+ for ( size_t n = 0; n < WXSIZEOF(utf7data); n++ )
+ {
+ const StringConversionData& d = utf7data[n];
+
+ // converting to/from UTF-7 using iconv() currently doesn't work
+ // because of several problems:
+ // - GetMBNulLen() doesn't return correct result (iconv converts L'\0'
+ // to an incomplete and anyhow nonsensical "+AA" string)
+ // - iconv refuses to convert "+-" (although it converts "+-\n" just
+ // fine, go figure)
+ //
+ // I have no idea how to fix this so just disable the test for now
+#if 0
+ d.Test(n, wxCSConv("utf-7"));
+#endif
+ d.Test(n, 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];
+ d.Test(n, conv);
+ d.Test(n, 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];
+ d.Test(n, 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];
+ d.Test(n, 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 );
+}
+
+void UnicodeTestCase::IsConvOk()
+{
+ CPPUNIT_ASSERT( wxCSConv(wxFONTENCODING_SYSTEM).IsOk() );
+ CPPUNIT_ASSERT( wxCSConv("US-ASCII").IsOk() );
+ CPPUNIT_ASSERT( wxCSConv("UTF-8").IsOk() );
+ CPPUNIT_ASSERT( !wxCSConv("NoSuchConversion").IsOk() );
+
+#ifdef __WINDOWS__
+ CPPUNIT_ASSERT( wxCSConv("WINDOWS-437").IsOk() );
+#endif
+}
+
+#if wxUSE_UNICODE
+void UnicodeTestCase::Iteration()
+{
+ // "czech" in Czech ("cestina"):
+ static const char *textUTF8 = "\304\215e\305\241tina";
+ static const wchar_t textUTF16[] = {0x10D, 0x65, 0x161, 0x74, 0x69, 0x6E, 0x61, 0};
+
+ wxString text(wxString::FromUTF8(textUTF8));
+ CPPUNIT_ASSERT( wxStrcmp(text.wc_str(), textUTF16) == 0 );
+
+ // verify the string was decoded correctly:
+ {
+ size_t idx = 0;
+ for ( wxString::const_iterator i = text.begin(); i != text.end(); ++i, ++idx )
+ {
+ CPPUNIT_ASSERT( *i == textUTF16[idx] );
+ }
+ }
+
+ // overwrite the string with something that is shorter in UTF-8:
+ {
+ for ( wxString::iterator i = text.begin(); i != text.end(); ++i )
+ *i = 'x';
+ }
+
+ // restore the original text now:
+ {
+ wxString::iterator end1 = text.end();
+ wxString::const_iterator end2 = text.end();
+
+ size_t idx = 0;
+ for ( wxString::iterator i = text.begin(); i != text.end(); ++i, ++idx )
+ {
+ *i = textUTF16[idx];
+
+ CPPUNIT_ASSERT( end1 == text.end() );
+ CPPUNIT_ASSERT( end2 == text.end() );
+ }
+
+ CPPUNIT_ASSERT( end1 == text.end() );
+ CPPUNIT_ASSERT( end2 == text.end() );
+ }
+
+ // and verify it again:
projects / wxWidgets.git / blobdiff