Dear GCC,

     This patch adds macros to the general preprocessor that allow
users to understand what the execution and wide execution charsets
are, which are used for "bark" and L"meow" literals in C-family
languages.

     The goal of this is to enable individuals in capable languages
like C and C++ to determine the encoding of string literals and,
hopefully, transcode those literals to other encodings. For example,
data stored in `char[]` on an IBM machine with EBCDIC can be
(losslessly, or with a compile-time warning) transcoded to UTF-8 at
constant-expression time in C++ with this patch, without having to use
a series of complicated architecture and platform definitions to
figure out what the encoding of string literals and wide character
literals might be.

     The names are meaningful because they are tied directly to iconv,
which means there is a strong reference between the name and the code
that encodes/decodes a proper sequence. Therefore, we just present the
names.

     Does this sound useful?

Sincerely,
JeanHeyd

Patch notes: the strings passed to the charset creation routines are
all static and/or allocated far at the beginning of the program and
never deallocated until the end of the invocation, so it seems safe to
just store a normal pointer to it.

2020-10-08  JeanHeyd "ThePhD" Meneide  <phdofthehouse@gmail.com>

        * gcc/c-family/c-cppbuiltin.c: Add predefined macro
definitions for charsets
        * gcc/doc/cpp.texi: Document new predefined macro.
        * gcc/testsuite/c-c++-common/cpp/wide-narrow-predef-macros.c (new):
          New test for macro definitions to always exist.
        * libcpp/include/cpplib.h: Add functions declarations for
          retrieving charset names
        * libcpp/directives.c: Add function definitions to retrieve charset
          names.
        * libcpp/internal.h: Add to/from name preservations