This TDPL book example still fails:

unittest {
int[] a = [0, 10, 20, 30, 40, 50, 60, 70];
auto b = a[4 .. $];
a = a[0 .. 4];
// At this point a and b are adjacent
a ~= [0, 0, 0, 0];
assert(b == [40, 50, 60, 70]); // passes; a got reallocated
}



because the agreed upon array append changes to druntime haven't been done yet. I can't recall who was doing that, or indeed if anyone had taken responsibility for getting it done.

Steve was working on this. Steve, did you commit your changes to phobos? (I think Steve is on the list, but I'm adding him just to be sure.)

Andrei

Walter Bright wrote:
> This TDPL book example still fails:
> 
> unittest {
> int[] a = [0, 10, 20, 30, 40, 50, 60, 70];
> auto b = a[4 .. $];
> a = a[0 .. 4];
> // At this point a and b are adjacent
> a ~= [0, 0, 0, 0];
> assert(b == [40, 50, 60, 70]); // passes; a got reallocated
> }
> 
> 
> 
> because the agreed upon array append changes to druntime haven't been
> done yet. I can't recall who was doing that, or indeed if anyone had
> taken responsibility for getting it done.
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

Yes, it's me.

Through discussion with Sean, the changes will be added (I will post it within the next couple days).  I need to recreate the patch with the newest druntime directory structure.  It's my fault it's not in yet, Sean told me he would add it late january, but I was on vacation  the first week of Feb, and the last week has been very busy for me.

There is still the issue of the special functions that append a dchar to a char or wchar array.  I don't have the typeinfo passed in so I cannot tell whether the array is shared or not.  However, I think it's better to add what I have now and worry about fixing shared stuff later (I'm pretty sure shared is not done exactly correct right now).  I haven't had time to read the latest installment of tdpl, but I'm confident that we can solve the problems of shared array appending later.  Fixing the stomping problem is important to get out there for people to play with.

-Steve



----- Original Message ----
> From: Andrei Alexandrescu <andrei at erdani.com>
> To: Discuss the phobos library for D <phobos at puremagic.com>; Steven Schveighoffer <schveiguy at yahoo.com>
> Sent: Wed, February 17, 2010 11:05:31 AM
> Subject: Re: [phobos] array appending changes to druntime
> 
> Steve was working on this. Steve, did you commit your changes to phobos? (I think Steve is on the list, but I'm adding him just to be sure.)
> 
> Andrei
> 
> Walter Bright wrote:
> > This TDPL book example still fails:
> > 
> > unittest {
> > int[] a = [0, 10, 20, 30, 40, 50, 60, 70];
> > auto b = a[4 .. $];
> > a = a[0 .. 4];
> > // At this point a and b are adjacent
> > a ~= [0, 0, 0, 0];
> > assert(b == [40, 50, 60, 70]); // passes; a got reallocated
> > }
> > 
> > 
> > 
> > because the agreed upon array append changes to druntime haven't been done
> yet. I can't recall who was doing that, or indeed if anyone had taken responsibility for getting it done.
> > _______________________________________________
> > phobos mailing list
> > phobos at puremagic.com
> > http://lists.puremagic.com/mailman/listinfo/phobos

Perfect. Thank you!

Andrei

Steve Schveighoffer wrote:
> Yes, it's me.
> 
> Through discussion with Sean, the changes will be added (I will post it within the next couple days).  I need to recreate the patch with the newest druntime directory structure.  It's my fault it's not in yet, Sean told me he would add it late january, but I was on vacation  the first week of Feb, and the last week has been very busy for me.
> 
> There is still the issue of the special functions that append a dchar to a char or wchar array.  I don't have the typeinfo passed in so I cannot tell whether the array is shared or not.  However, I think it's better to add what I have now and worry about fixing shared stuff later (I'm pretty sure shared is not done exactly correct right now).  I haven't had time to read the latest installment of tdpl, but I'm confident that we can solve the problems of shared array appending later.  Fixing the stomping problem is important to get out there for people to play with.
> 
> -Steve
> 
> 
> 
> ----- Original Message ----
>> From: Andrei Alexandrescu <andrei at erdani.com>
>> To: Discuss the phobos library for D <phobos at puremagic.com>; Steven Schveighoffer <schveiguy at yahoo.com>
>> Sent: Wed, February 17, 2010 11:05:31 AM
>> Subject: Re: [phobos] array appending changes to druntime
>>
>> Steve was working on this. Steve, did you commit your changes to phobos? (I think Steve is on the list, but I'm adding him just to be sure.)
>>
>> Andrei
>>
>> Walter Bright wrote:
>>> This TDPL book example still fails:
>>>
>>> unittest {
>>> int[] a = [0, 10, 20, 30, 40, 50, 60, 70];
>>> auto b = a[4 .. $];
>>> a = a[0 .. 4];
>>> // At this point a and b are adjacent
>>> a ~= [0, 0, 0, 0];
>>> assert(b == [40, 50, 60, 70]); // passes; a got reallocated
>>> }
>>>
>>>
>>>
>>> because the agreed upon array append changes to druntime haven't been done
>> yet. I can't recall who was doing that, or indeed if anyone had taken responsibility for getting it done.
>>> _______________________________________________
>>> 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

I agree, add what you have in now along with a comment about improvements that also need doing, so they won't get forgotten.

