Hi list,

This post is my first post at Phobos ML :)

I have one question about hasLength.

Current implementation from std.range:
-----
template hasLength(R)
{
       enum bool hasLength = is(typeof(R.init.length) : ulong) &&
           !isNarrowString!R;
}
-----
This implementation has following issues.

http://d.puremagic.com/issues/show_bug.cgi?id=2873 http://d.puremagic.com/issues/show_bug.cgi?id=3508

But, hasLength returns true when length method is marked as @property.

We should fix the above issue or length method should be marked as
@property?
If latter, some ranges in phobos are incomplete.

I think the length should be a property.

Andrei

Masahiro Nakagawa wrote:
> Hi list,
> 
> This post is my first post at Phobos ML :)
> 
> I have one question about hasLength.
> 
> Current implementation from std.range:
> -----
> template hasLength(R)
> {
>       enum bool hasLength = is(typeof(R.init.length) : ulong) &&
>           !isNarrowString!R;
> }
> -----
> This implementation has following issues.
> 
> http://d.puremagic.com/issues/show_bug.cgi?id=2873 http://d.puremagic.com/issues/show_bug.cgi?id=3508
> 
> But, hasLength returns true when length method is marked as @property.
> 
> We should fix the above issue or length method should be marked as
> @property?
> If latter, some ranges in phobos are incomplete.
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Length should be a property.  When I filed my initial bug report, there was no such thing as @property.  If that fixes this bug, them IMHO it's fixed.

On Tue, May 4, 2010 at 3:28 PM, Andrei Alexandrescu <andrei at erdani.com>wrote:

> I think the length should be a property.
>
> Andrei
>
>
> Masahiro Nakagawa wrote:
>
>> Hi list,
>>
>> This post is my first post at Phobos ML :)
>>
>> I have one question about hasLength.
>>
>> Current implementation from std.range:
>> -----
>> template hasLength(R)
>> {
>>      enum bool hasLength = is(typeof(R.init.length) : ulong) &&
>>          !isNarrowString!R;
>> }
>> -----
>> This implementation has following issues.
>>
>> http://d.puremagic.com/issues/show_bug.cgi?id=2873 http://d.puremagic.com/issues/show_bug.cgi?id=3508
>>
>> But, hasLength returns true when length method is marked as @property.
>>
>> We should fix the above issue or length method should be marked as
>> @property?
>> If latter, some ranges in phobos are incomplete.
>> _______________________________________________
>> phobos mailing list
>> phobos at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/phobos
>>
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.puremagic.com/pipermail/phobos/attachments/20100504/96009068/attachment.html>

I agree. But, length of some range objects in Phobos isn't a property. Is this a bug? If so, I will commit.


On Wed, 05 May 2010 04:28:06 +0900, Andrei Alexandrescu <andrei at erdani.com> wrote:

> I think the length should be a property.
>
> Andrei
>
> Masahiro Nakagawa wrote:
>> Hi list,
>>  This post is my first post at Phobos ML :)
>>  I have one question about hasLength.
>>  Current implementation from std.range:
>> -----
>> template hasLength(R)
>> {
>>       enum bool hasLength = is(typeof(R.init.length) : ulong) &&
>>           !isNarrowString!R;
>> }
>> -----
>> This implementation has following issues.
>>  http://d.puremagic.com/issues/show_bug.cgi?id=2873
>> http://d.puremagic.com/issues/show_bug.cgi?id=3508
>>  But, hasLength returns true when length method is marked as @property.
>>  We should fix the above issue or length method should be marked as
>> @property?
>> If latter, some ranges in phobos are incomplete.

Oh, I forgot D's history...


On Wed, 05 May 2010 04:31:51 +0900, David Simcha <dsimcha at gmail.com> wrote:

