wxApp destructor is virtual [ patch 1283503 ]

[wxWidgets.git] / docs / latex / wx / mbcnvfil.tex

diff --git a/docs/latex/wx/mbcnvfil.tex b/docs/latex/wx/mbcnvfil.tex
index e5e04f8b7a91a0d1e49fc9b8bdcb8eaab7c072e1..fc3f3d8388703d1a23035df2cf122212033e4806 100644 (file)
--- a/docs/latex/wx/mbcnvfil.tex
+++ b/docs/latex/wx/mbcnvfil.tex
@@ -1,17 +1,22 @@
-%
-% automatically generated by HelpGen from
-% ../include/wx/strconv.h at 25/Mar/00 10:20:56
-%
-
 \section{\class{wxMBConvFile}}\label{wxmbconvfile}
 
-This class converts file names between filesystem multibyte encoding and
-Unicode. It has one predefined instance, {\bf wxConvFileName}.
-Since some platforms (e.g. Win32) use Unicode in the filenames,
-and others (e.g. Unix) use multibyte encodings, this class should only
+This class used to define the class instance 
+{\bf wxConvFileName}, but nowadays {\bf wxConvFileName} is
+either of type wxConvLibc (on most platforms) or wxConvUTF8
+(on MacOS X). {\bf wxConvFileName} converts filenames between 
+filesystem multibyte encoding and Unicode. {\bf wxConvFileName} 
+can also be set to a something else at run-time which is used 
+e.g. by wxGTK to use a class which checks the environment 
+variable {\bf G\_FILESYSTEM\_ENCODING} indicating that filenames 
+should not be interpreted as UTF8 and also for converting 
+invalid UTF8 characters (e.g. if there is a filename in iso8859\_1)
+to strings with octal values. 
+
+Since some platforms (such as Win32) use Unicode in the filenames,
+and others (such as Unix) use multibyte encodings, this class should only
 be used directly if wxMBFILES is defined to 1. A convenience macro,
-wxFNCONV, is defined to wxConvFile.cWX2MB in this case. You could use it
-like this:
+wxFNCONV, is defined to wxConvFileName->cWX2MB in this case. You could
+use it like this:
 
 \begin{verbatim}
 wxChar *name = wxT("rawfile.doc");