1// © 2016 and later: Unicode, Inc. and others.
2// License & terms of use: http://www.unicode.org/copyright.html
3/*
4*******************************************************************************
5* Copyright (C) 2010-2015, International Business Machines
6* Corporation and others. All Rights Reserved.
7*******************************************************************************
8* file name: uts46.cpp
9* encoding: UTF-8
10* tab size: 8 (not used)
11* indentation:4
12*
13* created on: 2010mar09
14* created by: Markus W. Scherer
15*/
16
17#include "unicode/utypes.h"
18
19#if !UCONFIG_NO_IDNA
20
21#include "unicode/idna.h"
22#include "unicode/normalizer2.h"
23#include "unicode/uscript.h"
24#include "unicode/ustring.h"
25#include "unicode/utf16.h"
26#include "cmemory.h"
27#include "cstring.h"
28#include "punycode.h"
29#include "ubidi_props.h"
30#include "ustr_imp.h"
31
32// Note about tests for UIDNA_ERROR_DOMAIN_NAME_TOO_LONG:
33//
34// The domain name length limit is 255 octets in an internal DNS representation
35// where the last ("root") label is the empty label
36// represented by length byte 0 alone.
37// In a conventional string, this translates to 253 characters, or 254
38// if there is a trailing dot for the root label.
39
40U_NAMESPACE_BEGIN
41
42// Severe errors which usually result in a U+FFFD replacement character in the result string.
43const uint32_t severeErrors=
44 UIDNA_ERROR_LEADING_COMBINING_MARK|
45 UIDNA_ERROR_DISALLOWED|
46 UIDNA_ERROR_PUNYCODE|
47 UIDNA_ERROR_LABEL_HAS_DOT|
48 UIDNA_ERROR_INVALID_ACE_LABEL;
49
50static inline UBool
51isASCIIString(const UnicodeString &dest) {
52 const char16_t *s=dest.getBuffer();
53 const char16_t *limit=s+dest.length();
54 while(s<limit) {
55 if(*s++>0x7f) {
56 return false;
57 }
58 }
59 return true;
60}
61
62static UBool
63isASCIIOkBiDi(const char16_t *s, int32_t length);
64
65static UBool
66isASCIIOkBiDi(const char *s, int32_t length);
67
68// IDNA class default implementations -------------------------------------- ***
69
70IDNA::~IDNA() {}
71
72void
73IDNA::labelToASCII_UTF8(StringPiece label, ByteSink &dest,
74 IDNAInfo &info, UErrorCode &errorCode) const {
75 if(U_SUCCESS(errorCode)) {
76 UnicodeString destString;
77 labelToASCII(UnicodeString::fromUTF8(label), destString,
78 info, errorCode).toUTF8(dest);
79 }
80}
81
82void
83IDNA::labelToUnicodeUTF8(StringPiece label, ByteSink &dest,
84 IDNAInfo &info, UErrorCode &errorCode) const {
85 if(U_SUCCESS(errorCode)) {
86 UnicodeString destString;
87 labelToUnicode(UnicodeString::fromUTF8(label), destString,
88 info, errorCode).toUTF8(dest);
89 }
90}
91
92void
93IDNA::nameToASCII_UTF8(StringPiece name, ByteSink &dest,
94 IDNAInfo &info, UErrorCode &errorCode) const {
95 if(U_SUCCESS(errorCode)) {
96 UnicodeString destString;
97 nameToASCII(UnicodeString::fromUTF8(name), destString,
98 info, errorCode).toUTF8(dest);
99 }
100}
101
102void
103IDNA::nameToUnicodeUTF8(StringPiece name, ByteSink &dest,
104 IDNAInfo &info, UErrorCode &errorCode) const {
105 if(U_SUCCESS(errorCode)) {
106 UnicodeString destString;
107 nameToUnicode(UnicodeString::fromUTF8(name), destString,
108 info, errorCode).toUTF8(dest);
109 }
110}
111
112// UTS46 class declaration ------------------------------------------------- ***
113
114class UTS46 : public IDNA {
115public:
116 UTS46(uint32_t options, UErrorCode &errorCode);
117 virtual ~UTS46();
118
119 virtual UnicodeString &
120 labelToASCII(const UnicodeString &label, UnicodeString &dest,
121 IDNAInfo &info, UErrorCode &errorCode) const override;
122
123 virtual UnicodeString &
124 labelToUnicode(const UnicodeString &label, UnicodeString &dest,
125 IDNAInfo &info, UErrorCode &errorCode) const override;
126
127 virtual UnicodeString &
128 nameToASCII(const UnicodeString &name, UnicodeString &dest,
129 IDNAInfo &info, UErrorCode &errorCode) const override;
130
131 virtual UnicodeString &
132 nameToUnicode(const UnicodeString &name, UnicodeString &dest,
133 IDNAInfo &info, UErrorCode &errorCode) const override;
134
135 virtual void
136 labelToASCII_UTF8(StringPiece label, ByteSink &dest,
137 IDNAInfo &info, UErrorCode &errorCode) const override;
138
139 virtual void
140 labelToUnicodeUTF8(StringPiece label, ByteSink &dest,
141 IDNAInfo &info, UErrorCode &errorCode) const override;
142
143 virtual void
144 nameToASCII_UTF8(StringPiece name, ByteSink &dest,
145 IDNAInfo &info, UErrorCode &errorCode) const override;
146
147 virtual void
148 nameToUnicodeUTF8(StringPiece name, ByteSink &dest,
149 IDNAInfo &info, UErrorCode &errorCode) const override;
150
151private:
152 UnicodeString &
153 process(const UnicodeString &src,
154 UBool isLabel, UBool toASCII,
155 UnicodeString &dest,
156 IDNAInfo &info, UErrorCode &errorCode) const;
157
158 void
159 processUTF8(StringPiece src,
160 UBool isLabel, UBool toASCII,
161 ByteSink &dest,
162 IDNAInfo &info, UErrorCode &errorCode) const;
163
164 UnicodeString &
165 processUnicode(const UnicodeString &src,
166 int32_t labelStart, int32_t mappingStart,
167 UBool isLabel, UBool toASCII,
168 UnicodeString &dest,
169 IDNAInfo &info, UErrorCode &errorCode) const;
170
171 // returns the new dest.length()
172 int32_t
173 mapDevChars(UnicodeString &dest, int32_t labelStart, int32_t mappingStart,
174 UErrorCode &errorCode) const;
175
176 // returns the new label length
177 int32_t
178 processLabel(UnicodeString &dest,
179 int32_t labelStart, int32_t labelLength,
180 UBool toASCII,
181 IDNAInfo &info, UErrorCode &errorCode) const;
182 int32_t
183 markBadACELabel(UnicodeString &dest,
184 int32_t labelStart, int32_t labelLength,
185 UBool toASCII, IDNAInfo &info, UErrorCode &errorCode) const;
186
187 void
188 checkLabelBiDi(const char16_t *label, int32_t labelLength, IDNAInfo &info) const;
189
190 UBool
191 isLabelOkContextJ(const char16_t *label, int32_t labelLength) const;
192
193 void
194 checkLabelContextO(const char16_t *label, int32_t labelLength, IDNAInfo &info) const;
195
196 const Normalizer2 &uts46Norm2; // uts46.nrm
197 uint32_t options;
198};
199
200IDNA *
201IDNA::createUTS46Instance(uint32_t options, UErrorCode &errorCode) {
202 if(U_SUCCESS(errorCode)) {
203 IDNA *idna=new UTS46(options, errorCode);
204 if(idna==nullptr) {
205 errorCode=U_MEMORY_ALLOCATION_ERROR;
206 } else if(U_FAILURE(errorCode)) {
207 delete idna;
208 idna=nullptr;
209 }
210 return idna;
211 } else {
212 return nullptr;
213 }
214}
215
216// UTS46 implementation ---------------------------------------------------- ***
217
218UTS46::UTS46(uint32_t opt, UErrorCode &errorCode)
219 : uts46Norm2(*Normalizer2::getInstance(nullptr, "uts46", UNORM2_COMPOSE, errorCode)),
