config.d: use enums instead of structs for __c_long, etc.

[WalterBright]

needs dlang/dmd#8152

[dlang-bot]

Thanks for your pull request, @WalterBright!

Bugzilla references

Your PR doesn't reference any Bugzilla issue.

If your PR contains non-trivial changes, please reference a Bugzilla issue or create a manual changelog.

Testing this PR locally

If you don't have a local development environment setup, you can use Digger to test this PR:

dub fetch digger
dub run digger -- build "master + druntime#2160"

[WalterBright]

Fix https://issues.dlang.org/show_bug.cgi?id=17890
Fix https://issues.dlang.org/show_bug.cgi?id=16289

[jmdavis]

I had no clue that it was even legal to declare an enum with no members. That seems very weird. Is the point of doing this instead of using an alias so that it's a distinct type from int? I guess that that would mean that this technique is a typedef of sorts.

[jacob-carlborg]

Is the point of doing this instead of using an alias so that it's a distinct type from int?

Yes, so that the compiler can do special mangling for that type. Although I don't see how it's any better than a struct. I proposed to add support for adding pragma(mangle) to alias declarations but Walter doesn't seem to like that.

[WalterBright]

If you go down that path, you'll discover you've reinvented enums.

[jacob-carlborg]

@jmdavis BTW, one use case of declaring an enum without any members is to use it as a UDA that doesn't contain any data.

[WalterBright]

This is a Vision issue, as it resolves a long-running sore in compatibility with C++.

[jacob-carlborg]

What happens if we do the other way around, i.e. alias size_t = cpp_ulong in object.d? This would be to avoid having to use cpp_size_t instead of size_t.

[WalterBright]

I plan to do that separately.

[jacob-carlborg]

I think there should be a test somewhere (in druntime or in DMD) that verifies that the correct mangling is used.

[WalterBright]

I think there should be a test somewhere (in druntime or in DMD) that verifies that the correct mangling is used.

dlang/dmd#8152

[WalterBright]

@CyberShadow I don't understand this error message:

Test the D Language specification
../dmd/generated/linux/release/64/dmd -run tools/dspec_tester.d --compiler=../dmd/generated/linux/release/64/dmd
---
void main() {
import std.stdio;
import std.algorithm;
int[] s = [1, 2, 3, 4];

copy(s[1..3], s[0..2]);
assert(s == [2, 3, 3, 4]);

}
---
gcc: error: __stdin.o: No such file or directory
Error: linker exited with status 1
struct.dd: 35 examples found
traits.dd: 1 examples found
template.dd: 1 examples found
type.dd: 4 examples found
arrays.dd: 31 examples found
function.dd: 36 examples found
posix.mak:811: recipe for target 'test_dspec' failed

[WalterBright]

DAutoTest appears to fail the same way with other PRs, so is not caused by this PR.

[CyberShadow]

The problem is in DMD's handling of programs from stdin:
https://issues.dlang.org/show_bug.cgi?id=18762#c1

[WalterBright]

@CyberShadow thanks for the quick fix on that!

[ibuclaw]

Let's move on with this then.