patch to convert input character set to source character set

[Zack Weinberg]

Chandra Chavva <cchavva@redhat.com> writes:

> I have attached the patch which converters the input character set to
> source character set, UTF-8 or UTF-EDCDIC depending on basic execution
> character set. Also introduced new option --input-character= to
> specify the default character set of the source file. The input
> character set is set to "ISO-8859-1" by default.
>
> Zack has already approved this patch.

I have *NOT* approved this patch. I told you that I couldn't find any
problems but you had to get Neil to sign off on it.

> + /* The first 256 characters of ISO-8859-1 are similar to the first 256
> +    characters of Unicode, therefore the input value can be passed directly
> +    to one_cppchar_to_utf8. */

This comment is still misleading.  It should read

/* The first 256 code points of ISO 8859.1 have the same numeric
   values as the first 256 code points of Unicode, therefore the
   incoming ISO 8859.1 character can be passed directly to
   one_cppchar_to_utf8 (which expects a Unicode value).  */

zw