220 options(opt) {}
221
222UTS46::~UTS46() {}
223
224UnicodeString &
225UTS46::labelToASCII(const UnicodeString &label, UnicodeString &dest,
226 IDNAInfo &info, UErrorCode &errorCode) const {
227 return process(label, true, true, dest, info, errorCode);
228}
229
230UnicodeString &
231UTS46::labelToUnicode(const UnicodeString &label, UnicodeString &dest,
232 IDNAInfo &info, UErrorCode &errorCode) const {
233 return process(label, true, false, dest, info, errorCode);
234}
235
236UnicodeString &
237UTS46::nameToASCII(const UnicodeString &name, UnicodeString &dest,
238 IDNAInfo &info, UErrorCode &errorCode) const {
239 process(name, false, true, dest, info, errorCode);
240 if( dest.length()>=254 && (info.errors&UIDNA_ERROR_DOMAIN_NAME_TOO_LONG)==0 &&
241 isASCIIString(dest) &&
242 (dest.length()>254 || dest[253]!=0x2e)
243 ) {
244 info.errors|=UIDNA_ERROR_DOMAIN_NAME_TOO_LONG;
245 }
246 return dest;
247}
248
249UnicodeString &
250UTS46::nameToUnicode(const UnicodeString &name, UnicodeString &dest,
251 IDNAInfo &info, UErrorCode &errorCode) const {
252 return process(name, false, false, dest, info, errorCode);
253}
254
255void
256UTS46::labelToASCII_UTF8(StringPiece label, ByteSink &dest,
257 IDNAInfo &info, UErrorCode &errorCode) const {
258 processUTF8(label, true, true, dest, info, errorCode);
259}
260
261void
262UTS46::labelToUnicodeUTF8(StringPiece label, ByteSink &dest,
263 IDNAInfo &info, UErrorCode &errorCode) const {
264 processUTF8(label, true, false, dest, info, errorCode);
265}
266
267void
268UTS46::nameToASCII_UTF8(StringPiece name, ByteSink &dest,
269 IDNAInfo &info, UErrorCode &errorCode) const {
270 processUTF8(name, false, true, dest, info, errorCode);
271}
272
273void
274UTS46::nameToUnicodeUTF8(StringPiece name, ByteSink &dest,
275 IDNAInfo &info, UErrorCode &errorCode) const {
276 processUTF8(name, false, false, dest, info, errorCode);
277}
278
279// UTS #46 data for ASCII characters.
280// The normalizer (using uts46.nrm) maps uppercase ASCII letters to lowercase
281// and passes through all other ASCII characters.
282// If UIDNA_USE_STD3_RULES is set, then non-LDH characters are disallowed
283// using this data.
284// The ASCII fastpath also uses this data.
285// Values: -1=disallowed 0==valid 1==mapped (lowercase)
286static const int8_t asciiData[128]={
287 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
288 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,
289 // 002D..002E; valid # HYPHEN-MINUS..FULL STOP
290 -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 0, 0, -1,
291 // 0030..0039; valid # DIGIT ZERO..DIGIT NINE
292 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1, -1,
293 // 0041..005A; mapped # LATIN CAPITAL LETTER A..LATIN CAPITAL LETTER Z
294 -1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
295 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, -1, -1, -1, -1, -1,
296 // 0061..007A; valid # LATIN SMALL LETTER A..LATIN SMALL LETTER Z
297 -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
298 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, -1, -1, -1, -1, -1
299};
300
301UnicodeString &
302UTS46::process(const UnicodeString &src,
303 UBool isLabel, UBool toASCII,
304 UnicodeString &dest,
305 IDNAInfo &info, UErrorCode &errorCode) const {
306 // uts46Norm2.normalize() would do all of this error checking and setup,
307 // but with the ASCII fastpath we do not always call it, and do not
308 // call it first.
309 if(U_FAILURE(errorCode)) {
310 dest.setToBogus();
311 return dest;
312 }
313 const char16_t *srcArray=src.getBuffer();
314 if(&dest==&src || srcArray==nullptr) {
315 errorCode=U_ILLEGAL_ARGUMENT_ERROR;
316 dest.setToBogus();
317 return dest;
318 }
319 // Arguments are fine, reset output values.
320 dest.remove();
321 info.reset();
322 int32_t srcLength=src.length();
323 if(srcLength==0) {
324 info.errors|=UIDNA_ERROR_EMPTY_LABEL;
325 return dest;
326 }
327 char16_t *destArray=dest.getBuffer(srcLength);
328 if(destArray==nullptr) {
329 errorCode=U_MEMORY_ALLOCATION_ERROR;
330 return dest;
331 }
332 // ASCII fastpath
333 UBool disallowNonLDHDot=(options&UIDNA_USE_STD3_RULES)!=0;
334 int32_t labelStart=0;
335 int32_t i;
336 for(i=0;; ++i) {
337 if(i==srcLength) {
338 if(toASCII) {
339 if((i-labelStart)>63) {
340 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
341 }
342 // There is a trailing dot if labelStart==i.
343 if(!isLabel && i>=254 && (i>254 || labelStart<i)) {
344 info.errors|=UIDNA_ERROR_DOMAIN_NAME_TOO_LONG;
345 }
346 }
347 info.errors|=info.labelErrors;
348 dest.releaseBuffer(i);
349 return dest;
350 }
351 char16_t c=srcArray[i];
352 if(c>0x7f) {
353 break;
354 }
355 int cData=asciiData[c];
356 if(cData>0) {
357 destArray[i]=c+0x20; // Lowercase an uppercase ASCII letter.
358 } else if(cData<0 && disallowNonLDHDot) {
359 break; // Replacing with U+FFFD can be complicated for toASCII.
360 } else {
361 destArray[i]=c;
362 if(c==0x2d) { // hyphen
363 if(i==(labelStart+3) && srcArray[i-1]==0x2d) {
364 // "??--..." is Punycode or forbidden.
365 ++i; // '-' was copied to dest already
366 break;
367 }
368 if(i==labelStart) {
369 // label starts with "-"
370 info.labelErrors|=UIDNA_ERROR_LEADING_HYPHEN;
371 }
372 if((i+1)==srcLength || srcArray[i+1]==0x2e) {
373 // label ends with "-"
374 info.labelErrors|=UIDNA_ERROR_TRAILING_HYPHEN;
375 }
376 } else if(c==0x2e) { // dot
377 if(isLabel) {
378 // Replacing with U+FFFD can be complicated for toASCII.
379 ++i; // '.' was copied to dest already
380 break;
381 }
382 if(i==labelStart) {
383 info.labelErrors|=UIDNA_ERROR_EMPTY_LABEL;
384 }
385 if(toASCII && (i-labelStart)>63) {
386 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
387 }
388 info.errors|=info.labelErrors;
389 info.labelErrors=0;
390 labelStart=i+1;
391 }
392 }
393 }
394 info.errors|=info.labelErrors;
395 dest.releaseBuffer(i);
396 processUnicode(src, labelStart, i, isLabel, toASCII, dest, info, errorCode);
397 if( info.isBiDi && U_SUCCESS(errorCode) && (info.errors&severeErrors)==0 &&
398 (!info.isOkBiDi || (labelStart>0 && !isASCIIOkBiDi(dest.getBuffer(), labelStart)))
399 ) {
400 info.errors|=UIDNA_ERROR_BIDI;
401 }
402 return dest;
403}
404
405void
406UTS46::processUTF8(StringPiece src,
407 UBool isLabel, UBool toASCII,
408 ByteSink &dest,
409 IDNAInfo &info, UErrorCode &errorCode) const {
410 if(U_FAILURE(errorCode)) {
411 return;
412 }
413 const char *srcArray=src.data();
414 int32_t srcLength=src.length();
415 if(srcArray==nullptr && srcLength!=0) {
416 errorCode=U_ILLEGAL_ARGUMENT_ERROR;
417 return;
418 }
419 // Arguments are fine, reset output values.
