Remove most mentions of wxUSE_UNICODE from the documentation

Still document this symbol itself, but only as being defined as 1.

docs/doxygen/mainpages/const_cpp.h

@@ -368,7 +368,8 @@ more details.
         and so the usual idiom which allows to support both cases is to first include
         @c wx/wxprec.h} and then, inside <tt>\#ifndef WX_PRECOMP</tt>, individual
         headers you need.}
-@itemdef{_UNICODE and UNICODE, both are defined if wxUSE_UNICODE is set to @c 1}
+@itemdef{_UNICODE and UNICODE, are both always defined in current wxWidgets
+    versions.}
 @itemdef{wxUSE_GUI,
         this particular feature test macro is defined to 1
         when compiling or using the library with the GUI features activated,

docs/doxygen/mainpages/const_wxusedef.h

@@ -34,8 +34,8 @@ library:
     particular is implicitly convertible to @c char* and not std::[w]string).}
 @itemdef{wxUSE_STD_IOSTREAM, Standard C++ classes are used instead of or in
     addition to wx stream classes.}
-@itemdef{wxUSE_UNICODE, Compiled with Unicode support (default in wxWidgets
-    3.0, non-Unicode build is deprecated and will be removed in the future).}
+@itemdef{wxUSE_UNICODE, Always defined as 1 in wxWidgets 3.3 and later, only
+    exists for compatibility.}
 @itemdef{wxUSE_UNICODE_WCHAR, wxString uses wchar_t buffer for internal storage
     (default).}
 @itemdef{wxUSE_UNICODE_UTF8, wxString uses UTF-8 for internal storage (mostly

docs/doxygen/overviews/string.h

@@ -402,12 +402,9 @@ difference the change to @c EXTRA_ALLOC makes to your program.
 
 @section overview_string_settings wxString Related Compilation Settings
 
-The main option affecting wxString is @c wxUSE_UNICODE which is now always
-defined as @c 1 by default to indicate Unicode support. You may set it to 0 to
-disable Unicode support in wxString and elsewhere in wxWidgets but this is @e
-strongly not recommended.
-
-Another option affecting wxWidgets is @c wxUSE_UNICODE_WCHAR which is also 1 by
+wxString always supports Unicode in wxWidgets 3.3 and later, but it may use
+either UTF-8 or `wchar_t` (which, in turn, may use either UTF-16 or UTF-32)
+internally. It uses the latter if @c wxUSE_UNICODE_WCHAR is set, which is the case by
 default. You may want to set it to 0 and set @c wxUSE_UNICODE_UTF8 to 1 instead
 to use UTF-8 internally. wxString still provides the same API in this case, but
 using UTF-8 has performance implications as explained in @ref

docs/doxygen/overviews/unicode.h

@@ -215,9 +215,7 @@ size and removes the need for conversions in more cases.
 
 @subsection overview_unicode_settings Unicode Related Preprocessor Symbols
 
-@c wxUSE_UNICODE is defined as 1 now to indicate Unicode support. It can be
-explicitly set to 0 in @c setup.h under MSW or you can use @c \--disable-unicode
-under Unix but doing this is strongly discouraged. By default, @c
+@c wxUSE_UNICODE is always defined as 1 in wxWidgets 3.3 or later. By default, @c
 wxUSE_UNICODE_WCHAR is also defined as 1, however in UTF-8 build (described in
 the previous section), it is set to 0 and @c wxUSE_UNICODE_UTF8, which is
 usually 0, is set to 1 instead. In the latter case, @c wxUSE_UTF8_LOCALE_ONLY

interface/wx/chartype.h

@@ -98,36 +98,29 @@
 #define _T(string)
 
 /**
-    wxChar is defined to be
-    \- @c char when <tt>wxUSE_UNICODE==0</tt>
-    \- @c wchar_t when <tt>wxUSE_UNICODE==1</tt> (the default).
+    wxChar is a compatibility typedef always defined as @c wchar_t now.
+
+    Note that it is not affected by @c wxUSE_UNICODE_UTF8 option.
 */
-typedef wxUSE_UNICODE_dependent wxChar;
+typedef wchar_t wxChar;
 
 /**
-    wxSChar is defined to be
-    \- <tt>signed char</tt> when <tt>wxUSE_UNICODE==0</tt>
-    \- @c wchar_t when <tt>wxUSE_UNICODE==1</tt> (the default).
+    wxSChar is a compatibility typedef always defined as @c wchar_t now.
 */
-typedef wxUSE_UNICODE_dependent wxSChar;
+typedef wchar_t wxSChar;
 
