047 sal_Size nSize;
3048 sal_Char * pBuffer = convertFromUnicode(aUTF16, pUTF16End, eEncoding,
3049 nSize);
3050 if (!pBuffer)
3051 return false;
3052 DBG_ASSERT(nSize == 1,
3053 "INetMIME::translateUTF8Char(): Bad conversion");
3054 rCharacter = *pBuffer;
3055 delete[] pBuffer;
3056 }
3057 rBegin = p;
3058 return true;
3059 }
3060
3061 //============================================================================
3062 // static
3063 ByteString INetMIME::decodeUTF8(const ByteString & rText,
3064 rtl_TextEncoding eEncoding)
3065 {
3066 const sal_Char * p = rText.GetBuffer();
3067 const sal_Char * pEnd = p + rText.Len();
3068 ByteString sDecoded;
3069 while (p != pEnd)
3070 {
3071 // the value is initialized just to avoid warning: ‘nCharacter’ is used uninitialized in this function
3072 // there seems to be a bug in gcc-4.1 because this value is used only if it is initialized
3073 sal_uInt32 nCharacter=0;
3074 if (translateUTF8Char(p, pEnd, eEncoding, nCharacter))
3075 sDecoded += sal_Char(nCharacter);
3076 else
3077 sDecoded += sal_Char(*p++);
3078 }
3079 return sDecoded;
3080 }
3081
3082 //============================================================================
3083 // static
3084 UniString INetMIME::decodeHeaderFieldBody(HeaderFieldType eType,
3085 const ByteString & rBody)
3086 {
3087 // Due to a bug in INetCoreRFC822MessageStream::ConvertTo7Bit(), old
3088 // versions of StarOffice send mails with header fields where encoded
3089 // words can be preceded by '=', ',', '.', '"', or '(', and followed by
3090 // '=', ',', '.', '"', ')', without any required white space in between.
3091 // And there appear to exist some broken mailers that only encode single
3092 // letters within words, like "Appel
3093 // =?iso-8859-1?Q?=E0?=t=?iso-8859-1?Q?=E9?=moin", so it seems best to
3094 // detect encoded words even when not propperly surrounded by white space.
3095 //
3096 // Non US-ASCII characters in rBody are treated as ISO-8859-1.
3097 //
3098 // encoded-word = "=?"
3099 // 1*(%x21 / %x23-27 / %x2A-2B / %x2D / %30-39 / %x41-5A / %x5E-7E)
3100 // ["*" 1*8ALPHA *("-" 1*8ALPHA)] "?"
3101 // ("B?" *(4base64) (4base64 / 3base64 "=" / 2base64 "==")
3102 // / "Q?" 1*(%x21-3C / %x3E / %x40-7E / "=" 2HEXDIG))
3103 // "?="
3104 //
3105 // base64 = ALPHA / DIGIT / "+" / "/"
3106
3107 const sal_Char * pBegin = rBody.GetBuffer();
3108 const sal_Char * pEnd = pBegin + rBody.Len();
3109
3110 UniString sDecoded;
3111 const sal_Char * pCopyBegin = pBegin;
3112
3113 /* bool bStartEncodedWord = true; */
3114 const sal_Char * pWSPBegin = pBegin;
3115 UniString sEncodedText;
3116 bool bQuotedEncodedText = false;
3117 sal_uInt32 nCommentLevel = 0;
3118
3119 for (const sal_Char * p = pBegin; p != pEnd;)
3120 {
3121 if (p != pEnd && *p == '=' /* && bStartEncodedWord */)
3122 {
3123 const sal_Char * q = p + 1;
3124 bool bEncodedWord = q != pEnd && *q++ == '?';
3125
3126 rtl_TextEncoding eCharsetEncoding = RTL_TEXTENCODING_DONTKNOW;
3127 if (bEncodedWord)
3128 {
3129 const sal_Char * pCharsetBegin = q;
3130 const sal_Char * pLanguageBegin = 0;
3131 int nAlphaCount = 0;
3132 for (bool bDone = false; !bDone;)
3133 if (q == pEnd)
3134 {
3135 bEncodedWord = false;
3136 bDone = true;
3137 }
3138 else
3139 {
3140 sal_Char cChar = *q++;
3141 switch (cChar)
3142 {
3143 case '*':
3144 pLanguageBegin = q - 1;
3145 nAlphaCount = 0;
3146 break;
3147
3148 case '-':
3149 if (pLanguageBegin != 0)
3150 {
3151 if (nAlphaCount == 0)
3152 pLanguageBegin = 0;
3153 else
3154 nAlphaCount = 0;
3155 }
3156 break;
3157
3158 case '?':
3159 if (pCharsetBegin == q - 1)
3160 bEncodedWord = false;
3161 else
3162 {
3163 eCharsetEncoding
3164 = getCharsetEncoding(
3165 pCharsetBegin,
3166 pLanguageBegin == 0
3167 || nAlphaCount == 0 ?
3168 q - 1 : pLanguageBegin);
3169 bEncodedWord = isMIMECharsetEncoding(
3170 eCharsetEncoding);
3171 eCharsetEncoding
3172 = translateFromMIME(eCharsetEncoding);
3173 }
3174 bDone = true;
3175 break;
3176
3177 default:
3178 if (pLanguageBegin != 0
3179 && (!isAlpha(cChar) || ++nAlphaCount > 8))
3180 pLanguageBegin = 0;
3181 break;
3182 }
3183 }
3184 }
3185
3186 bool bEncodingB = false;
3187 if (bEncodedWord)
3188 {
3189 if (q == pEnd)
3190 bEncodedWord = false;
3191 else
3192 {
3193 switch (*q++)
3194 {
3195 case 'B':
3196 case 'b':
3197 bEncodingB = true;
3198 break;
3199
3200 case 'Q':
3201 case 'q':
3202 bEncodingB = false;
3203 break;
3204
3205 default:
3206 bEncodedWord = false;
3207 break;
3208 }
3209 }
3210 }
3211
3212 bEncodedWord = bEncodedWord && q != pEnd && *q++ == '?';
3213
3214 ByteString sText;
3215 if (bEncodedWord)
3216 {
3217 if (bEncodingB)
3218 {
3219 for (bool bDone = false; !bDone;)
3220 {
3221 if (pEnd - q < 4)
3222 {
3223 bEncodedWord = false;
3224 bDone = true;
3225 }
3226 else
3227 {
3228 bool bFinal = false;
3229 int nCount = 3;
3230 sal_uInt32 nValue = 0;
3231 for (int nShift = 18; nShift >= 0; nShift -= 6)
3232 {
3233 int nWeight = getBase64Weight(*q++);
3234 if (nWeight == -2)
3235 {
3236 bEncodedWord = false;
3237 bDone = true;
3238 break;
3239 }
3240 if (nWeight == -1)
3241 {
3242 if (!bFinal)
3243 {
3244 if (nShift >= 12)
3245 {
3246 bEncodedWord = false;
3247 bDone = true;
3248 break;
3249 }
3250 bFinal = true;
3251 nCount = nShift == 6 ? 1 : 2;
3252 }
3253 }
3254 else
3255 nValue |= nWeight << nShift;
3256 }
3257 if (bEncodedWord)
3258 {
3259 for (int nShift = 16; nCount-- > 0;
3260 nShift -= 8)
3261 sText += sal_Char(nValue >> nShift
3262 & 0xFF);
3263 if (*q == '?')