420 info.reset();
421 if(srcLength==0) {
422 info.errors|=UIDNA_ERROR_EMPTY_LABEL;
423 dest.Flush();
424 return;
425 }
426 UnicodeString destString;
427 int32_t labelStart=0;
428 if(srcLength<=256) { // length of stackArray[]
429 // ASCII fastpath
430 char stackArray[256];
431 int32_t destCapacity;
432 char *destArray=dest.GetAppendBuffer(srcLength, srcLength+20,
433 stackArray, UPRV_LENGTHOF(stackArray), &destCapacity);
434 UBool disallowNonLDHDot=(options&UIDNA_USE_STD3_RULES)!=0;
435 int32_t i;
436 for(i=0;; ++i) {
437 if(i==srcLength) {
438 if(toASCII) {
439 if((i-labelStart)>63) {
440 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
441 }
442 // There is a trailing dot if labelStart==i.
443 if(!isLabel && i>=254 && (i>254 || labelStart<i)) {
444 info.errors|=UIDNA_ERROR_DOMAIN_NAME_TOO_LONG;
445 }
446 }
447 info.errors|=info.labelErrors;
448 dest.Append(destArray, i);
449 dest.Flush();
450 return;
451 }
452 char c=srcArray[i];
453 if((int8_t)c<0) { // (uint8_t)c>0x7f
454 break;
455 }
456 int cData=asciiData[(int)c]; // Cast: gcc warns about indexing with a char.
457 if(cData>0) {
458 destArray[i]=c+0x20; // Lowercase an uppercase ASCII letter.
459 } else if(cData<0 && disallowNonLDHDot) {
460 break; // Replacing with U+FFFD can be complicated for toASCII.
461 } else {
462 destArray[i]=c;
463 if(c==0x2d) { // hyphen
464 if(i==(labelStart+3) && srcArray[i-1]==0x2d) {
465 // "??--..." is Punycode or forbidden.
466 break;
467 }
468 if(i==labelStart) {
469 // label starts with "-"
470 info.labelErrors|=UIDNA_ERROR_LEADING_HYPHEN;
471 }
472 if((i+1)==srcLength || srcArray[i+1]==0x2e) {
473 // label ends with "-"
474 info.labelErrors|=UIDNA_ERROR_TRAILING_HYPHEN;
475 }
476 } else if(c==0x2e) { // dot
477 if(isLabel) {
478 break; // Replacing with U+FFFD can be complicated for toASCII.
479 }
480 if(i==labelStart) {
481 info.labelErrors|=UIDNA_ERROR_EMPTY_LABEL;
482 }
483 if(toASCII && (i-labelStart)>63) {
484 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
485 }
486 info.errors|=info.labelErrors;
487 info.labelErrors=0;
488 labelStart=i+1;
489 }
490 }
491 }
492 info.errors|=info.labelErrors;
493 // Convert the processed ASCII prefix of the current label to UTF-16.
494 int32_t mappingStart=i-labelStart;
495 destString=UnicodeString::fromUTF8(StringPiece(destArray+labelStart, mappingStart));
496 // Output the previous ASCII labels and process the rest of src in UTF-16.
497 dest.Append(destArray, labelStart);
498 processUnicode(UnicodeString::fromUTF8(StringPiece(src, labelStart)), 0, mappingStart,
499 isLabel, toASCII,
500 destString, info, errorCode);
501 } else {
502 // src is too long for the ASCII fastpath implementation.
503 processUnicode(UnicodeString::fromUTF8(src), 0, 0,
504 isLabel, toASCII,
505 destString, info, errorCode);
506 }
507 destString.toUTF8(dest); // calls dest.Flush()
508 if(toASCII && !isLabel) {
509 // length==labelStart==254 means that there is a trailing dot (ok) and
510 // destString is empty (do not index at 253-labelStart).
511 int32_t length=labelStart+destString.length();
512 if( length>=254 && isASCIIString(destString) &&
513 (length>254 ||
514 (labelStart<254 && destString[253-labelStart]!=0x2e))
515 ) {
516 info.errors|=UIDNA_ERROR_DOMAIN_NAME_TOO_LONG;
517 }
518 }
519 if( info.isBiDi && U_SUCCESS(errorCode) && (info.errors&severeErrors)==0 &&
520 (!info.isOkBiDi || (labelStart>0 && !isASCIIOkBiDi(srcArray, labelStart)))
521 ) {
522 info.errors|=UIDNA_ERROR_BIDI;
523 }
524}
525
526UnicodeString &
527UTS46::processUnicode(const UnicodeString &src,
528 int32_t labelStart, int32_t mappingStart,
529 UBool isLabel, UBool toASCII,
530 UnicodeString &dest,
531 IDNAInfo &info, UErrorCode &errorCode) const {
532 if(mappingStart==0) {
533 uts46Norm2.normalize(src, dest, errorCode);
534 } else {
535 uts46Norm2.normalizeSecondAndAppend(dest, src.tempSubString(mappingStart), errorCode);
536 }
537 if(U_FAILURE(errorCode)) {
538 return dest;
539 }
540 UBool doMapDevChars=
541 toASCII ? (options&UIDNA_NONTRANSITIONAL_TO_ASCII)==0 :
542 (options&UIDNA_NONTRANSITIONAL_TO_UNICODE)==0;
543 const char16_t *destArray=dest.getBuffer();
544 int32_t destLength=dest.length();
545 int32_t labelLimit=labelStart;
546 while(labelLimit<destLength) {
547 char16_t c=destArray[labelLimit];
548 if(c==0x2e && !isLabel) {
549 int32_t labelLength=labelLimit-labelStart;
550 int32_t newLength=processLabel(dest, labelStart, labelLength,
551 toASCII, info, errorCode);
552 info.errors|=info.labelErrors;
553 info.labelErrors=0;
554 if(U_FAILURE(errorCode)) {
555 return dest;
556 }
557 destArray=dest.getBuffer();
558 destLength+=newLength-labelLength;
559 labelLimit=labelStart+=newLength+1;
560 continue;
561 } else if(c<0xdf) {
562 // pass
563 } else if(c<=0x200d && (c==0xdf || c==0x3c2 || c>=0x200c)) {
564 info.isTransDiff=true;
565 if(doMapDevChars) {
566 destLength=mapDevChars(dest, labelStart, labelLimit, errorCode);
567 if(U_FAILURE(errorCode)) {
568 return dest;
569 }
570 destArray=dest.getBuffer();
571 // All deviation characters have been mapped, no need to check for them again.
572 doMapDevChars=false;
573 // Do not increment labelLimit in case c was removed.
574 continue;
575 }
576 } else if(U16_IS_SURROGATE(c)) {
577 if(U16_IS_SURROGATE_LEAD(c) ?
578 (labelLimit+1)==destLength || !U16_IS_TRAIL(destArray[labelLimit+1]) :
579 labelLimit==labelStart || !U16_IS_LEAD(destArray[labelLimit-1])) {
580 // Map an unpaired surrogate to U+FFFD before normalization so that when
581 // that removes characters we do not turn two unpaired ones into a pair.
582 info.labelErrors|=UIDNA_ERROR_DISALLOWED;
583 dest.setCharAt(labelLimit, 0xfffd);
584 destArray=dest.getBuffer();
585 }
586 }
587 ++labelLimit;
588 }
589 // Permit an empty label at the end (0<labelStart==labelLimit==destLength is ok)
590 // but not an empty label elsewhere nor a completely empty domain name.
591 // processLabel() sets UIDNA_ERROR_EMPTY_LABEL when labelLength==0.
592 if(0==labelStart || labelStart<labelLimit) {
593 processLabel(dest, labelStart, labelLimit-labelStart,
594 toASCII, info, errorCode);
595 info.errors|=info.labelErrors;
596 }
597 return dest;
598}
599
600int32_t
601UTS46::mapDevChars(UnicodeString &dest, int32_t labelStart, int32_t mappingStart,
602 UErrorCode &errorCode) const {
603 if(U_FAILURE(errorCode)) {
604 return 0;
605 }
606 int32_t length=dest.length();
607 char16_t *s=dest.getBuffer(dest[mappingStart]==0xdf ? length+1 : length);
608 if(s==nullptr) {
609 errorCode=U_MEMORY_ALLOCATION_ERROR;
610 return length;
611 }
612 int32_t capacity=dest.getCapacity();
613 UBool didMapDevChars=false;
614 int32_t readIndex=mappingStart, writeIndex=mappingStart;
615 do {
616 char16_t c=s[readIndex++];
617 switch(c) {
618 case 0xdf:
619 // Map sharp s to ss.