 /**
-    wxUChar is defined to be
-    \- <tt>unsigned char</tt> when <tt>wxUSE_UNICODE==0</tt>
-    \- @c wchar_t when <tt>wxUSE_UNICODE==1</tt> (the default).
+    wxUChar is a compatibility typedef always defined as @c wchar_t now.
 */
-typedef wxUSE_UNICODE_dependent wxUChar;
+typedef wchar_t wxUChar;
 
 /**
     wxStringCharType is defined to be:
-    \- @c char when <tt>wxUSE_UNICODE==0</tt>
-    \- @c char when <tt>wxUSE_UNICODE_WCHAR==0</tt> and <tt>wxUSE_UNICODE==1</tt>
-    \- @c wchar_t when <tt>wxUSE_UNICODE_WCHAR==1</tt> and <tt>wxUSE_UNICODE==1</tt>
+    \- @c char when <tt>wxUSE_UNICODE_WCHAR==0</tt>
+    \- @c wchar_t when <tt>wxUSE_UNICODE_WCHAR==1</tt>
 
-    The @c wxUSE_UNICODE_WCHAR symbol is defined to @c 1 when building on
-    Windows while it's defined to @c 0 when building on Unix, Linux or macOS.
-    (Note that @c wxUSE_UNICODE_UTF8 symbol is defined as the opposite of
-    @c wxUSE_UNICODE_WCHAR.)
+    The @c wxUSE_UNICODE_WCHAR symbol is defined by default, but may be turned
+    off in which case @c wxUSE_UNICODE_UTF8 is turned on.
 
     Note that wxStringCharType (as the name says) is the type used by wxString
     for internal storage of the characters.

interface/wx/dataobj.h

@@ -753,18 +753,16 @@ public:
     /**
         Returns 2 under wxMac and wxGTK, where text data coming from the
         clipboard may be provided as ANSI (@c wxDF_TEXT) or as Unicode text
-        (@c wxDF_UNICODETEXT, but only when @c wxUSE_UNICODE==1).
+        (@c wxDF_UNICODETEXT).
 
-        Returns 1 under other platforms (e.g. wxMSW) or when building in ANSI mode
-        (@c wxUSE_UNICODE==0).
+        Returns 1 under other platforms (e.g. wxMSW).
     */
     virtual size_t GetFormatCount(wxDataObject::Direction dir = wxDataObject::Get) const;
 
     /**
         Returns the preferred format supported by this object.
 
-        This is @c wxDF_TEXT or @c wxDF_UNICODETEXT depending on the platform
-        and from the build mode (i.e. from @c wxUSE_UNICODE).
+        This is @c wxDF_TEXT or @c wxDF_UNICODETEXT depending on the platform.
     */
     const wxDataFormat& GetFormat() const;
 

interface/wx/event.h

@@ -1632,9 +1632,6 @@ public:
         If the key pressed doesn't have any character value (e.g. a cursor key)
         this method will return @c WXK_NONE. In this case you should use
         GetKeyCode() to retrieve the value of the key.
-
-        This function is only available in Unicode build, i.e. when
-        @c wxUSE_UNICODE is 1.
     */
     wxChar GetUnicodeKey() const;
 

interface/wx/strconv.h

@@ -231,18 +231,10 @@ public:
      */
     wxWCharBuffer cMB2WC(const wxCharBuffer& buf) const;
 
-    ///@{
     /**
-        Converts from multibyte encoding to the current wxChar type (which
-        depends on whether wxUSE_UNICODE is set to 1).
-
-        If wxChar is char, it returns the parameter unaltered. If wxChar is
-        wchar_t, it returns the result in a wxWCharBuffer. The macro wxMB2WXbuf
-        is defined as the correct return type (without const).
+        Converts from multibyte encoding to `wchar_t`.
     */
-    const char* cMB2WX(const char* psz) const;
     wxWCharBuffer cMB2WX(const char* psz) const;
-    ///@}
 
     /**
         Converts from Unicode to multibyte encoding by calling FromWChar() and

interface/wx/xml/xml.h

@@ -55,9 +55,6 @@ enum wxXmlNodeType
     The @c wxXML_DOCUMENT_TYPE_NODE is not implemented at this time. Instead,
     you should get and set the DOCTYPE values using the wxXmlDocument class.
 
-    If @c wxUSE_UNICODE is 0, all strings are encoded in the encoding given to
-    wxXmlDocument::Load (default is UTF-8).
-
     @note
     Once a wxXmlNode has been added to a wxXmlDocument it becomes owned by the
     document and this has two implications. Firstly, the wxXmlDocument takes