May 12, 2009
Jason House wrote:
> There really is only three reasons I'm aware of that stop D2 Tango
> from existing today:
> [...]
> 2.  There are bugs that limit how easily Tango can be ported to D2


I've heard this now and then, along with a bugzilla number or two. I've fixed every one those problems brought to my attention weeks (months?) ago.

If there are other bugzilla issues that prevent Tango from working with D2, please let me know what they are.

May 12, 2009
"Walter Bright" <newshound1@digitalmars.com> wrote in message news:guafcv$1tmg$3@digitalmars.com...
> Ameer Armaly wrote:
>>> It has been bashed out for D2. Doing such changes to D1 will break everyone's D1 code, which destroys its mission of being stable.
>> Really? I was under the impression that Tango will be ported to D2 to continue the battle of standard libs but if I'm mistaken, then it's a lot better than I originally figured.
>
> D2 has the "druntime" core which is designed to be the common root of Phobos and Tango. Switching to it involved a lot of changes to Phobos that broke existing user code.
I see. What about the actual "standard library bits" for common tasks like file I/O, strings, date and time, filesystem manipulation, etc. My main gripe is having two regularly accepted libraries that do the same extremely common functions; it makes us look unpolished, especially when there is the "community" versus "official" difference I mentioned earlier. Why not just merge the two libraries into the one D standard library and be done with it? I have yet to see a language with two standard libraries like we have now. [1] Thanks though for what's already been done.

Ameer

[1] I haven't really seen that many though- probably about a dozen.


May 12, 2009
Walter Bright Wrote:

> Jason House wrote:
> > There really is only three reasons I'm aware of that stop D2 Tango from existing today:
>  > [...]
> > 2.  There are bugs that limit how easily Tango can be ported to D2
> 
> 
> I've heard this now and then, along with a bugzilla number or two. I've fixed every one those problems brought to my attention weeks (months?) ago.
> 
> If there are other bugzilla issues that prevent Tango from working with D2, please let me know what they are.
> 

Obviously, Tango devs should answer, but...
IIRC, the whole "return const" thing is still an issue


May 12, 2009
== Quote from Jason House (jason.james.house@gmail.com)'s article
> Walter Bright Wrote:
> > Jason House wrote:
> > > There really is only three reasons I'm aware of that stop D2 Tango from existing today:
> >  > [...]
> > > 2.  There are bugs that limit how easily Tango can be ported to D2
> >
> >
> > I've heard this now and then, along with a bugzilla number or two. I've fixed every one those problems brought to my attention weeks (months?) ago.
> >
> > If there are other bugzilla issues that prevent Tango from working with D2, please let me know what they are.
> >
> Obviously, Tango devs should answer, but...
> IIRC, the whole "return const" thing is still an issue

Yeah, I can see why noone wanted to design a half-baked solution to this before shared was implemented, since that's going to need a similar fix.  Shared is coming next release.  Therefore, I think it will soon be high time to address this issue.
May 12, 2009
Ameer Armaly wrote:
> I see. What about the actual "standard library bits" for common tasks like file I/O, strings, date and time, filesystem manipulation, etc. My main gripe is having two regularly accepted libraries that do the same extremely common functions; it makes us look unpolished, especially when there is the "community" versus "official" difference I mentioned earlier. Why not just merge the two libraries into the one D standard library and be done with it?

That's being done for D2. It is not for D1, as that would break everyone's code.
May 12, 2009
"Walter Bright" <newshound1@digitalmars.com> wrote in message news:guakkp$285n$1@digitalmars.com...
> Ameer Armaly wrote:
>> I see. What about the actual "standard library bits" for common tasks like file I/O, strings, date and time, filesystem manipulation, etc. My main gripe is having two regularly accepted libraries that do the same extremely common functions; it makes us look unpolished, especially when there is the "community" versus "official" difference I mentioned earlier. Why not just merge the two libraries into the one D standard library and be done with it?
>
> That's being done for D2. It is not for D1, as that would break everyone's code.
Ah. If we're aiming for one standard library, then I'll shut up and try and understand const.


May 12, 2009
Walter Bright wrote:
> Ameer Armaly wrote:
>> I see. What about the actual "standard library bits" for common tasks like file I/O, strings, date and time, filesystem manipulation, etc. My main gripe is having two regularly accepted libraries that do the same extremely common functions; it makes us look unpolished, especially when there is the "community" versus "official" difference I mentioned earlier. Why not just merge the two libraries into the one D standard library and be done with it?
> 
> That's being done for D2. It is not for D1, as that would break everyone's code.

Well, one runtime library.  Anything beyond that isn't at all clear, is it?

There's some rather fundamental design differences -- neither right nor wrong -- between Phobos and Tango.  I don't see those converging at this point.

Regardless, able to sit side by side is still a big step forward.

Later,
Brad
May 12, 2009
Leandro Lucarella wrote:
> Isn't the demangle code taken from DMDFE?

Why would DMDFE need a demangler? It can just not mangle in the first place :).
May 12, 2009
== Quote from Frits van Bommel (fvbommel@REMwOVExCAPSs.nl)'s article
> Leandro Lucarella wrote:
> > Isn't the demangle code taken from DMDFE?
> Why would DMDFE need a demangler? It can just not mangle in the first place :).

You need name mangling for templates and function overloading to work.
May 12, 2009
Frits van Bommel, el 12 de mayo a las 16:28 me escribiste:
> Leandro Lucarella wrote:
> >Isn't the demangle code taken from DMDFE?
> 
> Why would DMDFE need a demangler? It can just not mangle in the first place :).

Yes, yes, yes, I get the point. I meant Phobos...

¬¬

-- 
Leandro Lucarella (luca) | Blog colectivo: http://www.mazziblog.com.ar/blog/
----------------------------------------------------------------------------
GPG Key: 5F5A8D05 (F8CD F9A7 BF00 5431 4145  104C 949E BFB6 5F5A 8D05)
----------------------------------------------------------------------------