620 didMapDevChars=true;
621 s[writeIndex++]=0x73; // Replace sharp s with first s.
622 // Insert second s and account for possible buffer reallocation.
623 if(writeIndex==readIndex) {
624 if(length==capacity) {
625 dest.releaseBuffer(length);
626 s=dest.getBuffer(length+1);
627 if(s==nullptr) {
628 errorCode=U_MEMORY_ALLOCATION_ERROR;
629 return length;
630 }
631 capacity=dest.getCapacity();
632 }
633 u_memmove(s+writeIndex+1, s+writeIndex, length-writeIndex);
634 ++readIndex;
635 }
636 s[writeIndex++]=0x73;
637 ++length;
638 break;
639 case 0x3c2: // Map final sigma to nonfinal sigma.
640 didMapDevChars=true;
641 s[writeIndex++]=0x3c3;
642 break;
643 case 0x200c: // Ignore/remove ZWNJ.
644 case 0x200d: // Ignore/remove ZWJ.
645 didMapDevChars=true;
646 --length;
647 break;
648 default:
649 // Only really necessary if writeIndex was different from readIndex.
650 s[writeIndex++]=c;
651 break;
652 }
653 } while(writeIndex<length);
654 dest.releaseBuffer(length);
655 if(didMapDevChars) {
656 // Mapping deviation characters might have resulted in an un-NFC string.
657 // We could use either the NFC or the UTS #46 normalizer.
658 // By using the UTS #46 normalizer again, we avoid having to load a second .nrm data file.
659 UnicodeString normalized;
660 uts46Norm2.normalize(dest.tempSubString(labelStart), normalized, errorCode);
661 if(U_SUCCESS(errorCode)) {
662 dest.replace(labelStart, 0x7fffffff, normalized);
663 if(dest.isBogus()) {
664 errorCode=U_MEMORY_ALLOCATION_ERROR;
665 }
666 return dest.length();
667 }
668 }
669 return length;
670}
671
672// Some non-ASCII characters are equivalent to sequences with
673// non-LDH ASCII characters. To find them:
674// grep disallowed_STD3_valid IdnaMappingTable.txt (or uts46.txt)
675static inline UBool
676isNonASCIIDisallowedSTD3Valid(UChar32 c) {
677 return c==0x2260 || c==0x226E || c==0x226F;
678}
679
680// Replace the label in dest with the label string, if the label was modified.
681// If &label==&dest then the label was modified in-place and labelLength
682// is the new label length, different from label.length().
683// If &label!=&dest then labelLength==label.length().
684// Returns labelLength (= the new label length).
685static int32_t
686replaceLabel(UnicodeString &dest, int32_t destLabelStart, int32_t destLabelLength,
687 const UnicodeString &label, int32_t labelLength, UErrorCode &errorCode) {
688 if(U_FAILURE(errorCode)) {
689 return 0;
690 }
691 if(&label!=&dest) {
692 dest.replace(destLabelStart, destLabelLength, label);
693 if(dest.isBogus()) {
694 errorCode=U_MEMORY_ALLOCATION_ERROR;
695 return 0;
696 }
697 }
698 return labelLength;
699}
700
701int32_t
702UTS46::processLabel(UnicodeString &dest,
703 int32_t labelStart, int32_t labelLength,
704 UBool toASCII,
705 IDNAInfo &info, UErrorCode &errorCode) const {
706 if(U_FAILURE(errorCode)) {
707 return 0;
708 }
709 UnicodeString fromPunycode;
710 UnicodeString *labelString;
711 const char16_t *label=dest.getBuffer()+labelStart;
712 int32_t destLabelStart=labelStart;
713 int32_t destLabelLength=labelLength;
714 UBool wasPunycode;
715 if(labelLength>=4 && label[0]==0x78 && label[1]==0x6e && label[2]==0x2d && label[3]==0x2d) {
716 // Label starts with "xn--", try to un-Punycode it.
717 // In IDNA2008, labels like "xn--" (decodes to an empty string) and
718 // "xn--ASCII-" (decodes to just "ASCII") fail the round-trip validation from
719 // comparing the ToUnicode input with the back-to-ToASCII output.
720 // They are alternate encodings of the respective ASCII labels.
721 // Ignore "xn---" here: It will fail Punycode.decode() which logically comes before
722 // the round-trip verification.
723 if(labelLength==4 || (labelLength>5 && label[labelLength-1]==u'-')) {
724 info.labelErrors|=UIDNA_ERROR_INVALID_ACE_LABEL;
725 return markBadACELabel(dest, labelStart, labelLength, toASCII, info, errorCode);
726 }
727 wasPunycode=true;
728 char16_t *unicodeBuffer=fromPunycode.getBuffer(-1); // capacity==-1: most labels should fit
729 if(unicodeBuffer==nullptr) {
730 // Should never occur if we used capacity==-1 which uses the internal buffer.
731 errorCode=U_MEMORY_ALLOCATION_ERROR;
732 return labelLength;
733 }
734 UErrorCode punycodeErrorCode=U_ZERO_ERROR;
735 int32_t unicodeLength=u_strFromPunycode(label+4, labelLength-4,
736 unicodeBuffer, fromPunycode.getCapacity(),
737 nullptr, &punycodeErrorCode);
738 if(punycodeErrorCode==U_BUFFER_OVERFLOW_ERROR) {
739 fromPunycode.releaseBuffer(0);
740 unicodeBuffer=fromPunycode.getBuffer(unicodeLength);
741 if(unicodeBuffer==nullptr) {
742 errorCode=U_MEMORY_ALLOCATION_ERROR;
743 return labelLength;
744 }
745 punycodeErrorCode=U_ZERO_ERROR;
746 unicodeLength=u_strFromPunycode(label+4, labelLength-4,
747 unicodeBuffer, fromPunycode.getCapacity(),
748 nullptr, &punycodeErrorCode);
749 }
750 fromPunycode.releaseBuffer(unicodeLength);
751 if(U_FAILURE(punycodeErrorCode)) {
752 info.labelErrors|=UIDNA_ERROR_PUNYCODE;
753 return markBadACELabel(dest, labelStart, labelLength, toASCII, info, errorCode);
754 }
755 // Check for NFC, and for characters that are not
756 // valid or deviation characters according to the normalizer.
757 // If there is something wrong, then the string will change.
758 // Note that the normalizer passes through non-LDH ASCII and deviation characters.
759 // Deviation characters are ok in Punycode even in transitional processing.
760 // In the code further below, if we find non-LDH ASCII and we have UIDNA_USE_STD3_RULES
761 // then we will set UIDNA_ERROR_INVALID_ACE_LABEL there too.
762 UBool isValid=uts46Norm2.isNormalized(fromPunycode, errorCode);
763 if(U_FAILURE(errorCode)) {
764 return labelLength;
765 }
766 if(!isValid) {
767 info.labelErrors|=UIDNA_ERROR_INVALID_ACE_LABEL;
768 return markBadACELabel(dest, labelStart, labelLength, toASCII, info, errorCode);
769 }
770 labelString=&fromPunycode;
771 label=fromPunycode.getBuffer();
772 labelStart=0;
773 labelLength=fromPunycode.length();
774 } else {
775 wasPunycode=false;
776 labelString=&dest;
777 }
778 // Validity check
779 if(labelLength==0) {
780 info.labelErrors|=UIDNA_ERROR_EMPTY_LABEL;
781 return replaceLabel(dest, destLabelStart, destLabelLength,
782 *labelString, labelLength, errorCode);
783 }
784 // labelLength>0
785 if(labelLength>=4 && label[2]==0x2d && label[3]==0x2d) {
786 // label starts with "??--"
787 info.labelErrors|=UIDNA_ERROR_HYPHEN_3_4;
788 }
789 if(label[0]==0x2d) {
790 // label starts with "-"
791 info.labelErrors|=UIDNA_ERROR_LEADING_HYPHEN;
792 }
793 if(label[labelLength-1]==0x2d) {
794 // label ends with "-"
795 info.labelErrors|=UIDNA_ERROR_TRAILING_HYPHEN;
796 }
797 // If the label was not a Punycode label, then it was the result of
798 // mapping, normalization and label segmentation.