> Length should be a property.  When I filed my initial bug report, there
> was
> no such thing as @property.  If that fixes this bug, them IMHO it's
> fixed.
>
> On Tue, May 4, 2010 at 3:28 PM, Andrei Alexandrescu <andrei at erdani.com>wrote:
>
>> I think the length should be a property.
>>
>> Andrei
>>
>>
>> Masahiro Nakagawa wrote:
>>
>>> Hi list,
>>>
>>> This post is my first post at Phobos ML :)
>>>
>>> I have one question about hasLength.
>>>
>>> Current implementation from std.range:
>>> -----
>>> template hasLength(R)
>>> {
>>>      enum bool hasLength = is(typeof(R.init.length) : ulong) &&
>>>          !isNarrowString!R;
>>> }
>>> -----
>>> This implementation has following issues.
>>>
>>> http://d.puremagic.com/issues/show_bug.cgi?id=2873 http://d.puremagic.com/issues/show_bug.cgi?id=3508
>>>
>>> But, hasLength returns true when length method is marked as @property.
>>>
>>> We should fix the above issue or length method should be marked as
>>> @property?
>>> If latter, some ranges in phobos are incomplete.
>>> _______________________________________________

It is a bug, if you could fix that would be great. There are many places in which we don't use @property but we must.

Andrei

Masahiro Nakagawa wrote:
> I agree. But, length of some range objects in Phobos isn't a property. Is this a bug? If so, I will commit.
> 
> 
> On Wed, 05 May 2010 04:28:06 +0900, Andrei Alexandrescu <andrei at erdani.com> wrote:
> 
>> I think the length should be a property.
>>
>> Andrei
>>
>> Masahiro Nakagawa wrote:
>>> Hi list,
>>>  This post is my first post at Phobos ML :)
>>>  I have one question about hasLength.
>>>  Current implementation from std.range:
>>> -----
>>> template hasLength(R)
>>> {
>>>       enum bool hasLength = is(typeof(R.init.length) : ulong) &&
>>>           !isNarrowString!R;
>>> }
>>> -----
>>> This implementation has following issues.
>>>  http://d.puremagic.com/issues/show_bug.cgi?id=2873
>>> http://d.puremagic.com/issues/show_bug.cgi?id=3508
>>>  But, hasLength returns true when length method is marked as @property.
>>>  We should fix the above issue or length method should be marked as
>>> @property?
>>> If latter, some ranges in phobos are incomplete.
> 
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

http://www.dsource.org/projects/phobos/changeset/1495 Committed.


On Wed, 05 May 2010 06:32:46 +0900, Andrei Alexandrescu <andrei at erdani.com> wrote:

> It is a bug, if you could fix that would be great. There are many places in which we don't use @property but we must.
>
> Andrei
>
> Masahiro Nakagawa wrote:
>> I agree. But, length of some range objects in Phobos isn't a property.
>> Is this a bug? If so, I will commit.
>>   On Wed, 05 May 2010 04:28:06 +0900, Andrei Alexandrescu
>> <andrei at erdani.com> wrote:
>>
>>> I think the length should be a property.
>>>
>>> Andrei
>>>
>>> Masahiro Nakagawa wrote:
>>>> Hi list,
>>>>  This post is my first post at Phobos ML :)
>>>>  I have one question about hasLength.
>>>>  Current implementation from std.range:
>>>> -----
>>>> template hasLength(R)
>>>> {
>>>>       enum bool hasLength = is(typeof(R.init.length) : ulong) &&
>>>>           !isNarrowString!R;
>>>> }
>>>> -----
>>>> This implementation has following issues.
>>>>  http://d.puremagic.com/issues/show_bug.cgi?id=2873
>>>> http://d.puremagic.com/issues/show_bug.cgi?id=3508
>>>>  But, hasLength returns true when length method is marked as
>>>> @property.
>>>>  We should fix the above issue or length method should be marked as
>>>> @property?
>>>> If latter, some ranges in phobos are incomplete.
>>  _______________________________________________
>> phobos mailing list
>> phobos at puremagic.com
>> http://lists.puremagic.com/mailman/listinfo/phobos
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Don't forget to update the changelog, noting the issue number if there is one.  They're located in the phobos project.

