1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /*
4 *******************************************************************************
5 *   Copyright (C) 2010-2014, International Business Machines
6 *   Corporation and others.  All Rights Reserved.
7 *******************************************************************************
8 *   file name:  uts46test.cpp
9 *   encoding:   UTF-8
10 *   tab size:   8 (not used)
11 *   indentation:4
12 *
13 *   created on: 2010may05
14 *   created by: Markus W. Scherer
15 */
16 
17 #include "unicode/utypes.h"
18 
19 #if !UCONFIG_NO_IDNA
20 
21 #include <string.h>
22 #include "unicode/bytestream.h"
23 #include "unicode/idna.h"
24 #include "unicode/localpointer.h"
25 #include "unicode/std_string.h"
26 #include "unicode/stringpiece.h"
27 #include "unicode/uidna.h"
28 #include "unicode/unistr.h"
29 #include "charstr.h"
30 #include "cmemory.h"
31 #include "intltest.h"
32 #include "uparse.h"
33 
34 class UTS46Test : public IntlTest {
35 public:
UTS46Test()36     UTS46Test() : trans(NULL), nontrans(NULL) {}
37     virtual ~UTS46Test();
38 
39     void runIndexedTest(int32_t index, UBool exec, const char *&name, char *par=NULL);
40     void TestAPI();
41     void TestNotSTD3();
42     void TestSomeCases();
43     void IdnaTest();
44 
45     void checkIdnaTestResult(const char *line, const char *type,
46                              const UnicodeString &expected, const UnicodeString &result,
47                              const char *status, const IDNAInfo &info);
48     void idnaTestOneLine(char *fields[][2], UErrorCode &errorCode);
49 
50 private:
51     IDNA *trans, *nontrans;
52 };
53 
createUTS46Test()54 extern IntlTest *createUTS46Test() {
55     return new UTS46Test();
56 }
57 
~UTS46Test()58 UTS46Test::~UTS46Test() {
59     delete trans;
60     delete nontrans;
61 }
62 
runIndexedTest(int32_t index,UBool exec,const char * & name,char *)63 void UTS46Test::runIndexedTest(int32_t index, UBool exec, const char *&name, char * /*par*/) {
64     if(exec) {
65         logln("TestSuite UTS46Test: ");
66         if(trans==NULL) {
67             IcuTestErrorCode errorCode(*this, "init/createUTS46Instance()");
68             uint32_t commonOptions=
69                 UIDNA_USE_STD3_RULES|UIDNA_CHECK_BIDI|
70                 UIDNA_CHECK_CONTEXTJ|UIDNA_CHECK_CONTEXTO;
71             trans=IDNA::createUTS46Instance(commonOptions, errorCode);
72             nontrans=IDNA::createUTS46Instance(
73                 commonOptions|
74                 UIDNA_NONTRANSITIONAL_TO_ASCII|UIDNA_NONTRANSITIONAL_TO_UNICODE,
75                 errorCode);
76             if(errorCode.errDataIfFailureAndReset("createUTS46Instance()")) {
77                 name="";
78                 return;
79             }
80         }
81     }
82     TESTCASE_AUTO_BEGIN;
83     TESTCASE_AUTO(TestAPI);
84     TESTCASE_AUTO(TestNotSTD3);
85     TESTCASE_AUTO(TestSomeCases);
86     TESTCASE_AUTO(IdnaTest);
87     TESTCASE_AUTO_END;
88 }
89 
90 const uint32_t severeErrors=
91     UIDNA_ERROR_LEADING_COMBINING_MARK|
92     UIDNA_ERROR_DISALLOWED|
93     UIDNA_ERROR_PUNYCODE|
94     UIDNA_ERROR_LABEL_HAS_DOT|
95     UIDNA_ERROR_INVALID_ACE_LABEL;
96 
isASCII(const UnicodeString & str)97 static UBool isASCII(const UnicodeString &str) {
98     const UChar *s=str.getBuffer();
99     int32_t length=str.length();
100     for(int32_t i=0; i<length; ++i) {
101         if(s[i]>=0x80) {
102             return FALSE;
103         }
104     }
105     return TRUE;
106 }
107 
108 class TestCheckedArrayByteSink : public CheckedArrayByteSink {
109 public:
TestCheckedArrayByteSink(char * outbuf,int32_t capacity)110     TestCheckedArrayByteSink(char* outbuf, int32_t capacity)
111             : CheckedArrayByteSink(outbuf, capacity), calledFlush(FALSE) {}
Reset()112     virtual CheckedArrayByteSink& Reset() {
113         CheckedArrayByteSink::Reset();
114         calledFlush = FALSE;
115         return *this;
116     }
Flush()117     virtual void Flush() { calledFlush = TRUE; }
118     UBool calledFlush;
119 };
120 
TestAPI()121 void UTS46Test::TestAPI() {
122     UErrorCode errorCode=U_ZERO_ERROR;
123     UnicodeString result;
124     IDNAInfo info;
125     UnicodeString input=UNICODE_STRING_SIMPLE("www.eXample.cOm");
126     UnicodeString expected=UNICODE_STRING_SIMPLE("www.example.com");
127     trans->nameToASCII(input, result, info, errorCode);
128     if(U_FAILURE(errorCode) || info.hasErrors() || result!=expected) {
129         errln("T.nameToASCII(www.example.com) info.errors=%04lx result matches=%d %s",
130               (long)info.getErrors(), result==expected, u_errorName(errorCode));
131     }
132     errorCode=U_USELESS_COLLATOR_ERROR;
133     trans->nameToUnicode(input, result, info, errorCode);
134     if(errorCode!=U_USELESS_COLLATOR_ERROR || !result.isBogus()) {
135         errln("T.nameToUnicode(U_FAILURE) did not preserve the errorCode "
136               "or not result.setToBogus() - %s",
137               u_errorName(errorCode));
138     }
139     errorCode=U_ZERO_ERROR;
140     input.setToBogus();
141     result=UNICODE_STRING_SIMPLE("quatsch");
142     nontrans->labelToASCII(input, result, info, errorCode);
143     if(errorCode!=U_ILLEGAL_ARGUMENT_ERROR || !result.isBogus()) {
144         errln("N.labelToASCII(bogus) did not set illegal-argument-error "
145               "or not result.setToBogus() - %s",
146               u_errorName(errorCode));
147     }
148     errorCode=U_ZERO_ERROR;
149     input=UNICODE_STRING_SIMPLE("xn--bcher.de-65a");
150     expected=UNICODE_STRING_SIMPLE("xn--bcher\\uFFFDde-65a").unescape();
151     nontrans->labelToASCII(input, result, info, errorCode);
152     if( U_FAILURE(errorCode) ||
153         info.getErrors()!=(UIDNA_ERROR_LABEL_HAS_DOT|UIDNA_ERROR_INVALID_ACE_LABEL) ||
154         result!=expected
155     ) {
156         errln("N.labelToASCII(label-with-dot) failed with errors %04lx - %s",
157               info.getErrors(), u_errorName(errorCode));
158     }
159     // UTF-8
160     char buffer[100];
161     TestCheckedArrayByteSink sink(buffer, UPRV_LENGTHOF(buffer));
162     errorCode=U_ZERO_ERROR;
163     nontrans->labelToUnicodeUTF8(StringPiece(NULL, 5), sink, info, errorCode);
164     if(errorCode!=U_ILLEGAL_ARGUMENT_ERROR || sink.NumberOfBytesWritten()!=0) {
165         errln("N.labelToUnicodeUTF8(StringPiece(NULL, 5)) did not set illegal-argument-error ",
166               "or did output something - %s",
167               u_errorName(errorCode));
168     }
169 
170     sink.Reset();
171     errorCode=U_ZERO_ERROR;
172     nontrans->nameToASCII_UTF8(StringPiece(), sink, info, errorCode);
173     if(U_FAILURE(errorCode) || sink.NumberOfBytesWritten()!=0 || !sink.calledFlush) {
174         errln("N.nameToASCII_UTF8(empty) failed - %s",
