Using .require for struct types

Sep 10, 2022

Erdem Demir

Sep 10, 2022

Ali Çehreli

Sep 13, 2022

Sep 10, 2022

Sep 13, 2022

Sep 11, 2022

Sep 13, 2022

struct A { double val; bool isBig; } void main() { alias DListOfA = DList!A; DListOfA[string] temp; A a = {2.0, true}; DListOfA returnVal = temp.require("a", DListOfA());--> I wish I could use ref DListOfA here returnVal.insert(a); writeln(temp); }

On 9/10/22 09:33, Erdem Demir wrote: > DListOfA returnVal = temp.require("a", DListOfA());--> I wish I > could use ref DListOfA here But keeping a reference to a temporary would not work because the life of that temporary ends by the end of that expression (practically, at the semicolon). An option is to allocate the object dynamically with new (and store DListOfA* in the associative array). Then the GC would keep it alive as long its pointer was in the associative arrray. But a better option is to just forget about it because D already takes care of rvalues by blitting (bit-level copying) them by default. Everything just works... :) It is not expensive either. For example, your struct is very cheap to copy. But I am probably missing the reason why you want a ref there. Perhaps there are even better options. Ali

import object, std.container; struct A { double val; bool isBig; } void main() { alias DListOfA = DList!A; DListOfA returnVal; //DListOfA[string] temp;/* DListOfA[string] temp = [ "a": DListOfA( A(0) ) ];//*/ auto a = A(6, true); // replacement element temp.update("a", { return DListOfA( A(0) ); // not updated: unsucceeded but initialized }, (ref DListOfA v) { v = DListOfA( a ); // existing v has been replaced returnVal = v; assert(returnVal.front == a); }); assert(is(typeof(temp["a"]) == DList!A)); }

On Saturday, 10 September 2022 at 18:38:40 UTC, Ali Çehreli wrote: > On 9/10/22 09:33, Erdem Demir wrote: > > > DListOfA returnVal = temp.require("a", > DListOfA());--> I wish I > > could use ref DListOfA here > > But keeping a reference to a temporary would not work because the life of that temporary ends by the end of that expression (practically, at the semicolon). > > An option is to allocate the object dynamically with new (and store DListOfA* in the associative array). Then the GC would keep it alive as long its pointer was in the associative arrray. > > But a better option is to just forget about it because D already takes care of rvalues by blitting (bit-level copying) them by default. Everything just works... :) It is not expensive either. For example, your struct is very cheap to copy. > > But I am probably missing the reason why you want a ref there. Perhaps there are even better options. > > Ali In the code sample I posted I need returnVal.insert(a); have an effect on DListOfA[string] temp; but if returnVal is a rvalue copy I can't accomplish this right? I see your point about ref would have been assign to a temp val, thanks for pointing out.

September 13, 2022

Re: Using .require for struct types

Posted by Erdem
in reply to Salih Dincer

Permalink

Erdem

Posted in reply to Salih Dincer

Permalink

On Sunday, 11 September 2022 at 21:01:27 UTC, Salih Dincer wrote:

On Saturday, 10 September 2022 at 16:33:03 UTC, Erdem Demir wrote:

I wish I could use ref DListOfA returnVal = .... but we can't in D.

Can you please suggest alternatives?

I think you should try advanced update. Your flexibility and what you can do are limited by your dreams. A couple delicious code:

import object, std.container;

struct A
{
    double val;
    bool isBig;
}

void main()
{
    alias  DListOfA = DList!A;
    DListOfA returnVal;
    //DListOfA[string] temp;/*
    DListOfA[string] temp = [
        "a": DListOfA( A(0) )
    ];//*/
    
    auto a = A(6, true); // replacement element
    temp.update("a", {
        return DListOfA( A(0) ); // not updated: unsucceeded but initialized
    }, (ref DListOfA v) {
        v = DListOfA( a ); // existing v has been replaced
        returnVal = v;
        assert(returnVal.front == a);
    });
    assert(is(typeof(temp["a"]) == DList!A));
}

SDB@79

I really like the possibilities of update as well but the overhead is too much especially in this example. I am sorry to say I found old "in" usage less complicated than that even I think "in" is complicated. I liked the Steven's solution all this code can be done via 2 lines only.

Forums