1 ///////////////////////////////////////////////////////////////////////////////
2 // Name: tests/strings/strings.cpp
3 // Purpose: wxString unit test
4 // Author: Vadim Zeitlin, Wlodzimierz ABX Skiba
7 // Copyright: (c) 2004 Vadim Zeitlin, Wlodzimierz Skiba
8 ///////////////////////////////////////////////////////////////////////////////
10 // ----------------------------------------------------------------------------
12 // ----------------------------------------------------------------------------
24 #include "wx/tokenzr.h"
26 // ----------------------------------------------------------------------------
28 // ----------------------------------------------------------------------------
30 class StringTestCase
: public CppUnit::TestCase
36 CPPUNIT_TEST_SUITE( StringTestCase
);
37 CPPUNIT_TEST( String
);
38 CPPUNIT_TEST( PChar
);
39 CPPUNIT_TEST( Format
);
40 CPPUNIT_TEST( Constructors
);
42 CPPUNIT_TEST( ConstructorsWithConversion
);
43 CPPUNIT_TEST( Conversion
);
44 CPPUNIT_TEST( ConversionUTF7
);
45 CPPUNIT_TEST( ConversionUTF8
);
46 #endif // wxUSE_WCHAR_T
47 CPPUNIT_TEST( Extraction
);
49 CPPUNIT_TEST( Tokenizer
);
50 CPPUNIT_TEST( TokenizerGetPosition
);
51 CPPUNIT_TEST( Replace
);
52 CPPUNIT_TEST( Match
);
53 CPPUNIT_TEST( CaseChanges
);
54 CPPUNIT_TEST( Compare
);
55 CPPUNIT_TEST( CompareNoCase
);
56 CPPUNIT_TEST_SUITE_END();
63 void ConstructorsWithConversion();
65 void ConversionUTF7();
66 void ConversionUTF8();
67 #endif // wxUSE_WCHAR_T
70 void SingleTokenizerTest( wxChar
*str
, wxChar
*delims
, size_t count
, wxStringTokenizerMode mode
);
72 void TokenizerGetPosition();
80 // test if converting s using the given encoding gives ws and vice versa
82 // if either of the first 2 arguments is NULL, the conversion is supposed
84 void DoTestConversion(const char *s
, const wchar_t *w
, wxCSConv
& conv
);
85 #endif // wxUSE_WCHAR_T
87 DECLARE_NO_COPY_CLASS(StringTestCase
)
90 // register in the unnamed registry so that these tests are run by default
91 CPPUNIT_TEST_SUITE_REGISTRATION( StringTestCase
);
93 // also include in it's own registry so that these tests can be run alone
94 CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( StringTestCase
, "StringTestCase" );
96 StringTestCase::StringTestCase()
100 void StringTestCase::String()
108 for (int i
= 0; i
< 2; ++i
)
112 c
= _T("! How'ya doin'?");
115 c
= _T("Hello world! What's up?");
116 CPPUNIT_ASSERT( c
!= a
);
120 void StringTestCase::PChar()
126 for (int i
= 0; i
< 2; ++i
)
128 wxStrcpy (a
, _T("Hello"));
129 wxStrcpy (b
, _T(" world"));
130 wxStrcpy (c
, _T("! How'ya doin'?"));
133 wxStrcpy (c
, _T("Hello world! What's up?"));
134 CPPUNIT_ASSERT( wxStrcmp (c
, a
) != 0 );
138 void StringTestCase::Format()
141 s1
.Printf(_T("%03d"), 18);
142 CPPUNIT_ASSERT( s1
== wxString::Format(_T("%03d"), 18) );
143 s2
.Printf(_T("Number 18: %s\n"), s1
.c_str());
144 CPPUNIT_ASSERT( s2
== wxString::Format(_T("Number 18: %s\n"), s1
.c_str()) );
147 void StringTestCase::Constructors()
149 #define TEST_CTOR(args, res) \
152 CPPUNIT_ASSERT( s == res ); \
155 TEST_CTOR((_T('Z'), 4), _T("ZZZZ"));
156 TEST_CTOR((_T("Hello"), 4), _T("Hell"));
157 TEST_CTOR((_T("Hello"), 5), _T("Hello"));
159 static const wxChar
*s
= _T("?really!");
160 const wxChar
*start
= wxStrchr(s
, _T('r'));
161 const wxChar
*end
= wxStrchr(s
, _T('!'));
162 TEST_CTOR((start
, end
), _T("really"));
166 void StringTestCase::ConstructorsWithConversion()