175               u_errorName(errorCode));
176     }
177 
178     static const char s[]={ 0x61, (char)0xc3, (char)0x9f };
179     sink.Reset();
180     errorCode=U_USELESS_COLLATOR_ERROR;
181     nontrans->nameToUnicodeUTF8(StringPiece(s, 3), sink, info, errorCode);
182     if(errorCode!=U_USELESS_COLLATOR_ERROR || sink.NumberOfBytesWritten()!=0) {
183         errln("N.nameToUnicode_UTF8(U_FAILURE) did not preserve the errorCode "
184               "or did output something - %s",
185               u_errorName(errorCode));
186     }
187 
188     sink.Reset();
189     errorCode=U_ZERO_ERROR;
190     trans->labelToUnicodeUTF8(StringPiece(s, 3), sink, info, errorCode);
191     if( U_FAILURE(errorCode) || sink.NumberOfBytesWritten()!=3 ||
192         buffer[0]!=0x61 || buffer[1]!=0x73 || buffer[2]!=0x73 ||
193         !sink.calledFlush
194     ) {
195         errln("T.labelToUnicodeUTF8(a sharp-s) failed - %s",
196               u_errorName(errorCode));
197     }
198 
199     sink.Reset();
200     errorCode=U_ZERO_ERROR;
201     // "eXampLe.cOm"
202     static const char eX[]={ 0x65, 0x58, 0x61, 0x6d, 0x70, 0x4c, 0x65, 0x2e, 0x63, 0x4f, 0x6d, 0 };
203     // "example.com"
204     static const char ex[]={ 0x65, 0x78, 0x61, 0x6d, 0x70, 0x6c, 0x65, 0x2e, 0x63, 0x6f, 0x6d };
205     trans->nameToUnicodeUTF8(eX, sink, info, errorCode);
206     if( U_FAILURE(errorCode) || sink.NumberOfBytesWritten()!=11 ||
207         0!=memcmp(ex, buffer, 11) || !sink.calledFlush
208     ) {
209         errln("T.nameToUnicodeUTF8(eXampLe.cOm) failed - %s",
210               u_errorName(errorCode));
211     }
212 }
213 
TestNotSTD3()214 void UTS46Test::TestNotSTD3() {
215     IcuTestErrorCode errorCode(*this, "TestNotSTD3()");
216     char buffer[400];
217     LocalPointer<IDNA> not3(IDNA::createUTS46Instance(UIDNA_CHECK_BIDI, errorCode));
218     if(errorCode.isFailure()) {
219         return;
220     }
221     UnicodeString input=UNICODE_STRING_SIMPLE("\\u0000A_2+2=4\\u000A.e\\u00DFen.net").unescape();
222     UnicodeString result;
223     IDNAInfo info;
224     if( not3->nameToUnicode(input, result, info, errorCode)!=
225             UNICODE_STRING_SIMPLE("\\u0000a_2+2=4\\u000A.essen.net").unescape() ||
226         info.hasErrors()
227     ) {
228         prettify(result).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
229         errln("notSTD3.nameToUnicode(non-LDH ASCII) unexpected errors %04lx string %s",
230               (long)info.getErrors(), buffer);
231     }
232     // A space (BiDi class WS) is not allowed in a BiDi domain name.
233     input=UNICODE_STRING_SIMPLE("a z.xn--4db.edu");
234     not3->nameToASCII(input, result, info, errorCode);
235     if(result!=input || info.getErrors()!=UIDNA_ERROR_BIDI) {
236         errln("notSTD3.nameToASCII(ASCII-with-space.alef.edu) failed");
237     }
238     // Characters that are canonically equivalent to sequences with non-LDH ASCII.
239     input=UNICODE_STRING_SIMPLE("a\\u2260b\\u226Ec\\u226Fd").unescape();
240     not3->nameToUnicode(input, result, info, errorCode);
241     if(result!=input || info.hasErrors()) {
242         prettify(result).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
243         errln("notSTD3.nameToUnicode(equiv to non-LDH ASCII) unexpected errors %04lx string %s",
244               (long)info.getErrors(), buffer);
245     }
246 }
247 
248 struct TestCase {
249     // Input string and options string (Nontransitional/Transitional/Both).
250     const char *s, *o;
251     // Expected Unicode result string.
252     const char *u;
253     uint32_t errors;
254 };
255 
256 static const TestCase testCases[]={
257     { "www.eXample.cOm", "B",  // all ASCII
258       "www.example.com", 0 },
259     { "B\\u00FCcher.de", "B",  // u-umlaut
260       "b\\u00FCcher.de", 0 },
261     { "\\u00D6BB", "B",  // O-umlaut
262       "\\u00F6bb", 0 },
263     { "fa\\u00DF.de", "N",  // sharp s
264       "fa\\u00DF.de", 0 },
265     { "fa\\u00DF.de", "T",  // sharp s
266       "fass.de", 0 },
267     { "XN--fA-hia.dE", "B",  // sharp s in Punycode
268       "fa\\u00DF.de", 0 },
269     { "\\u03B2\\u03CC\\u03BB\\u03BF\\u03C2.com", "N",  // Greek with final sigma
270       "\\u03B2\\u03CC\\u03BB\\u03BF\\u03C2.com", 0 },
271     { "\\u03B2\\u03CC\\u03BB\\u03BF\\u03C2.com", "T",  // Greek with final sigma
272       "\\u03B2\\u03CC\\u03BB\\u03BF\\u03C3.com", 0 },
273     { "xn--nxasmm1c", "B",  // Greek with final sigma in Punycode
274       "\\u03B2\\u03CC\\u03BB\\u03BF\\u03C2", 0 },
275     { "www.\\u0DC1\\u0DCA\\u200D\\u0DBB\\u0DD3.com", "N",  // "Sri" in "Sri Lanka" has a ZWJ
276       "www.\\u0DC1\\u0DCA\\u200D\\u0DBB\\u0DD3.com", 0 },
277     { "www.\\u0DC1\\u0DCA\\u200D\\u0DBB\\u0DD3.com", "T",  // "Sri" in "Sri Lanka" has a ZWJ
278       "www.\\u0DC1\\u0DCA\\u0DBB\\u0DD3.com", 0 },
279     { "www.xn--10cl1a0b660p.com", "B",  // "Sri" in Punycode
280       "www.\\u0DC1\\u0DCA\\u200D\\u0DBB\\u0DD3.com", 0 },
281     { "\\u0646\\u0627\\u0645\\u0647\\u200C\\u0627\\u06CC", "N",  // ZWNJ
282       "\\u0646\\u0627\\u0645\\u0647\\u200C\\u0627\\u06CC", 0 },
283     { "\\u0646\\u0627\\u0645\\u0647\\u200C\\u0627\\u06CC", "T",  // ZWNJ
284       "\\u0646\\u0627\\u0645\\u0647\\u0627\\u06CC", 0 },
285     { "xn--mgba3gch31f060k.com", "B",  // ZWNJ in Punycode
286       "\\u0646\\u0627\\u0645\\u0647\\u200C\\u0627\\u06CC.com", 0 },
287     { "a.b\\uFF0Ec\\u3002d\\uFF61", "B",
288       "a.b.c.d.", 0 },
289     { "U\\u0308.xn--tda", "B",  // U+umlaut.u-umlaut
290       "\\u00FC.\\u00FC", 0 },
291     { "xn--u-ccb", "B",  // u+umlaut in Punycode
292       "xn--u-ccb\\uFFFD", UIDNA_ERROR_INVALID_ACE_LABEL },
293     { "a\\u2488com", "B",  // contains 1-dot
294       "a\\uFFFDcom", UIDNA_ERROR_DISALLOWED },
295     { "xn--a-ecp.ru", "B",  // contains 1-dot in Punycode
296       "xn--a-ecp\\uFFFD.ru", UIDNA_ERROR_INVALID_ACE_LABEL },
297     { "xn--0.pt", "B",  // invalid Punycode
298       "xn--0\\uFFFD.pt", UIDNA_ERROR_PUNYCODE },
299     { "xn--a.pt", "B",  // U+0080
300       "xn--a\\uFFFD.pt", UIDNA_ERROR_INVALID_ACE_LABEL },
301     { "xn--a-\\u00C4.pt", "B",  // invalid Punycode
302       "xn--a-\\u00E4.pt", UIDNA_ERROR_PUNYCODE },
303     { "\\u65E5\\u672C\\u8A9E\\u3002\\uFF2A\\uFF30", "B",  // Japanese with fullwidth ".jp"
304       "\\u65E5\\u672C\\u8A9E.jp", 0 },
305     { "\\u2615", "B", "\\u2615", 0 },  // Unicode 4.0 HOT BEVERAGE
306     // some characters are disallowed because they are canonically equivalent
307     // to sequences with non-LDH ASCII
308     { "a\\u2260b\\u226Ec\\u226Fd", "B",
309       "a\\uFFFDb\\uFFFDc\\uFFFDd", UIDNA_ERROR_DISALLOWED },
310     // many deviation characters, test the special mapping code
311     { "1.a\\u00DF\\u200C\\u200Db\\u200C\\u200Dc\\u00DF\\u00DF\\u00DF\\u00DFd"
312       "\\u03C2\\u03C3\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFe"