799 // If the label was in Punycode, then we mapped it again above
800 // and checked its validity.
801 // Now we handle the STD3 restriction to LDH characters (if set)
802 // and we look for U+FFFD which indicates disallowed characters
803 // in a non-Punycode label or U+FFFD itself in a Punycode label.
804 // We also check for dots which can come from the input to a single-label function.
805 // Ok to cast away const because we own the UnicodeString.
806 char16_t *s=(char16_t *)label;
807 const char16_t *limit=label+labelLength;
808 char16_t oredChars=0;
809 // If we enforce STD3 rules, then ASCII characters other than LDH and dot are disallowed.
810 UBool disallowNonLDHDot=(options&UIDNA_USE_STD3_RULES)!=0;
811 do {
812 char16_t c=*s;
813 if(c<=0x7f) {
814 if(c==0x2e) {
815 info.labelErrors|=UIDNA_ERROR_LABEL_HAS_DOT;
816 *s=0xfffd;
817 } else if(disallowNonLDHDot && asciiData[c]<0) {
818 info.labelErrors|=UIDNA_ERROR_DISALLOWED;
819 *s=0xfffd;
820 }
821 } else {
822 oredChars|=c;
823 if(disallowNonLDHDot && isNonASCIIDisallowedSTD3Valid(c)) {
824 info.labelErrors|=UIDNA_ERROR_DISALLOWED;
825 *s=0xfffd;
826 } else if(c==0xfffd) {
827 info.labelErrors|=UIDNA_ERROR_DISALLOWED;
828 }
829 }
830 ++s;
831 } while(s<limit);
832 // Check for a leading combining mark after other validity checks
833 // so that we don't report UIDNA_ERROR_DISALLOWED for the U+FFFD from here.
834 UChar32 c;
835 int32_t cpLength=0;
836 // "Unsafe" is ok because unpaired surrogates were mapped to U+FFFD.
837 U16_NEXT_UNSAFE(label, cpLength, c);
838 if((U_GET_GC_MASK(c)&U_GC_M_MASK)!=0) {
839 info.labelErrors|=UIDNA_ERROR_LEADING_COMBINING_MARK;
840 labelString->replace(labelStart, cpLength, (char16_t)0xfffd);
841 label=labelString->getBuffer()+labelStart;
842 labelLength+=1-cpLength;
843 if(labelString==&dest) {
844 destLabelLength=labelLength;
845 }
846 }
847 if((info.labelErrors&severeErrors)==0) {
848 // Do contextual checks only if we do not have U+FFFD from a severe error
849 // because U+FFFD can make these checks fail.
850 if((options&UIDNA_CHECK_BIDI)!=0 && (!info.isBiDi || info.isOkBiDi)) {
851 checkLabelBiDi(label, labelLength, info);
852 }
853 if( (options&UIDNA_CHECK_CONTEXTJ)!=0 && (oredChars&0x200c)==0x200c &&
854 !isLabelOkContextJ(label, labelLength)
855 ) {
856 info.labelErrors|=UIDNA_ERROR_CONTEXTJ;
857 }
858 if((options&UIDNA_CHECK_CONTEXTO)!=0 && oredChars>=0xb7) {
859 checkLabelContextO(label, labelLength, info);
860 }
861 if(toASCII) {
862 if(wasPunycode) {
863 // Leave a Punycode label unchanged if it has no severe errors.
864 if(destLabelLength>63) {
865 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
866 }
867 return destLabelLength;
868 } else if(oredChars>=0x80) {
869 // Contains non-ASCII characters.
870 UnicodeString punycode;
871 char16_t *buffer=punycode.getBuffer(63); // 63==maximum DNS label length
872 if(buffer==nullptr) {
873 errorCode=U_MEMORY_ALLOCATION_ERROR;
874 return destLabelLength;
875 }
876 buffer[0]=0x78; // Write "xn--".
877 buffer[1]=0x6e;
878 buffer[2]=0x2d;
879 buffer[3]=0x2d;
880 int32_t punycodeLength=u_strToPunycode(label, labelLength,
881 buffer+4, punycode.getCapacity()-4,
882 nullptr, &errorCode);
883 if(errorCode==U_BUFFER_OVERFLOW_ERROR) {
884 errorCode=U_ZERO_ERROR;
885 punycode.releaseBuffer(4);
886 buffer=punycode.getBuffer(4+punycodeLength);
887 if(buffer==nullptr) {
888 errorCode=U_MEMORY_ALLOCATION_ERROR;
889 return destLabelLength;
890 }
891 punycodeLength=u_strToPunycode(label, labelLength,
892 buffer+4, punycode.getCapacity()-4,
893 nullptr, &errorCode);
894 }
895 punycodeLength+=4;
896 punycode.releaseBuffer(punycodeLength);
897 if(U_FAILURE(errorCode)) {
898 return destLabelLength;
899 }
900 if(punycodeLength>63) {
901 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
902 }
903 return replaceLabel(dest, destLabelStart, destLabelLength,
904 punycode, punycodeLength, errorCode);
905 } else {
906 // all-ASCII label
907 if(labelLength>63) {
908 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
909 }
910 }
911 }
912 } else {
913 // If a Punycode label has severe errors,
914 // then leave it but make sure it does not look valid.
915 if(wasPunycode) {
916 info.labelErrors|=UIDNA_ERROR_INVALID_ACE_LABEL;
917 return markBadACELabel(dest, destLabelStart, destLabelLength, toASCII, info, errorCode);
918 }
919 }
920 return replaceLabel(dest, destLabelStart, destLabelLength,
921 *labelString, labelLength, errorCode);
922}
923
924// Make sure an ACE label does not look valid.
925// Append U+FFFD if the label has only LDH characters.
926// If UIDNA_USE_STD3_RULES, also replace disallowed ASCII characters with U+FFFD.
927int32_t
928UTS46::markBadACELabel(UnicodeString &dest,
929 int32_t labelStart, int32_t labelLength,
930 UBool toASCII, IDNAInfo &info, UErrorCode &errorCode) const {
931 if(U_FAILURE(errorCode)) {
932 return 0;
933 }
934 UBool disallowNonLDHDot=(options&UIDNA_USE_STD3_RULES)!=0;
935 UBool isASCII=true;
936 UBool onlyLDH=true;
937 const char16_t *label=dest.getBuffer()+labelStart;
938 const char16_t *limit=label+labelLength;
939 // Start after the initial "xn--".
940 // Ok to cast away const because we own the UnicodeString.
