Velocity Reviews > Is there any good way to do it

# Is there any good way to do it

qianz99@gmail.com
Guest
Posts: n/a

 10-12-2005
Hello I am writing a big program.

I have define a struction

typedef struct
{
int Name;
int Year;
}birth;

Now I'd like to define a sequence of A, such as

Birth A = {1,1990}
Birth B = {1,1991}
.....

I hope to use #define so that A,B can be used as constant.
Can I?
and is it a good way of programming?

Thanks

Walter Roberson
Guest
Posts: n/a

 10-12-2005
In article <(E-Mail Removed). com>,
<(E-Mail Removed)> wrote:
>I have define a struction

>typedef struct
>{
> int Name;
> int Year;
>}birth;

>Now I'd like to define a sequence of A, such as

>Birth A = {1,1990}
>Birth B = {1,1991}

birth A = {1,1990};
birth B = {1,1991};

:I hope to use #define so that A,B can be used as constant.
:Can I?

No.

What you can do is

#define B90 {1,1990}
#define B91 {1,1991}

birth A = B90;
birth B = B91;

and use B90 or B91 wherever else in the code that you happen to need
to initialize new variables to those values.

But A and B will be writable. You cannot create an unwritable structure
with any particular contents: the closest you can get is, as in the
above, to define textual substituations that happen to expand to the
values you need.

You could also get fancier with functions that return const pointers
to structures. You could probably even have something like

const birth B90(void) { birth B90_temp = {1,1990}; return B90_temp; }

but returning whole structures tends to make old-time programmers
queasy.

--
Watch for our new, improved .signatures -- Wittier! Profounder! and
with less than 2 grams of Trite!

Keith Thompson
Guest
Posts: n/a

 10-12-2005
http://www.velocityreviews.com/forums/(E-Mail Removed)-cnrc.gc.ca (Walter Roberson) writes:
> In article <(E-Mail Removed). com>,
> <(E-Mail Removed)> wrote:
>>I have define a struction

>
>>typedef struct
>>{
>> int Name;
>> int Year;
>>}birth;

>
>>Now I'd like to define a sequence of A, such as

>
>>Birth A = {1,1990}
>>Birth B = {1,1991}

>
> birth A = {1,1990};
> birth B = {1,1991};
>
> :I hope to use #define so that A,B can be used as constant.
> :Can I?
>
> No.
>
> What you can do is
>
> #define B90 {1,1990}
> #define B91 {1,1991}
>
> birth A = B90;
> birth B = B91;
>
> and use B90 or B91 wherever else in the code that you happen to need
> to initialize new variables to those values.

Right, the {1,1990} construct can be used in an initializer, but it
can't be used as an expression. C99 has compound literals, but
they're not universally supported yet.

> But A and B will be writable. You cannot create an unwritable structure
> with any particular contents: the closest you can get is, as in the
> above, to define textual substituations that happen to expand to the
> values you need.

Of course you can:

const birth A = { 1, 1990 };
const birth B = { 2, 1991 };

A and B aren't true "constants"; they're best thought of as read-only
variables. But they're probably suitable for the OP's purposes.

--
Keith Thompson (The_Other_Keith) http://www.velocityreviews.com/forums/(E-Mail Removed) <http://www.ghoti.net/~kst>
San Diego Supercomputer Center <*> <http://users.sdsc.edu/~kst>
We must do something. This is something. Therefore, we must do this.

tanmoy87544@gmail.com
Guest
Posts: n/a

 10-13-2005

Walter Roberson wrote:
> birth A = {1,1990};
> birth B = {1,1991};
>
> :I hope to use #define so that A,B can be used as constant.
> :Can I?
>
> No.

The answer is correct for C89

> birth A = B90;
> birth B = B91;
>
> and use B90 or B91 wherever else in the code that you happen to need
> to initialize new variables to those values.
>
> But A and B will be writable. You cannot create an unwritable structure
> with any particular contents: the closest you can get is, as in the

Actually, you can define const objects. Except for odd uses of strchr
etc. or by use of casts, attempts to write to them always involves a
constraint violation; in the exceptional cases, it invokes undefined
behaviour. So, they are kind of unwritable.

>
> You could also get fancier with functions that return const pointers
> to structures. You could probably even have something like
>
> const birth B90(void) { birth B90_temp = {1,1990}; return B90_temp; }

A definition of a function type which is incompatible with birth(void),
but which is functionally identical.

>
> but returning whole structures tends to make old-time programmers
> queasy.
>

But using qualified values in C rather than qualified objects just to
make someone queasy!