I have added this bit of code to my dmd, it gives warnings in cases of integral signed-unsigned comparisons: https://github.com/D-Programming-Language/dmd/pull/119

Then I have compiled Phobos (with unittests) with the -wi switch, this has generated some of those signed-unsigned warning messages, maybe this list is of interest for someone:
http://codepad.org/Ku42j1SR

Note: that "ignored variadic arguments to the constructor" that lists no line number is generated at line 1766 of template ConstructorGeneratingPolicy() inside typecons.d Phobos module.

Bye,
bearophile

bearophile wrote:
> I have added this bit of code to my dmd, it gives warnings in cases of integral signed-unsigned comparisons:
> https://github.com/D-Programming-Language/dmd/pull/119
> 
> Then I have compiled Phobos (with unittests) with the -wi switch, this has generated some of those signed-unsigned warning messages, maybe this list is of interest for someone:
> http://codepad.org/Ku42j1SR
> 
> Note: that "ignored variadic arguments to the constructor" that lists no line number is generated at line 1766 of template ConstructorGeneratingPolicy() inside typecons.d Phobos module.
> 
> Bye,
> bearophile

I've had a look at a dozen or so of these, and they were all real. I didn't see any which require a cast to "make the compiler shut up".
That's pretty impressive. In C++ I find that such messages are nearly always false positives.

The one case where it's a bit annoying is this:

int [] x = new int[6]; // or x = some array literal.
for (int i = 0; i < x.length; ++i) {...}

Here is a suggestion for how we could eliminate such false positives.
http://d.puremagic.com/issues/show_bug.cgi?id=6478

Don <nospam@nospam.com> wrote:

> I've had a look at a dozen or so of these, and they were all real. I didn't see any which require a cast to "make the compiler shut up". That's pretty impressive. In C++ I find that such messages are nearly always false positives.
> 
> The one case where it's a bit annoying is this:
> 
> int [] x = new int[6]; // or x = some array literal.
> for (int i = 0; i < x.length; ++i) {...}
> 
> Here is a suggestion for how we could eliminate such false positives. http://d.puremagic.com/issues/show_bug.cgi?id=6478

Doesn't this require flow analysis? And the type of index 'i' should be 'size_t' anyway.

kennytm wrote:
> Don <nospam@nospam.com> wrote:
> 
>> I've had a look at a dozen or so of these, and they were all real. I
>> didn't see any which require a cast to "make the compiler shut up".
>> That's pretty impressive. In C++ I find that such messages are nearly
>> always false positives.
>>
>> The one case where it's a bit annoying is this:
>>
>> int [] x = new int[6]; // or x = some array literal.
>> for (int i = 0; i < x.length; ++i) {...}
>>
>> Here is a suggestion for how we could eliminate such false positives.
>> http://d.puremagic.com/issues/show_bug.cgi?id=6478
> 
> Doesn't this require flow analysis? 
Yes. See the bug report.

> And the type of index 'i' should be 'size_t' anyway.

Why? It will only ever be in the range 0..6.

Am 12.08.2011, 12:22 Uhr, schrieb kennytm <kennytm@gmail.com>:

> Don <nospam@nospam.com> wrote:
>
>> I've had a look at a dozen or so of these, and they were all real. I
>> didn't see any which require a cast to "make the compiler shut up".
>> That's pretty impressive. In C++ I find that such messages are nearly
>> always false positives.
>>
>> The one case where it's a bit annoying is this:
>>
>> int [] x = new int[6]; // or x = some array literal.
>> for (int i = 0; i < x.length; ++i) {...}
>>
>> Here is a suggestion for how we could eliminate such false positives.
>> http://d.puremagic.com/issues/show_bug.cgi?id=6478
>
> Doesn't this require flow analysis? And the type of index 'i' should be
> 'size_t' anyway.

I think I once shot myself in the foot with this when I used 'auto' for 'i' and the code wouldn't compile on x86_64, because I assigned the variable to an int or uint later on in the loop. You just have to be aware that this is an unsigned integer of machine word length. So I agree with kennytm on this.