3264 {
3265 ++q;
3266 bDone = true;
3267 }
3268 if (bFinal && !bDone)
3269 {
3270 bEncodedWord = false;
3271 bDone = true;
3272 }
3273 }
3274 }
3275 }
3276 }
3277 else
3278 {
3279 const sal_Char * pEncodedTextBegin = q;
3280 const sal_Char * pEncodedTextCopyBegin = q;
3281 for (bool bDone = false; !bDone;)
3282 if (q == pEnd)
3283 {
3284 bEncodedWord = false;
3285 bDone = true;
3286 }
3287 else
3288 {
3289 sal_uInt32 nChar = *q++;
3290 switch (nChar)
3291 {
3292 case '=':
3293 {
3294 if (pEnd - q < 2)
3295 {
3296 bEncodedWord = false;
3297 bDone = true;
3298 break;
3299 }
3300 int nDigit1 = getHexWeight(q[0]);
3301 int nDigit2 = getHexWeight(q[1]);
3302 if (nDigit1 < 0 || nDigit2 < 0)
3303 {
3304 bEncodedWord = false;
3305 bDone = true;
3306 break;
3307 }
3308 sText += rBody.Copy(
3309 static_cast< xub_StrLen >(
3310 pEncodedTextCopyBegin - pBegin),
3311 static_cast< xub_StrLen >(
3312 q - 1 - pEncodedTextCopyBegin));
3313 sText += sal_Char(nDigit1 << 4 | nDigit2);
3314 q += 2;
3315 pEncodedTextCopyBegin = q;
3316 break;
3317 }
3318
3319 case '?':
3320 if (q - pEncodedTextBegin > 1)
3321 sText += rBody.Copy(
3322 static_cast< xub_StrLen >(
3323 pEncodedTextCopyBegin - pBegin),
3324 static_cast< xub_StrLen >(
3325 q - 1 - pEncodedTextCopyBegin));
3326 else
3327 bEncodedWord = false;
3328 bDone = true;
3329 break;
3330
3331 case '_':
3332 sText += rBody.Copy(
3333 static_cast< xub_StrLen >(
3334 pEncodedTextCopyBegin - pBegin),
3335 static_cast< xub_StrLen >(
3336 q - 1 - pEncodedTextCopyBegin));
3337 sText += ' ';
3338 pEncodedTextCopyBegin = q;
3339 break;
3340
3341 default:
3342 if (!isVisible(nChar))
3343 {
3344 bEncodedWord = false;
3345 bDone = true;
3346 }
3347 break;
3348 }
3349 }
3350 }
3351 }
3352
3353 bEncodedWord = bEncodedWord && q != pEnd && *q++ == '=';
3354
3355 // if (bEncodedWord && q != pEnd)
3356 // switch (*q)
3357 // {
3358 // case '\t':
3359 // case ' ':
3360 // case '"':
3361 // case ')':
3362 // case ',':
3363 // case '.':
3364 // case '=':
3365 // break;
3366 //
3367 // default:
3368 // bEncodedWord = false;
3369 // break;
3370 // }
3371
3372 sal_Unicode * pUnicodeBuffer = 0;
3373 sal_Size nUnicodeSize = 0;
3374 if (bEncodedWord)
3375 {
3376 pUnicodeBuffer
3377 = convertToUnicode(sText.GetBuffer(),
3378 sText.GetBuffer() + sText.Len(),
3379 eCharsetEncoding, nUnicodeSize);
3380 if (pUnicodeBuffer == 0)
3381 bEncodedWord = false;
3382 }
3383
3384 if (bEncodedWord)
3385 {
3386 appendISO88591(sDecoded, pCopyBegin, pWSPBegin);
3387 if (eType == HEADER_FIELD_TEXT)
3388 sDecoded.Append(
3389 pUnicodeBuffer,
3390 static_cast< xub_StrLen >(nUnicodeSize));
3391 else if (nCommentLevel == 0)
3392 {
3393 sEncodedText.Append(
3394 pUnicodeBuffer,
3395 static_cast< xub_StrLen >(nUnicodeSize));
3396 if (!bQuotedEncodedText)
3397 {
3398 const sal_Unicode * pTextPtr = pUnicodeBuffer;
3399 const sal_Unicode * pTextEnd = pTextPtr
3400 + nUnicodeSize;
3401 for (; pTextPtr != pTextEnd; ++pTextPtr)
3402 if (!isEncodedWordTokenChar(*pTextPtr))
3403 {
3404 bQuotedEncodedText = true;
3405 break;
3406 }
3407 }
3408 }
3409 else
3410 {
3411 const sal_Unicode * pTextPtr = pUnicodeBuffer;
3412 const sal_Unicode * pTextEnd = pTextPtr + nUnicodeSize;
3413 for (; pTextPtr != pTextEnd; ++pTextPtr)
3414 {
3415 switch (*pTextPtr)
3416 {
3417 case '(':
3418 case ')':
3419 case '\\':
3420 case '\x0D':
3421 case '=':
3422 sDecoded += '\\';
3423 break;
3424 }
3425 sDecoded += *pTextPtr;
3426 }
3427 }
3428 delete[] pUnicodeBuffer;
3429 p = q;
3430 pCopyBegin = p;
3431
3432 pWSPBegin = p;
3433 while (p != pEnd && isWhiteSpace(*p))
3434 ++p;
3435 /* bStartEncodedWord = p != pWSPBegin; */
3436 continue;
3437 }
3438 }
3439
3440 if (sEncodedText.Len() != 0)
3441 {
3442 if (bQuotedEncodedText)
3443 {
3444 sDecoded += '"';
3445 const sal_Unicode * pTextPtr = sEncodedText.GetBuffer();
3446 const sal_Unicode * pTextEnd = pTextPtr + sEncodedText.Len();
3447 for (;pTextPtr != pTextEnd; ++pTextPtr)
3448 {
3449 switch (*pTextPtr)
3450 {
3451 case '"':
3452 case '\\':
3453 case '\x0D':
3454 sDecoded += '\\';
3455 break;
3456 }
3457 sDecoded += *pTextPtr;
3458 }
3459 sDecoded += '"';
3460 }
3461 else
3462 sDecoded += sEncodedText;
3463 sEncodedText.Erase();
3464 bQuotedEncodedText = false;
3465 }
3466
3467 if (p == pEnd)
3468 break;
3469
3470 switch (*p++)
3471 {
3472 // case '\t':
3473 // case ' ':
3474 // case ',':
3475 // case '.':
3476 // case '=':
3477 // bStartEncodedWord = true;
3478 // break;
3479
3480 case '"':
3481 if (eType != HEADER_FIELD_TEXT && nCommentLevel == 0)
3482 {
3483 const sal_Char * pQuotedStringEnd
3484 = skipQuotedString(p - 1, pEnd);
3485 p = pQuotedStringEnd == p - 1 ? pEnd : pQuotedStringEnd;
3486 }
3487 /* bStartEncodedWord = true; */
3488 break;
3489
3490 case '(':
3491 if (eType != HEADER_FIELD_TEXT)
3492 ++nCommentLevel;
3493 /* bStartEncodedWord = true; */
3494 break;
3495
3496 case ')':
3497 if (nCommentLevel > 0)
3498 --nCommentLevel;
3499 /* bStartEncodedWord = false; */
3500 break;
3501
3502 default:
3503 {
3504 const sal_Char * pUTF8Begin = p - 1;
3505 const sal_Char * pUTF8End = pUTF8Begin;
3506 // the value is initialized just to avoid warning: ‘nCharacter’ is used uninitialized in this function
3507 // there seems to be a bug in gcc-4.1 because this value is used only if it is initialized
3508 sal_uInt32 nCharacter=0;
3509 if (translateUTF8Char(pUTF8End, pEnd, RTL_TEXTENCODING_UCS4,
3510 nCharacter))
3511 {
3512 appendISO88591(sDecoded, pCopyBegin, p - 1);
3513 sal_Unicode aUTF16Buf[2];
3514 xub_StrLen nUTF16Len = static_cast< xub_StrLen >(
3515 putUTF32Character(aUTF16Buf, nCharacter) - aUTF16Buf);
3516 sDecoded.Append(aUTF16Buf, nUTF16Len);
3517 p = pUTF8End;
3518 pCopyBegin = p;
3519 }
3520 /* bStartEncodedWord = false; */
3521 break;
3522 }
3523 }
3524 pWSPBegin = p;
3525 }
3526
3527 appendISO88591(sDecoded, pCopyBegin, pEnd);
3528 return sDecoded;
3529 }
3530
3531 //============================================================================
