object construction recognized as function declaration

[tthunder]

Hi @all,

there are always details, why it isn't possible to create a consistent
framework :(
However, here some code:

//--------------

class foo
{
public:
foo(){}
};

class fooConst
{
public:
explicit fooConst(const foo &p_Foo){}
void doSomething() {}
};

void test()
{
fooConst myObject(foo());
myObject.doSomething(); // <--- Error
}

//--------------

This doen't work on all compilers I have tested, because "fooConst
myObject(foo());" is assumed to be a funtion declaration!

I know that I could write:

foo aFoo;
fooConst myObject(aFoo);

But I don't want to!

Is there another keyword, possibility,........ you can think of ???

Greetings,
Kirsten

[Alf P. Steinbach]

* tt******@gmx.de:

Hi @all,

there are always details, why it isn't possible to create a consistent
framework :(
However, here some code:

//--------------

class foo
{
public:
foo(){}
};

class fooConst
{
public:
explicit fooConst(const foo &p_Foo){}
void doSomething() {}
};

void test()
{
fooConst myObject(foo());
myObject.doSomething(); // <--- Error
}

//--------------

This doen't work on all compilers I have tested, because "fooConst
myObject(foo());" is assumed to be a funtion declaration!

I know that I could write:

foo aFoo;
fooConst myObject(aFoo);

But I don't want to!

Is there another keyword, possibility,........ you can think of ???

You can

* Declare the argument object as a local variable, or

* use extra parentheses, or

* use the old "=" initialization syntax.

You should

* Try to include fewer misdirections in your examples.

Hth.

--
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?

[tthunder]

> You can

* Declare the argument object as a local variable, or

As I already mentioned, this would be a nasty break in constistence. I
don't really want to.
* use extra parentheses, or

Where exactly?
* use the old "=" initialization syntax.

Not possible, because explicit
THX,
Kirsten

[Alf P. Steinbach]

* tt******@gmx.de:

You can

* Declare the argument object as a local variable, or

As I already mentioned, this would be a nasty break in constistence. I
don't really want to.

It may be for the best, for it wouldn't suprise me if otherwise you'll
end up with a dangling reference or pointer.

* use extra parentheses, or

Where exactly?

You can always parenthesize an expression, such as an actual argument,
and if you look this up in the standard you'll find just that as an
example on how to do this.

* use the old "=" initialization syntax.

Not possible, because explicit

I'm sorry, but that's incorrect. In some news-groups a good way to get
a concrete example is to say "that's impossible". In this group, you're
more likely to elicit responses such as "what's your textbook?" (this
one is common because there really are a lot of bad C++ textbooks).

Anyway, I'll let you figure out why your statement is incorrect.

Just ask again if that's, uh, "impossible"... ;-)

--
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?

[Alan Johnson]

tt******@gmx.de wrote:

You can

* Declare the argument object as a local variable, or

As I already mentioned, this would be a nasty break in constistence. I
don't really want to.

* use extra parentheses, or

Where exactly?

fooConst myObject( ( foo() ) );
^ ^

* use the old "=" initialization syntax.

Not possible, because explicit

fooConst myObject = myObject(foo());
--
Alan Johnson

[Alan Johnson]

Alan Johnson wrote:

* use the old "=" initialization syntax.

Not possible, because explicit

fooConst myObject = myObject(foo());

Sorry,

fooConst myObject = fooConst(foo());

--
Alan Johnson

[tthunder]

> >

Where exactly?

fooConst myObject( ( foo() ) );
^ ^

Sorry, this does not work on my compiler Borland C++ Builder 6.0...
I've already tried that.

* use the old "=" initialization syntax.

Not possible, because explicit

fooConst myObject = myObject(foo());

After I have posted my last reply I thought about this, right... but
I'm sorry, because I have forgotten an important fact in my example:
the assignment operator and copy constructor are hidden!

[red floyd]

tt******@gmx.de wrote:

Hi @all,

there are always details, why it isn't possible to create a consistent
framework :(
However, here some code:

//--------------

class foo
{
public:
foo(){}
};

class fooConst
{
public:
explicit fooConst(const foo &p_Foo){}
void doSomething() {}
};

void test()
{
fooConst myObject(foo());
myObject.doSomething(); // <--- Error
}

//--------------

This doen't work on all compilers I have tested, because "fooConst
myObject(foo());" is assumed to be a funtion declaration!

I know that I could write:

foo aFoo;
fooConst myObject(aFoo);

google for "most vexing parse"

[tthunder]

red floyd schrieb:

google for "most vexing parse"

Thank you for this nice headword!