[lnkForumImage]
TotalShareware - Download Free Software
Confronta i prezzi di migliaia di prodotti.
Asp Forum
 Home | Login | Register | Search 


 

Forums >

comp.lang.c++

Can constructor call another method?

Christopher

9/11/2008 1:02:00 AM

Its been awhile and I am rusty.

Can the constructor of my class call another method in the same class
if that other method does not change any member data?

I want to simply have a seperate method that returns a huge string,
that string containing code in another language, which is to be
compiled by a third party API when my object is being contructed.

class MyClass()
{
public:

MyClass()
{
m_effect = ThirdPartyAPIFunction( GetHLSL() );
}

private:

const std::string GetHLSL()
{
std::string hlsl = "";
/*1*/ hlsl += " // This is HLSL code";
/*2*/ hlsl += "float4 IntensityAmbient;";

return hlsl;
}

ThirdPartyAPIVaribale * m_effect;
};
8 Answers

Pete Becker

9/11/2008 2:07:00 AM

0

On 2008-09-10 21:02:24 -0400, Christopher <cpisz@austin.rr.com> said:

> Can the constructor of my class call another method in the same class
> if that other method does not change any member data?

Yes. And it can call another member function even if that member
function changes the values of data members. In fact, it's pretty
common to have an init() member function that does some of the hard
work, and gets called from various constructors, assignment operators,
etc.

--
Pete
Roundhouse Consulting, Ltd. (www.versatilecoding.com) Author of "The
Standard C++ Library Extensions: a Tutorial and Reference
(www.petebecker.com/tr1book)

asm23

9/11/2008 8:21:00 AM

0

Christopher wrote:
> Its been awhile and I am rusty.
>
> Can the constructor of my class call another method in the same class
> if that other method does not change any member data?

Yes, it can call another method in the same class. And there is no limit
to change member data or not.

Eric Johnson

9/11/2008 4:38:00 PM

0

On Sep 10, 6:02 pm, Christopher <cp...@austin.rr.com> wrote:
> Its been awhile and I am rusty.
>
> Can the constructor of my class call another method in the same class
> if that other method does not change any member data?
>

One other thing to watch out for -- you want to stay away from calling
virtual functions from within a constructor. Search in this list for
"Call virtual function in constructor" for more info on that topic.

Happy coding!
-Eric

Alf P. Steinbach

9/11/2008 5:16:00 PM

0

* Eric Johnson:
> On Sep 10, 6:02 pm, Christopher <cp...@austin.rr.com> wrote:
>> Its been awhile and I am rusty.
>>
>> Can the constructor of my class call another method in the same class
>> if that other method does not change any member data?
>>
>
> One other thing to watch out for -- you want to stay away from calling
> virtual functions from within a constructor.

There's no reason to.

In C++ it's safe to call virtual functions from a constructor.

Other languages (Java, C#) have problems with that, not C++.


Cheers & hth.,

- Alf

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

Pete Becker

9/11/2008 5:16:00 PM

0

On 2008-09-11 12:37:40 -0400, Eric Johnson <eric.eerriicc@gmail.com> said:

> On Sep 10, 6:02 pm, Christopher <cp...@austin.rr.com> wrote:
>> Its been awhile and I am rusty.
>>
>> Can the constructor of my class call another method in the same class
>> if that other method does not change any member data?
>>
>
> One other thing to watch out for -- you want to stay away from calling
> virtual functions from within a constructor. Search in this list for
> "Call virtual function in constructor" for more info on that topic.
>

Well, yes, if you don't understand the implications of doing that. But
there's nothing inherently dangerous in it.

--
Pete
Roundhouse Consulting, Ltd. (www.versatilecoding.com) Author of "The
Standard C++ Library Extensions: a Tutorial and Reference
(www.petebecker.com/tr1book)

James Kanze

9/11/2008 9:32:00 PM

0

On Sep 11, 7:15 pm, "Alf P. Steinbach" <al...@start.no> wrote:
> * Eric Johnson:

> > On Sep 10, 6:02 pm, Christopher <cp...@austin.rr.com> wrote:
> >> Its been awhile and I am rusty.