3532 //
3533 // INetMIMEOutputSink
3534 //
3535 //============================================================================
3536
3537 // virtual
3538 sal_Size INetMIMEOutputSink::writeSequence(const sal_Char * pSequence)
3539 {
3540 sal_Size nLength = rtl_str_getLength(pSequence);
3541 writeSequence(pSequence, pSequence + nLength);
3542 return nLength;
3543 }
3544
3545 //============================================================================
3546 // virtual
3547 void INetMIMEOutputSink::writeSequence(const sal_uInt32 * pBegin,
3548 const sal_uInt32 * pEnd)
3549 {
3550 DBG_ASSERT(pBegin && pBegin <= pEnd,
3551 "INetMIMEOutputSink::writeSequence(): Bad sequence");
3552
3553 sal_Char * pBufferBegin = new sal_Char[pEnd - pBegin];
3554 sal_Char * pBufferEnd = pBufferBegin;
3555 while (pBegin != pEnd)
3556 {
3557 DBG_ASSERT(*pBegin < 256,
3558 "INetMIMEOutputSink::writeSequence(): Bad octet");
3559 *pBufferEnd++ = sal_Char(*pBegin++);
3560 }
3561 writeSequence(pBufferBegin, pBufferEnd);
3562 delete[] pBufferBegin;
3563 }
3564
3565 //============================================================================
3566 // virtual
3567 void INetMIMEOutputSink::writeSequence(const sal_Unicode * pBegin,
3568 const sal_Unicode * pEnd)
3569 {
3570 DBG_ASSERT(pBegin && pBegin <= pEnd,
3571 "INetMIMEOutputSink::writeSequence(): Bad sequence");
3572
3573 sal_Char * pBufferBegin = new sal_Char[pEnd - pBegin];
3574 sal_Char * pBufferEnd = pBufferBegin;
3575 while (pBegin != pEnd)
3576 {
3577 DBG_ASSERT(*pBegin < 256,
3578 "INetMIMEOutputSink::writeSequence(): Bad octet");
3579 *pBufferEnd++ = sal_Char(*pBegin++);
3580 }
3581 writeSequence(pBufferBegin, pBufferEnd);
3582 delete[] pBufferBegin;
3583 }
3584
3585 //============================================================================
3586 // virtual
3587 ErrCode INetMIMEOutputSink::getError() const
3588 {
3589 return ERRCODE_NONE;
3590 }
3591
3592 //============================================================================
3593 void INetMIMEOutputSink::writeLineEnd()
3594 {
3595 static const sal_Char aCRLF[2] = { 0x0D, 0x0A };
3596 writeSequence(aCRLF, aCRLF + 2);
3597 m_nColumn = 0;
3598 }
3599
3600 //============================================================================
3601 //
3602 // INetMIMEStringOutputSink
3603 //
3604 //============================================================================
3605
3606 // virtual
3607 void INetMIMEStringOutputSink::writeSequence(const sal_Char * pBegin,
3608 const sal_Char * pEnd)
3609 {
3610 DBG_ASSERT(pBegin && pBegin <= pEnd,
3611 "INetMIMEStringOutputSink::writeSequence(): Bad sequence");
3612
3613 m_bOverflow = m_bOverflow
3614 || pEnd - pBegin > STRING_MAXLEN - m_aBuffer.Len();
3615 if (!m_bOverflow)
3616 m_aBuffer.Append(pBegin, static_cast< xub_StrLen >(pEnd - pBegin));
3617 }
3618
3619 //============================================================================
3620 // virtual
3621 ErrCode INetMIMEStringOutputSink::getError() const
3622 {
3623 return m_bOverflow ? ERRCODE_IO_OUTOFMEMORY : ERRCODE_NONE;
3624 }
3625
3626 //============================================================================
3627 //
3628 // INetMIMEUnicodeOutputSink
3629 //
3630 //============================================================================
3631
3632 // virtual
3633 void INetMIMEUnicodeOutputSink::writeSequence(const sal_Char * pBegin,
3634 const sal_Char * pEnd)
3635 {
3636 DBG_ASSERT(pBegin && pBegin <= pEnd,
3637 "INetMIMEUnicodeOutputSink::writeSequence(): Bad sequence");
3638
3639 sal_Unicode * pBufferBegin = new sal_Unicode[pEnd - pBegin];
3640 sal_Unicode * pBufferEnd = pBufferBegin;
3641 while (pBegin != pEnd)
3642 *pBufferEnd++ = sal_uChar(*pBegin++);
3643 writeSequence(pBufferBegin, pBufferEnd);
3644 delete[] pBufferBegin;
3645 }
3646
3647 //============================================================================
3648 // virtual
3649 void INetMIMEUnicodeOutputSink::writeSequence(const sal_uInt32 * pBegin,
3650 const sal_uInt32 * pEnd)
3651 {
3652 DBG_ASSERT(pBegin && pBegin <= pEnd,
3653 "INetMIMEUnicodeOutputSink::writeSequence(): Bad sequence");
3654
3655 sal_Unicode * pBufferBegin = new sal_Unicode[pEnd - pBegin];
3656 sal_Unicode * pBufferEnd = pBufferBegin;
3657 while (pBegin != pEnd)
3658 {
3659 DBG_ASSERT(*pBegin < 256,
3660 "INetMIMEOutputSink::writeSequence(): Bad octet");
3661 *pBufferEnd++ = sal_Unicode(*pBegin++);
3662 }
3663 writeSequence(pBufferBegin, pBufferEnd);
3664 delete[] pBufferBegin;
3665 }
3666
3667 //============================================================================
3668 // virtual
3669 void INetMIMEUnicodeOutputSink::writeSequence(const sal_Unicode * pBegin,
3670 const sal_Unicode * pEnd)
3671 {
3672 DBG_ASSERT(pBegin && pBegin <= pEnd,
3673 "INetMIMEUnicodeOutputSink::writeSequence(): Bad sequence");
3674
3675 m_bOverflow = m_bOverflow
3676 || pEnd - pBegin > STRING_MAXLEN - m_aBuffer.Len();
3677 if (!m_bOverflow)
3678 m_aBuffer.Append(pBegin, static_cast< xub_StrLen >(pEnd - pBegin));
3679 }
3680
3681 //============================================================================
3682 // virtual
3683 ErrCode INetMIMEUnicodeOutputSink::getError() const
3684 {
3685 return m_bOverflow ? ERRCODE_IO_OUTOFMEMORY : ERRCODE_NONE;
3686 }
3687
3688 //============================================================================
3689 //
3690 // INetMIMEEncodedWordOutputSink
3691 //
3692 //============================================================================
3693
3694 static const sal_Char aEscape[128]
3695 = { INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x00
3696 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x01
3697 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x02
3698 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x03
3699 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x04
3700 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x05
3701 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x06