Andrei Alexandrescu wrote:
> Perfect. Thank you!
>
> Andrei
>
> Steve Schveighoffer wrote:
>> Yes, it's me.
>>
>> Through discussion with Sean, the changes will be added (I will post it within the next couple days).  I need to recreate the patch with the newest druntime directory structure.  It's my fault it's not in yet, Sean told me he would add it late january, but I was on vacation  the first week of Feb, and the last week has been very busy for me.
>>
>> There is still the issue of the special functions that append a dchar to a char or wchar array.  I don't have the typeinfo passed in so I cannot tell whether the array is shared or not.  However, I think it's better to add what I have now and worry about fixing shared stuff later (I'm pretty sure shared is not done exactly correct right now).  I haven't had time to read the latest installment of tdpl, but I'm confident that we can solve the problems of shared array appending later.  Fixing the stomping problem is important to get out there for people to play with.
>>
>> -Steve
>>
>>
>>
>> ----- Original Message ----
>>> From: Andrei Alexandrescu <andrei at erdani.com>
>>> To: Discuss the phobos library for D <phobos at puremagic.com>; Steven
>>> Schveighoffer <schveiguy at yahoo.com>
>>> Sent: Wed, February 17, 2010 11:05:31 AM
>>> Subject: Re: [phobos] array appending changes to druntime
>>>
>>> Steve was working on this. Steve, did you commit your changes to phobos? (I think Steve is on the list, but I'm adding him just to be sure.)
>>>
>>> Andrei
>>>
>>> Walter Bright wrote:
>>>> This TDPL book example still fails:
>>>>
>>>> unittest {
>>>> int[] a = [0, 10, 20, 30, 40, 50, 60, 70];
>>>> auto b = a[4 .. $];
>>>> a = a[0 .. 4];
>>>> // At this point a and b are adjacent
>>>> a ~= [0, 0, 0, 0];
>>>> assert(b == [40, 50, 60, 70]); // passes; a got reallocated
>>>> }
>>>>
>>>>
>>>>
>>>> because the agreed upon array append changes to druntime haven't been done
>>> yet. I can't recall who was doing that, or indeed if anyone had taken responsibility for getting it done.
>>>> _______________________________________________
>>>> 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
>
>

On Wed, 17 Feb 2010, Walter Bright wrote:

> I agree, add what you have in now along with a comment about improvements that also need doing, so they won't get forgotten.

I'd suggest a bugzilla ticket rather than a comment.

Agreed, I'll do both :)

-Steve



----- Original Message ----
> From: Brad Roberts <braddr at puremagic.com>
> To: Discuss the phobos library for D <phobos at puremagic.com>
> Sent: Wed, February 17, 2010 4:45:01 PM
> Subject: Re: [phobos] array appending changes to druntime
> 
> On Wed, 17 Feb 2010, Walter Bright wrote:
> 
> > I agree, add what you have in now along with a comment about improvements that also need doing, so they won't get forgotten.
> 
> I'd suggest a bugzilla ticket rather than a comment.
> 
> 
> _______________________________________________
> phobos mailing list
> phobos at puremagic.com
> http://lists.puremagic.com/mailman/listinfo/phobos

All set:

http://d.puremagic.com/issues/show_bug.cgi?id=3637 http://d.puremagic.com/issues/show_bug.cgi?id=3811

We need to have a further discussion on replacing the "pre-allocate" feature.  I'll start a separate thread for that.

-Steve



----- Original Message ----
> From: Steve Schveighoffer <schveiguy at yahoo.com>
> To: Discuss the phobos library for D <phobos at puremagic.com>
> Sent: Wed, February 17, 2010 6:36:15 PM
> Subject: Re: [phobos] array appending changes to druntime
> 
> Agreed, I'll do both :)
> 
> -Steve
> 
> 
> 
> ----- Original Message ----
> > From: Brad Roberts
> > To: Discuss the phobos library for D
> > Sent: Wed, February 17, 2010 4:45:01 PM
> > Subject: Re: [phobos] array appending changes to druntime
> > 
> > On Wed, 17 Feb 2010, Walter Bright wrote:
> > 
> > > I agree, add what you have in now along with a comment about improvements
> that
> > > also need doing, so they won't get forgotten.
> > 
> > I'd suggest a bugzilla ticket rather than a comment.
> > 
> > 
> > _______________________________________________
> > 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

Perfect. Thanks, Steve!

Andrei

Steve Schveighoffer wrote:
> All set:
> 
> http://d.puremagic.com/issues/show_bug.cgi?id=3637 http://d.puremagic.com/issues/show_bug.cgi?id=3811
> 
> We need to have a further discussion on replacing the "pre-allocate" feature.  I'll start a separate thread for that.
> 
> -Steve
> 
> 
> 
> ----- Original Message ----
>> From: Steve Schveighoffer <schveiguy at yahoo.com>
>> To: Discuss the phobos library for D <phobos at puremagic.com>
>> Sent: Wed, February 17, 2010 6:36:15 PM
>> Subject: Re: [phobos] array appending changes to druntime
>>
>> Agreed, I'll do both :)
>>
>> -Steve
>>
>>
>>
>> ----- Original Message ----
>>> From: Brad Roberts
>>> To: Discuss the phobos library for D
>>> Sent: Wed, February 17, 2010 4:45:01 PM
>>> Subject: Re: [phobos] array appending changes to druntime
>>>
>>> On Wed, 17 Feb 2010, Walter Bright wrote:
>>>
>>>> I agree, add what you have in now along with a comment about improvements
>> that
>>>> also need doing, so they won't get forgotten.
>>> I'd suggest a bugzilla ticket rather than a comment.
>>>
>>>
>>> _______________________________________________
>>> 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

Steve Schveighoffer wrote:
> All set:
>
> 

The changes don't seem to be checked in, this still fails:

unittest {
#line 950 "0350-arrays.tex"
int[] a = [0, 10, 20, 30, 40, 50, 60, 70];
auto b = a[4 .. $];
a = a[0 .. 4];
// At this point a and b are adjacent
a ~= [0, 0, 0, 0];
assert(b == [40, 50, 60, 70]); // passes; a got reallocated
}

void main(){}