I've been wanting to write a D-styled data format module but just got
extremely discouraged :)
Most of the functionality I need for this is already available in D2's
std.conv.
Is it possible to add this functionality into D1's std.conv?

On Wed, May 27, 2009 at 7:38 PM, Saaa <empty@needmail.com> wrote:
> I've been wanting to write a D-styled data format module but just got
> extremely discouraged :)
> Most of the functionality I need for this is already available in D2's
> std.conv.
> Is it possible to add this functionality into D1's std.conv?
>
>
>

Have a look at std2.  It's a backporting of some features of Phobos 2, including some of the new std.conv, to D1:

http://www.dsource.org/projects/std2

>
> Have a look at std2.  It's a backporting of some features of Phobos 2, including some of the new std.conv, to D1:
>
> http://www.dsource.org/projects/std2

Looks nice!
Why can't this be in official Phobos1?
It's a ploy from the D2 camp, right?

> Why can't this be in official Phobos1?

Especially because the documentation gets more and more outdated the further D2 progresses from 2.008 :(

On Wed, May 27, 2009 at 8:57 PM, Saaa <empty@needmail.com> wrote:
>>
>> Have a look at std2.  It's a backporting of some features of Phobos 2, including some of the new std.conv, to D1:
>>
>> http://www.dsource.org/projects/std2
>
> Looks nice!
> Why can't this be in official Phobos1?
> It's a ploy from the D2 camp, right?

If by "ploy" you mean "Walter's official stance not to change D1 or Phobos 1," then yes, it's a ploy.  But that's a bit harsh, no?

Also, if you need the Phobos docs from 2.008, just download the 2.008 DMD zip.  They're in there.

> Why can't this be in official Phobos1?
> It's a ploy from the D2 camp, right? :)

>If by "ploy" you mean "Walter's official stance not to change D1 or Phobos 1," then yes, it's a ploy.  But that's a bit harsh, no?
 I fixed it to how I meant it.

But I do find it a tad strange.
What disadvantage is there? Additional bugs maybe..
Sorry, this is probably a very old and closed discussion :)

>Also, if you need the Phobos docs from 2.008, just download the 2.008 DMD zip.  They're in there.

Maybe the docs could be on the project page, as now the links are to the latest phobos.

Saaa wrote:
>> Why can't this be in official Phobos1?
>> It's a ploy from the D2 camp, right? :)
> 
>> If by "ploy" you mean "Walter's official stance not to change D1 or
>> Phobos 1," then yes, it's a ploy.  But that's a bit harsh, no?
>  I fixed it to how I meant it.
> 
> But I do find it a tad strange.
> What disadvantage is there? Additional bugs maybe..
> Sorry, this is probably a very old and closed discussion :)

I'm not sure that it's closed. Things generally cannot be removed from D1 phobos. And as long as D2 Phobos is in flux, porting D2 to D1 would be a problem -- D1 has to be stable, it can't have something added to it which is an unstable work-in-progress. The question of whether a stable part of Phobos D2 could be backported to Phobos1 is I think still open.

> 
>> Also, if you need the Phobos docs from 2.008, just download the 2.008
>> DMD zip.  They're in there.
> 
> Maybe the docs could be on the project page, as now the links are to the latest phobos. 
> 
> 

Don Wrote:

> Saaa wrote:
> >> Why can't this be in official Phobos1?
> >> It's a ploy from the D2 camp, right? :)
> > 
> >> If by "ploy" you mean "Walter's official stance not to change D1 or Phobos 1," then yes, it's a ploy.  But that's a bit harsh, no?
> >  I fixed it to how I meant it.
> > 
> > But I do find it a tad strange.
> > What disadvantage is there? Additional bugs maybe..
> > Sorry, this is probably a very old and closed discussion :)
> 
> I'm not sure that it's closed. Things generally cannot be removed from D1 phobos. And as long as D2 Phobos is in flux, porting D2 to D1 would be a problem -- D1 has to be stable, it can't have something added to it which is an unstable work-in-progress. The question of whether a stable part of Phobos D2 could be backported to Phobos1 is I think still open.

I would be surprised to see a D2 Phobos officially ported to and distributed with D1. I think someone has done a partial port already.

Given the upcoming finalizetion of D2 it'd be great to see if there could be some reflection on lessons learned from D1. There's a fine line between stable and stale.

On Wed, May 27, 2009 at 4:38 PM, Saaa<empty@needmail.com> wrote:
> I've been wanting to write a D-styled data format module but just got
> extremely discouraged :)
> Most of the functionality I need for this is already available in D2's
> std.conv.
> Is it possible to add this functionality into D1's std.conv?

The dsource std2 project has D1 ports of some parts of D2's phobos.

I wrote on the main page that it has std.conv.   But it's been a while since I did any updates to it, so the version might be out of date with what's currently in D2 std.conv.

--bb

On Wed, Jun 10, 2009 at 11:08 AM, Bill Baxter<wbaxter@gmail.com> wrote:
> On Wed, May 27, 2009 at 4:38 PM, Saaa<empty@needmail.com> wrote:
>> I've been wanting to write a D-styled data format module but just got
>> extremely discouraged :)
>> Most of the functionality I need for this is already available in D2's
>> std.conv.
>> Is it possible to add this functionality into D1's std.conv?
>
> The dsource std2 project has D1 ports of some parts of D2's phobos.
>
> I wrote on the main page that it has std.conv.

Just to clarify, the std2 ports were done by me, but it was a while back.  The fact that I listed "std.conv" among the ported modules on the dsource page there must mean that I thought it was pretty much ported at the time made that page.

URL: http://www.dsource.org/projects/std2

--bb