digitalmars.D - Copy constructor hell

9il (20/20) May 07 2019 Hi,

9il (2/3) May 07 2019 struct S // just a non-generic type
RazvanN (3/23) May 08 2019 What is the signature of the copy constructor that you have

9il (26/58) May 08 2019 The copy constructor is autogenerated for S.

9il (12/43) May 08 2019 It introduces regression for Phobos as well. For example,

RazvanN (5/49) May 08 2019 The problems seems to be that the default constructor takes the

Paul Backus (23/27) May 08 2019 Isn't the issue that the presence of a constructor disables the

9il (2/13) May 08 2019 Yes, exactly. --Ilya

RazvanN (2/17) May 14 2019 PR: https://github.com/dlang/dmd/pull/9790
RazvanN (9/24) May 14 2019 Thanks for the report. Please notify me of any other bugs

Atila Neves (3/16) May 14 2019 I just found this one:

Atila Neves (3/21) May 14 2019 ...And another one:

RazvanN (2/25) May 14 2019 This will be fixed by the patch I just mentioned.

RazvanN (2/20) May 14 2019 I posted a patch for it: https://github.com/dlang/dmd/pull/9792

9il <ilyayaroshenko gmail.com> writes:

Hi,

Just tried to upgrade mir-algorithm with the copy constructor.

Mir has only a few types that should define copy constructors 
explicitly.

However, Slice (main structure ) and more than 30 internal 
generic types like Iterators and Fields should be upgraded now as 
well. The compiler adds an automatically generated constructor. 
So, default structure construction that worked before, would not 
work anymore:

struct S(T)
{
     T t;
}

T oneT;

S(oneT); // would not work anymore if T has a copy constructor.

So, it is a big breaking change for user code as well.

Can we fix it at least for types that have compiler generated 
copy-constructors?

Best,
Ilya

May 07 2019

9il <ilyayaroshenko gmail.com> writes:

On Wednesday, 8 May 2019 at 03:18:14 UTC, 9il wrote:
 struct S(T)

struct S // just a non-generic type

May 07 2019

RazvanN <razvan.nitu1305 gmail.com> writes:

On Wednesday, 8 May 2019 at 03:18:14 UTC, 9il wrote:
 Hi,

 Just tried to upgrade mir-algorithm with the copy constructor.

 Mir has only a few types that should define copy constructors 
 explicitly.

 However, Slice (main structure ) and more than 30 internal 
 generic types like Iterators and Fields should be upgraded now 
 as well. The compiler adds an automatically generated 
 constructor. So, default structure construction that worked 
 before, would not work anymore:

 struct S(T)
 {
     T t;
 }

 T oneT;

 S(oneT); // would not work anymore if T has a copy constructor.

 So, it is a big breaking change for user code as well.

 Can we fix it at least for types that have compiler generated 
 copy-constructors?

 Best,
 Ilya

What is the signature of the copy constructor that you have 
defined? Have you tried the this(inout typeof(this)) inout one?

May 08 2019

9il <ilyayaroshenko gmail.com> writes:

On Wednesday, 8 May 2019 at 09:05:48 UTC, RazvanN wrote:
 On Wednesday, 8 May 2019 at 03:18:14 UTC, 9il wrote:
 Hi,

 Just tried to upgrade mir-algorithm with the copy constructor.

 Mir has only a few types that should define copy constructors 
 explicitly.

 However, Slice (main structure ) and more than 30 internal 
 generic types like Iterators and Fields should be upgraded now 
 as well. The compiler adds an automatically generated 
 constructor. So, default structure construction that worked 
 before, would not work anymore:

 struct S(T)
 {
     T t;
 }

 T oneT;

 S(oneT); // would not work anymore if T has a copy constructor.

 So, it is a big breaking change for user code as well.

 Can we fix it at least for types that have compiler generated 
 copy-constructors?

 Best,
 Ilya

 What is the signature of the copy constructor that you have 
 defined? Have you tried the this(inout typeof(this)) inout one?

The copy constructor is autogenerated for S.

Check: https://run.dlang.io/is/3sJTrf
-------
struct T
{
     int i;
     this(ref return scope inout typeof(this) src)
         inout  safe pure nothrow  nogc
     {
         i = src.i;
     }
}

struct S
{
     T t;
}

auto bar(T a)
{
	S(a);
}
-------

onlineapp.d(18): Error: copy constructor onlineapp.S.this(ref 
inout(S) p) inout is not callable using argument types (T)
onlineapp.d(18):        cannot pass argument a of type T to 
parameter ref inout(S) p

May 08 2019

9il <ilyayaroshenko gmail.com> writes:

On Wednesday, 8 May 2019 at 11:19:02 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 09:05:48 UTC, RazvanN wrote:
 On Wednesday, 8 May 2019 at 03:18:14 UTC, 9il wrote:
 [...]