313       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFx"
314       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFy"
315       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u0302\\u00DFz", "N",
316       "1.a\\u00DF\\u200C\\u200Db\\u200C\\u200Dc\\u00DF\\u00DF\\u00DF\\u00DFd"
317       "\\u03C2\\u03C3\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFe"
318       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFx"
319       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFy"
320       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u0302\\u00DFz",
321       UIDNA_ERROR_LABEL_TOO_LONG|UIDNA_ERROR_CONTEXTJ },
322     { "1.a\\u00DF\\u200C\\u200Db\\u200C\\u200Dc\\u00DF\\u00DF\\u00DF\\u00DFd"
323       "\\u03C2\\u03C3\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFe"
324       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFx"
325       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DFy"
326       "\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u00DF\\u0302\\u00DFz", "T",
327       "1.assbcssssssssd"
328       "\\u03C3\\u03C3sssssssssssssssse"
329       "ssssssssssssssssssssx"
330       "ssssssssssssssssssssy"
331       "sssssssssssssss\\u015Dssz", UIDNA_ERROR_LABEL_TOO_LONG },
332     // "xn--bss" with deviation characters
333     { "\\u200Cx\\u200Dn\\u200C-\\u200D-b\\u00DF", "N",
334       "\\u200Cx\\u200Dn\\u200C-\\u200D-b\\u00DF", UIDNA_ERROR_CONTEXTJ },
335     { "\\u200Cx\\u200Dn\\u200C-\\u200D-b\\u00DF", "T",
336       "\\u5919", 0 },
337     // "xn--bssffl" written as:
338     // 02E3 MODIFIER LETTER SMALL X
339     // 034F COMBINING GRAPHEME JOINER (ignored)
340     // 2115 DOUBLE-STRUCK CAPITAL N
341     // 200B ZERO WIDTH SPACE (ignored)
342     // FE63 SMALL HYPHEN-MINUS
343     // 00AD SOFT HYPHEN (ignored)
344     // FF0D FULLWIDTH HYPHEN-MINUS
345     // 180C MONGOLIAN FREE VARIATION SELECTOR TWO (ignored)
346     // 212C SCRIPT CAPITAL B
347     // FE00 VARIATION SELECTOR-1 (ignored)
348     // 017F LATIN SMALL LETTER LONG S
349     // 2064 INVISIBLE PLUS (ignored)
350     // 1D530 MATHEMATICAL FRAKTUR SMALL S
351     // E01EF VARIATION SELECTOR-256 (ignored)
352     // FB04 LATIN SMALL LIGATURE FFL
353     { "\\u02E3\\u034F\\u2115\\u200B\\uFE63\\u00AD\\uFF0D\\u180C"
354       "\\u212C\\uFE00\\u017F\\u2064\\U0001D530\\U000E01EF\\uFB04", "B",
355       "\\u5921\\u591E\\u591C\\u5919", 0 },
356     { "123456789012345678901234567890123456789012345678901234567890123."
357       "123456789012345678901234567890123456789012345678901234567890123."
358       "123456789012345678901234567890123456789012345678901234567890123."
359       "1234567890123456789012345678901234567890123456789012345678901", "B",
360       "123456789012345678901234567890123456789012345678901234567890123."
361       "123456789012345678901234567890123456789012345678901234567890123."
362       "123456789012345678901234567890123456789012345678901234567890123."
363       "1234567890123456789012345678901234567890123456789012345678901", 0 },
364     { "123456789012345678901234567890123456789012345678901234567890123."
365       "123456789012345678901234567890123456789012345678901234567890123."
366       "123456789012345678901234567890123456789012345678901234567890123."
367       "1234567890123456789012345678901234567890123456789012345678901.", "B",
368       "123456789012345678901234567890123456789012345678901234567890123."
369       "123456789012345678901234567890123456789012345678901234567890123."
370       "123456789012345678901234567890123456789012345678901234567890123."
371       "1234567890123456789012345678901234567890123456789012345678901.", 0 },
372     // Domain name >256 characters, forces slow path in UTF-8 processing.
373     { "123456789012345678901234567890123456789012345678901234567890123."
374       "123456789012345678901234567890123456789012345678901234567890123."
375       "123456789012345678901234567890123456789012345678901234567890123."
376       "123456789012345678901234567890123456789012345678901234567890123."
377       "12345678901234567890123456789012345678901234567890123456789012", "B",
378       "123456789012345678901234567890123456789012345678901234567890123."
379       "123456789012345678901234567890123456789012345678901234567890123."
380       "123456789012345678901234567890123456789012345678901234567890123."
381       "123456789012345678901234567890123456789012345678901234567890123."
382       "12345678901234567890123456789012345678901234567890123456789012",
383       UIDNA_ERROR_DOMAIN_NAME_TOO_LONG },
384     { "123456789012345678901234567890123456789012345678901234567890123."
385       "123456789012345678901234567890123456789012345678901234567890123."
386       "123456789012345678901234567890123456789012345678901234567890123."
387       "123456789012345678901234567890123456789012345678901234567890123."
388       "1234567890123456789012345678901234567890123456789\\u05D0", "B",
389       "123456789012345678901234567890123456789012345678901234567890123."
390       "123456789012345678901234567890123456789012345678901234567890123."
391       "123456789012345678901234567890123456789012345678901234567890123."
392       "123456789012345678901234567890123456789012345678901234567890123."
393       "1234567890123456789012345678901234567890123456789\\u05D0",
394       UIDNA_ERROR_DOMAIN_NAME_TOO_LONG|UIDNA_ERROR_BIDI },
395     { "123456789012345678901234567890123456789012345678901234567890123."
396       "1234567890123456789012345678901234567890123456789012345678901234."
397       "123456789012345678901234567890123456789012345678901234567890123."
398       "123456789012345678901234567890123456789012345678901234567890", "B",
399       "123456789012345678901234567890123456789012345678901234567890123."
400       "1234567890123456789012345678901234567890123456789012345678901234."
401       "123456789012345678901234567890123456789012345678901234567890123."
402       "123456789012345678901234567890123456789012345678901234567890",
403       UIDNA_ERROR_LABEL_TOO_LONG },
404     { "123456789012345678901234567890123456789012345678901234567890123."
405       "1234567890123456789012345678901234567890123456789012345678901234."
406       "123456789012345678901234567890123456789012345678901234567890123."
407       "123456789012345678901234567890123456789012345678901234567890.", "B",
408       "123456789012345678901234567890123456789012345678901234567890123."
409       "1234567890123456789012345678901234567890123456789012345678901234."
410       "123456789012345678901234567890123456789012345678901234567890123."
411       "123456789012345678901234567890123456789012345678901234567890.",
412       UIDNA_ERROR_LABEL_TOO_LONG },
413     { "123456789012345678901234567890123456789012345678901234567890123."