3702 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x07
3703 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x08
3704 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x09
3705 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x0A
3706 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x0B
3707 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x0C
3708 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x0D
3709 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x0E
3710 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x0F
3711 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x10
3712 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x11
3713 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x12
3714 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x13
3715 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x14
3716 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x15
3717 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x16
3718 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x17
3719 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x18
3720 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x19
3721 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x1A
3722 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x1B
3723 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x1C
3724 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x1D
3725 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x1E
3726 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // 0x1F
3727 0, // ' '
3728 0, // '!'
3729 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '"'
3730 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '#'
3731 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '$'
3732 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '%'
3733 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '&'
3734 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '''
3735 INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '('
3736 INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // ')'
3737 0, // '*'
3738 0, // '+'
3739 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // ','
3740 0, // '-'
3741 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '.'
3742 0, // '/'
3743 0, // '0'
3744 0, // '1'
3745 0, // '2'
3746 0, // '3'
3747 0, // '4'
3748 0, // '5'
3749 0, // '6'
3750 0, // '7'
3751 0, // '8'
3752 0, // '9'
3753 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // ':'
3754 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // ';'
3755 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '<'
3756 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '='
3757 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '>'
3758 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '?'
3759 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '@'
3760 0, // 'A'
3761 0, // 'B'
3762 0, // 'C'
3763 0, // 'D'
3764 0, // 'E'
3765 0, // 'F'
3766 0, // 'G'
3767 0, // 'H'
3768 0, // 'I'
3769 0, // 'J'
3770 0, // 'K'
3771 0, // 'L'
3772 0, // 'M'
3773 0, // 'N'
3774 0, // 'O'
3775 0, // 'P'
3776 0, // 'Q'
3777 0, // 'R'
3778 0, // 'S'
3779 0, // 'T'
3780 0, // 'U'
3781 0, // 'V'
3782 0, // 'W'
3783 0, // 'X'
3784 0, // 'Y'
3785 0, // 'Z'
3786 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '['
3787 INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '\'
3788 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // ']'
3789 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '^'
3790 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '_'
3791 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '`'
3792 0, // 'a'
3793 0, // 'b'
3794 0, // 'c'
3795 0, // 'd'
3796 0, // 'e'
3797 0, // 'f'
3798 0, // 'g'
3799 0, // 'h'
3800 0, // 'i'
3801 0, // 'j'
3802 0, // 'k'
3803 0, // 'l'
3804 0, // 'm'
3805 0, // 'n'
3806 0, // 'o'
3807 0, // 'p'
3808 0, // 'q'
3809 0, // 'r'
3810 0, // 's'
3811 0, // 't'
3812 0, // 'u'
3813 0, // 'v'
3814 0, // 'w'
3815 0, // 'x'
3816 0, // 'y'
3817 0, // 'z'
3818 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '{'
3819 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '|'
3820 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '}'
3821 INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE, // '~'
3822 INetMIMEEncodedWordOutputSink::CONTEXT_TEXT | INetMIMEEncodedWordOutputSink::CONTEXT_COMMENT | INetMIMEEncodedWordOutputSink::CONTEXT_PHRASE }; // DEL
3823
3824 inline bool
3825 INetMIMEEncodedWordOutputSink::needsEncodedWordEscape(sal_uInt32 nChar) const
3826 {
3827 return !INetMIME::isUSASCII(nChar) || aEscape[nChar] & m_eContext;
3828 }
3829
3830 //============================================================================
3831 void INetMIMEEncodedWordOutputSink::finish(bool bWriteTrailer)
3832 {
3833 if (m_eInitialSpace == SPACE_ALWAYS && m_nExtraSpaces == 0)
3834 m_nExtraSpaces = 1;
3835
3836 if (m_eEncodedWordState == STATE_SECOND_EQUALS)
3837 {
3838 // If the text is already an encoded word, copy it verbatim:
3839 sal_uInt32 nSize = m_pBufferEnd - m_pBuffer;
3840 switch (m_ePrevCoding)
3841 {
3842 case CODING_QUOTED:
3843 m_rSink << '"';
3844 case CODING_NONE:
3845 if (m_eInitialSpace == SPACE_ENCODED && m_nExtraSpaces == 0)
3846 m_nExtraSpaces = 1;
3847 for (; m_nExtraSpaces > 1; --m_nExtraSpaces)
3848 {
3849 if (m_rSink.getColumn() >= m_rSink.getLineLengthLimit())
3850 m_rSink << INetMIMEOutputSink::endl;
3851 m_rSink << ' ';
3852 }
3853 if (m_nExtraSpaces == 1)
3854 {
3855 if (m_rSink.getColumn() + nSize
3856 >= m_rSink.getLineLengthLimit())
3857 m_rSink << INetMIMEOutputSink::endl;
3858 m_rSink << ' ';
3859 }
3860 break;
3861
3862 case CODING_ENCODED:
3863 {
3864 const sal_Char * pCharsetName
3865 = INetMIME::getCharsetName(m_ePrevMIMEEncoding);
3866 while (m_nExtraSpaces-- > 0)
3867 {
3868 if (m_rSink.getColumn()
3869 > m_rSink.getLineLengthLimit() - 3)
3870 m_rSink << "?=" << INetMIMEOutputSink::endl << " =?"
