[PATCH libX11] Compose.man: fix escaped hexadecimal char description

[Ran Benita]

On Sun, Feb 16, 2014 at 07:53:33PM -0500, James Cloos wrote:
> >>>>> "RB" == Ran Benita <ran234 at gmail.com> writes:
> 
> RB> But the grammar in the parser and the implementation say:
> RB>     ESCAPED_CHAR  ::= ('\\' | '\"' | OCTAL | HEX )
> RB>     HEX           ::= '\' (x|X) HEX_CHAR [HEX_CHAR]]
> RB>     HEX_CHAR      ::= (0|1|2|3|4|5|6|7|8|9|A|B|C|D|E|F|a|b|c|d|e|f)
> 
> Given that the original code was written for eight-bit encodings, I
> suppose that is not surprising that the grammar is only cabable of
> octal and hex octets.
> 
> I presume a different parser must be used for the XLC_LOCALE files,
> as those have longer hex escapes (also using \x).

Yes, that's a different parser. I haven't looked closely at the
XLC_LOCALE files, they seem scary and should be largely irrelevant for
modern systems, no?

> Should we fix the doc or update the parser to handle larger hex?

I don't think so, currently it's just simple C-like char <-> escape
correspondence. And I'd hope there won't be any more non-UTF-8
multiple-bytes encoding that we need to support in the future.

> And on that front, should we add a \u escape to insert the utf-8 for
> a given code point? 

If the file is UTF-8 already, I think it makes more sense to just write
the string directly, without escaping it. But if you think this makes
sense I can try to add it.

Ran

> -JimC
> --
> James Cloos <cloos at jhcloos.com>         OpenPGP: 1024D/ED7DAEA6