414       "1234567890123456789012345678901234567890123456789012345678901234."
415       "123456789012345678901234567890123456789012345678901234567890123."
416       "1234567890123456789012345678901234567890123456789012345678901", "B",
417       "123456789012345678901234567890123456789012345678901234567890123."
418       "1234567890123456789012345678901234567890123456789012345678901234."
419       "123456789012345678901234567890123456789012345678901234567890123."
420       "1234567890123456789012345678901234567890123456789012345678901",
421       UIDNA_ERROR_LABEL_TOO_LONG|UIDNA_ERROR_DOMAIN_NAME_TOO_LONG },
422     // label length 63: xn--1234567890123456789012345678901234567890123456789012345-9te
423     { "\\u00E41234567890123456789012345678901234567890123456789012345", "B",
424       "\\u00E41234567890123456789012345678901234567890123456789012345", 0 },
425     { "1234567890\\u00E41234567890123456789012345678901234567890123456", "B",
426       "1234567890\\u00E41234567890123456789012345678901234567890123456", UIDNA_ERROR_LABEL_TOO_LONG },
427     { "123456789012345678901234567890123456789012345678901234567890123."
428       "1234567890\\u00E4123456789012345678901234567890123456789012345."
429       "123456789012345678901234567890123456789012345678901234567890123."
430       "1234567890123456789012345678901234567890123456789012345678901", "B",
431       "123456789012345678901234567890123456789012345678901234567890123."
432       "1234567890\\u00E4123456789012345678901234567890123456789012345."
433       "123456789012345678901234567890123456789012345678901234567890123."
434       "1234567890123456789012345678901234567890123456789012345678901", 0 },
435     { "123456789012345678901234567890123456789012345678901234567890123."
436       "1234567890\\u00E4123456789012345678901234567890123456789012345."
437       "123456789012345678901234567890123456789012345678901234567890123."
438       "1234567890123456789012345678901234567890123456789012345678901.", "B",
439       "123456789012345678901234567890123456789012345678901234567890123."
440       "1234567890\\u00E4123456789012345678901234567890123456789012345."
441       "123456789012345678901234567890123456789012345678901234567890123."
442       "1234567890123456789012345678901234567890123456789012345678901.", 0 },
443     { "123456789012345678901234567890123456789012345678901234567890123."
444       "1234567890\\u00E4123456789012345678901234567890123456789012345."
445       "123456789012345678901234567890123456789012345678901234567890123."
446       "12345678901234567890123456789012345678901234567890123456789012", "B",
447       "123456789012345678901234567890123456789012345678901234567890123."
448       "1234567890\\u00E4123456789012345678901234567890123456789012345."
449       "123456789012345678901234567890123456789012345678901234567890123."
450       "12345678901234567890123456789012345678901234567890123456789012",
451       UIDNA_ERROR_DOMAIN_NAME_TOO_LONG },
452     { "123456789012345678901234567890123456789012345678901234567890123."
453       "1234567890\\u00E41234567890123456789012345678901234567890123456."
454       "123456789012345678901234567890123456789012345678901234567890123."
455       "123456789012345678901234567890123456789012345678901234567890", "B",
456       "123456789012345678901234567890123456789012345678901234567890123."
457       "1234567890\\u00E41234567890123456789012345678901234567890123456."
458       "123456789012345678901234567890123456789012345678901234567890123."
459       "123456789012345678901234567890123456789012345678901234567890",
460       UIDNA_ERROR_LABEL_TOO_LONG },
461     { "123456789012345678901234567890123456789012345678901234567890123."
462       "1234567890\\u00E41234567890123456789012345678901234567890123456."
463       "123456789012345678901234567890123456789012345678901234567890123."
464       "123456789012345678901234567890123456789012345678901234567890.", "B",
465       "123456789012345678901234567890123456789012345678901234567890123."
466       "1234567890\\u00E41234567890123456789012345678901234567890123456."
467       "123456789012345678901234567890123456789012345678901234567890123."
468       "123456789012345678901234567890123456789012345678901234567890.",
469       UIDNA_ERROR_LABEL_TOO_LONG },
470     { "123456789012345678901234567890123456789012345678901234567890123."
471       "1234567890\\u00E41234567890123456789012345678901234567890123456."
472       "123456789012345678901234567890123456789012345678901234567890123."
473       "1234567890123456789012345678901234567890123456789012345678901", "B",
474       "123456789012345678901234567890123456789012345678901234567890123."
475       "1234567890\\u00E41234567890123456789012345678901234567890123456."
476       "123456789012345678901234567890123456789012345678901234567890123."
477       "1234567890123456789012345678901234567890123456789012345678901",
478       UIDNA_ERROR_LABEL_TOO_LONG|UIDNA_ERROR_DOMAIN_NAME_TOO_LONG },
479     // hyphen errors and empty-label errors
480     // Ticket #10883: ToUnicode also checks for empty labels.
481     { ".", "B", ".", UIDNA_ERROR_EMPTY_LABEL },
482     { "\\uFF0E", "B", ".", UIDNA_ERROR_EMPTY_LABEL },
483     // "xn---q----jra"=="-q--a-umlaut-"
484     { "a.b..-q--a-.e", "B", "a.b..-q--a-.e",
485       UIDNA_ERROR_EMPTY_LABEL|UIDNA_ERROR_LEADING_HYPHEN|UIDNA_ERROR_TRAILING_HYPHEN|
486       UIDNA_ERROR_HYPHEN_3_4 },
487     { "a.b..-q--\\u00E4-.e", "B", "a.b..-q--\\u00E4-.e",
488       UIDNA_ERROR_EMPTY_LABEL|UIDNA_ERROR_LEADING_HYPHEN|UIDNA_ERROR_TRAILING_HYPHEN|
489       UIDNA_ERROR_HYPHEN_3_4 },
490     { "a.b..xn---q----jra.e", "B", "a.b..-q--\\u00E4-.e",
491       UIDNA_ERROR_EMPTY_LABEL|UIDNA_ERROR_LEADING_HYPHEN|UIDNA_ERROR_TRAILING_HYPHEN|
492       UIDNA_ERROR_HYPHEN_3_4 },
493     { "a..c", "B", "a..c", UIDNA_ERROR_EMPTY_LABEL },
494     { "a.xn--.c", "B", "a..c", UIDNA_ERROR_EMPTY_LABEL },