3871 << pCharsetName << "?Q?";
3872 m_rSink << '_';
3873 }
3874 m_rSink << "?=";
3875 }
3876 case CODING_ENCODED_TERMINATED:
3877 if (m_rSink.getColumn() + nSize
3878 > m_rSink.getLineLengthLimit() - 1)
3879 m_rSink << INetMIMEOutputSink::endl;
3880 m_rSink << ' ';
3881 break;
3882 }
3883 m_rSink.write(m_pBuffer, m_pBufferEnd);
3884 m_eCoding = CODING_ENCODED_TERMINATED;
3885 }
3886 else
3887 {
3888 // If the text itself is too long to fit into a single line, make it
3889 // into multiple encoded words:
3890 switch (m_eCoding)
3891 {
3892 case CODING_NONE:
3893 if (m_nExtraSpaces == 0)
3894 {
3895 DBG_ASSERT(m_ePrevCoding == CODING_NONE
3896 || m_pBuffer == m_pBufferEnd,
3897 "INetMIMEEncodedWordOutputSink::finish():"
3898 " Bad state");
3899 if (m_rSink.getColumn() + (m_pBufferEnd - m_pBuffer)
3900 > m_rSink.getLineLengthLimit())
3901 m_eCoding = CODING_ENCODED;
3902 }
3903 else
3904 {
3905 OSL_ASSERT(m_pBufferEnd >= m_pBuffer);
3906 if (static_cast< std::size_t >(m_pBufferEnd - m_pBuffer)
3907 > m_rSink.getLineLengthLimit() - 1)
3908 {
3909 m_eCoding = CODING_ENCODED;
3910 }
3911 }
3912 break;
3913
3914 case CODING_QUOTED:
3915 if (m_nExtraSpaces == 0)
3916 {
3917 DBG_ASSERT(m_ePrevCoding == CODING_NONE,
3918 "INetMIMEEncodedWordOutputSink::finish():"
3919 " Bad state");
3920 if (m_rSink.getColumn() + (m_pBufferEnd - m_pBuffer)
3921 + m_nQuotedEscaped
3922 > m_rSink.getLineLengthLimit() - 2)
3923 m_eCoding = CODING_ENCODED;
3924 }
3925 else if ((m_pBufferEnd - m_pBuffer) + m_nQuotedEscaped
3926 > m_rSink.getLineLengthLimit() - 3)
3927 m_eCoding = CODING_ENCODED;
3928 break;
3929
3930 default:
3931 break;
3932 }
3933
3934 switch (m_eCoding)
3935 {
3936 case CODING_NONE:
3937 switch (m_ePrevCoding)
3938 {
3939 case CODING_QUOTED:
3940 if (m_rSink.getColumn() + m_nExtraSpaces
3941 + (m_pBufferEnd - m_pBuffer)
3942 < m_rSink.getLineLengthLimit())
3943 m_eCoding = CODING_QUOTED;
3944 else
3945 m_rSink << '"';
3946 break;
3947
3948 case CODING_ENCODED:
3949 m_rSink << "?=";
3950 break;
3951
3952 default:
3953 break;
3954 }
3955 for (; m_nExtraSpaces > 1; --m_nExtraSpaces)
3956 {
3957 if (m_rSink.getColumn() >= m_rSink.getLineLengthLimit())
3958 m_rSink << INetMIMEOutputSink::endl;
3959 m_rSink << ' ';
3960 }
3961 if (m_nExtraSpaces == 1)
3962 {
3963 if (m_rSink.getColumn() + (m_pBufferEnd - m_pBuffer)
3964 >= m_rSink.getLineLengthLimit())
3965 m_rSink << INetMIMEOutputSink::endl;
3966 m_rSink << ' ';
3967 }
3968 m_rSink.write(m_pBuffer, m_pBufferEnd);
3969 if (m_eCoding == CODING_QUOTED && bWriteTrailer)
3970 {
3971 m_rSink << '"';
3972 m_eCoding = CODING_NONE;
3973 }
3974 break;
3975
3976 case CODING_QUOTED:
3977 {
3978 bool bInsertLeadingQuote = true;
3979 sal_uInt32 nSize = (m_pBufferEnd - m_pBuffer)
3980 + m_nQuotedEscaped + 2;
3981 switch (m_ePrevCoding)
3982 {
3983 case CODING_QUOTED:
3984 if (m_rSink.getColumn() + m_nExtraSpaces + nSize - 1
3985 < m_rSink.getLineLengthLimit())
3986 {
3987 bInsertLeadingQuote = false;
3988 --nSize;
3989 }
3990 else
3991 m_rSink << '"';
3992 break;
3993
3994 case CODING_ENCODED:
3995 m_rSink << "?=";
3996 break;
3997
3998 default:
3999 break;
4000 }
4001 for (; m_nExtraSpaces > 1; --m_nExtraSpaces)
4002 {
4003 if (m_rSink.getColumn() >= m_rSink.getLineLengthLimit())
4004 m_rSink << INetMIMEOutputSink::endl;
4005 m_rSink << ' ';
4006 }
4007 if (m_nExtraSpaces == 1)
4008 {
4009 if (m_rSink.getColumn() + nSize
4010 >= m_rSink.getLineLengthLimit())
4011 m_rSink << INetMIMEOutputSink::endl;
4012 m_rSink << ' ';
4013 }
4014 if (bInsertLeadingQuote)
4015 m_rSink << '"';
4016 for (const sal_Unicode * p = m_pBuffer; p != m_pBufferEnd;
4017 ++p)
4018 {
4019 if (INetMIME::needsQuotedStringEscape(*p))
4020 m_rSink << '\\';
4021 m_rSink << sal_Char(*p);
4022 }
4023 if (bWriteTrailer)
4024 {
4025 m_rSink << '"';
4026 m_eCoding = CODING_NONE;
4027 }
4028 break;
4029 }
4030
4031 case CODING_ENCODED:
4032 {
4033 rtl_TextEncoding eCharsetEncoding
4034 = m_pEncodingList->
4035 getPreferredEncoding(RTL_TEXTENCODING_UTF8);
4036 rtl_TextEncoding eMIMEEncoding
4037 = INetMIME::translateToMIME(eCharsetEncoding);
4038
4039 // The non UTF-8 code will only work for stateless single byte
4040 // character encodings (see also below):
4041 sal_Char * pTargetBuffer = NULL;
4042 sal_Size nTargetSize = 0;
4043 sal_uInt32 nSize;
4044 if (eMIMEEncoding == RTL_TEXTENCODING_UTF8)
4045 {
4046 nSize = 0;
4047 for (sal_Unicode const * p = m_pBuffer;
4048 p != m_pBufferEnd;)
4049 {
4050 sal_uInt32 nUTF32
4051 = INetMIME::getUTF32Character(p, m_pBufferEnd);
4052 nSize += needsEncodedWordEscape(nUTF32) ?