On Wed, 5 May 2010, Masahiro Nakagawa wrote:

> Date: Wed, 05 May 2010 07:06:56 +0900
> From: Masahiro Nakagawa <repeatedly at gmail.com>
> Reply-To: Discuss the phobos library for D <phobos at puremagic.com>
> To: Discuss the phobos library for D <phobos at puremagic.com>
> Subject: Re: [phobos] length implementation for Range
> 
> http://www.dsource.org/projects/phobos/changeset/1495 Committed.
> 
> 
> On Wed, 05 May 2010 06:32:46 +0900, Andrei Alexandrescu <andrei at erdani.com> wrote:
> 
> > It is a bug, if you could fix that would be great. There are many places in which we don't use @property but we must.
> > 
> > Andrei
> > 
> > Masahiro Nakagawa wrote:
> > > I agree. But, length of some range objects in Phobos isn't a property.
> > > Is this a bug? If so, I will commit.
> > >  On Wed, 05 May 2010 04:28:06 +0900, Andrei Alexandrescu
> > > <andrei at erdani.com> wrote:
> > > 
> > > > I think the length should be a property.
> > > > 
> > > > Andrei
> > > > 
> > > > Masahiro Nakagawa wrote:
> > > > > Hi list,
> > > > > This post is my first post at Phobos ML :)
> > > > > I have one question about hasLength.
> > > > > Current implementation from std.range:
> > > > > -----
> > > > > template hasLength(R)
> > > > > {
> > > > >      enum bool hasLength = is(typeof(R.init.length) : ulong) &&
> > > > >          !isNarrowString!R;
> > > > > }
> > > > > -----
> > > > > This implementation has following issues.
> > > > > http://d.puremagic.com/issues/show_bug.cgi?id=2873
> > > > > http://d.puremagic.com/issues/show_bug.cgi?id=3508
> > > > > But, hasLength returns true when length method is marked as @property.
> > > > > We should fix the above issue or length method should be marked as
> > > > > @property?
> > > > > If latter, some ranges in phobos are incomplete.
> > > _______________________________________________
> > > phobos mailing list
> > > phobos at puremagic.com
> > > http://lists.puremagic.com/mailman/listinfo/phobos
> > _______________________________________________
> > phobos mailing list
> > phobos at puremagic.com
> > http://lists.puremagic.com/mailman/listinfo/phobos
> 
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Thanks for your advice.

changeset 1495 isn't bug fixing, so I listed the summery at the WHATSNEW section.


On Wed, 05 May 2010 08:17:11 +0900, Brad Roberts <braddr at puremagic.com> wrote:

> Don't forget to update the changelog, noting the issue number if there is one.  They're located in the phobos project.
>
> On Wed, 5 May 2010, Masahiro Nakagawa wrote:
>
>> Date: Wed, 05 May 2010 07:06:56 +0900
>> From: Masahiro Nakagawa <repeatedly at gmail.com>
>> Reply-To: Discuss the phobos library for D <phobos at puremagic.com>
>> To: Discuss the phobos library for D <phobos at puremagic.com>
>> Subject: Re: [phobos] length implementation for Range
>>
>> http://www.dsource.org/projects/phobos/changeset/1495 Committed.
>>
>>
>> On Wed, 05 May 2010 06:32:46 +0900, Andrei Alexandrescu
>> <andrei at erdani.com>
>> wrote:
>>
>> > It is a bug, if you could fix that would be great. There are many
>> places in
>> > which we don't use @property but we must.
>> >
>> > Andrei
>> >
>> > Masahiro Nakagawa wrote:
>> > > I agree. But, length of some range objects in Phobos isn't a
>> property.
>> > > Is this a bug? If so, I will commit.
>> > >  On Wed, 05 May 2010 04:28:06 +0900, Andrei Alexandrescu
>> > > <andrei at erdani.com> wrote:
>> > >
>> > > > I think the length should be a property.
>> > > >
>> > > > Andrei
>> > > >
>> > > > Masahiro Nakagawa wrote:
>> > > > > Hi list,
>> > > > > This post is my first post at Phobos ML :)
>> > > > > I have one question about hasLength.
>> > > > > Current implementation from std.range:
>> > > > > -----
>> > > > > template hasLength(R)
>> > > > > {
>> > > > >      enum bool hasLength = is(typeof(R.init.length) : ulong) &&
>> > > > >          !isNarrowString!R;
>> > > > > }
>> > > > > -----
>> > > > > This implementation has following issues.
>> > > > > http://d.puremagic.com/issues/show_bug.cgi?id=2873
>> > > > > http://d.puremagic.com/issues/show_bug.cgi?id=3508
>> > > > > But, hasLength returns true when length method is marked as
>> @property.
>> > > > > We should fix the above issue or length method should be marked
>> as
>> > > > > @property?
>> > > > > If latter, some ranges in phobos are incomplete.