495     { "a.-b.", "B", "a.-b.", UIDNA_ERROR_LEADING_HYPHEN },
496     { "a.b-.c", "B", "a.b-.c", UIDNA_ERROR_TRAILING_HYPHEN },
497     { "a.-.c", "B", "a.-.c", UIDNA_ERROR_LEADING_HYPHEN|UIDNA_ERROR_TRAILING_HYPHEN },
498     { "a.bc--de.f", "B", "a.bc--de.f", UIDNA_ERROR_HYPHEN_3_4 },
499     { "\\u00E4.\\u00AD.c", "B", "\\u00E4..c", UIDNA_ERROR_EMPTY_LABEL },
500     { "\\u00E4.xn--.c", "B", "\\u00E4..c", UIDNA_ERROR_EMPTY_LABEL },
501     { "\\u00E4.-b.", "B", "\\u00E4.-b.", UIDNA_ERROR_LEADING_HYPHEN },
502     { "\\u00E4.b-.c", "B", "\\u00E4.b-.c", UIDNA_ERROR_TRAILING_HYPHEN },
503     { "\\u00E4.-.c", "B", "\\u00E4.-.c", UIDNA_ERROR_LEADING_HYPHEN|UIDNA_ERROR_TRAILING_HYPHEN },
504     { "\\u00E4.bc--de.f", "B", "\\u00E4.bc--de.f", UIDNA_ERROR_HYPHEN_3_4 },
505     { "a.b.\\u0308c.d", "B", "a.b.\\uFFFDc.d", UIDNA_ERROR_LEADING_COMBINING_MARK },
506     { "a.b.xn--c-bcb.d", "B",
507       "a.b.xn--c-bcb\\uFFFD.d", UIDNA_ERROR_LEADING_COMBINING_MARK|UIDNA_ERROR_INVALID_ACE_LABEL },
508     // BiDi
509     { "A0", "B", "a0", 0 },
510     { "0A", "B", "0a", 0 },  // all-LTR is ok to start with a digit (EN)
511     { "0A.\\u05D0", "B",  // ASCII label does not start with L/R/AL
512       "0a.\\u05D0", UIDNA_ERROR_BIDI },
513     { "c.xn--0-eha.xn--4db", "B",  // 2nd label does not start with L/R/AL
514       "c.0\\u00FC.\\u05D0", UIDNA_ERROR_BIDI },
515     { "b-.\\u05D0", "B",  // label does not end with L/EN
516       "b-.\\u05D0", UIDNA_ERROR_TRAILING_HYPHEN|UIDNA_ERROR_BIDI },
517     { "d.xn----dha.xn--4db", "B",  // 2nd label does not end with L/EN
518       "d.\\u00FC-.\\u05D0", UIDNA_ERROR_TRAILING_HYPHEN|UIDNA_ERROR_BIDI },
519     { "a\\u05D0", "B", "a\\u05D0", UIDNA_ERROR_BIDI },  // first dir != last dir
520     { "\\u05D0\\u05C7", "B", "\\u05D0\\u05C7", 0 },
521     { "\\u05D09\\u05C7", "B", "\\u05D09\\u05C7", 0 },
522     { "\\u05D0a\\u05C7", "B", "\\u05D0a\\u05C7", UIDNA_ERROR_BIDI },  // first dir != last dir
523     { "\\u05D0\\u05EA", "B", "\\u05D0\\u05EA", 0 },
524     { "\\u05D0\\u05F3\\u05EA", "B", "\\u05D0\\u05F3\\u05EA", 0 },
525     { "a\\u05D0Tz", "B", "a\\u05D0tz", UIDNA_ERROR_BIDI },  // mixed dir
526     { "\\u05D0T\\u05EA", "B", "\\u05D0t\\u05EA", UIDNA_ERROR_BIDI },  // mixed dir
527     { "\\u05D07\\u05EA", "B", "\\u05D07\\u05EA", 0 },
528     { "\\u05D0\\u0667\\u05EA", "B", "\\u05D0\\u0667\\u05EA", 0 },  // Arabic 7 in the middle
529     { "a7\\u0667z", "B", "a7\\u0667z", UIDNA_ERROR_BIDI },  // AN digit in LTR
530     { "a7\\u0667", "B", "a7\\u0667", UIDNA_ERROR_BIDI },  // AN digit in LTR
531     { "\\u05D07\\u0667\\u05EA", "B",  // mixed EN/AN digits in RTL
532       "\\u05D07\\u0667\\u05EA", UIDNA_ERROR_BIDI },
533     { "\\u05D07\\u0667", "B",  // mixed EN/AN digits in RTL
534       "\\u05D07\\u0667", UIDNA_ERROR_BIDI },
535     // ZWJ
536     { "\\u0BB9\\u0BCD\\u200D", "N", "\\u0BB9\\u0BCD\\u200D", 0 },  // Virama+ZWJ
537     { "\\u0BB9\\u200D", "N", "\\u0BB9\\u200D", UIDNA_ERROR_CONTEXTJ },  // no Virama
538     { "\\u200D", "N", "\\u200D", UIDNA_ERROR_CONTEXTJ },  // no Virama
539     // ZWNJ
540     { "\\u0BB9\\u0BCD\\u200C", "N", "\\u0BB9\\u0BCD\\u200C", 0 },  // Virama+ZWNJ
541     { "\\u0BB9\\u200C", "N", "\\u0BB9\\u200C", UIDNA_ERROR_CONTEXTJ },  // no Virama
542     { "\\u200C", "N", "\\u200C", UIDNA_ERROR_CONTEXTJ },  // no Virama
543     { "\\u0644\\u0670\\u200C\\u06ED\\u06EF", "N",  // Joining types D T ZWNJ T R
544       "\\u0644\\u0670\\u200C\\u06ED\\u06EF", 0 },
545     { "\\u0644\\u0670\\u200C\\u06EF", "N",  // D T ZWNJ R
546       "\\u0644\\u0670\\u200C\\u06EF", 0 },
547     { "\\u0644\\u200C\\u06ED\\u06EF", "N",  // D ZWNJ T R
548       "\\u0644\\u200C\\u06ED\\u06EF", 0 },
549     { "\\u0644\\u200C\\u06EF", "N",  // D ZWNJ R
550       "\\u0644\\u200C\\u06EF", 0 },
551     { "\\u0644\\u0670\\u200C\\u06ED", "N",  // D T ZWNJ T
552       "\\u0644\\u0670\\u200C\\u06ED", UIDNA_ERROR_BIDI|UIDNA_ERROR_CONTEXTJ },
553     { "\\u06EF\\u200C\\u06EF", "N",  // R ZWNJ R
554       "\\u06EF\\u200C\\u06EF", UIDNA_ERROR_CONTEXTJ },
555     { "\\u0644\\u200C", "N",  // D ZWNJ
556       "\\u0644\\u200C", UIDNA_ERROR_BIDI|UIDNA_ERROR_CONTEXTJ },
557     { "\\u0660\\u0661", "B",  // Arabic-Indic Digits alone
558       "\\u0660\\u0661", UIDNA_ERROR_BIDI },
559     { "\\u06F0\\u06F1", "B",  // Extended Arabic-Indic Digits alone
560       "\\u06F0\\u06F1", 0 },
561     { "\\u0660\\u06F1", "B",  // Mixed Arabic-Indic Digits
562       "\\u0660\\u06F1", UIDNA_ERROR_CONTEXTO_DIGITS|UIDNA_ERROR_BIDI },
563     // All of the CONTEXTO "Would otherwise have been DISALLOWED" characters
564     // in their correct contexts,
565     // then each in incorrect context.
566     { "l\\u00B7l\\u4E00\\u0375\\u03B1\\u05D0\\u05F3\\u05F4\\u30FB", "B",
567       "l\\u00B7l\\u4E00\\u0375\\u03B1\\u05D0\\u05F3\\u05F4\\u30FB", UIDNA_ERROR_BIDI },
568     { "l\\u00B7", "B",
569       "l\\u00B7", UIDNA_ERROR_CONTEXTO_PUNCTUATION },
570     { "\\u00B7l", "B",
571       "\\u00B7l", UIDNA_ERROR_CONTEXTO_PUNCTUATION },
572     { "\\u0375", "B",
573       "\\u0375", UIDNA_ERROR_CONTEXTO_PUNCTUATION },
574     { "\\u03B1\\u05F3", "B",
575       "\\u03B1\\u05F3", UIDNA_ERROR_CONTEXTO_PUNCTUATION|UIDNA_ERROR_BIDI },
576     { "\\u05F4", "B",
577       "\\u05F4", UIDNA_ERROR_CONTEXTO_PUNCTUATION },
578     { "l\\u30FB", "B",
579       "l\\u30FB", UIDNA_ERROR_CONTEXTO_PUNCTUATION },
580     // Ticket #8137: UTS #46 toUnicode() fails with non-ASCII labels that turn
581     // into 15 characters (UChars).
582     // The bug was in u_strFromPunycode() which did not write the last character
583     // if it just so fit into the end of the destination buffer.
584     // The UTS #46 code gives a default-capacity UnicodeString as the destination buffer,
585     // and the internal UnicodeString capacity is currently 15 UChars on 64-bit machines
586     // but 13 on 32-bit machines.