 What is the signature of the copy constructor that you have 
 defined? Have you tried the this(inout typeof(this)) inout one?

 The copy constructor is autogenerated for S.

 Check: https://run.dlang.io/is/3sJTrf
 -------
 struct T
 {
     int i;
     this(ref return scope inout typeof(this) src)
         inout  safe pure nothrow  nogc
     {
         i = src.i;
     }
 }

 struct S
 {
     T t;
 }

 auto bar(T a)
 {
 	S(a);
 }
 -------

 onlineapp.d(18): Error: copy constructor onlineapp.S.this(ref 
 inout(S) p) inout is not callable using argument types (T)
 onlineapp.d(18):        cannot pass argument a of type T to 
 parameter ref inout(S) p

It introduces regression for Phobos as well. For example,

`DontCallDestructorT(arg)` is no longer possible for T arg.

---
struct Nullable(T)
{
     private union DontCallDestructorT
     {
         T payload;
     }
  ....
}

May 08 2019

RazvanN <razvan.nitu1305 gmail.com> writes:

On Wednesday, 8 May 2019 at 12:47:10 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 11:19:02 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 09:05:48 UTC, RazvanN wrote:
 [...]

 The copy constructor is autogenerated for S.

 Check: https://run.dlang.io/is/3sJTrf
 -------
 struct T
 {
     int i;
     this(ref return scope inout typeof(this) src)
         inout  safe pure nothrow  nogc
     {
         i = src.i;
     }
 }

 struct S
 {
     T t;
 }

 auto bar(T a)
 {
 	S(a);
 }
 -------

 onlineapp.d(18): Error: copy constructor onlineapp.S.this(ref 
 inout(S) p) inout is not callable using argument types (T)
 onlineapp.d(18):        cannot pass argument a of type T to 
 parameter ref inout(S) p

 It introduces regression for Phobos as well. For example,

 `DontCallDestructorT(arg)` is no longer possible for T arg.

 ---
 struct Nullable(T)
 {
     private union DontCallDestructorT
     {
         T payload;
     }
  ....
 }

The problems seems to be that the default constructor takes the 
arguments by value so a copy needs to be performed, hence the 
copy constructor use. This is pretty nasty. I will return with 
the details after I sort this out.

May 08 2019

Paul Backus <snarwin gmail.com> writes:

On Wednesday, 8 May 2019 at 13:19:24 UTC, RazvanN wrote:
 The problems seems to be that the default constructor takes the 
 arguments by value so a copy needs to be performed, hence the 
 copy constructor use. This is pretty nasty. I will return with 
 the details after I sort this out.

Isn't the issue that the presence of a constructor disables the 
use of "struct literal" syntax [1] for construction? The compiler 
is trying to find a constructor that matches the given argument, 
but since the only constructor is the auto-generated copy 
constructor, it can't.

If you add an explicit constructor to S, the problem goes away:

struct HasCopyCtor
{
     this(ref HasCopyCtor other) {}
}

struct S
{
     HasCopyCtor member;
     // Comment out the line below and compilation fails
     this(HasCopyCtor arg) { member = arg; }
}

void main()
{
     S s = S(HasCopyCtor());
}

Runnable: https://run.dlang.io/is/F4tHky

[1] https://dlang.org/spec/struct.html#struct-literal

May 08 2019

9il <ilyayaroshenko gmail.com> writes:

On Wednesday, 8 May 2019 at 18:01:34 UTC, Paul Backus wrote:
 On Wednesday, 8 May 2019 at 13:19:24 UTC, RazvanN wrote:
 The problems seems to be that the default constructor takes 
 the arguments by value so a copy needs to be performed, hence 
 the copy constructor use. This is pretty nasty. I will return 
 with the details after I sort this out.

 Isn't the issue that the presence of a constructor disables the 
 use of "struct literal" syntax [1] for construction? The 
 compiler is trying to find a constructor that matches the given 
 argument, but since the only constructor is the auto-generated 
 copy constructor, it can't.

Yes, exactly. --Ilya

May 08 2019

RazvanN <razvan.nitu1305 gmail.com> writes:

On Thursday, 9 May 2019 at 04:58:14 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 18:01:34 UTC, Paul Backus wrote:
 On Wednesday, 8 May 2019 at 13:19:24 UTC, RazvanN wrote:
 The problems seems to be that the default constructor takes 
 the arguments by value so a copy needs to be performed, hence 
 the copy constructor use. This is pretty nasty. I will return 
 with the details after I sort this out.

 Isn't the issue that the presence of a constructor disables 
 the use of "struct literal" syntax [1] for construction? The 
 compiler is trying to find a constructor that matches the 
 given argument, but since the only constructor is the 
 auto-generated copy constructor, it can't.

