• Russel Winder via Digitalmars-d-learn (20/20) Jun 03 2017 A stripped down problem to avoid fluff. The C macro:
• ketmar (6/7) Jun 03 2017 ..use templates! ;-)
• Nicholas Wilson (3/14) Jun 03 2017 I think an alias template parameter will work here as aliases
• Jacob Carlborg (14/16) Jun 03 2017 No, it doesn't work for types:
• ketmar (2/4) Jun 03 2017 yeah. that's what Phobos does, for example.
• David Nadlinger (5/7) Jun 03 2017 We could also finally fix the frontend to get around this. At
• Jacob Carlborg (4/6) Jun 03 2017 That would be nice.
• Jonathan M Davis via Digitalmars-d-learn (9/15) Jun 04 2017 He did agree, but AFAIK, no one has ever actually done the work, and I
• David Nadlinger (9/10) Jun 03 2017 It's a limitation of DStep – for that use case, it would need to

Russel Winder via Digitalmars-d-learn <digitalmars-d-learn puremagic.com> writes:

A stripped down problem to avoid fluff. The C macro:

    #define FLOB(t) (sizeof(t))

Can be used in another macro:

    #define THINGY(a, b) (_THING(a, FLOB(b)))

We can use this as in:

    THINGY(10, __u32)

Now the D Way says use functions not macros. except that=20

    size_t FLOB(T)(auto ref T t) {=C2=A0return t.sizeof; }

which is the result of DStep doesn't allow passing a type name as t.

Is this a problem in D or a problem in DStep?

--=20
Russel.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder ekiga.n=
et
41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel winder.org.uk
London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder

ketmar <ketmar ketmar.no-ip.org> writes:

Russel Winder wrote:

 Now the D Way says...

..use templates! ;-)

it is not really possible to guess what macro author means (to do that, 
dstep should be able to actually *understand* the code), so it tries to do 
what is used more often.

that is, it's dstep failed guess.

Nicholas Wilson <iamthewilsonator hotmail.com> writes:

On Saturday, 3 June 2017 at 13:17:46 UTC, Russel Winder wrote:
 A stripped down problem to avoid fluff. The C macro:

     #define FLOB(t) (sizeof(t))

 Can be used in another macro:

     #define THINGY(a, b) (_THING(a, FLOB(b)))

 We can use this as in:

     THINGY(10, __u32)

 Now the D Way says use functions not macros. except that

     size_t FLOB(T)(auto ref T t) { return t.sizeof; }

 which is the result of DStep doesn't allow passing a type name 
 as t.

 Is this a problem in D or a problem in DStep?

I think an alias template parameter will work here as aliases 
take anything(types, literals, symbols).

Jacob Carlborg <doob me.com> writes:

On 2017-06-03 16:03, Nicholas Wilson wrote:

 I think an alias template parameter will work here as aliases take
 anything(types, literals, symbols).

No, it doesn't work for types:

void foo(alias a)() {}

void main()
{
     foo!(int)();
}

Results in:

Error: template instance foo!int does not match template declaration 
foo(alias a)()

Perhaps using the variadic template with a constraint on one element 
trick will work. Ugly, but I think that will work.

-- 
/Jacob Carlborg

ketmar <ketmar ketmar.no-ip.org> writes:

Jacob Carlborg wrote:

 Perhaps using the variadic template with a constraint on one element 
 trick will work. Ugly, but I think that will work.

yeah. that's what Phobos does, for example.

David Nadlinger <code klickverbot.at> writes:

On Saturday, 3 June 2017 at 14:19:00 UTC, Jacob Carlborg wrote:
 Perhaps using the variadic template with a constraint on one 
 element trick will work. Ugly, but I think that will work.

We could also finally fix the frontend to get around this. At 
DConf 2015, Walter officially agreed that this is a bug that 
needs fixing. ;)

  — David

Jacob Carlborg <doob me.com> writes:

On 2017-06-03 16:22, David Nadlinger wrote:

 We could also finally fix the frontend to get around this. At DConf
 2015, Walter officially agreed that this is a bug that needs fixing. ;)

That would be nice.

-- 
/Jacob Carlborg

Jonathan M Davis via Digitalmars-d-learn writes:

On Saturday, June 03, 2017 14:22:11 David Nadlinger via Digitalmars-d-learn 
wrote:
 On Saturday, 3 June 2017 at 14:19:00 UTC, Jacob Carlborg wrote:
 Perhaps using the variadic template with a constraint on one
 element trick will work. Ugly, but I think that will work.

 We could also finally fix the frontend to get around this. At
 DConf 2015, Walter officially agreed that this is a bug that
 needs fixing. ;)

He did agree, but AFAIK, no one has ever actually done the work, and I
suspect that unless Walter gets frustrated over the problem himself, he
won't get around to fixing it himself, but I don't know. I also have no idea
how easy or hard the implementation would be. It really should be fixed at
some point though, since it's clearly causing problems and consistently
forcing us to have workarounds in the standard library.

- Jonathan M Davis

David Nadlinger <code klickverbot.at> writes:

On Saturday, 3 June 2017 at 13:17:46 UTC, Russel Winder wrote:
 Is this a problem in D or a problem in DStep?

It's a limitation of DStep – for that use case, it would need to 
transform one of the macro arguments into a template argument 
rather than a runtime function parameter.

If you need to make the code work as-is, I suppose you could 
create some aliases like `enum u32 = __u32.init;` and pass these 
instead of the types – using runtime values just to convey their 
type.

  — David