Re: Loops (was Re: do { quit; } else { })

On 23/04/2025 19:43, Kaz Kylheku wrote:

On 2025-04-23, bart <bc@freeuk.com> wrote:

On 23/04/2025 16:31, David Brown wrote:

On 22/04/2025 22:03, bart wrote:

>

Too few levels of functions and/or macros (there is no semantic
difference between macros and functions in this matter)

>
There is a great deal of difference. Functions tend to be well-formed in
their inputs and outputs.
>
Macros take some abitrary blocks of syntax and return another arbitrary
block of syntax:
>
      #define INDEX(a, b, y) a y b
      INDEX(a, i, [) ];

 While that's terrible, I've never seen anything like it in the wild.
 Macros which open a brace, relying on the invocation of a paired
macro to close it, are reasonably common.

So you /have/ seen things like that. There are entire libraries built solely with the C processor, some even implement languages; they will use every trick they can.
My example was only for illustrating my claim that macros work with blobs of syntax, that may be malformed and unbalanced.
They can also work at the sub-token level:
     #define O(x) 0##x
     printf("%d\n", 100);             # 100
     printf("%d\n", O(100));          #  64

 The macro we had been discussing whereby the for (..) part of
a loop is also an example of something which generates a syntactic
fragment, which must be completed by something which follows
the macro, namely a statement.
 Speaking purely in terms of formal syntqx, this is exactly what is going
on in your example with the completing closing bracket. In that
situation, it gives us a useful technique whose use is clear.

Such libraries as I mentioned for embedded languages or syntax wrapping are clever, but usually impractical, unwieldy and inefficient.
If you want a new language, do it properly! Or use a more appropriate base language.