168 // the string "Déjà" in UTF-8 and wchar_t:
169 const unsigned char utf8Buf
[] = {0x44,0xC3,0xA9,0x6A,0xC3,0xA0,0};
170 const wchar_t wchar
[] = {0x44,0xE9,0x6A,0xE0,0};
171 const unsigned char utf8subBuf
[] = {0x44,0xC3,0xA9,0x6A,0}; // just "Déj"
172 const char *utf8
= (char *)utf8Buf
;
173 const char *utf8sub
= (char *)utf8subBuf
;
175 wxString
s1(utf8
, wxConvUTF8
);
176 wxString
s2(wchar
, wxConvUTF8
);
179 CPPUNIT_ASSERT( s1
== wchar
);
180 CPPUNIT_ASSERT( s2
== wchar
);
182 CPPUNIT_ASSERT( s1
== utf8
);
183 CPPUNIT_ASSERT( s2
== utf8
);
186 wxString
sub(utf8sub
, wxConvUTF8
); // "Dej" substring
187 wxString
s3(utf8
, wxConvUTF8
, 4);
188 wxString
s4(wchar
, wxConvUTF8
, 3);
190 CPPUNIT_ASSERT( s3
== sub
);
191 CPPUNIT_ASSERT( s4
== sub
);
194 CPPUNIT_ASSERT ( wxString("\t[pl]open.format.Sformatuj dyskietkê=gfloppy %f",
195 wxConvUTF8
) == wxT("") ); //should stop at pos 35
199 void StringTestCase::Conversion()
202 wxString
szTheString(L
"The\0String", wxConvLibc
, 10);
203 wxCharBuffer theBuffer
= szTheString
.mb_str();
205 CPPUNIT_ASSERT( memcmp(theBuffer
.data(), "The\0String", 11) == 0 );
207 wxString
szTheString2("The\0String", wxConvLocal
, 10);
208 CPPUNIT_ASSERT( szTheString2
.length() == 11 );
209 CPPUNIT_ASSERT( wxTmemcmp(szTheString2
.c_str(), L
"The\0String", 11) == 0 );
211 wxString
szTheString(wxT("TheString"));
212 szTheString
.insert(3, 1, '\0');
213 wxWCharBuffer theBuffer
= szTheString
.wc_str(wxConvLibc
);
215 CPPUNIT_ASSERT( memcmp(theBuffer
.data(), L
"The\0String", 11 * sizeof(wchar_t)) == 0 );
217 wxString
szLocalTheString(wxT("TheString"));
218 szLocalTheString
.insert(3, 1, '\0');
219 wxWCharBuffer theLocalBuffer
= szLocalTheString
.wc_str(wxConvLocal
);
221 CPPUNIT_ASSERT( memcmp(theLocalBuffer
.data(), L
"The\0String", 11 * sizeof(wchar_t)) == 0 );
226 StringTestCase::DoTestConversion(const char *s
,
233 wxCharBuffer
buf(wxString(ws
).mb_str(conv
));
235 CPPUNIT_ASSERT( strcmp(buf
, s
) == 0 );
237 #else // wxUSE_UNICODE
240 wxWCharBuffer
wbuf(wxString(s
).wc_str(conv
));
243 CPPUNIT_ASSERT( wcscmp(wbuf
, ws
) == 0 );
245 CPPUNIT_ASSERT( !*wbuf
);
247 #endif // wxUSE_UNICODE/!wxUSE_UNICODE
250 struct StringConversionData
256 void StringTestCase::ConversionUTF7()
258 static const StringConversionData utf7data
[] =
262 //\u isn't recognized on MSVC 6
263 #if !defined(_MSC_VER)
264 #if !defined(__GNUC__) || (__GNUC__ >= 3)
265 { "+AKM-", L
"\u00a3" },
268 // Windows accepts invalid UTF-7 strings and so does our UTF-7
269 // conversion code -- this is wrong IMO but the way it is for now
271 // notice that converting "+" still behaves as expected because the
272 // result is just an empty string, i.e. the same as if there were an
273 // error, but converting "a+" results in "a" while it really should
279 wxCSConv
conv(_T("utf-7"));
280 for ( size_t n
= 0; n
< WXSIZEOF(utf7data
); n
++ )
282 const StringConversionData
& d
= utf7data
[n
];
283 DoTestConversion(d
.str
, d
.wcs
, conv
);
287 void StringTestCase::ConversionUTF8()
289 static const StringConversionData utf8data
[] =
291 //\u isn't recognized on MSVC 6
292 #if !defined(_MSC_VER)
293 #if !defined(__GNUC__) || (__GNUC__ >= 3)
294 { "\xc2\xa3", L
"\u00a3" },
300 wxCSConv
conv(_T("utf-8"));
301 for ( size_t n
= 0; n
< WXSIZEOF(utf8data
); n
++ )
303 const StringConversionData
& d
= utf8data
[n
];
304 DoTestConversion(d
.str
, d
.wcs
, conv
);
308 #endif // wxUSE_WCHAR_T
311 void StringTestCase::Extraction()
313 wxString
