www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - Really nooB question - property

reply "Eric" <eric makechip.com> writes:
There are a lot of discussions in the forums about how  property
should or could be implemented.  But I can't seem to find anything
that explains why or when I should use  property with the current
compiler.  Can anyone explain why and when I should use the 
 property tag?

Thx.
Eric
Jul 20 2014
next sibling parent "Gary Willoughby" <dev nomad.so> writes:
On Sunday, 20 July 2014 at 16:35:52 UTC, Eric wrote:

 There are a lot of discussions in the forums about how  property
 should or could be implemented.  But I can't seem to find 
 anything
 that explains why or when I should use  property with the 
 current
 compiler.  Can anyone explain why and when I should use the 
  property tag?

 Thx.
 Eric


I wondered the same:

http://forum.dlang.org/thread/uskutitmqgdfjeusrtfv forum.dlang.org
Jul 20 2014
prev sibling next sibling parent reply "John Colvin" <john.loughran.colvin gmail.com> writes:
On Sunday, 20 July 2014 at 16:35:52 UTC, Eric wrote:

 There are a lot of discussions in the forums about how  property
 should or could be implemented.  But I can't seem to find 
 anything
 that explains why or when I should use  property with the 
 current
 compiler.  Can anyone explain why and when I should use the 
  property tag?

 Thx.
 Eric


Use  property when you want a pseudo-variable or something that 
might be conceptually considered a "property" of the object, i.e. 
to do this:

auto blah = thing.someProperty;
thing.someProperty = blahblah;

or

struct myArray(T)
{
     T[] arr;
      property size_t memSize() { return arr.length * T.sizeof; }
}

Other than that, don't use it. Ordinary functions can be called 
without parenthesis anyway.


Using those ideas you shouldn't run in to any surprises.
Jul 20 2014
next sibling parent "John Colvin" <john.loughran.colvin gmail.com> writes:
On Sunday, 20 July 2014 at 17:59:07 UTC, John Colvin wrote:

 On Sunday, 20 July 2014 at 16:35:52 UTC, Eric wrote:

 There are a lot of discussions in the forums about how 
  property
 should or could be implemented.  But I can't seem to find 
 anything
 that explains why or when I should use  property with the 
 current
 compiler.  Can anyone explain why and when I should use the 
  property tag?

 Thx.
 Eric


 Use  property when you want a pseudo-variable or something that 
 might be conceptually considered a "property" of the object, 
 i.e. to do this:

 auto blah = thing.someProperty;
 thing.someProperty = blahblah;

 or

 struct myArray(T)
 {
     T[] arr;
      property size_t memSize() { return arr.length * T.sizeof; }
 }

 Other than that, don't use it. Ordinary functions can be called 
 without parenthesis anyway.


 Using those ideas you shouldn't run in to any surprises.


Oh, but don't expect the compiler to enforce this and please 
don't use the -property flag, it will only cause you pain.
Jul 20 2014
prev sibling parent reply "Eric" <eric makechip.com> writes:
 Use  property when you want a pseudo-variable or something that 
 might be conceptually considered a "property" of the object, 
 i.e. to do this:

 auto blah = thing.someProperty;
 thing.someProperty = blahblah;


This is basically what I suspected.  But why
write:

 property int getValue() { return(value); }

When you could just have a public field:

int value;

That lets you set and get the value without the parens anyways?

thanks,

Eric
Jul 20 2014
next sibling parent =?UTF-8?B?QWxpIMOHZWhyZWxp?= <acehreli yahoo.com> writes:
On 07/20/2014 11:14 AM, Eric wrote:

 Use  property when you want a pseudo-variable or something that might
 be conceptually considered a "property" of the object, i.e. to do this:

 auto blah = thing.someProperty;
 thing.someProperty = blahblah;


 This is basically what I suspected.  But why
 write:

  property int getValue() { return(value); }

 When you could just have a public field:

 int value;

 That lets you set and get the value without the parens anyways?


Freely setting a member makes sense only in limited cases where that 
member does not take any part in any invariant of the object. For 
example, if a Rectangle class has .length, .width, and .area, it would 
be an error to set either of them.

Additionally, properties enable one to make it look like a type has such 
a member:

struct Rectangle
{
     // ...

      property int area()
     {
         return length * width;
     }
}

Ali
Jul 20 2014
prev sibling next sibling parent Mike Parker <aldacron gmail.com> writes:
On 7/21/2014 3:14 AM, Eric wrote:

 Use  property when you want a pseudo-variable or something that might
 be conceptually considered a "property" of the object, i.e. to do this:

 auto blah = thing.someProperty;
 thing.someProperty = blahblah;


 This is basically what I suspected.  But why
 write:

  property int getValue() { return(value); }

 When you could just have a public field:

 int value;

 That lets you set and get the value without the parens anyways?

 thanks,

 Eric


In addition to what others have pointed out, it's useful for read-only 
values. You don't always want to allow clients to be able to set an 
objects members.

---
This email is free from viruses and malware because avast! Antivirus protection
is active.
http://www.avast.com
Jul 20 2014
prev sibling next sibling parent "John Colvin" <john.loughran.colvin gmail.com> writes:
On Sunday, 20 July 2014 at 18:14:29 UTC, Eric wrote:

 Use  property when you want a pseudo-variable or something 
 that might be conceptually considered a "property" of the 
 object, i.e. to do this:

 auto blah = thing.someProperty;
 thing.someProperty = blahblah;


 This is basically what I suspected.  But why
 write:

  property int getValue() { return(value); }

 When you could just have a public field:

 int value;

 That lets you set and get the value without the parens anyways?

 thanks,

 Eric


It allows you to add extra logic on read/write, without breaking 
any user code.
Jul 21 2014
prev sibling parent "Kagamin" <spam here.lot> writes:
On Sunday, 20 July 2014 at 18:14:29 UTC, Eric wrote:

 Use  property when you want a pseudo-variable or something 
 that might be conceptually considered a "property" of the 
 object, i.e. to do this:

 auto blah = thing.someProperty;
 thing.someProperty = blahblah;


 This is basically what I suspected.  But why
 write:

  property int getValue() { return(value); }


You should use nouns to name properties and verbs to name 
methods. D has optional parentheses, so you can write x=getValue;
So  property is generally not needed except for the case when the 
method returns a delegate, which in its turn can be implicitly 
called, so there's an ambiguity, which  property was meant to 
solve (but AFAIK in the end it didn't). It also serves as a 
documentation, that the method should be viewed as a property. 
Initially optional parentheses were meant as an easy 
implementation of properties, and  property wasn't meant to exist 
until the ambiguity with the delegate was understood.
Jul 21 2014
prev sibling parent "Danyal Zia" <catofdanyal yahoo.com> writes:
On Sunday, 20 July 2014 at 16:35:52 UTC, Eric wrote:

 There are a lot of discussions in the forums about how  property
 should or could be implemented.  But I can't seem to find 
 anything
 that explains why or when I should use  property with the 
 current
 compiler.  Can anyone explain why and when I should use the 
  property tag?


Consider the following struct:

struct Vector4 {
      property float x() { return vec[0]; }
     .....
     .....
      property void x(float _x) { vec[0] = _x }
     .....
     .....
private:
     float[4] vec;
}

Here I like to use static array for storing components of vector 
because I want to use special operations on arrays while taking 
advantage of auto-vectorization if possible. In this case, it is 
very convenient to use properties.
Jul 21 2014