4053 3 * INetMIME::getUTF8OctetCount(nUTF32) :
4054 1;
4055 // only US-ASCII characters (that are converted to
4056 // a single byte by UTF-8) need no encoded word
4057 // escapes...
4058 }
4059 }
4060 else
4061 {
4062 rtl_UnicodeToTextConverter hConverter
4063 = rtl_createUnicodeToTextConverter(eCharsetEncoding);
4064 rtl_UnicodeToTextContext hContext
4065 = rtl_createUnicodeToTextContext(hConverter);
4066 for (sal_Size nBufferSize = m_pBufferEnd - m_pBuffer;;
4067 nBufferSize += nBufferSize / 3 + 1)
4068 {
4069 pTargetBuffer = new sal_Char[nBufferSize];
4070 sal_uInt32 nInfo;
4071 sal_Size nSrcCvtBytes;
4072 nTargetSize
4073 = rtl_convertUnicodeToText(
4074 hConverter, hContext, m_pBuffer,
4075 m_pBufferEnd - m_pBuffer, pTargetBuffer,
4076 nBufferSize,
4077 RTL_UNICODETOTEXT_FLAGS_UNDEFINED_IGNORE
4078 | RTL_UNICODETOTEXT_FLAGS_INVALID_IGNORE,
4079 &nInfo, &nSrcCvtBytes);
4080 if (!(nInfo
4081 & RTL_UNICODETOTEXT_INFO_DESTBUFFERTOSMALL))
4082 break;
4083 delete[] pTargetBuffer;
4084 pTargetBuffer = NULL;
4085 rtl_resetUnicodeToTextContext(hConverter, hContext);
4086 }
4087 rtl_destroyUnicodeToTextContext(hConverter, hContext);
4088 rtl_destroyUnicodeToTextConverter(hConverter);
4089
4090 nSize = nTargetSize;
4091 for (sal_Size k = 0; k < nTargetSize; ++k)
4092 if (needsEncodedWordEscape(sal_uChar(
4093 pTargetBuffer[k])))
4094 nSize += 2;
4095 }
4096
4097 const sal_Char * pCharsetName
4098 = INetMIME::getCharsetName(eMIMEEncoding);
4099 sal_uInt32 nWrapperSize = rtl_str_getLength(pCharsetName) + 7;
4100 // '=?', '?Q?', '?='
4101
4102 switch (m_ePrevCoding)
4103 {
4104 case CODING_QUOTED:
4105 m_rSink << '"';
4106 case CODING_NONE:
4107 if (m_eInitialSpace == SPACE_ENCODED
4108 && m_nExtraSpaces == 0)
4109 m_nExtraSpaces = 1;
4110 nSize += nWrapperSize;
4111 for (; m_nExtraSpaces > 1; --m_nExtraSpaces)
4112 {
4113 if (m_rSink.getColumn()
4114 >= m_rSink.getLineLengthLimit())
4115 m_rSink << INetMIMEOutputSink::endl;
4116 m_rSink << ' ';
4117 }
4118 if (m_nExtraSpaces == 1)
4119 {
4120 if (m_rSink.getColumn() + nSize
4121 >= m_rSink.getLineLengthLimit())
4122 m_rSink << INetMIMEOutputSink::endl;
4123 m_rSink << ' ';
4124 }
4125 m_rSink << "=?" << pCharsetName << "?Q?";
4126 break;
4127
4128 case CODING_ENCODED:
4129 if (m_ePrevMIMEEncoding != eMIMEEncoding
4130 || m_rSink.getColumn() + m_nExtraSpaces + nSize
4131 > m_rSink.getLineLengthLimit() - 2)
4132 {
4133 m_rSink << "?=";
4134 if (m_rSink.getColumn() + nWrapperSize
4135 + m_nExtraSpaces + nSize
4136 > m_rSink.getLineLengthLimit() - 1)
4137 m_rSink << INetMIMEOutputSink::endl;
4138 m_rSink << " =?" << pCharsetName << "?Q?";
4139 }
4140 while (m_nExtraSpaces-- > 0)
4141 {
4142 if (m_rSink.getColumn()
4143 > m_rSink.getLineLengthLimit() - 3)
4144 m_rSink << "?=" << INetMIMEOutputSink::endl
4145 << " =?" << pCharsetName << "?Q?";
4146 m_rSink << '_';
4147 }
4148 break;
4149
4150 case CODING_ENCODED_TERMINATED:
4151 if (m_rSink.getColumn() + nWrapperSize
4152 + m_nExtraSpaces + nSize
4153 > m_rSink.getLineLengthLimit() - 1)
4154 m_rSink << INetMIMEOutputSink::endl;
4155 m_rSink << " =?" << pCharsetName << "?Q?";
4156 while (m_nExtraSpaces-- > 0)
4157 {
4158 if (m_rSink.getColumn()
4159 > m_rSink.getLineLengthLimit() - 3)
4160 m_rSink << "?=" << INetMIMEOutputSink::endl
4161 << " =?" << pCharsetName << "?Q?";
4162 m_rSink << '_';
4163 }
4164 break;
4165 }
4166
4167 // The non UTF-8 code will only work for stateless single byte
4168 // character encodings (see also above):
4169 if (eMIMEEncoding == RTL_TEXTENCODING_UTF8)
4170 {
4171 bool bInitial = true;
4172 for (sal_Unicode const * p = m_pBuffer;
4173 p != m_pBufferEnd;)
4174 {
4175 sal_uInt32 nUTF32
4176 = INetMIME::getUTF32Character(p, m_pBufferEnd);
4177 bool bEscape = needsEncodedWordEscape(nUTF32);
4178 sal_uInt32 nWidth
4179 = bEscape ?
4180 3 * INetMIME::getUTF8OctetCount(nUTF32) : 1;
4181 // only US-ASCII characters (that are converted to
4182 // a single byte by UTF-8) need no encoded word
4183 // escapes...