587     // Label with 15 UChars, for 64-bit-machine testing:
588     { "aaaaaaaaaaaaa\\u00FCa.de", "B", "aaaaaaaaaaaaa\\u00FCa.de", 0 },
589     { "xn--aaaaaaaaaaaaaa-ssb.de", "B", "aaaaaaaaaaaaa\\u00FCa.de", 0 },
590     { "abschlu\\u00DFpr\\u00FCfung.de", "N", "abschlu\\u00DFpr\\u00FCfung.de", 0 },
591     { "xn--abschluprfung-hdb15b.de", "B", "abschlu\\u00DFpr\\u00FCfung.de", 0 },
592     // Label with 13 UChars, for 32-bit-machine testing:
593     { "xn--aaaaaaaaaaaa-nlb.de", "B", "aaaaaaaaaaa\\u00FCa.de", 0 },
594     { "xn--schluprfung-z6a39a.de", "B", "schlu\\u00DFpr\\u00FCfung.de", 0 },
595     // { "", "B",
596     //   "", 0 },
597 };
598 
TestSomeCases()599 void UTS46Test::TestSomeCases() {
600     IcuTestErrorCode errorCode(*this, "TestSomeCases");
601     char buffer[400], buffer2[400];
602     int32_t i;
603     for(i=0; i<UPRV_LENGTHOF(testCases); ++i) {
604         const TestCase &testCase=testCases[i];
605         UnicodeString input(ctou(testCase.s));
606         UnicodeString expected(ctou(testCase.u));
607         // ToASCII/ToUnicode, transitional/nontransitional
608         UnicodeString aT, uT, aN, uN;
609         IDNAInfo aTInfo, uTInfo, aNInfo, uNInfo;
610         trans->nameToASCII(input, aT, aTInfo, errorCode);
611         trans->nameToUnicode(input, uT, uTInfo, errorCode);
612         nontrans->nameToASCII(input, aN, aNInfo, errorCode);
613         nontrans->nameToUnicode(input, uN, uNInfo, errorCode);
614         if(errorCode.errIfFailureAndReset("first-level processing [%d/%s] %s",
615                                           (int)i, testCase.o, testCase.s)
616         ) {
617             continue;
618         }
619         // ToUnicode does not set length-overflow errors.
620         uint32_t uniErrors=testCase.errors&~
621             (UIDNA_ERROR_LABEL_TOO_LONG|
622              UIDNA_ERROR_DOMAIN_NAME_TOO_LONG);
623         char mode=testCase.o[0];
624         if(mode=='B' || mode=='N') {
625             if(uNInfo.getErrors()!=uniErrors) {
626                 errln("N.nameToUnicode([%d] %s) unexpected errors %04lx",
627                       (int)i, testCase.s, (long)uNInfo.getErrors());
628                 continue;
629             }
630             if(uN!=expected) {
631                 prettify(uN).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
632                 errln("N.nameToUnicode([%d] %s) unexpected string %s",
633                       (int)i, testCase.s, buffer);
634                 continue;
635             }
636             if(aNInfo.getErrors()!=testCase.errors) {
637                 errln("N.nameToASCII([%d] %s) unexpected errors %04lx",
638                       (int)i, testCase.s, (long)aNInfo.getErrors());
639                 continue;
640             }
641         }
642         if(mode=='B' || mode=='T') {
643             if(uTInfo.getErrors()!=uniErrors) {
644                 errln("T.nameToUnicode([%d] %s) unexpected errors %04lx",
645                       (int)i, testCase.s, (long)uTInfo.getErrors());
646                 continue;
647             }
648             if(uT!=expected) {
649                 prettify(uT).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
650                 errln("T.nameToUnicode([%d] %s) unexpected string %s",
651                       (int)i, testCase.s, buffer);
652                 continue;
653             }
654             if(aTInfo.getErrors()!=testCase.errors) {
655                 errln("T.nameToASCII([%d] %s) unexpected errors %04lx",
656                       (int)i, testCase.s, (long)aTInfo.getErrors());
657                 continue;
658             }
659         }
660         // ToASCII is all-ASCII if no severe errors
661         if((aNInfo.getErrors()&severeErrors)==0 && !isASCII(aN)) {
662             prettify(aN).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
663             errln("N.nameToASCII([%d] %s) (errors %04lx) result is not ASCII %s",
664                   (int)i, testCase.s, aNInfo.getErrors(), buffer);
665             continue;
666         }
667         if((aTInfo.getErrors()&severeErrors)==0 && !isASCII(aT)) {
668             prettify(aT).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
669             errln("T.nameToASCII([%d] %s) (errors %04lx) result is not ASCII %s",
670                   (int)i, testCase.s, aTInfo.getErrors(), buffer);
671             continue;
672         }
673         if(verbose) {
674             char m= mode=='B' ? mode : 'N';
675             prettify(aN).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
676             logln("%c.nameToASCII([%d] %s) (errors %04lx) result string: %s",
677                   m, (int)i, testCase.s, aNInfo.getErrors(), buffer);
678             if(mode!='B') {
679                 prettify(aT).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
680                 logln("T.nameToASCII([%d] %s) (errors %04lx) result string: %s",
681                       (int)i, testCase.s, aTInfo.getErrors(), buffer);
682             }
683         }
684         // second-level processing
685         UnicodeString aTuN, uTaN, aNuN, uNaN;
686         IDNAInfo aTuNInfo, uTaNInfo, aNuNInfo, uNaNInfo;
687         nontrans->nameToUnicode(aT, aTuN, aTuNInfo, errorCode);
688         nontrans->nameToASCII(uT, uTaN, uTaNInfo, errorCode);
689         nontrans->nameToUnicode(aN, aNuN, aNuNInfo, errorCode);
690         nontrans->nameToASCII(uN, uNaN, uNaNInfo, errorCode);
691         if(errorCode.errIfFailureAndReset("second-level processing [%d/%s] %s",
692                                           (int)i, testCase.o, testCase.s)
693         ) {
694             continue;
695         }
696         if(aN!=uNaN) {
697             prettify(aN).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
698             prettify(uNaN).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
699             errln("N.nameToASCII([%d] %s)!=N.nameToUnicode().N.nameToASCII() "
700                   "(errors %04lx) %s vs. %s",
701                   (int)i, testCase.s, aNInfo.getErrors(), buffer, buffer2);
702             continue;
703         }
704         if(aT!=uTaN) {
705             prettify(aT).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
706             prettify(uTaN).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
707             errln("T.nameToASCII([%d] %s)!=T.nameToUnicode().N.nameToASCII() "
708                   "(errors %04lx) %s vs. %s",
709                   (int)i, testCase.s, aNInfo.getErrors(), buffer, buffer2);
710             continue;
711         }
712         if(uN!=aNuN) {
713             prettify(uN).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
714             prettify(aNuN).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
715             errln("N.nameToUnicode([%d] %s)!=N.nameToASCII().N.nameToUnicode() "
716                   "(errors %04lx) %s vs. %s",
717                   (int)i, testCase.s, uNInfo.getErrors(), buffer, buffer2);
718             continue;
719         }
720         if(uT!=aTuN) {
721             prettify(uT).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
722             prettify(aTuN).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
723             errln("T.nameToUnicode([%d] %s)!=T.nameToASCII().N.nameToUnicode() "
724                   "(errors %04lx) %s vs. %s",
725                   (int)i, testCase.s, uNInfo.getErrors(), buffer, buffer2);
726             continue;
727         }
728         // labelToUnicode
729         UnicodeString aTL, uTL, aNL, uNL;
730         IDNAInfo aTLInfo, uTLInfo, aNLInfo, uNLInfo;
731         trans->labelToASCII(input, aTL, aTLInfo, errorCode);
732         trans->labelToUnicode(input, uTL, uTLInfo, errorCode);
733         nontrans->labelToASCII(input, aNL, aNLInfo, errorCode);
