Re: Language Choice and Coding style

[Mark Eichin <eichin@cygnus.com>]

Intriguing.

>     p one, p two: char const *;

Interestingly enough, this eliminates the common mistake of using

	char const *p1, p2; /* A */
when 
	char const *p1, *p2; /* B */

is what is intended. (Yes, I know you'll get a compiler error with A
in most actual uses, but you can construct cases where you don't...)

One obvious nit is the C bitfield specifier... but it turns out that
it is easy to leave unchanged from context (by putting it in the type,
not the variable... ie:
	int j1:1;
becomes
	j one: int: 1;
with no ambiguity.)

The conclusion
>    it is not our *compiler technology* that forbids embedded blanks
>    in identifiers, it is our *programming language standards*.

is interesting, because the C standard was explicitly an attempt to
codify existing practice in what had begun as a low-level language
design, and not redesign it.  C++ is correspondingly limited to the
same thing by it's C heritage.  Ada doesn't have these particular
constraints... so were spaces in identifiers simply not considered
[since noone else had used them in 30 years] or where they
deliberately omitted?

					_Mark_ <eichin@cygnus.com>
					Cygnus Support, Eastern USA