941 for(char16_t *s=const_cast<char16_t *>(label+4); s<limit; ++s) {
942 char16_t c=*s;
943 if(c<=0x7f) {
944 if(c==0x2e) {
945 info.labelErrors|=UIDNA_ERROR_LABEL_HAS_DOT;
946 *s=0xfffd;
947 isASCII=onlyLDH=false;
948 } else if(asciiData[c]<0) {
949 onlyLDH=false;
950 if(disallowNonLDHDot) {
951 *s=0xfffd;
952 isASCII=false;
953 }
954 }
955 } else {
956 isASCII=onlyLDH=false;
957 }
958 }
959 if(onlyLDH) {
960 dest.insert(labelStart+labelLength, (char16_t)0xfffd);
961 if(dest.isBogus()) {
962 errorCode=U_MEMORY_ALLOCATION_ERROR;
963 return 0;
964 }
965 ++labelLength;
966 } else {
967 if(toASCII && isASCII && labelLength>63) {
968 info.labelErrors|=UIDNA_ERROR_LABEL_TOO_LONG;
969 }
970 }
971 return labelLength;
972}
973
974const uint32_t L_MASK=U_MASK(U_LEFT_TO_RIGHT);
975const uint32_t R_AL_MASK=U_MASK(U_RIGHT_TO_LEFT)|U_MASK(U_RIGHT_TO_LEFT_ARABIC);
976const uint32_t L_R_AL_MASK=L_MASK|R_AL_MASK;
977
978const uint32_t R_AL_AN_MASK=R_AL_MASK|U_MASK(U_ARABIC_NUMBER);
979
980const uint32_t EN_AN_MASK=U_MASK(U_EUROPEAN_NUMBER)|U_MASK(U_ARABIC_NUMBER);
981const uint32_t R_AL_EN_AN_MASK=R_AL_MASK|EN_AN_MASK;
982const uint32_t L_EN_MASK=L_MASK|U_MASK(U_EUROPEAN_NUMBER);
983
984const uint32_t ES_CS_ET_ON_BN_NSM_MASK=
985 U_MASK(U_EUROPEAN_NUMBER_SEPARATOR)|
986 U_MASK(U_COMMON_NUMBER_SEPARATOR)|
987 U_MASK(U_EUROPEAN_NUMBER_TERMINATOR)|
988 U_MASK(U_OTHER_NEUTRAL)|
989 U_MASK(U_BOUNDARY_NEUTRAL)|
990 U_MASK(U_DIR_NON_SPACING_MARK);
991const uint32_t L_EN_ES_CS_ET_ON_BN_NSM_MASK=L_EN_MASK|ES_CS_ET_ON_BN_NSM_MASK;
992const uint32_t R_AL_AN_EN_ES_CS_ET_ON_BN_NSM_MASK=R_AL_MASK|EN_AN_MASK|ES_CS_ET_ON_BN_NSM_MASK;
993
994// We scan the whole label and check both for whether it contains RTL characters
995// and whether it passes the BiDi Rule.
996// In a BiDi domain name, all labels must pass the BiDi Rule, but we might find
997// that a domain name is a BiDi domain name (has an RTL label) only after
998// processing several earlier labels.
999void
1000UTS46::checkLabelBiDi(const char16_t *label, int32_t labelLength, IDNAInfo &info) const {
1001 // IDNA2008 BiDi rule
1002 // Get the directionality of the first character.
1003 UChar32 c;
1004 int32_t i=0;
1005 U16_NEXT_UNSAFE(label, i, c);
1006 uint32_t firstMask=U_MASK(u_charDirection(c));
1007 // 1. The first character must be a character with BIDI property L, R
1008 // or AL. If it has the R or AL property, it is an RTL label; if it
1009 // has the L property, it is an LTR label.
1010 if((firstMask&~L_R_AL_MASK)!=0) {
1011 info.isOkBiDi=false;
1012 }
1013 // Get the directionality of the last non-NSM character.
1014 uint32_t lastMask;
1015 for(;;) {
1016 if(i>=labelLength) {
1017 lastMask=firstMask;
1018 break;
1019 }
1020 U16_PREV_UNSAFE(label, labelLength, c);
1021 UCharDirection dir=u_charDirection(c);
1022 if(dir!=U_DIR_NON_SPACING_MARK) {
1023 lastMask=U_MASK(dir);
1024 break;
1025 }
1026 }
1027 // 3. In an RTL label, the end of the label must be a character with
1028 // BIDI property R, AL, EN or AN, followed by zero or more
1029 // characters with BIDI property NSM.
1030 // 6. In an LTR label, the end of the label must be a character with
1031 // BIDI property L or EN, followed by zero or more characters with
1032 // BIDI property NSM.
1033 if( (firstMask&L_MASK)!=0 ?
1034 (lastMask&~L_EN_MASK)!=0 :
1035 (lastMask&~R_AL_EN_AN_MASK)!=0
1036 ) {
1037 info.isOkBiDi=false;
1038 }
1039 // Add the directionalities of the intervening characters.
1040 uint32_t mask=firstMask|lastMask;
1041 while(i<labelLength) {
1042 U16_NEXT_UNSAFE(label, i, c);
1043 mask|=U_MASK(u_charDirection(c));
1044 }
1045 if(firstMask&L_MASK) {
1046 // 5. In an LTR label, only characters with the BIDI properties L, EN,
1047 // ES, CS, ET, ON, BN and NSM are allowed.
1048 if((mask&~L_EN_ES_CS_ET_ON_BN_NSM_MASK)!=0) {
1049 info.isOkBiDi=false;
1050 }
1051 } else {
1052 // 2. In an RTL label, only characters with the BIDI properties R, AL,
1053 // AN, EN, ES, CS, ET, ON, BN and NSM are allowed.
1054 if((mask&~R_AL_AN_EN_ES_CS_ET_ON_BN_NSM_MASK)!=0) {
1055 info.isOkBiDi=false;
1056 }
1057 // 4. In an RTL label, if an EN is present, no AN may be present, and
1058 // vice versa.
1059 if((mask&EN_AN_MASK)==EN_AN_MASK) {
1060 info.isOkBiDi=false;
1061 }
1062 }
1063 // An RTL label is a label that contains at least one character of type
1064 // R, AL or AN. [...]
1065 // A "BIDI domain name" is a domain name that contains at least one RTL
1066 // label. [...]
1067 // The following rule, consisting of six conditions, applies to labels
1068 // in BIDI domain names.
1069 if((mask&R_AL_AN_MASK)!=0) {
1070 info.isBiDi=true;
1071 }
1072}
1073
1074// Special code for the ASCII prefix of a BiDi domain name.
1075// The ASCII prefix is all-LTR.
1076
1077// IDNA2008 BiDi rule, parts relevant to ASCII labels:
1078// 1. The first character must be a character with BIDI property L [...]
1079// 5. In an LTR label, only characters with the BIDI properties L, EN,
1080// ES, CS, ET, ON, BN and NSM are allowed.
1081// 6. In an LTR label, the end of the label must be a character with
1082// BIDI property L or EN [...]
1083
1084// UTF-16 version, called for mapped ASCII prefix.
1085// Cannot contain uppercase A-Z.
1086// s[length-1] must be the trailing dot.
1087static UBool
1088isASCIIOkBiDi(const char16_t *s, int32_t length) {
1089 int32_t labelStart=0;
1090 for(int32_t i=0; i<length; ++i) {
1091 char16_t c=s[i];
1092 if(c==0x2e) { // dot
1093 if(i>labelStart) {
1094 c=s[i-1];
1095 if(!(0x61<=c && c<=0x7a) && !(0x30<=c && c<=0x39)) {
1096 // Last character in the label is not an L or EN.
1097 return false;
1098 }
1099 }
1100 labelStart=i+1;
1101 } else if(i==labelStart) {
1102 if(!(0x61<=c && c<=0x7a)) {
1103 // First character in the label is not an L.
1104 return false;
1105 }
1106 } else {
1107 if(c<=0x20 && (c>=0x1c || (9<=c && c<=0xd))) {
1108 // Intermediate character in the label is a B, S or WS.
1109 return false;
1110 }
1111 }
1112 }
1113 return true;
1114}
1115
1116// UTF-8 version, called for source ASCII prefix.
1117// Can contain uppercase A-Z.
1118// s[length-1] must be the trailing dot.
1119static UBool
1120isASCIIOkBiDi(const char *s, int32_t length) {
1121 int32_t labelStart=0;
1122 for(int32_t i=0; i<length; ++i) {
1123 char c=s[i];
1124 if(c==0x2e) { // dot
1125 if(i>labelStart) {
1126 c=s[i-1];
1127 if(!(0x61<=c && c<=0x7a) && !(0x41<=c && c<=0x5a) && !(0x30<=c && c<=0x39)) {
1128 // Last character in the label is not an L or EN.
1129 return false;
1130 }
1131 }
1132 labelStart=i+1;
1133 } else if(i==labelStart) {
1134 if(!(0x61<=c && c<=0x7a) && !(0x41<=c && c<=0x5a)) {
1135 // First character in the label is not an L.