734         nontrans->labelToUnicode(input, uNL, uNLInfo, errorCode);
735         if(errorCode.errIfFailureAndReset("labelToXYZ processing [%d/%s] %s",
736                                           (int)i, testCase.o, testCase.s)
737         ) {
738             continue;
739         }
740         if(aN.indexOf((UChar)0x2e)<0) {
741             if(aN!=aNL || aNInfo.getErrors()!=aNLInfo.getErrors()) {
742                 prettify(aN).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
743                 prettify(aNL).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
744                 errln("N.nameToASCII([%d] %s)!=N.labelToASCII() "
745                       "(errors %04lx vs %04lx) %s vs. %s",
746                       (int)i, testCase.s, aNInfo.getErrors(), aNLInfo.getErrors(), buffer, buffer2);
747                 continue;
748             }
749         } else {
750             if((aNLInfo.getErrors()&UIDNA_ERROR_LABEL_HAS_DOT)==0) {
751                 errln("N.labelToASCII([%d] %s) errors %04lx missing UIDNA_ERROR_LABEL_HAS_DOT",
752                       (int)i, testCase.s, (long)aNLInfo.getErrors());
753                 continue;
754             }
755         }
756         if(aT.indexOf((UChar)0x2e)<0) {
757             if(aT!=aTL || aTInfo.getErrors()!=aTLInfo.getErrors()) {
758                 prettify(aT).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
759                 prettify(aTL).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
760                 errln("T.nameToASCII([%d] %s)!=T.labelToASCII() "
761                       "(errors %04lx vs %04lx) %s vs. %s",
762                       (int)i, testCase.s, aTInfo.getErrors(), aTLInfo.getErrors(), buffer, buffer2);
763                 continue;
764             }
765         } else {
766             if((aTLInfo.getErrors()&UIDNA_ERROR_LABEL_HAS_DOT)==0) {
767                 errln("T.labelToASCII([%d] %s) errors %04lx missing UIDNA_ERROR_LABEL_HAS_DOT",
768                       (int)i, testCase.s, (long)aTLInfo.getErrors());
769                 continue;
770             }
771         }
772         if(uN.indexOf((UChar)0x2e)<0) {
773             if(uN!=uNL || uNInfo.getErrors()!=uNLInfo.getErrors()) {
774                 prettify(uN).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
775                 prettify(uNL).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
776                 errln("N.nameToUnicode([%d] %s)!=N.labelToUnicode() "
777                       "(errors %04lx vs %04lx) %s vs. %s",
778                       (int)i, testCase.s, uNInfo.getErrors(), uNLInfo.getErrors(), buffer, buffer2);
779                 continue;
780             }
781         } else {
782             if((uNLInfo.getErrors()&UIDNA_ERROR_LABEL_HAS_DOT)==0) {
783                 errln("N.labelToUnicode([%d] %s) errors %04lx missing UIDNA_ERROR_LABEL_HAS_DOT",
784                       (int)i, testCase.s, (long)uNLInfo.getErrors());
785                 continue;
786             }
787         }
788         if(uT.indexOf((UChar)0x2e)<0) {
789             if(uT!=uTL || uTInfo.getErrors()!=uTLInfo.getErrors()) {
790                 prettify(uT).extract(0, 0x7fffffff, buffer, UPRV_LENGTHOF(buffer));
791                 prettify(uTL).extract(0, 0x7fffffff, buffer2, UPRV_LENGTHOF(buffer2));
792                 errln("T.nameToUnicode([%d] %s)!=T.labelToUnicode() "
793                       "(errors %04lx vs %04lx) %s vs. %s",
794                       (int)i, testCase.s, uTInfo.getErrors(), uTLInfo.getErrors(), buffer, buffer2);
795                 continue;
796             }
797         } else {
798             if((uTLInfo.getErrors()&UIDNA_ERROR_LABEL_HAS_DOT)==0) {
799                 errln("T.labelToUnicode([%d] %s) errors %04lx missing UIDNA_ERROR_LABEL_HAS_DOT",
800                       (int)i, testCase.s, (long)uTLInfo.getErrors());
801                 continue;
802             }
803         }
804         // Differences between transitional and nontransitional processing
805         if(mode=='B') {
806             if( aNInfo.isTransitionalDifferent() ||
807                 aTInfo.isTransitionalDifferent() ||
808                 uNInfo.isTransitionalDifferent() ||
809                 uTInfo.isTransitionalDifferent() ||
810                 aNLInfo.isTransitionalDifferent() ||
811                 aTLInfo.isTransitionalDifferent() ||
812                 uNLInfo.isTransitionalDifferent() ||
813                 uTLInfo.isTransitionalDifferent()
814             ) {
815                 errln("B.process([%d] %s) isTransitionalDifferent()", (int)i, testCase.s);
816                 continue;
817             }
818             if( aN!=aT || uN!=uT || aNL!=aTL || uNL!=uTL ||
819                 aNInfo.getErrors()!=aTInfo.getErrors() || uNInfo.getErrors()!=uTInfo.getErrors() ||
820                 aNLInfo.getErrors()!=aTLInfo.getErrors() || uNLInfo.getErrors()!=uTLInfo.getErrors()
821             ) {
822                 errln("N.process([%d] %s) vs. T.process() different errors or result strings",
823                       (int)i, testCase.s);
824                 continue;
825             }
826         } else {
827             if( !aNInfo.isTransitionalDifferent() ||
828                 !aTInfo.isTransitionalDifferent() ||
829                 !uNInfo.isTransitionalDifferent() ||
830                 !uTInfo.isTransitionalDifferent() ||
831                 !aNLInfo.isTransitionalDifferent() ||
832                 !aTLInfo.isTransitionalDifferent() ||
833                 !uNLInfo.isTransitionalDifferent() ||
834                 !uTLInfo.isTransitionalDifferent()
835             ) {
836                 errln("%s.process([%d] %s) !isTransitionalDifferent()",
837                       testCase.o, (int)i, testCase.s);
838                 continue;
839             }
840             if(aN==aT || uN==uT || aNL==aTL || uNL==uTL) {
841                 errln("N.process([%d] %s) vs. T.process() same result strings",
842                       (int)i, testCase.s);
843                 continue;
844             }
845         }
846         // UTF-8
847         std::string input8, aT8, uT8, aN8, uN8;
848         StringByteSink<std::string> aT8Sink(&aT8), uT8Sink(&uT8), aN8Sink(&aN8), uN8Sink(&uN8);
849         IDNAInfo aT8Info, uT8Info, aN8Info, uN8Info;
850         input.toUTF8String(input8);
851         trans->nameToASCII_UTF8(input8, aT8Sink, aT8Info, errorCode);
852         trans->nameToUnicodeUTF8(input8, uT8Sink, uT8Info, errorCode);
853         nontrans->nameToASCII_UTF8(input8, aN8Sink, aN8Info, errorCode);
854         nontrans->nameToUnicodeUTF8(input8, uN8Sink, uN8Info, errorCode);
855         if(errorCode.errIfFailureAndReset("UTF-8 processing [%d/%s] %s",
856                                           (int)i, testCase.o, testCase.s)
857         ) {
858             continue;
859         }
860         UnicodeString aT16(UnicodeString::fromUTF8(aT8));
861         UnicodeString uT16(UnicodeString::fromUTF8(uT8));
862         UnicodeString aN16(UnicodeString::fromUTF8(aN8));
863         UnicodeString uN16(UnicodeString::fromUTF8(uN8));
864         if( aN8Info.getErrors()!=aNInfo.getErrors() ||
865             uN8Info.getErrors()!=uNInfo.getErrors()
866         ) {
867             errln("N.xyzUTF8([%d] %s) vs. UTF-16 processing different errors %04lx vs. %04lx",
868                   (int)i, testCase.s,
869                   (long)aN8Info.getErrors(), (long)aNInfo.getErrors());
870             continue;
871         }
872         if( aT8Info.getErrors()!=aTInfo.getErrors() ||
873             uT8Info.getErrors()!=uTInfo.getErrors()
874         ) {
875             errln("T.xyzUTF8([%d] %s) vs. UTF-16 processing different errors %04lx vs. %04lx",
876                   (int)i, testCase.s,