4184 if (!bInitial
4185 && m_rSink.getColumn() + nWidth + 2
4186 > m_rSink.getLineLengthLimit())
4187 m_rSink << "?=" << INetMIMEOutputSink::endl
4188 << " =?" << pCharsetName << "?Q?";
4189 if (bEscape)
4190 {
4191 DBG_ASSERT(
4192 nUTF32 < 0x10FFFF,
4193 "INetMIMEEncodedWordOutputSink::finish():"
4194 " Bad char");
4195 if (nUTF32 < 0x80)
4196 INetMIME::writeEscapeSequence(m_rSink,
4197 nUTF32);
4198 else if (nUTF32 < 0x800)
4199 {
4200 INetMIME::writeEscapeSequence(m_rSink,
4201 (nUTF32 >> 6)
4202 | 0xC0);
4203 INetMIME::writeEscapeSequence(m_rSink,
4204 (nUTF32 & 0x3F)
4205 | 0x80);
4206 }
4207 else if (nUTF32 < 0x10000)
4208 {
4209 INetMIME::writeEscapeSequence(m_rSink,
4210 (nUTF32 >> 12)
4211 | 0xE0);
4212 INetMIME::writeEscapeSequence(m_rSink,
4213 ((nUTF32 >> 6)
4214 & 0x3F)
4215 | 0x80);
4216 INetMIME::writeEscapeSequence(m_rSink,
4217 (nUTF32 & 0x3F)
4218 | 0x80);
4219 }
4220 else
4221 {
4222 INetMIME::writeEscapeSequence(m_rSink,
4223 (nUTF32 >> 18)
4224 | 0xF0);
4225 INetMIME::writeEscapeSequence(m_rSink,
4226 ((nUTF32 >> 12)
4227 & 0x3F)
4228 | 0x80);
4229 INetMIME::writeEscapeSequence(m_rSink,
4230 ((nUTF32 >> 6)
4231 & 0x3F)
4232 | 0x80);
4233 INetMIME::writeEscapeSequence(m_rSink,
4234 (nUTF32 & 0x3F)
4235 | 0x80);
4236 }
4237 }
4238 else
4239 m_rSink << sal_Char(nUTF32);
4240 bInitial = false;
4241 }
4242 }
4243 else
4244 {
4245 for (sal_Size k = 0; k < nTargetSize; ++k)
4246 {
4247 sal_uInt32 nUCS4 = sal_uChar(pTargetBuffer[k]);
4248 bool bEscape = needsEncodedWordEscape(nUCS4);
4249 if (k > 0
4250 && m_rSink.getColumn() + (bEscape ? 5 : 3)
4251 > m_rSink.getLineLengthLimit())
4252 m_rSink << "?=" << INetMIMEOutputSink::endl
4253 << " =?" << pCharsetName << "?Q?";
4254 if (bEscape)
4255 INetMIME::writeEscapeSequence(m_rSink, nUCS4);
4256 else
4257 m_rSink << sal_Char(nUCS4);
4258 }
4259 delete[] pTargetBuffer;
4260 }
4261
4262 if (bWriteTrailer)
4263 {
4264 m_rSink << "?=";
4265 m_eCoding = CODING_ENCODED_TERMINATED;
4266 }
4267
4268 m_ePrevMIMEEncoding = eMIMEEncoding;
4269 break;
4270 }
4271
4272 default:
4273 OSL_ASSERT(false);
4274 break;
4275 }
4276 }
4277
4278 m_eInitialSpace = SPACE_NO;
4279 m_nExtraSpaces = 0;
4280 m_pEncodingList->reset();
4281 m_pBufferEnd = m_pBuffer;
4282 m_ePrevCoding = m_eCoding;
4283 m_eCoding = CODING_NONE;
4284 m_nQuotedEscaped = 0;
4285 m_eEncodedWordState = STATE_INITIAL;
4286 }
4287
4288 //============================================================================
4289 INetMIMEEncodedWordOutputSink::~INetMIMEEncodedWordOutputSink()
4290 {
4291 rtl_freeMemory(m_pBuffer);
4292 delete m_pEncodingList;
4293 }
4294
4295 //============================================================================
4296 INetMIMEEncodedWordOutputSink &
4297 INetMIMEEncodedWordOutputSink::operator <<(sal_uInt32 nChar)
4298 {
4299 if (nChar == ' ')
4300 {
4301 if (m_pBufferEnd != m_pBuffer)
4302 finish(false);
4303 ++m_nExtraSpaces;
4304 }
4305 else
4306 {
4307 // Check for an already encoded word:
4308 switch (m_eEncodedWordState)
4309 {
4310 case STATE_INITIAL:
4311 if (nChar == '=')
4312 m_eEncodedWordState = STATE_FIRST_EQUALS;
4313 else
4314 m_eEncodedWordState = STATE_BAD;
4315 break;
4316
4317 case STATE_FIRST_EQUALS:
4318 if (nChar == '?')
4319 m_eEncodedWordState = STATE_FIRST_EQUALS;
4320 else
4321 m_eEncodedWordState = STATE_BAD;
4322 break;
4323
4324 case STATE_FIRST_QUESTION:
4325 if (INetMIME::isEncodedWordTokenChar(nChar))
4326 m_eEncodedWordState = STATE_CHARSET;
4327 else
4328 m_eEncodedWordState = STATE_BAD;
4329 break;
4330
4331 case STATE_CHARSET:
4332 if (nChar == '?')
4333 m_eEncodedWordState = STATE_SECOND_QUESTION;
4334 else if (!INetMIME::isEncodedWordTokenChar(nChar))
4335 m_eEncodedWordState = STATE_BAD;
4336 break;
4337
4338 case STATE_SECOND_QUESTION:
4339 if (nChar == 'B' || nChar == 'Q'
4340 || nChar == 'b' || nChar == 'q')
4341 m_eEncodedWordState = STATE_ENCODING;
4342 else
4343 m_eEncodedWordState = STATE_BAD;
4344 break;
4345
4346 case STATE_ENCODING:
4347 if (nChar == '?')
4348 m_eEncodedWordState = STATE_THIRD_QUESTION;
4349 else
4350 m_eEncodedWordState = STATE_BAD;
4351 break;
4352
4353 case STATE_THIRD_QUESTION:
4354 if (INetMIME::isVisible(nChar) && nChar != '?')
4355 m_eEncodedWordState = STATE_ENCODED_TEXT;
4356 else
4357 m_eEncodedWordState = STATE_BAD;
4358 break;
4359
4360 case STATE_ENCODED_TEXT:
4361 if (nChar == '?')