Just remember that reverse loops are written like this:

for (size_t i = x.length; i-- > 0; ) {...}

Marco Leise:

> Just remember that reverse loops are written like this:
> 
> for (size_t i = x.length; i-- > 0; ) {...}

Thankfully in D there is foreach_reverse :-)

import std.stdio;

void main() {
    auto array = [10, 20, 30];

    for (size_t i = array.length; i-- > 0; )
        writeln(i, " ", array[i]);

    foreach_reverse (i, item; array)
        writeln(i, " ", item);
}

Bye,
bearophile

On 08/12/2011 03:08 PM, bearophile wrote:
> Marco Leise:
>
>> Just remember that reverse loops are written like this:
>>
>> for (size_t i = x.length; i-->  0; ) {...}
>
> Thankfully in D there is foreach_reverse :-)
>
> import std.stdio;
>
> void main() {
>      auto array = [10, 20, 30];
>
>      for (size_t i = array.length; i-->  0; )
>          writeln(i, " ", array[i]);
>
>      foreach_reverse (i, item; array)
>          writeln(i, " ", item);
> }
>
> Bye,
> bearophile

But some people seem to strongly dislike it though. It is a bit unfortunate that you don't currently get the same functionality by using retro.

On 2011-08-12 14:55, Marco Leise wrote:
> Am 12.08.2011, 12:22 Uhr, schrieb kennytm <kennytm@gmail.com>:
>
>> Don <nospam@nospam.com> wrote:
>>
>>> I've had a look at a dozen or so of these, and they were all real. I
>>> didn't see any which require a cast to "make the compiler shut up".
>>> That's pretty impressive. In C++ I find that such messages are nearly
>>> always false positives.
>>>
>>> The one case where it's a bit annoying is this:
>>>
>>> int [] x = new int[6]; // or x = some array literal.
>>> for (int i = 0; i < x.length; ++i) {...}
>>>
>>> Here is a suggestion for how we could eliminate such false positives.
>>> http://d.puremagic.com/issues/show_bug.cgi?id=6478
>>
>> Doesn't this require flow analysis? And the type of index 'i' should be
>> 'size_t' anyway.
>
> I think I once shot myself in the foot with this when I used 'auto' for
> 'i' and the code wouldn't compile on x86_64, because I assigned the
> variable to an int or uint later on in the loop. You just have to be
> aware that this is an unsigned integer of machine word length. So I
> agree with kennytm on this.
>
> Just remember that reverse loops are written like this:
>
> for (size_t i = x.length; i-- > 0; ) {...}

I've done something similar myself.

-- 
/Jacob Carlborg

Am 12.08.2011, 15:08 Uhr, schrieb bearophile <bearophileHUGS@lycos.com>:

> Marco Leise:
>
>> Just remember that reverse loops are written like this:
>>
>> for (size_t i = x.length; i-- > 0; ) {...}
>
> Thankfully in D there is foreach_reverse :-)
>
> import std.stdio;
>
> void main() {
>     auto array = [10, 20, 30];
>
>     for (size_t i = array.length; i-- > 0; )
>         writeln(i, " ", array[i]);
>
>     foreach_reverse (i, item; array)
>         writeln(i, " ", item);
> }
>
> Bye,
> bearophile

Sure, but we were talking about for, not foreach. Maybe these cases in Phobos cannot use foreach, because they modify 'i' or they don't require 'item'.

On 08/12/2011 03:33 PM, Marco Leise wrote:
> Am 12.08.2011, 15:08 Uhr, schrieb bearophile <bearophileHUGS@lycos.com>:
>
>> Marco Leise:
>>
>>> Just remember that reverse loops are written like this:
>>>
>>> for (size_t i = x.length; i-- > 0; ) {...}
>>
>> Thankfully in D there is foreach_reverse :-)
>>
>> import std.stdio;
>>
>> void main() {
>> auto array = [10, 20, 30];
>>
>> for (size_t i = array.length; i-- > 0; )
>> writeln(i, " ", array[i]);
>>
>> foreach_reverse (i, item; array)
>> writeln(i, " ", item);
>> }
>>
>> Bye,
>> bearophile
>
> Sure, but we were talking about for, not foreach. Maybe these cases in
> Phobos cannot use foreach, because they modify 'i' or they don't require
> 'item'.

foreach_reverse(i;0..array.length) writeln(array[i]),i--;