 Yes, exactly. --Ilya

PR: https://github.com/dlang/dmd/pull/9790

May 14 2019

RazvanN <razvan.nitu1305 gmail.com> writes:

On Thursday, 9 May 2019 at 04:58:14 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 18:01:34 UTC, Paul Backus wrote:
 On Wednesday, 8 May 2019 at 13:19:24 UTC, RazvanN wrote:
 The problems seems to be that the default constructor takes 
 the arguments by value so a copy needs to be performed, hence 
 the copy constructor use. This is pretty nasty. I will return 
 with the details after I sort this out.

 Isn't the issue that the presence of a constructor disables 
 the use of "struct literal" syntax [1] for construction? The 
 compiler is trying to find a constructor that matches the 
 given argument, but since the only constructor is the 
 auto-generated copy constructor, it can't.

 Yes, exactly. --Ilya

Thanks for the report. Please notify me of any other bugs 
regarding the copy constructor that you find. After you finish 
the transition of mir from postblit
to copy constructor maybe we can get Mike to add a blog post with 
your experience
so that others migrate to the copy constructor as well,

Thanks,
RazvanN

May 14 2019

Atila Neves <atila.neves gmail.com> writes:

On Tuesday, 14 May 2019 at 12:42:15 UTC, RazvanN wrote:
 On Thursday, 9 May 2019 at 04:58:14 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 18:01:34 UTC, Paul Backus wrote:
[...]

 Yes, exactly. --Ilya

 Thanks for the report. Please notify me of any other bugs 
 regarding the copy constructor that you find. After you finish 
 the transition of mir from postblit
 to copy constructor maybe we can get Mike to add a blog post 
 with your experience
 so that others migrate to the copy constructor as well,

 Thanks,
 RazvanN

I just found this one:

https://issues.dlang.org/show_bug.cgi?id=19871

May 14 2019

Atila Neves <atila.neves gmail.com> writes:

On Tuesday, 14 May 2019 at 14:28:35 UTC, Atila Neves wrote:
 On Tuesday, 14 May 2019 at 12:42:15 UTC, RazvanN wrote:
 On Thursday, 9 May 2019 at 04:58:14 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 18:01:34 UTC, Paul Backus wrote:
[...]

 Yes, exactly. --Ilya

 Thanks for the report. Please notify me of any other bugs 
 regarding the copy constructor that you find. After you finish 
 the transition of mir from postblit
 to copy constructor maybe we can get Mike to add a blog post 
 with your experience
 so that others migrate to the copy constructor as well,

 Thanks,
 RazvanN

 I just found this one:

 https://issues.dlang.org/show_bug.cgi?id=19871

...And another one:

https://issues.dlang.org/show_bug.cgi?id=19872

May 14 2019

RazvanN <razvan.nitu1305 gmail.com> writes:

On Tuesday, 14 May 2019 at 14:45:31 UTC, Atila Neves wrote:
 On Tuesday, 14 May 2019 at 14:28:35 UTC, Atila Neves wrote:
 On Tuesday, 14 May 2019 at 12:42:15 UTC, RazvanN wrote:
 On Thursday, 9 May 2019 at 04:58:14 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 18:01:34 UTC, Paul Backus wrote:
[...]

 Yes, exactly. --Ilya

 Thanks for the report. Please notify me of any other bugs 
 regarding the copy constructor that you find. After you 
 finish the transition of mir from postblit
 to copy constructor maybe we can get Mike to add a blog post 
 with your experience
 so that others migrate to the copy constructor as well,

 Thanks,
 RazvanN

 I just found this one:

 https://issues.dlang.org/show_bug.cgi?id=19871

 ...And another one:

 https://issues.dlang.org/show_bug.cgi?id=19872

This will be fixed by the patch I just mentioned.

May 14 2019

RazvanN <razvan.nitu1305 gmail.com> writes:

On Tuesday, 14 May 2019 at 14:28:35 UTC, Atila Neves wrote:
 On Tuesday, 14 May 2019 at 12:42:15 UTC, RazvanN wrote:
 On Thursday, 9 May 2019 at 04:58:14 UTC, 9il wrote:
 On Wednesday, 8 May 2019 at 18:01:34 UTC, Paul Backus wrote:
[...]

 Yes, exactly. --Ilya

 Thanks for the report. Please notify me of any other bugs 
 regarding the copy constructor that you find. After you finish 
 the transition of mir from postblit
 to copy constructor maybe we can get Mike to add a blog post 
 with your experience
 so that others migrate to the copy constructor as well,

 Thanks,
 RazvanN

 I just found this one:

 https://issues.dlang.org/show_bug.cgi?id=19871

I posted a patch for it: https://github.com/dlang/dmd/pull/9792

May 14 2019

D Programming

C/C++ Programming

Other

digitalmars.D - Copy constructor hell