added wxString::FromUTF8/ToUTF8/utf8_str()

git-svn-id: https://svn.wxwidgets.org/svn/wx/wxWidgets/trunk@45842 c3d73ce0-8a6f-49c7-b76d-6d57e0e08775

diff --git a/docs/latex/wx/wxstring.tex b/docs/latex/wx/wxstring.tex
index 3a35293c249189d2edd8b697b431f99050101ff9..20b9f19f698990986422dcc27fe52fba16dbaba2 100644 (file)
--- a/docs/latex/wx/wxstring.tex
+++ b/docs/latex/wx/wxstring.tex
@@ -779,6 +779,19 @@ Use \helpref{wxString constructors}{wxstringconstruct} if you
 need to convert from another charset.
 
 
 need to convert from another charset.
 
 

+\membersection{wxString::FromUTF8}\label{wxstringfromutf8}
+
+\func{static wxString }{FromUTF8}{\param{const char*}{ s}}
+
+\func{static wxString }{FromUTF8}{\param{const char*}{ s}, \param{size\_t}{ len}}
+
+Converts C string encoded in UTF-8 to wxString.
+
+Note that this method assumes that \arg{s} is a valid UTF-8 sequence and
+doesn't do any validation in release builds, it's validity is only checked in
+debug builds.
+
+

 \membersection{wxString::GetChar}\label{wxstringgetchar}
 
 \constfunc{wxChar}{GetChar}{\param{size\_t}{ n}}
 \membersection{wxString::GetChar}\label{wxstringgetchar}
 
 \constfunc{wxChar}{GetChar}{\param{size\_t}{ n}}


@@ -1166,12 +1179,10 @@ This is a convenience method useful when storing binary data in wxString.
 
 \constfunc{const char*}{ToAscii}{\void}
 
 
 \constfunc{const char*}{ToAscii}{\void}
 

-Converts the string to an ASCII, 7-bit string (ANSI builds only).
-

 \constfunc{const wxCharBuffer}{ToAscii}{\void}
 
 Converts the string to an ASCII, 7-bit string in the form of
 \constfunc{const wxCharBuffer}{ToAscii}{\void}
 
 Converts the string to an ASCII, 7-bit string in the form of

-a wxCharBuffer (Unicode builds only).
+a wxCharBuffer (Unicode builds only) or a C string (ANSI builds).

 
 Note that this conversion only works if the string contains only ASCII
 characters. The \helpref{mb\_str}{wxstringmbstr} method provides more
 
 Note that this conversion only works if the string contains only ASCII
 characters. The \helpref{mb\_str}{wxstringmbstr} method provides more


@@ -1263,6 +1274,15 @@ bit integer numbers.
 Please see \helpref{ToLongLong}{wxstringtolonglong} for additional remarks.
 
 
 Please see \helpref{ToLongLong}{wxstringtolonglong} for additional remarks.
 
 

+\membersection{wxString::ToUTF8}\label{wxstringtoutf8}
+
+\constfunc{const char*}{ToUTF8}{\void}
+
+\constfunc{const wxCharBuffer}{ToUF8}{\void}
+
+Same as \helpref{utf8\_str}{wxstringutf8str}.
+
+

 \membersection{wxString::Trim}\label{wxstringtrim}
 
 \func{wxString\&}{Trim}{\param{bool}{ fromRight = true}}
 \membersection{wxString::Trim}\label{wxstringtrim}
 
 \func{wxString\&}{Trim}{\param{bool}{ fromRight = true}}


@@ -1316,6 +1336,18 @@ The same as MakeUpper.
 This is a wxWidgets 1.xx compatibility function; you should not use it in new code.
 
 
 This is a wxWidgets 1.xx compatibility function; you should not use it in new code.
 
 

+\membersection{wxString::utf8\_str}\label{wxstringutf8str}
+
+\constfunc{const char*}{utf8\_str}{\void}
+
+\constfunc{const wxCharBuffer}{utf8\_str}{\void}
+
+Converts the strings contents to UTF-8 and returns it either as a temporary
+wxCharBuffer object or as a pointer to the internal string contents in
+UTF-8 build.
+% FIXME-UTF8: link to a topic explaining UTF-8 build here
+
+

 \membersection{wxString::wc\_str}\label{wxstringwcstr}
 
 \constfunc{const wchar\_t*}{wc\_str}{\param{wxMBConv\&}{ conv}}
 \membersection{wxString::wc\_str}\label{wxstringwcstr}
 
 \constfunc{const wchar\_t*}{wc\_str}{\param{wxMBConv\&}{ conv}}

diff --git a/include/wx/string.h b/include/wx/string.h
index 3da1f2f10d494b1bb59beef69c1dd8e7ab576484..6a47d751a09d423ca50dbe2ccd57fa291b8637fa 100644 (file)
--- a/include/wx/string.h
+++ b/include/wx/string.h
@@ -1129,6 +1129,36 @@ public:
     const char *ToAscii() const { return c_str(); }
 #endif // Unicode/!Unicode
 
     const char *ToAscii() const { return c_str(); }
 #endif // Unicode/!Unicode
 

+    // conversion to/from UTF-8:
+#if wxUSE_UNICODE_UTF8
+    static wxString FromUTF8(const char *utf8)
+    {
+      wxASSERT( wxStringOperations::IsValidUtf8String(utf8) );
+      return FromImpl(wxStringImpl(utf8));
+    }
+    static wxString FromUTF8(const char *utf8, size_t len)
+    {
+      wxASSERT( wxStringOperations::IsValidUtf8String(utf8, len) );
+      return FromImpl(wxStringImpl(utf8, len));
+    }
+    const char* utf8_str() const { return wx_str(); }
+    const char* ToUTF8() const { return wx_str(); }
+#elif wxUSE_UNICODE_WCHAR
+    static wxString FromUTF8(const char *utf8)
+      { return wxString(utf8, wxConvUTF8); }
+    static wxString FromUTF8(const char *utf8, size_t len)
+      { return wxString(utf8, wxConvUTF8, len); }
+    const wxCharBuffer utf8_str() const { return mb_str(wxConvUTF8); }
+    const wxCharBuffer ToUTF8() const { return utf8_str(); }
+#else // ANSI
+    static wxString FromUTF8(const char *utf8)
+      { return wxString(wxConvUTF8.cMB2WC(utf8)); }
+    static wxString FromUTF8(const char *utf8, size_t len)
+      { return wxString(wxConvUTF8.cMB2WC(utf8, len == npos ? wxNO_LEN : len)); }
+    const wxCharBuffer utf8_str() const { return wxConvUTF8.cWC2MB(wc_str()); }
+    const wxCharBuffer ToUTF8() const { return utf8_str(); }
+#endif
+

     // functions for storing binary data in wxString:
 #if wxUSE_UNICODE
     static wxString From8BitData(const char *data, size_t len)
     // functions for storing binary data in wxString:
 #if wxUSE_UNICODE
     static wxString From8BitData(const char *data, size_t len)