1136 return false;
1137 }
1138 } else {
1139 if(c<=0x20 && (c>=0x1c || (9<=c && c<=0xd))) {
1140 // Intermediate character in the label is a B, S or WS.
1141 return false;
1142 }
1143 }
1144 }
1145 return true;
1146}
1147
1148UBool
1149UTS46::isLabelOkContextJ(const char16_t *label, int32_t labelLength) const {
1150 // [IDNA2008-Tables]
1151 // 200C..200D ; CONTEXTJ # ZERO WIDTH NON-JOINER..ZERO WIDTH JOINER
1152 for(int32_t i=0; i<labelLength; ++i) {
1153 if(label[i]==0x200c) {
1154 // Appendix A.1. ZERO WIDTH NON-JOINER
1155 // Rule Set:
1156 // False;
1157 // If Canonical_Combining_Class(Before(cp)) .eq. Virama Then True;
1158 // If RegExpMatch((Joining_Type:{L,D})(Joining_Type:T)*\u200C
1159 // (Joining_Type:T)*(Joining_Type:{R,D})) Then True;
1160 if(i==0) {
1161 return false;
1162 }
1163 UChar32 c;
1164 int32_t j=i;
1165 U16_PREV_UNSAFE(label, j, c);
1166 if(uts46Norm2.getCombiningClass(c)==9) {
1167 continue;
1168 }
1169 // check precontext (Joining_Type:{L,D})(Joining_Type:T)*
1170 for(;;) {
1171 UJoiningType type=ubidi_getJoiningType(c);
1172 if(type==U_JT_TRANSPARENT) {
1173 if(j==0) {
1174 return false;
1175 }
1176 U16_PREV_UNSAFE(label, j, c);
1177 } else if(type==U_JT_LEFT_JOINING || type==U_JT_DUAL_JOINING) {
1178 break; // precontext fulfilled
1179 } else {
1180 return false;
1181 }
1182 }
1183 // check postcontext (Joining_Type:T)*(Joining_Type:{R,D})
1184 for(j=i+1;;) {
1185 if(j==labelLength) {
1186 return false;
1187 }
1188 U16_NEXT_UNSAFE(label, j, c);
1189 UJoiningType type=ubidi_getJoiningType(c);
1190 if(type==U_JT_TRANSPARENT) {
1191 // just skip this character
1192 } else if(type==U_JT_RIGHT_JOINING || type==U_JT_DUAL_JOINING) {
1193 break; // postcontext fulfilled
1194 } else {
1195 return false;
1196 }
1197 }
1198 } else if(label[i]==0x200d) {
1199 // Appendix A.2. ZERO WIDTH JOINER (U+200D)
1200 // Rule Set:
1201 // False;
1202 // If Canonical_Combining_Class(Before(cp)) .eq. Virama Then True;
1203 if(i==0) {
1204 return false;
1205 }
1206 UChar32 c;
1207 int32_t j=i;
1208 U16_PREV_UNSAFE(label, j, c);
1209 if(uts46Norm2.getCombiningClass(c)!=9) {
1210 return false;
1211 }
1212 }
1213 }
1214 return true;
1215}
1216
1217void
1218UTS46::checkLabelContextO(const char16_t *label, int32_t labelLength, IDNAInfo &info) const {
1219 int32_t labelEnd=labelLength-1; // inclusive
1220 int32_t arabicDigits=0; // -1 for 066x, +1 for 06Fx
1221 for(int32_t i=0; i<=labelEnd; ++i) {
1222 UChar32 c=label[i];
1223 if(c<0xb7) {
1224 // ASCII fastpath
1225 } else if(c<=0x6f9) {
1226 if(c==0xb7) {
1227 // Appendix A.3. MIDDLE DOT (U+00B7)
1228 // Rule Set:
1229 // False;
1230 // If Before(cp) .eq. U+006C And
1231 // After(cp) .eq. U+006C Then True;
1232 if(!(0<i && label[i-1]==0x6c &&
1233 i<labelEnd && label[i+1]==0x6c)) {
1234 info.labelErrors|=UIDNA_ERROR_CONTEXTO_PUNCTUATION;
1235 }
1236 } else if(c==0x375) {
1237 // Appendix A.4. GREEK LOWER NUMERAL SIGN (KERAIA) (U+0375)
1238 // Rule Set:
1239 // False;
1240 // If Script(After(cp)) .eq. Greek Then True;
1241 UScriptCode script=USCRIPT_INVALID_CODE;
1242 if(i<labelEnd) {
1243 UErrorCode errorCode=U_ZERO_ERROR;
1244 int32_t j=i+1;
1245 U16_NEXT(label, j, labelLength, c);
1246 script=uscript_getScript(c, &errorCode);
1247 }
1248 if(script!=USCRIPT_GREEK) {
1249 info.labelErrors|=UIDNA_ERROR_CONTEXTO_PUNCTUATION;
1250 }
1251 } else if(c==0x5f3 || c==0x5f4) {
1252 // Appendix A.5. HEBREW PUNCTUATION GERESH (U+05F3)
1253 // Rule Set:
1254 // False;
1255 // If Script(Before(cp)) .eq. Hebrew Then True;
1256 //
1257 // Appendix A.6. HEBREW PUNCTUATION GERSHAYIM (U+05F4)
1258 // Rule Set:
1259 // False;
1260 // If Script(Before(cp)) .eq. Hebrew Then True;
1261 UScriptCode script=USCRIPT_INVALID_CODE;
1262 if(0<i) {
1263 UErrorCode errorCode=U_ZERO_ERROR;
1264 int32_t j=i;
1265 U16_PREV(label, 0, j, c);
1266 script=uscript_getScript(c, &errorCode);
1267 }
1268 if(script!=USCRIPT_HEBREW) {
1269 info.labelErrors|=UIDNA_ERROR_CONTEXTO_PUNCTUATION;
1270 }
1271 } else if(0x660<=c /* && c<=0x6f9 */) {
1272 // Appendix A.8. ARABIC-INDIC DIGITS (0660..0669)
1273 // Rule Set:
1274 // True;
1275 // For All Characters:
1276 // If cp .in. 06F0..06F9 Then False;
1277 // End For;
1278 //
1279 // Appendix A.9. EXTENDED ARABIC-INDIC DIGITS (06F0..06F9)
1280 // Rule Set:
1281 // True;
1282 // For All Characters:
1283 // If cp .in. 0660..0669 Then False;
1284 // End For;
1285 if(c<=0x669) {
1286 if(arabicDigits>0) {
1287 info.labelErrors|=UIDNA_ERROR_CONTEXTO_DIGITS;
1288 }
1289 arabicDigits=-1;
1290 } else if(0x6f0<=c) {
1291 if(arabicDigits<0) {
1292 info.labelErrors|=UIDNA_ERROR_CONTEXTO_DIGITS;
1293 }
1294 arabicDigits=1;
1295 }
1296 }
1297 } else if(c==0x30fb) {
1298 // Appendix A.7. KATAKANA MIDDLE DOT (U+30FB)
1299 // Rule Set:
1300 // False;
1301 // For All Characters:
1302 // If Script(cp) .in. {Hiragana, Katakana, Han} Then True;
1303 // End For;
1304 UErrorCode errorCode=U_ZERO_ERROR;
1305 for(int j=0;;) {
1306 if(j>labelEnd) {
1307 info.labelErrors|=UIDNA_ERROR_CONTEXTO_PUNCTUATION;
1308 break;
1309 }
1310 U16_NEXT(label, j, labelLength, c);
1311 UScriptCode script=uscript_getScript(c, &errorCode);
1312 if(script==USCRIPT_HIRAGANA || script==USCRIPT_KATAKANA || script==USCRIPT_HAN) {
1313 break;
1314 }
1315 }
1316 }
1317 }
1318}
1319
1320U_NAMESPACE_END
1321
1322// C API ------------------------------------------------------------------- ***
1323
1324U_NAMESPACE_USE
1325
1326U_CAPI UIDNA * U_EXPORT2
1327uidna_openUTS46(uint32_t options, UErrorCode *pErrorCode) {
1328 return reinterpret_cast<UIDNA *>(IDNA::createUTS46Instance(options, *pErrorCode));
1329}
1330
1331U_CAPI void U_EXPORT2
1332uidna_close(UIDNA *idna) {
1333 delete reinterpret_cast<IDNA *>(idna);
1334}
1335
1336static UBool
1337checkArgs(const void *label, int32_t length,
1338 void *dest, int32_t capacity,
1339 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1340 if(U_FAILURE(*pErrorCode)) {
1341 return false;
1342 }
1343 // sizeof(UIDNAInfo)=16 in the first API version.
