[C PATCH]: tidy some code

[Bonzini]

> The point was that some other part of the sentence may need to change to
> agree with the gender of 'struct foo' / 'union foo' / 'enum foo', not that
> the keyword is translated.

Yes, In Italian I say 'una struct' (where una is the feminine article) but
'un enum' (where un is the masculine article) because in Italian `structure'
is feminine but `type' is masculine (enum is then masculine because it is
short for enumerative type).

Another case is where you use apostrophed forms.  For example, my patch also
separates cases for `increment' and `decrement' (i say `l'incremento' and
`il decremento' in Italian even though both are masculine).

In this particular case, Italian uses the same gender and article for struct
and union, but since I separated the two "just in case".  I agree that you
don't translate keywords, but it is a matter of fact that some languages are
stricter than English in applying genders and articles.

> Hi Nathan, I'm not clear about why this conditional is done here.
> I think the diagnostic machinary is able to figure out how to print
> name with the appropriate class-key.  What else am I missing?

Here I'm missing something.  Does the C front-end use %D and %T like the C++
front-end does?

Also note that the C++ front-end does suffer from big internationalization
problems.  I tried to fix them with a patch in June 2002, but the patch was
said to be done wrong and dismissed.  It might be worthwhile to add this to
beginners.html.

Paolo