> >> Can the constructor of my class call another method in the
> >> same class if that other method does not change any member
> >> data?

> > One other thing to watch out for -- you want to stay away
> > from calling virtual functions from within a constructor.

> There's no reason to.

> In C++ it's safe to call virtual functions from a constructor.

> Other languages (Java, C#) have problems with that, not C++.

I don't know about C#, but it's "safe" in Java as well. Just
safe in a different (IMHO less useful) way. In both cases, you
have to know what it means. And for obvious reasons, in neither
case can it behave "intuitively".

--
James Kanze (GABI Software) email:james.kanze@gmail.com
Conseils en informatique orientée objet/
Beratung in objektorientierter Datenverarbeitung
9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34

Alf P. Steinbach

9/11/2008 10:24:00 PM

0

* James Kanze:
> On Sep 11, 7:15 pm, "Alf P. Steinbach" <al...@start.no> wrote:
>> * Eric Johnson:
>
>>> On Sep 10, 6:02 pm, Christopher <cp...@austin.rr.com> wrote:
>>>> Its been awhile and I am rusty.
>
>>>> Can the constructor of my class call another method in the
>>>> same class if that other method does not change any member
>>>> data?
>
>>> One other thing to watch out for -- you want to stay away
>>> from calling virtual functions from within a constructor.
>
>> There's no reason to.
>
>> In C++ it's safe to call virtual functions from a constructor.
>
>> Other languages (Java, C#) have problems with that, not C++.
>
> I don't know about C#, but it's "safe" in Java as well. Just
> safe in a different (IMHO less useful) way.

I don't know what you mean by that. In fact, unless my memory is playing tricks
on me, you have earlier remarked on the unsafety of Java in that respect,
maintaining that it's a major cause of errors in Java programs. Anyway, virtual
calls from constructors are not type safe in Java.


> In both cases, you
> have to know what it means.

I think that goes for any language feature. :-)


> And for obvious reasons, in neither
> case can it behave "intuitively".

Hm, the C++ rules are very intuitive -- unless one's background is from Java.


Cheers,

- Alf

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

James Kanze

9/12/2008 7:56:00 AM

0

On Sep 12, 12:23 am, "Alf P. Steinbach" <al...@start.no> wrote:
> * James Kanze:

[...]
> >> In C++ it's safe to call virtual functions from a constructor.

> >> Other languages (Java, C#) have problems with that, not C++.

> > I don't know about C#, but it's "safe" in Java as well. Just
> > safe in a different (IMHO less useful) way.

> I don't know what you mean by that. In fact, unless my memory
> is playing tricks on me, you have earlier remarked on the
> unsafety of Java in that respect, maintaining that it's a
> major cause of errors in Java programs. Anyway, virtual calls
> from constructors are not type safe in Java.

It's safe in the sense that the behavior is well defined, and
you can take precautions and handle it. (All of the member
variables of the derived class are "zero initialized" before the
base class constructor runs.) That doesn't mean that it isn't a
major cause of errors; just because you can handle it correctly
doesn't mean that programmers think to do so. It's definitely a
case of poor language design.

(You'll note that I put safe in quotes. To indicate that I was
using the word in a very special, and in this case, restricted
sense.)

> > In both cases, you have to know what it means.

> I think that goes for any language feature. :-)

Yes. But some are more intuitive than others: I've yet to
encounter a language where + meant anything but addition. Where
as I can't really think of an intuitive meaning in this case:
there's nothing intuitive about what C++ does, and what Java
does is even worse. But a language has to define something.

> > And for obvious reasons, in neither case can it behave
> > "intuitively".

> Hm, the C++ rules are very intuitive -- unless one's
> background is from Java.

I disagree. I don't think that there is any real "intuitive"
behavior here. Neither in C++ nor in Java. We're beyond the
realm of intuition.

--
James Kanze (GABI Software) email:james.kanze@gmail.com
Conseils en informatique orientée objet/
Beratung in objektorientierter Datenverarbeitung
9 place Sémard, 78210 St.-Cyr-l'École, France, +33 (0)1 30 23 00 34