@david_chisnall you can use `typeof` for this too

#define Const(x) typeof(const typeof(x))
#define Volatile(x) typeof(volatile typeof(x))

Const(int *) pointer_that_is_const;
Const(int) * pointer_to_const;

@gloriouscow my money is on timing. can you #define every printf into a usleep() or such?

@SRAZKVT @ska

#define expand_ns(ns) namespace_ ## ns
#define concat2(lhs, rhs) lhs ## _ ## rhs
#define concat(lhs, rhs) concat2(lhs, rhs)
#define to(in, symbol) concat(expand_ns(in), symbol)

#define namespace_session ext_image_copy_capture_session_v1
#define create_frame to(session, create_frame)

#define namespace_frame ext_image_copy_capture_frame_v1
#define image_frame namespace_frame
#define frame_destroy to(frame, destroy)
#define frame_attach to(frame, attach_buffer)
#define frame_add_listener to(frame, add_listener)

struct image_frame *frame = create_frame(capture_session);
frame_add_listener(frame, &frame_listener, NULL);

i spent way too much time on this

also, cursed, i don’t want to see that again

@eniko You mean as a shorthand for #undef and #define in a row ? Or as a "I've undefed this, now restore the previous version"?

So now I have a bunch of this in my code

#undef EXCEPTION_HANDLER
#define EXCEPTION_HANDLER myhandler

Then a block of code followed by

#undef EXCEPTION_HANDLER
#define EXCEPTION_HANDLER default_exception_handler

Is it jank? Yes, extremely. Does it do what I need it to do? Also yes. Do I feel embarrassed about this in any way? Not at all

(adds a bunch of #define at the start of my C code to make it look weird) C IS NOT REAL

@whitequark thing is, i think you could actually make this work

```
enum class old_t { OLD };

#define old new (old_t::OLD)

class Oldable {
static void *operator new(size_t size, align_val_t align, old_t);
}

Oldable *foo() { return old Oldable; }
```

@TheConversationUS

#Define SMALLTOWN ANYTOWN

@horenmar #define __builtin_unreachable __builtin_trap

I'm going to start adding 8086/V30 support.

The easiest thing to do would be to have a #define for it, but that would require reuploading the sketch every time you wanted to switch CPUs. I don't like that, so I think I should auto-detect what CPU you have.

@aras @msinilo think of using as of a #define, maybe that'll help.

@ariadne the CFLAGS bit didn't seem to work. The output is nearly the same but with an extra warning:

#13 444.0 /usr/lib/perl5/core_perl/CORE/config.h:5262: warning: "USE_THREADS" redefined
#13 444.0 5262 | #define USE_THREADS /**/
#13 444.0 |
#13 444.0 <command-line>: note: this is the location of the previous definition

Not sure we can use our own modified version as you suggested. May try refactoring away Moops to ditch this module instead. Thanks for trying!

@thephd
This works to define rsize_t:

#define __STDC_WANT_LIB_EXT1__ 1
#include <string.h>

use “xcrun clang -dD -E file.c”, do the same thing with the broken configuration, see where things go wrong. If you do find there’s a problem with the macOS SDK, that sounds fixable