877                   (long)aT8Info.getErrors(), (long)aTInfo.getErrors());
878             continue;
879         }
880         if(aT16!=aT || uT16!=uT || aN16!=aN || uN16!=uN) {
881             errln("%s.xyzUTF8([%d] %s) vs. UTF-16 processing different string results",
882                   testCase.o, (int)i, testCase.s, (long)aTInfo.getErrors());
883             continue;
884         }
885         if( aT8Info.isTransitionalDifferent()!=aTInfo.isTransitionalDifferent() ||
886             uT8Info.isTransitionalDifferent()!=uTInfo.isTransitionalDifferent() ||
887             aN8Info.isTransitionalDifferent()!=aNInfo.isTransitionalDifferent() ||
888             uN8Info.isTransitionalDifferent()!=uNInfo.isTransitionalDifferent()
889         ) {
890             errln("%s.xyzUTF8([%d] %s) vs. UTF-16 processing different isTransitionalDifferent()",
891                   testCase.o, (int)i, testCase.s);
892             continue;
893         }
894     }
895 }
896 
897 namespace {
898 
899 const int32_t kNumFields = 7;
900 
901 void U_CALLCONV
idnaTestLineFn(void * context,char * fields[][2],int32_t,UErrorCode * pErrorCode)902 idnaTestLineFn(void *context,
903                char *fields[][2], int32_t /* fieldCount */,
904                UErrorCode *pErrorCode) {
905     reinterpret_cast<UTS46Test *>(context)->idnaTestOneLine(fields, *pErrorCode);
906 }
907 
s16FromField(char * (& field)[2])908 UnicodeString s16FromField(char *(&field)[2]) {
909     int32_t length = (int32_t)(field[1] - field[0]);
910     return UnicodeString::fromUTF8(StringPiece(field[0], length)).trim().unescape();
911 }
912 
statusFromField(char * (& field)[2])913 std::string statusFromField(char *(&field)[2]) {
914     const char *start = u_skipWhitespace(field[0]);
915     std::string status;
916     if (start != field[1]) {
917         int32_t length = (int32_t)(field[1] - start);
918         while (length > 0 && (start[length - 1] == u' ' || start[length - 1] == u'\t')) {
919             --length;
920         }
921         status.assign(start, length);
922     }
923     return status;
924 }
925 
926 }  // namespace
927 
checkIdnaTestResult(const char * line,const char * type,const UnicodeString & expected,const UnicodeString & result,const char * status,const IDNAInfo & info)928 void UTS46Test::checkIdnaTestResult(const char *line, const char *type,
929                                     const UnicodeString &expected, const UnicodeString &result,
930                                     const char *status, const IDNAInfo &info) {
931     // An error in toUnicode or toASCII is indicated by a value in square brackets,
932     // such as "[B5 B6]".
933     UBool expectedHasErrors = FALSE;
934     if (*status != 0) {
935         if (*status != u'[') {
936             errln("%s  status field does not start with '[': %s\n    %s", type, status, line);
937         }
938         if (strcmp(status, u8"[]") != 0) {
939             expectedHasErrors = TRUE;
940         }
941     }
942     if (expectedHasErrors != info.hasErrors()) {
943         errln("%s  expected errors %s %d != %d = actual has errors: %04lx\n    %s",
944               type, status, expectedHasErrors, info.hasErrors(), (long)info.getErrors(), line);
945     }
946     if (!expectedHasErrors && expected != result) {
947         errln("%s  expected != actual\n    %s", type, line);
948         errln(UnicodeString(u"    ") + expected);
949         errln(UnicodeString(u"    ") + result);
950     }
951 }
952 
idnaTestOneLine(char * fields[][2],UErrorCode & errorCode)953 void UTS46Test::idnaTestOneLine(char *fields[][2], UErrorCode &errorCode) {
954     // IdnaTestV2.txt (since Unicode 11)
955     // Column 1: source
956     // The source string to be tested
957     UnicodeString source = s16FromField(fields[0]);
958 
959     // Column 2: toUnicode
960     // The result of applying toUnicode to the source, with Transitional_Processing=false.
961     // A blank value means the same as the source value.
962     UnicodeString toUnicode = s16FromField(fields[1]);
963     if (toUnicode.isEmpty()) {
964         toUnicode = source;
965     }
966 
967     // Column 3: toUnicodeStatus
968     // A set of status codes, each corresponding to a particular test.
969     // A blank value means [].
970     std::string toUnicodeStatus = statusFromField(fields[2]);
971 
972     // Column 4: toAsciiN
973     // The result of applying toASCII to the source, with Transitional_Processing=false.
974     // A blank value means the same as the toUnicode value.
975     UnicodeString toAsciiN = s16FromField(fields[3]);
976     if (toAsciiN.isEmpty()) {
977         toAsciiN = toUnicode;
978     }
979 
980     // Column 5: toAsciiNStatus
981     // A set of status codes, each corresponding to a particular test.
982     // A blank value means the same as the toUnicodeStatus value.
983     std::string toAsciiNStatus = statusFromField(fields[4]);
984     if (toAsciiNStatus.empty()) {
985         toAsciiNStatus = toUnicodeStatus;
986     }
987 
988     // Column 6: toAsciiT
989     // The result of applying toASCII to the source, with Transitional_Processing=true.
990     // A blank value means the same as the toAsciiN value.
991     UnicodeString toAsciiT = s16FromField(fields[5]);
992     if (toAsciiT.isEmpty()) {
993         toAsciiT = toAsciiN;
994     }
995 
996     // Column 7: toAsciiTStatus
997     // A set of status codes, each corresponding to a particular test.
998     // A blank value means the same as the toAsciiNStatus value.
999     std::string toAsciiTStatus = statusFromField(fields[6]);
1000     if (toAsciiTStatus.empty()) {
1001         toAsciiTStatus = toAsciiNStatus;
1002     }
1003 
1004     // ToASCII/ToUnicode, transitional/nontransitional
1005     UnicodeString uN, aN, aT;
1006     IDNAInfo uNInfo, aNInfo, aTInfo;
1007     nontrans->nameToUnicode(source, uN, uNInfo, errorCode);
1008     checkIdnaTestResult(fields[0][0], "toUnicodeNontrans", toUnicode, uN,
1009                         toUnicodeStatus.c_str(), uNInfo);
1010     nontrans->nameToASCII(source, aN, aNInfo, errorCode);
1011     checkIdnaTestResult(fields[0][0], "toASCIINontrans", toAsciiN, aN,
1012                         toAsciiNStatus.c_str(), aNInfo);
1013     trans->nameToASCII(source, aT, aTInfo, errorCode);
1014     checkIdnaTestResult(fields[0][0], "toASCIITrans", toAsciiT, aT,
1015                         toAsciiTStatus.c_str(), aTInfo);
1016 }
1017 
1018 // TODO: de-duplicate
1019 U_DEFINE_LOCAL_OPEN_POINTER(LocalStdioFilePointer, FILE, fclose);
1020 
1021 // http://www.unicode.org/Public/idna/latest/IdnaTest.txt
IdnaTest()1022 void UTS46Test::IdnaTest() {
1023     IcuTestErrorCode errorCode(*this, "IdnaTest");
1024     const char *sourceTestDataPath = getSourceTestData(errorCode);
1025     if (errorCode.errIfFailureAndReset("unable to find the source/test/testdata "
1026                                        "folder (getSourceTestData())")) {
1027         return;
1028     }
1029     CharString path(sourceTestDataPath, errorCode);
1030     path.appendPathPart("IdnaTestV2.txt", errorCode);
1031     LocalStdioFilePointer idnaTestFile(fopen(path.data(), "r"));
1032     if (idnaTestFile.isNull()) {
1033         errln("unable to open %s", path.data());
1034         return;
1035     }
1036 
1037     // Columns (c1, c2,...) are separated by semicolons.
1038     // Leading and trailing spaces and tabs in each column are ignored.
1039     // Comments are indicated with hash marks.
1040     char *fields[kNumFields][2];
1041     u_parseDelimitedFile(path.data(), ';', fields, kNumFields, idnaTestLineFn, this, errorCode);
1042     if (errorCode.errIfFailureAndReset("error parsing IdnaTest.txt")) {
1043         return;
1044     }
1045 }
1046 
1047 #endif  // UCONFIG_NO_IDNA
1048