4362 m_eEncodedWordState = STATE_FOURTH_QUESTION;
4363 else if (!INetMIME::isVisible(nChar))
4364 m_eEncodedWordState = STATE_BAD;
4365 break;
4366
4367 case STATE_FOURTH_QUESTION:
4368 if (nChar == '=')
4369 m_eEncodedWordState = STATE_SECOND_EQUALS;
4370 else
4371 m_eEncodedWordState = STATE_BAD;
4372 break;
4373
4374 case STATE_SECOND_EQUALS:
4375 m_eEncodedWordState = STATE_BAD;
4376 break;
4377
4378 case STATE_BAD:
4379 break;
4380 }
4381
4382 // Update encoding:
4383 m_pEncodingList->includes(nChar);
4384
4385 // Update coding:
4386 enum { TENQ = 1, // CONTEXT_TEXT, CODING_ENCODED
4387 CENQ = 2, // CONTEXT_COMMENT, CODING_ENCODED
4388 PQTD = 4, // CONTEXT_PHRASE, CODING_QUOTED
4389 PENQ = 8 }; // CONTEXT_PHRASE, CODING_ENCODED
4390 static const sal_Char aMinimal[128]
4391 = { TENQ | CENQ | PENQ, // 0x00
4392 TENQ | CENQ | PENQ, // 0x01
4393 TENQ | CENQ | PENQ, // 0x02
4394 TENQ | CENQ | PENQ, // 0x03
4395 TENQ | CENQ | PENQ, // 0x04
4396 TENQ | CENQ | PENQ, // 0x05
4397 TENQ | CENQ | PENQ, // 0x06
4398 TENQ | CENQ | PENQ, // 0x07
4399 TENQ | CENQ | PENQ, // 0x08
4400 TENQ | CENQ | PENQ, // 0x09
4401 TENQ | CENQ | PENQ, // 0x0A
4402 TENQ | CENQ | PENQ, // 0x0B
4403 TENQ | CENQ | PENQ, // 0x0C
4404 TENQ | CENQ | PENQ, // 0x0D
4405 TENQ | CENQ | PENQ, // 0x0E
4406 TENQ | CENQ | PENQ, // 0x0F
4407 TENQ | CENQ | PENQ, // 0x10
4408 TENQ | CENQ | PENQ, // 0x11
4409 TENQ | CENQ | PENQ, // 0x12
4410 TENQ | CENQ | PENQ, // 0x13
4411 TENQ | CENQ | PENQ, // 0x14
4412 TENQ | CENQ | PENQ, // 0x15
4413 TENQ | CENQ | PENQ, // 0x16
4414 TENQ | CENQ | PENQ, // 0x17
4415 TENQ | CENQ | PENQ, // 0x18
4416 TENQ | CENQ | PENQ, // 0x19
4417 TENQ | CENQ | PENQ, // 0x1A
4418 TENQ | CENQ | PENQ, // 0x1B
4419 TENQ | CENQ | PENQ, // 0x1C
4420 TENQ | CENQ | PENQ, // 0x1D
4421 TENQ | CENQ | PENQ, // 0x1E
4422 TENQ | CENQ | PENQ, // 0x1F
4423 0, // ' '
4424 0, // '!'
4425 PQTD , // '"'
4426 0, // '#'
4427 0, // '$'
4428 0, // '%'
4429 0, // '&'
4430 0, // '''
4431 CENQ | PQTD , // '('
4432 CENQ | PQTD , // ')'
4433 0, // '*'
4434 0, // '+'
4435 PQTD , // ','
4436 0, // '-'
4437 PQTD , // '.'
4438 0, // '/'
4439 0, // '0'
4440 0, // '1'
4441 0, // '2'
4442 0, // '3'
4443 0, // '4'
4444 0, // '5'
4445 0, // '6'
4446 0, // '7'
4447 0, // '8'
4448 0, // '9'
4449 PQTD , // ':'
4450 PQTD , // ';'
4451 PQTD , // '<'
4452 0, // '='
4453 PQTD , // '>'
4454 0, // '?'
4455 PQTD , // '@'
4456 0, // 'A'
4457 0, // 'B'
4458 0, // 'C'
4459 0, // 'D'
4460 0, // 'E'
4461 0, // 'F'
4462 0, // 'G'
4463 0, // 'H'
4464 0, // 'I'
4465 0, // 'J'
4466 0, // 'K'
4467 0, // 'L'
4468 0, // 'M'
4469 0, // 'N'
4470 0, // 'O'
4471 0, // 'P'
4472 0, // 'Q'
4473 0, // 'R'
4474 0, // 'S'
4475 0, // 'T'
4476 0, // 'U'
4477 0, // 'V'
4478 0, // 'W'
4479 0, // 'X'
4480 0, // 'Y'
4481 0, // 'Z'
4482 PQTD , // '['
4483 CENQ | PQTD , // '\'
4484 PQTD , // ']'
4485 0, // '^'
4486 0, // '_'
4487 0, // '`'
4488 0, // 'a'
4489 0, // 'b'
4490 0, // 'c'
4491 0, // 'd'
4492 0, // 'e'
4493 0, // 'f'
4494 0, // 'g'
4495 0, // 'h'
4496 0, // 'i'
4497 0, // 'j'
4498 0, // 'k'
4499 0, // 'l'
4500 0, // 'm'
4501 0, // 'n'
4502 0, // 'o'
4503 0, // 'p'
4504 0, // 'q'
4505 0, // 'r'
4506 0, // 's'
4507 0, // 't'
4508 0, // 'u'
4509 0, // 'v'
4510 0, // 'w'
4511 0, // 'x'
4512 0, // 'y'
4513 0, // 'z'
4514 0, // '{'
4515 0, // '|'
4516 0, // '}'
4517 0, // '~'
4518 TENQ | CENQ | PENQ }; // DEL
4519 Coding eNewCoding = !INetMIME::isUSASCII(nChar) ? CODING_ENCODED :
4520 m_eContext == CONTEXT_PHRASE ?
4521 Coding(aMinimal[nChar] >> 2) :
4522 aMinimal[nChar] & m_eContext ? CODING_ENCODED :
4523 CODING_NONE;
4524 if (eNewCoding > m_eCoding)
4525 m_eCoding = eNewCoding;
4526 if (m_eCoding == CODING_QUOTED
4527 && INetMIME::needsQuotedStringEscape(nChar))
4528 ++m_nQuotedEscaped;
4529
4530 // Append to buffer:
4531 if (sal_uInt32(m_pBufferEnd - m_pBuffer) == m_nBufferSize)
4532 {
4533 m_pBuffer
4534 = static_cast< sal_Unicode * >(
4535 rtl_reallocateMemory(m_pBuffer,
4536 (m_nBufferSize + BUFFER_SIZE)
4537 * sizeof (sal_Unicode)));
4538 m_pBufferEnd = m_pBuffer + m_nBufferSize;
4539 m_nBufferSize += BUFFER_SIZE;
4540 }
4541 *m_pBufferEnd++ = sal_Unicode(nChar);
4542 }
4543 return *this;
4544 }
4545
4546 //============================================================================
4547 //
4548 // INetContentTypeParameterList
4549 //
4550 //============================================================================
4551
4552 void INetContentTypeParameterList::Clear()
4553 {
4554 while (Count() > 0)
4555 delete static_cast< INetContentTypeParameter * >(Remove(Count() - 1));
4556 }
4557
4558 //============================================================================
4559 const INetContentTypeParameter *
4560 INetContentTypeParameterList::find(const ByteString & rAttribute) const
4561 {
4562 for (ULONG i = 0; i < Count(); ++i)
4563 {
4564 const INetContentTypeParameter * pParameter = GetObject(i);
4565 if (pParameter->m_sAttribute.EqualsIgnoreCaseAscii(rAttribute))
4566 return pParameter;
4567 }
4568 return 0;
4569 }
4570
vba interoperability for OOo