Merge branch 'utf8-stdstring-interop' of https://github.com/minoki/wxWidgets

Make it easier to interoperate with the code using UTF-8-encoded std::strings.

Closes #17461.

docs/changes.txt

@@ -62,6 +62,7 @@ All:
 - wxLogInfo() now logs messages if the log level is high enough, even without
   wxLog::SetVerbose() which now only affects wxLogVerbose().
 - Add wxFileType::GetExpandedCommand() (troelsk).
+- Make it easier to convert to/from UTF-8-encoded std::string (ARATA Mizuki).
 
 All (GUI):
 

include/wx/string.h

@@ -1241,12 +1241,17 @@ public:
     // wxStringImpl is std::string in the encoding we want
     #define wxStringToStdStringRetType const std::string&
     const std::string& ToStdString() const { return m_impl; }
+    std::string ToStdString(const wxMBConv& conv) const
+    {
+        wxScopedCharBuffer buf(mb_str(conv));
+        return std::string(buf.data(), buf.length());
+    }
   #else
     // wxStringImpl is either not std::string or needs conversion
     #define wxStringToStdStringRetType std::string
-    std::string ToStdString() const
+    std::string ToStdString(const wxMBConv& conv = wxConvLibc) const
     {
-        wxScopedCharBuffer buf(mb_str());
+        wxScopedCharBuffer buf(mb_str(conv));
         return std::string(buf.data(), buf.length());
     }
   #endif
@@ -1611,6 +1616,24 @@ public:
         return FromImpl(wxStringImpl(utf8, len));
     }
 
+#if wxUSE_STD_STRING
+    static wxString FromUTF8Unchecked(const std::string& utf8)
+    {
+        wxASSERT( wxStringOperations::IsValidUtf8String(utf8.c_str(), utf8.length()) );
+        /*
+          Note that, under wxUSE_UNICODE_UTF8 and wxUSE_STD_STRING, wxStringImpl can be
+          initialized with a std::string whether wxUSE_STL_BASED_WXSTRING is 1 or not.
+        */
+        return FromImpl(utf8);
+    }
+    static wxString FromUTF8(const std::string& utf8)
+    {
+        if ( utf8.empty() || !wxStringOperations::IsValidUtf8String(utf8.c_str(), utf8.length()) )
+            return wxString();
+        return FromImpl(utf8);
+    }
+#endif
+
     const wxScopedCharBuffer utf8_str() const
         { return wxCharBuffer::CreateNonOwned(m_impl.c_str(), m_impl.length()); }
 
@@ -1627,6 +1650,12 @@ public:
                       "string must be valid UTF-8" );
         return s;
     }
+#if wxUSE_STD_STRING
+    static wxString FromUTF8(const std::string& utf8)
+      { return FromUTF8(utf8.c_str(), utf8.length()); }
+    static wxString FromUTF8Unchecked(const std::string& utf8)
+      { return FromUTF8Unchecked(utf8.c_str(), utf8.length()); }
+#endif
     const wxScopedCharBuffer utf8_str() const { return mb_str(wxMBConvUTF8()); }
 #else // ANSI
     static wxString FromUTF8(const char *utf8)
@@ -1654,6 +1683,12 @@ public:
 
         return wxString(buf.data(), wlen);
     }
+#if wxUSE_STD_STRING
+    static wxString FromUTF8(const std::string& utf8)
+      { return FromUTF8(utf8.c_str(), utf8.length()); }
+    static wxString FromUTF8Unchecked(const std::string& utf8)
+      { return FromUTF8Unchecked(utf8.c_str(), utf8.length()); }
+#endif
     const wxScopedCharBuffer utf8_str() const
       { return wxMBConvUTF8().cWC2MB(wc_str()); }
 #endif

interface/wx/string.h

@@ -85,8 +85,8 @@
         - String in UTF-8 encoding using wxString::utf8_str().
         - String in any given encoding using mb_str() with the appropriate
         wxMBConv object. This is also a potentially destructive operation.
-        - Standard @c std::string using wxString::ToStdString(). The contents
-        of the returned string use the current locale encoding, so this
+        - Standard @c std::string using wxString::ToStdString(). The encoding
+        of the returned string is specified with a wxMBConv object, so this
         conversion is potentially destructive as well.
         - Wide C string using wxString::wc_str().
         - Standard @c std::wstring using wxString::ToStdWstring().
@@ -745,10 +745,10 @@ public:
     const TYPE ToAscii(char replaceWith = '_') const;
 
     /**
-        Return the string as an std::string in current locale encoding.
+        Return the string as an std::string using @e conv's wxMBConv::cWC2MB method.
 
-        Note that if the conversion of (Unicode) string contents to the current
-        locale fails, the return string will be empty. Be sure to check for
+        Note that if the conversion of (Unicode) string contents using @e conv
+        fails, the return string will be empty. Be sure to check for
         this to avoid silent data loss.
 
         Instead of using this function it's also possible to write
@@ -760,9 +760,12 @@ public:
         @endcode
         but using ToStdString() may make the code more clear.
 
+        @param conv
+            The converter to be used. This parameter is new in wxWidgets 3.1.1.
+
         @since 2.9.1
     */
-    std::string ToStdString() const;
+    std::string ToStdString(const wxMBConv& conv = wxConvLibc) const;
 
     /**
         Return the string as an std::wstring.
@@ -1813,10 +1816,14 @@ public:
         alternative to this function called FromUTF8Unchecked() which, unlike
         this one, doesn't check that the input string is valid.
 
+        The overload taking @c std::string is only available starting with
+        wxWidgets 3.1.1.
+
         @since 2.8.4
     */
     static wxString FromUTF8(const char* s);
     static wxString FromUTF8(const char* s, size_t len);
+    static wxString FromUTF8(const std::string& s);
     //@}
 
     //@{
@@ -1833,10 +1840,14 @@ public:
         string to this function will result in creating a corrupted wxString
         and all the subsequent operations on it will be undefined.
 
+        The overload taking @c std::string is only available starting with
+        wxWidgets 3.1.1.
+
         @since 2.8.9
     */
     static wxString FromUTF8Unchecked(const char* s);
     static wxString FromUTF8Unchecked(const char* s, size_t len);
+    static wxString FromUTF8Unchecked(const std::string& s);
     //@}
 };
 

tests/strings/stdstrings.cpp

@@ -608,5 +608,12 @@ void StdStringTestCase::StdConversion()
 
     wxStdWideString s8(s4);
     CPPUNIT_ASSERT( s8 == "hello" );
+
+    std::string s9("\xF0\x9F\x90\xB1\0\xE7\x8C\xAB", 9); /* U+1F431 U+0000 U+732B */
+    wxString s10 = wxString::FromUTF8(s9);
+    CPPUNIT_ASSERT_EQUAL( s9, s10.ToStdString(wxConvUTF8) );
+
+    std::string s11("xyz\0\xFF", 5); /* an invalid UTF-8 sequence */
+    CPPUNIT_ASSERT_EQUAL( wxString::FromUTF8(s11), "" );
 }
 #endif // wxUSE_STD_STRING