s(_T("Hello, world!"));
315 CPPUNIT_ASSERT( wxStrcmp( s
.c_str() , _T("Hello, world!") ) == 0 );
316 CPPUNIT_ASSERT( wxStrcmp( s
.Left(5).c_str() , _T("Hello") ) == 0 );
317 CPPUNIT_ASSERT( wxStrcmp( s
.Right(6).c_str() , _T("world!") ) == 0 );
318 CPPUNIT_ASSERT( wxStrcmp( s(3, 5).c_str() , _T("lo, w") ) == 0 );
319 CPPUNIT_ASSERT( wxStrcmp( s
.Mid(3).c_str() , _T("lo, world!") ) == 0 );
320 CPPUNIT_ASSERT( wxStrcmp( s
.substr(3, 5).c_str() , _T("lo, w") ) == 0 );
321 CPPUNIT_ASSERT( wxStrcmp( s
.substr(3).c_str() , _T("lo, world!") ) == 0 );
325 #define TEST_STARTS_WITH( prefix , correct_rest, result ) \
327 ( s.StartsWith( prefix, &rest ) == result ) && \
328 ( ( result == false ) || ( wxStrcmp( correct_rest , rest ) == 0 ) ) \
331 TEST_STARTS_WITH( _T("Hello"), _T(", world!"), true );
332 TEST_STARTS_WITH( _T("Hello, "), _T("world!"), true );
333 TEST_STARTS_WITH( _T("Hello, world!"), _T(""), true );
334 TEST_STARTS_WITH( _T("Hello, world!!!"), _T(""), false );
335 TEST_STARTS_WITH( _T(""), _T("Hello, world!"), true );
336 TEST_STARTS_WITH( _T("Goodbye"), _T(""), false );
337 TEST_STARTS_WITH( _T("Hi"), _T(""), false );
339 #undef TEST_STARTS_WITH
342 void StringTestCase::Find()
344 #define TEST_FIND( str , start , result ) \
345 CPPUNIT_ASSERT( wxString(str).find(_T("ell"), start) == result );
347 TEST_FIND( _T("Well, hello world"), 0, 1 );
348 TEST_FIND( _T("Well, hello world"), 6, 7 );
349 TEST_FIND( _T("Well, hello world"), 9, wxString::npos
);
354 void StringTestCase::SingleTokenizerTest( wxChar
*str
, wxChar
*delims
, size_t count
, wxStringTokenizerMode mode
)
356 wxStringTokenizer
tkz( str
, delims
, mode
);
357 CPPUNIT_ASSERT( tkz
.CountTokens() == count
);
359 wxChar
*buf
, *s
= NULL
, *last
;
361 if ( tkz
.GetMode() == wxTOKEN_STRTOK
)
363 buf
= new wxChar
[wxStrlen(str
) + 1];
365 s
= wxStrtok(buf
, delims
, &last
);
373 while ( tkz
.HasMoreTokens() )
375 wxString token
= tkz
.GetNextToken();
378 CPPUNIT_ASSERT( token
== s
);
379 s
= wxStrtok(NULL
, delims
, &last
);
384 CPPUNIT_ASSERT( count2
== count
);
391 void StringTestCase::Tokenizer()
393 SingleTokenizerTest( _T(""), _T(" "), 0, wxTOKEN_DEFAULT
);
394 SingleTokenizerTest( _T("Hello, world"), _T(" "), 2, wxTOKEN_DEFAULT
);
395 SingleTokenizerTest( _T("Hello, world "), _T(" "), 2, wxTOKEN_DEFAULT
);
396 SingleTokenizerTest( _T("Hello, world"), _T(","), 2, wxTOKEN_DEFAULT
);
397 SingleTokenizerTest( _T("Hello, world!"), _T(",!"), 2, wxTOKEN_DEFAULT
);
398 SingleTokenizerTest( _T("Hello,, world!"), _T(",!"), 3, wxTOKEN_DEFAULT
);
399 SingleTokenizerTest( _T("Hello, world!"), _T(",!"), 3, wxTOKEN_RET_EMPTY_ALL
);
400 SingleTokenizerTest( _T("username:password:uid:gid:gecos:home:shell"), _T(":"), 7, wxTOKEN_DEFAULT
);
401 SingleTokenizerTest( _T("1 \t3\t4 6 "), wxDEFAULT_DELIMITERS
, 4, wxTOKEN_DEFAULT
);
402 SingleTokenizerTest( _T("1 \t3\t4 6 "), wxDEFAULT_DELIMITERS
, 6, wxTOKEN_RET_EMPTY
);
403 SingleTokenizerTest( _T("1 \t3\t4 6 "), wxDEFAULT_DELIMITERS
, 9, wxTOKEN_RET_EMPTY_ALL
);
404 SingleTokenizerTest( _T("01/02/99"), _T("/-"), 3, wxTOKEN_DEFAULT
);
405 SingleTokenizerTest( _T("01-02/99"), _T("/-"), 3, wxTOKEN_RET_DELIMS
);
408 // call this with the string to tokenize, delimeters to use and the expected
409 // positions (i.e. results of GetPosition()) after each GetNextToken() call,
410 // terminate positions with 0
411 static void DoTokenizerGetPosition(const wxChar
*s
,
412 const wxChar
*delims
, int pos
, ...)
