Amebis/wxWidgets
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'pr1312-no-unsafe-wxstring-conv'

Browse Source 
Add a way to optionally disallow all implicit conversions between
wxString and "const char*".

See https://github.com/wxWidgets/wxWidgets/pull/1312

See #18113.
This commit is contained in:
Vadim Zeitlin
2020-07-20 15:47:17 +02:00
parent 4beccf8883 05cce8d89d
commit 4552009805
380 changed files with 4734 additions and 1249 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
docs/doxygen/mainpages/const_cpp.h
View File

@@ -375,6 +375,11 @@ more details.
set to 1 for compatibility reasons as @c -DwxNO_UNSAFE_WXSTRING_CONV
can be used only compiling the application code, without rebuilding the
library. Support for this option appeared in wxWidgets 3.1.1.}
@itemdef{wxNO_IMPLICIT_WXSTRING_ENCODING,
this symbol is not defined by wxWidgets itself, but can be defined by
the applications using the library to disable implicit
conversions from and to <tt>const char*</tt> in wxString class.
Support for this option appeared in wxWidgets 3.1.4.}
@itemdef{WXMAKINGDLL_XXX,
used internally and defined when building the
library @c XXX as a DLL; when a monolithic wxWidgets build is used only a

30
docs/doxygen/overviews/string.h
View File

@@ -236,12 +236,40 @@ arguments should take <tt>const wxString&</tt> (this makes assignment to the
strings inside the function faster) and all functions returning strings
should return wxString - this makes it safe to return local variables.
Finally note that wxString uses the current locale encoding to convert any C string
Note that wxString uses by default the current locale encoding to convert any C string
literal to Unicode. The same is done for converting to and from @c std::string
and for the return value of c_str().
For this conversion, the @a wxConvLibc class instance is used.
See wxCSConv and wxMBConv.
It is also possible to disable any automatic conversions from C
strings to Unicode. This can be useful when the @a wxConvLibc encoding
is not appropriate for the current software and platform. The macro @c
wxNO_IMPLICIT_WXSTRING_ENCODING disables all implicit conversions, and
forces the code to explicitly indicate the encoding of all C strings.
Finally note that encodings, either implicitly or explicitly selected,
may not be able to represent all the string's characters. The result
in this case is undefined: the string may be empty, or the
unrepresentable characters may be missing or wrong.
@code
wxString s;
// s = "world"; does not compile with wxNO_IMPLICIT_WXSTRING_ENCODING
s = wxString::FromAscii("world"); // Always compiles
s = wxASCII_STR("world"); // shorthand for the above
s = wxString::FromUTF8("world"); // Always compiles
s = wxString("world", wxConvLibc); // Always compiles, explicit encoding
s = wxASCII_STR("Grüße"); // Always compiles but encoding fails
const char *c;
// c = s.c_str(); does not compile with wxNO_IMPLICIT_WXSTRING_ENCODING
// c = s.mb_str(); does not compile with wxNO_IMPLICIT_WXSTRING_ENCODING
c = s.ToAscii(); // Always compiles, encoding may fail
c = s.ToUTF8(); // Always compiles, encoding never fails
c = s.utf8_str(); // Alias for the above
c = s.mb_str(wxConvLibc); // Always compiles, explicit encoding
@endcode
@subsection overview_string_iterating Iterating wxString Characters