Assuming that the idea behind the "-w" switch is to help coders locate the assumptions that the compiler is making on our behalf, I think that it would be useful to add an "optional error" when the compiler detects a potential loss of data.

Example 1:
import std.stdio;
void main()
{
    uint a;
    int b;

    b = -3;
    a = b; // A warning here would be appreciated.

    writefln("a = ", a, ", b = ", b);
    // result: a = 4294967293, b = -3
}

Example 2:
void Foo()
{
    int a;
    long b;

    b = SomeFunction();
    a = b;  // A warning here would be appreciated.

}

and there are many other examples too.
-- 
Derek
(skype: derek.j.parnell)
Melbourne, Australia
"Down with mediocracy!"
7/02/2006 11:16:27 AM

Derek Parnell wrote:
> Assuming that the idea behind the "-w" switch is to help coders locate the
> assumptions that the compiler is making on our behalf, I think that it
> would be useful to add an "optional error" when the compiler detects a
> potential loss of data.

Agreed.  I'd like the compiler to warn me about narrowing conversions that don't use an explicit cast.


Sean

"Sean Kelly" <sean@f4.ca> wrote in message news:ds8pmm$ups$1@digitaldaemon.com...
> Derek Parnell wrote:
> > Assuming that the idea behind the "-w" switch is to help coders locate
the
> > assumptions that the compiler is making on our behalf, I think that it would be useful to add an "optional error" when the compiler detects a potential loss of data.
>
> Agreed.  I'd like the compiler to warn me about narrowing conversions that don't use an explicit cast.

Wistfully ROFL!

Derek Parnell wrote:
> Assuming that the idea behind the "-w" switch is to help coders locate the
> assumptions that the compiler is making on our behalf, I think that it
> would be useful to add an "optional error" when the compiler detects a
> potential loss of data.

What is an "optional error"?  Either a piece of code is legal, or it isn't, surely?

<snip>
> Example 2:
> void Foo()
> {
>     int a;
>     long b;
>         b = SomeFunction();
>     a = b;  // A warning here would be appreciated.
>     }
<snip>

I'd like this to be an error.

Stewart.

-- 
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GCS/M d- s:- C++@ a->--- UB@ P+ L E@ W++@ N+++ o K-@ w++@ O? M V? PS- PE- Y? PGP- t- 5? X? R b DI? D G e++>++++ h-- r-- !y
------END GEEK CODE BLOCK------

My e-mail is valid but not my primary mailbox.  Please keep replies on the 'group where everyone may benefit.

In article <dsa2nv$1v8q$1@digitaldaemon.com>, Stewart Gordon says...
>
>What is an "optional error"?  Either a piece of code is legal, or it isn't, surely?
>
><snip>
>> Example 2:
>> void Foo()
>> {
>>     int a;
>>     long b;
>> 
>>     b = SomeFunction();
>>     a = b;  // A warning here would be appreciated.
>> 
>> }
><snip>
>
>I'd like this to be an error.
>
>Stewart.
>
Yep this is error - some rocket can fall down because of this!

--
UB

On Tue, 07 Feb 2006 23:10:39 +1100, Stewart Gordon <smjg_1998@yahoo.com> wrote:

> Derek Parnell wrote:
>> Assuming that the idea behind the "-w" switch is to help coders locate the
>> assumptions that the compiler is making on our behalf, I think that it
>> would be useful to add an "optional error" when the compiler detects a
>> potential loss of data.
>
> What is an "optional error"?  Either a piece of code is legal, or it isn't, surely?

It was from something that Walter said about "-w" warnings ...

'Think of them as "optional errors" instead of warnings <g>.' - Walter



-- 
Derek Parnell
Melbourne, Australia

Derek Parnell wrote:
> On Tue, 07 Feb 2006 23:10:39 +1100, Stewart Gordon <smjg_1998@yahoo.com>  wrote:
> 
>> Derek Parnell wrote:
>>
>>> Assuming that the idea behind the "-w" switch is to help coders locate  the
>>> assumptions that the compiler is making on our behalf, I think that it
>>> would be useful to add an "optional error" when the compiler detects a
>>> potential loss of data.
>>
>>
>> What is an "optional error"?  Either a piece of code is legal, or it  isn't, surely?
> 
> 
> It was from something that Walter said about "-w" warnings ...
> 
> 'Think of them as "optional errors" instead of warnings <g>.' - Walter
> 

"pedantic errors" might be a better term.

I think that what you're doing is telling the compiler something about your coding style (and perhaps, what kinds of bugs you create).