414 wxStringTokenizer
tkz(s
, delims
);
416 CPPUNIT_ASSERT( tkz
.GetPosition() == 0 );
425 CPPUNIT_ASSERT( !tkz
.HasMoreTokens() );
431 CPPUNIT_ASSERT( tkz
.GetPosition() == (size_t)pos
);
433 pos
= va_arg(ap
, int);
439 void StringTestCase::TokenizerGetPosition()
441 DoTokenizerGetPosition(_T("foo"), _T("_"), 3, 0);
442 DoTokenizerGetPosition(_T("foo_bar"), _T("_"), 4, 7, 0);
443 DoTokenizerGetPosition(_T("foo_bar_"), _T("_"), 4, 8, 0);
446 void StringTestCase::Replace()
448 #define TEST_REPLACE( original , pos , len , replacement , result ) \
450 wxString s = original; \
451 s.replace( pos , len , replacement ); \
452 CPPUNIT_ASSERT( s == result ); \
455 TEST_REPLACE( _T("012-AWORD-XYZ"), 4, 5, _T("BWORD"), _T("012-BWORD-XYZ") );
456 TEST_REPLACE( _T("increase"), 0, 2, _T("de"), _T("decrease") );
457 TEST_REPLACE( _T("wxWindow"), 8, 0, _T("s"), _T("wxWindows") );
458 TEST_REPLACE( _T("foobar"), 3, 0, _T("-"), _T("foo-bar") );
459 TEST_REPLACE( _T("barfoo"), 0, 6, _T("foobar"), _T("foobar") );
462 #define TEST_NULLCHARREPLACE( o , olen, pos , len , replacement , r, rlen ) \
464 wxString s(o,olen); \
465 s.replace( pos , len , replacement ); \
466 CPPUNIT_ASSERT( s == wxString(r,rlen) ); \
469 TEST_NULLCHARREPLACE( _T("null\0char"), 9, 5, 1, _T("d"),
470 _T("null\0dhar"), 9 );
472 #define TEST_WXREPLACE( o , olen, olds, news, all, r, rlen ) \
474 wxString s(o,olen); \
475 s.Replace( olds, news, all ); \
476 CPPUNIT_ASSERT( s == wxString(r,rlen) ); \
479 TEST_WXREPLACE( _T("null\0char"), 9, _T("c"), _T("de"), true,
480 _T("null\0dehar"), 10 );
482 TEST_WXREPLACE( _T("null\0dehar"), 10, _T("de"), _T("c"), true,
483 _T("null\0char"), 9 );
485 #undef TEST_WXREPLACE
486 #undef TEST_NULLCHARREPLACE
490 void StringTestCase::Match()
492 #define TEST_MATCH( s1 , s2 , result ) \
493 CPPUNIT_ASSERT( wxString(s1).Matches(s2) == result )
495 TEST_MATCH( _T("foobar"), _T("foo*"), true );
496 TEST_MATCH( _T("foobar"), _T("*oo*"), true );
497 TEST_MATCH( _T("foobar"), _T("*bar"), true );
498 TEST_MATCH( _T("foobar"), _T("??????"), true );
499 TEST_MATCH( _T("foobar"), _T("f??b*"), true );
500 TEST_MATCH( _T("foobar"), _T("f?b*"), false );
501 TEST_MATCH( _T("foobar"), _T("*goo*"), false );
502 TEST_MATCH( _T("foobar"), _T("*foo"), false );
503 TEST_MATCH( _T("foobarfoo"), _T("*foo"), true );
504 TEST_MATCH( _T(""), _T("*"), true );
505 TEST_MATCH( _T(""), _T("?"), false );
511 void StringTestCase::CaseChanges()
513 wxString
s1(_T("Hello!"));
522 CPPUNIT_ASSERT( s1u
== _T("HELLO!") );
523 CPPUNIT_ASSERT( s1l