1344 if(pInfo==nullptr || pInfo->size<16) {
1345 *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
1346 return false;
1347 }
1348 if( (label==nullptr ? length!=0 : length<-1) ||
1349 (dest==nullptr ? capacity!=0 : capacity<0) ||
1350 (dest==label && label!=nullptr)
1351 ) {
1352 *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
1353 return false;
1354 }
1355 // Set all *pInfo bytes to 0 except for the size field itself.
1356 uprv_memset(&pInfo->size+1, 0, pInfo->size-sizeof(pInfo->size));
1357 return true;
1358}
1359
1360static void
1361idnaInfoToStruct(IDNAInfo &info, UIDNAInfo *pInfo) {
1362 pInfo->isTransitionalDifferent=info.isTransitionalDifferent();
1363 pInfo->errors=info.getErrors();
1364}
1365
1366U_CAPI int32_t U_EXPORT2
1367uidna_labelToASCII(const UIDNA *idna,
1368 const char16_t *label, int32_t length,
1369 char16_t *dest, int32_t capacity,
1370 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1371 if(!checkArgs(label, length, dest, capacity, pInfo, pErrorCode)) {
1372 return 0;
1373 }
1374 UnicodeString src((UBool)(length<0), label, length);
1375 UnicodeString destString(dest, 0, capacity);
1376 IDNAInfo info;
1377 reinterpret_cast<const IDNA *>(idna)->labelToASCII(src, destString, info, *pErrorCode);
1378 idnaInfoToStruct(info, pInfo);
1379 return destString.extract(dest, capacity, *pErrorCode);
1380}
1381
1382U_CAPI int32_t U_EXPORT2
1383uidna_labelToUnicode(const UIDNA *idna,
1384 const char16_t *label, int32_t length,
1385 char16_t *dest, int32_t capacity,
1386 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1387 if(!checkArgs(label, length, dest, capacity, pInfo, pErrorCode)) {
1388 return 0;
1389 }
1390 UnicodeString src((UBool)(length<0), label, length);
1391 UnicodeString destString(dest, 0, capacity);
1392 IDNAInfo info;
1393 reinterpret_cast<const IDNA *>(idna)->labelToUnicode(src, destString, info, *pErrorCode);
1394 idnaInfoToStruct(info, pInfo);
1395 return destString.extract(dest, capacity, *pErrorCode);
1396}
1397
1398U_CAPI int32_t U_EXPORT2
1399uidna_nameToASCII(const UIDNA *idna,
1400 const char16_t *name, int32_t length,
1401 char16_t *dest, int32_t capacity,
1402 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1403 if(!checkArgs(name, length, dest, capacity, pInfo, pErrorCode)) {
1404 return 0;
1405 }
1406 UnicodeString src((UBool)(length<0), name, length);
1407 UnicodeString destString(dest, 0, capacity);
1408 IDNAInfo info;
1409 reinterpret_cast<const IDNA *>(idna)->nameToASCII(src, destString, info, *pErrorCode);
1410 idnaInfoToStruct(info, pInfo);
1411 return destString.extract(dest, capacity, *pErrorCode);
1412}
1413
1414U_CAPI int32_t U_EXPORT2
1415uidna_nameToUnicode(const UIDNA *idna,
1416 const char16_t *name, int32_t length,
1417 char16_t *dest, int32_t capacity,
1418 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1419 if(!checkArgs(name, length, dest, capacity, pInfo, pErrorCode)) {
1420 return 0;
1421 }
1422 UnicodeString src((UBool)(length<0), name, length);
1423 UnicodeString destString(dest, 0, capacity);
1424 IDNAInfo info;
1425 reinterpret_cast<const IDNA *>(idna)->nameToUnicode(src, destString, info, *pErrorCode);
1426 idnaInfoToStruct(info, pInfo);
1427 return destString.extract(dest, capacity, *pErrorCode);
1428}
1429
1430U_CAPI int32_t U_EXPORT2
1431uidna_labelToASCII_UTF8(const UIDNA *idna,
1432 const char *label, int32_t length,
1433 char *dest, int32_t capacity,
1434 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1435 if(!checkArgs(label, length, dest, capacity, pInfo, pErrorCode)) {
1436 return 0;
1437 }
1438 StringPiece src(label, length<0 ? static_cast<int32_t>(uprv_strlen(label)) : length);
1439 CheckedArrayByteSink sink(dest, capacity);
1440 IDNAInfo info;
1441 reinterpret_cast<const IDNA *>(idna)->labelToASCII_UTF8(src, sink, info, *pErrorCode);
1442 idnaInfoToStruct(info, pInfo);
1443 return u_terminateChars(dest, capacity, sink.NumberOfBytesAppended(), pErrorCode);
1444}
1445
1446U_CAPI int32_t U_EXPORT2
1447uidna_labelToUnicodeUTF8(const UIDNA *idna,
1448 const char *label, int32_t length,
1449 char *dest, int32_t capacity,
1450 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1451 if(!checkArgs(label, length, dest, capacity, pInfo, pErrorCode)) {
1452 return 0;
1453 }
1454 StringPiece src(label, length<0 ? static_cast<int32_t>(uprv_strlen(label)) : length);
1455 CheckedArrayByteSink sink(dest, capacity);
1456 IDNAInfo info;
1457 reinterpret_cast<const IDNA *>(idna)->labelToUnicodeUTF8(src, sink, info, *pErrorCode);
1458 idnaInfoToStruct(info, pInfo);
1459 return u_terminateChars(dest, capacity, sink.NumberOfBytesAppended(), pErrorCode);
1460}
1461
1462U_CAPI int32_t U_EXPORT2
1463uidna_nameToASCII_UTF8(const UIDNA *idna,
1464 const char *name, int32_t length,
1465 char *dest, int32_t capacity,
1466 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1467 if(!checkArgs(name, length, dest, capacity, pInfo, pErrorCode)) {
1468 return 0;
1469 }
1470 StringPiece src(name, length<0 ? static_cast<int32_t>(uprv_strlen(name)) : length);
1471 CheckedArrayByteSink sink(dest, capacity);
1472 IDNAInfo info;
1473 reinterpret_cast<const IDNA *>(idna)->nameToASCII_UTF8(src, sink, info, *pErrorCode);
1474 idnaInfoToStruct(info, pInfo);
1475 return u_terminateChars(dest, capacity, sink.NumberOfBytesAppended(), pErrorCode);
1476}
1477
1478U_CAPI int32_t U_EXPORT2
1479uidna_nameToUnicodeUTF8(const UIDNA *idna,
1480 const char *name, int32_t length,
1481 char *dest, int32_t capacity,
1482 UIDNAInfo *pInfo, UErrorCode *pErrorCode) {
1483 if(!checkArgs(name, length, dest, capacity, pInfo, pErrorCode)) {
1484 return 0;
1485 }
1486 StringPiece src(name, length<0 ? static_cast<int32_t>(uprv_strlen(name)) : length);
1487 CheckedArrayByteSink sink(dest, capacity);
1488 IDNAInfo info;
1489 reinterpret_cast<const IDNA *>(idna)->nameToUnicodeUTF8(src, sink, info, *pErrorCode);
1490 idnaInfoToStruct(info, pInfo);
1491 return u_terminateChars(dest, capacity, sink.NumberOfBytesAppended(), pErrorCode);
1492}
1493
1494#endif // UCONFIG_NO_IDNA
1495