X-Git-Url: https://git.saurik.com/wxWidgets.git/blobdiff_plain/ede838eb4d96c83963498e28480175c1375d2901..d924939be81c8a836516cded30f1392cd9b6baf5:/tests/strings/unicode.cpp diff --git a/tests/strings/unicode.cpp b/tests/strings/unicode.cpp index 5e17bd3a74..775fd3eedf 100644 --- a/tests/strings/unicode.cpp +++ b/tests/strings/unicode.cpp @@ -20,6 +20,25 @@ #ifndef WX_PRECOMP #endif // WX_PRECOMP +// ---------------------------------------------------------------------------- +// local functions +// ---------------------------------------------------------------------------- + +#if wxUSE_WCHAR_T && !wxUSE_UNICODE + +// in case wcscmp is missing +static int wx_wcscmp(const wchar_t *s1, const wchar_t *s2) +{ + while (*s1 == *s2 && *s1 != 0) + { + s1++; + s2++; + } + return *s1 - *s2; +} + +#endif // wxUSE_WCHAR_T && !wxUSE_UNICODE + // ---------------------------------------------------------------------------- // test class // ---------------------------------------------------------------------------- @@ -32,9 +51,34 @@ public: private: CPPUNIT_TEST_SUITE( UnicodeTestCase ); CPPUNIT_TEST( ToFromAscii ); +#if wxUSE_WCHAR_T + CPPUNIT_TEST( ConstructorsWithConversion ); + CPPUNIT_TEST( ConversionEmpty ); + CPPUNIT_TEST( ConversionWithNULs ); + CPPUNIT_TEST( ConversionUTF7 ); + CPPUNIT_TEST( ConversionUTF8 ); + CPPUNIT_TEST( ConversionUTF16 ); + CPPUNIT_TEST( ConversionUTF32 ); +#endif // wxUSE_WCHAR_T CPPUNIT_TEST_SUITE_END(); void ToFromAscii(); +#if wxUSE_WCHAR_T + void ConstructorsWithConversion(); + void ConversionEmpty(); + void ConversionWithNULs(); + void ConversionUTF7(); + void ConversionUTF8(); + void ConversionUTF16(); + void ConversionUTF32(); + + // test if converting s using the given encoding gives ws and vice versa + // + // if either of the first 2 arguments is NULL, the conversion is supposed + // to fail + void DoTestConversion(const char *s, const wchar_t *w, wxMBConv& conv); +#endif // wxUSE_WCHAR_T + DECLARE_NO_COPY_CLASS(UnicodeTestCase) }; @@ -43,7 +87,7 @@ private: CPPUNIT_TEST_SUITE_REGISTRATION( UnicodeTestCase ); // also include in it's own registry so that these tests can be run alone -CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( UnicodeTestCase, "UnicodeTestCase" ); +CPPUNIT_TEST_SUITE_NAMED_REGISTRATION( UnicodeTestCase, "Unicode" ); UnicodeTestCase::UnicodeTestCase() { @@ -63,3 +107,215 @@ void UnicodeTestCase::ToFromAscii() TEST_TO_FROM_ASCII( "additional \" special \t test \\ component \n :-)" ); } +#if wxUSE_WCHAR_T +void UnicodeTestCase::ConstructorsWithConversion() +{ + // the string "Déjà" in UTF-8 and wchar_t: + const unsigned char utf8Buf[] = {0x44,0xC3,0xA9,0x6A,0xC3,0xA0,0}; + const wchar_t wchar[] = {0x44,0xE9,0x6A,0xE0,0}; + const unsigned char utf8subBuf[] = {0x44,0xC3,0xA9,0x6A,0}; // just "Déj" + const char *utf8 = (char *)utf8Buf; + const char *utf8sub = (char *)utf8subBuf; + + wxString s1(utf8, wxConvUTF8); + wxString s2(wchar, wxConvUTF8); + +#if wxUSE_UNICODE + CPPUNIT_ASSERT( s1 == wchar ); + CPPUNIT_ASSERT( s2 == wchar ); +#else + CPPUNIT_ASSERT( s1 == utf8 ); + CPPUNIT_ASSERT( s2 == utf8 ); +#endif + + wxString sub(utf8sub, wxConvUTF8); // "Dej" substring + wxString s3(utf8, wxConvUTF8, 4); + wxString s4(wchar, wxConvUTF8, 3); + + CPPUNIT_ASSERT( s3 == sub ); + CPPUNIT_ASSERT( s4 == sub ); + +#if wxUSE_UNICODE + CPPUNIT_ASSERT ( wxString("\t[pl]open.format.Sformatuj dyskietkê=gfloppy %f", + wxConvUTF8) == wxT("") ); //should stop at pos 35 +#endif +} + +void UnicodeTestCase::ConversionEmpty() +{ + size_t len; + +#if wxUSE_UNICODE + wxCharBuffer buf = wxConvLibc.cWC2MB(L"", 0, &len); +#else // !wxUSE_UNICODE + wxWCharBuffer wbuf = wxConvLibc.cMB2WC("", 0, &len); +#endif // wxUSE_UNICODE/!wxUSE_UNICODE + + CPPUNIT_ASSERT_EQUAL(0u, len); +} + +void UnicodeTestCase::ConversionWithNULs() +{ +#if wxUSE_UNICODE + static const size_t lenNulString = 10; + + wxString szTheString(L"The\0String", wxConvLibc, lenNulString); + wxCharBuffer theBuffer = szTheString.mb_str(); + + CPPUNIT_ASSERT( memcmp(theBuffer.data(), "The\0String", + lenNulString + 1) == 0 ); + + wxString szTheString2("The\0String", wxConvLocal, lenNulString); + CPPUNIT_ASSERT_EQUAL( lenNulString, szTheString2.length() ); + CPPUNIT_ASSERT( wxTmemcmp(szTheString2.c_str(), L"The\0String", + lenNulString + 1) == 0 ); +#else // !wxUSE_UNICODE + wxString szTheString(wxT("TheString")); + szTheString.insert(3, 1, '\0'); + wxWCharBuffer theBuffer = szTheString.wc_str(wxConvLibc); + + CPPUNIT_ASSERT( memcmp(theBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 ); + + wxString szLocalTheString(wxT("TheString")); + szLocalTheString.insert(3, 1, '\0'); + wxWCharBuffer theLocalBuffer = szLocalTheString.wc_str(wxConvLocal); + + CPPUNIT_ASSERT( memcmp(theLocalBuffer.data(), L"The\0String", 11 * sizeof(wchar_t)) == 0 ); +#endif // wxUSE_UNICODE/!wxUSE_UNICODE +} + +void +UnicodeTestCase::DoTestConversion(const char *s, + const wchar_t *ws, + wxMBConv& conv) +{ +#if wxUSE_UNICODE + if ( ws ) + { + wxCharBuffer buf = conv.cWC2MB(ws, (size_t)-1, NULL); + + CPPUNIT_ASSERT( strcmp(buf, s) == 0 ); + } +#else // wxUSE_UNICODE + if ( s ) + { + wxWCharBuffer wbuf = conv.cMB2WC(s, (size_t)-1, NULL); + + if ( ws ) + { + CPPUNIT_ASSERT( wbuf.data() ); + CPPUNIT_ASSERT( wx_wcscmp(wbuf, ws) == 0 ); + } + else // conversion is supposed to fail + { + CPPUNIT_ASSERT_EQUAL( (wchar_t *)NULL, wbuf.data() ); + } + } +#endif // wxUSE_UNICODE/!wxUSE_UNICODE +} + +struct StringConversionData +{ + const char *str; + const wchar_t *wcs; +}; + +void UnicodeTestCase::ConversionUTF7() +{ + static const StringConversionData utf7data[] = + { + { "+-", L"+" }, + { "+--", L"+-" }, + +#ifdef wxHAVE_U_ESCAPE + { "+AKM-", L"\u00a3" }, +#endif // wxHAVE_U_ESCAPE + + // the following are invalid UTF-7 sequences + { "+", NULL }, + { "+~", NULL }, + { "a+", NULL }, + }; + + wxCSConv conv(_T("utf-7")); + for ( size_t n = 0; n < WXSIZEOF(utf7data); n++ ) + { + const StringConversionData& d = utf7data[n]; + DoTestConversion(d.str, d.wcs, conv); + DoTestConversion(d.str, d.wcs, wxConvUTF7); + } +} + +void UnicodeTestCase::ConversionUTF8() +{ + static const StringConversionData utf8data[] = + { +#ifdef wxHAVE_U_ESCAPE + { "\xc2\xa3", L"\u00a3" }, +#endif + { "\xc2", NULL }, + }; + + wxCSConv conv(_T("utf-8")); + for ( size_t n = 0; n < WXSIZEOF(utf8data); n++ ) + { + const StringConversionData& d = utf8data[n]; + DoTestConversion(d.str, d.wcs, conv); + DoTestConversion(d.str, d.wcs, wxConvUTF8); + } +} + +void UnicodeTestCase::ConversionUTF16() +{ + static const StringConversionData utf16data[] = + { +#ifdef wxHAVE_U_ESCAPE + { "\x04\x1f\x04\x40\x04\x38\x04\x32\x04\x35\x04\x42\0\0", + L"\u041f\u0440\u0438\u0432\u0435\u0442" }, + { "\x01\0\0b\x01\0\0a\x01\0\0r\0\0", L"\u0100b\u0100a\u0100r" }, +#endif + { "\0f\0o\0o\0\0", L"foo" }, + }; + + wxCSConv conv(wxFONTENCODING_UTF16BE); + for ( size_t n = 0; n < WXSIZEOF(utf16data); n++ ) + { + const StringConversionData& d = utf16data[n]; + DoTestConversion(d.str, d.wcs, conv); + } + + // special case: this string has consecutive NULs inside it which don't + // terminate the string, this exposed a bug in our conversion code which + // got confused in this case + size_t len; + wxWCharBuffer wbuf(conv.cMB2WC("\x01\0\0B\0C" /* A macron BC */, 6, &len)); + CPPUNIT_ASSERT_EQUAL( (size_t)3, len ); +} + +void UnicodeTestCase::ConversionUTF32() +{ + static const StringConversionData utf32data[] = + { +#ifdef wxHAVE_U_ESCAPE + { + "\0\0\x04\x1f\0\0\x04\x40\0\0\x04\x38\0\0\x04\x32\0\0\x04\x35\0\0\x04\x42\0\0\0\0", + L"\u041f\u0440\u0438\u0432\u0435\u0442" }, +#endif + { "\0\0\0f\0\0\0o\0\0\0o\0\0\0\0", L"foo" }, + }; + + wxCSConv conv(wxFONTENCODING_UTF32BE); + for ( size_t n = 0; n < WXSIZEOF(utf32data); n++ ) + { + const StringConversionData& d = utf32data[n]; + DoTestConversion(d.str, d.wcs, conv); + } + + size_t len; + wxWCharBuffer wbuf(conv.cMB2WC("\0\0\x01\0\0\0\0B\0\0\0C" /* A macron BC */, + 12, &len)); + CPPUNIT_ASSERT_EQUAL( (size_t)3, len ); +} + +#endif // wxUSE_WCHAR_T +