== _T("hello!") );
524 CPPUNIT_ASSERT( s2u
== wxEmptyString
);
525 CPPUNIT_ASSERT( s2l
== wxEmptyString
);
528 wxLocale
locRu(wxLANGUAGE_RUSSIAN
, 0 /* flags */);
531 // try upper casing 8bit strings
532 wxString
sUpper("\xdf"),
535 CPPUNIT_ASSERT( sUpper
.Lower() == sLower
);
536 CPPUNIT_ASSERT( sLower
.Upper() == sUpper
);
538 #endif // !wxUSE_UNICODE
541 void StringTestCase::Compare()
543 wxString s1
= wxT("AHH");
544 wxString eq
= wxT("AHH");
545 wxString neq1
= wxT("HAH");
546 wxString neq2
= wxT("AH");
547 wxString neq3
= wxT("AHHH");
548 wxString neq4
= wxT("AhH");
550 CPPUNIT_ASSERT( s1
== eq
);
551 CPPUNIT_ASSERT( s1
!= neq1
);
552 CPPUNIT_ASSERT( s1
!= neq2
);
553 CPPUNIT_ASSERT( s1
!= neq3
);
554 CPPUNIT_ASSERT( s1
!= neq4
);
556 // wxString _s1 = wxT("A\0HH");
557 // wxString _eq = wxT("A\0HH");
558 // wxString _neq1 = wxT("H\0AH");
559 // wxString _neq2 = wxT("A\0H");
560 // wxString _neq3 = wxT("A\0HHH");
561 // wxString _neq4 = wxT("A\0hH");
564 neq1
.insert(1,1,'\0');
565 neq2
.insert(1,1,'\0');
566 neq3
.insert(1,1,'\0');
567 neq4
.insert(1,1,'\0');
569 CPPUNIT_ASSERT( s1
== eq
);
570 CPPUNIT_ASSERT( s1
!= neq1
);
571 CPPUNIT_ASSERT( s1
!= neq2
);
572 CPPUNIT_ASSERT( s1
!= neq3
);
573 CPPUNIT_ASSERT( s1
!= neq4
);
576 void StringTestCase::CompareNoCase()
578 wxString s1
= wxT("AHH");
579 wxString eq
= wxT("AHH");
580 wxString eq2
= wxT("AhH");
581 wxString eq3
= wxT("ahh");
582 wxString neq
= wxT("HAH");
583 wxString neq2
= wxT("AH");
584 wxString neq3
= wxT("AHHH");
586 #define CPPUNIT_CNCEQ_ASSERT(s1, s2) CPPUNIT_ASSERT( s1.CmpNoCase(s2) == 0)
587 #define CPPUNIT_CNCNEQ_ASSERT(s1, s2) CPPUNIT_ASSERT( s1.CmpNoCase(s2) != 0)
589 CPPUNIT_CNCEQ_ASSERT( s1
, eq
);
590 CPPUNIT_CNCEQ_ASSERT( s1
, eq2
);
591 CPPUNIT_CNCEQ_ASSERT( s1
, eq3
);
593 CPPUNIT_CNCNEQ_ASSERT( s1
, neq
);
594 CPPUNIT_CNCNEQ_ASSERT( s1
, neq2
);
595 CPPUNIT_CNCNEQ_ASSERT( s1
, neq3
);
598 // wxString _s1 = wxT("A\0HH");
599 // wxString _eq = wxT("A\0HH");
600 // wxString _eq2 = wxT("A\0hH");
601 // wxString _eq3 = wxT("a\0hh");
602 // wxString _neq = wxT("H\0AH");
603 // wxString _neq2 = wxT("A\0H");
604 // wxString _neq3 = wxT("A\0HHH");
608 eq2
.insert(1,1,'\0');
609 eq3
.insert(1,1,'\0');
610 neq
.insert(1,1,'\0');
611 neq2
.insert(1,1,'\0');
612 neq3
.insert(1,1,'\0');
614 CPPUNIT_CNCEQ_ASSERT( s1
, eq
);
615 CPPUNIT_CNCEQ_ASSERT( s1
, eq2
);
616 CPPUNIT_CNCEQ_ASSERT( s1
, eq3
);
618 CPPUNIT_CNCNEQ_ASSERT( s1
, neq
);
619 CPPUNIT_CNCNEQ_ASSERT( s1
, neq2
);
620 CPPUNIT_CNCNEQ_ASSERT( s1
, neq3
);