• Bill Baxter (20/20) Sep 30 2007 I just decided to do a little experiment to see how hard it would be to
• Bruce Adams (6/32) Oct 01 2007 Good stuff. Have you posted the source anywhere?
• Bill Baxter (15/52) Oct 01 2007 No I mostly wanted to point out that the Tango containers are 95%
• Bruce Adams (3/11) Oct 01 2007 I dunno. Sounds a bit grim to me :).
• Janice Caron (3/7) Oct 01 2007 Thunderbolt and lightning, very very frightning me!
• Bruce Adams (2/11) Oct 01 2007 Galileo?
• Clay Smith (2/14) Oct 01 2007 Photango?
• Daniel Keep (4/19) Oct 01 2007 We are now the knights who say: ecky, ecky, ecky, PHTANGO zoop-boing!
• BLS (9/70) Oct 02 2007 Phandango is excellent reminds on a song from
• 0ffh (9/11) Oct 01 2007 Cool devices!
• Bill Baxter (5/19) Oct 01 2007 My guess is that the porting trouble spots all center around the modules...
• Janice Caron (4/8) Oct 02 2007 I suspect there are a lot of people who are happy enough with Phobos's
• Lars Ivar Igesund (10/19) Oct 02 2007 The differences in the runtime aren't really about the API's of those
• Janice Caron (9/12) Oct 02 2007 Well certainly not "all", because "all" includes me.
• Regan Heath (20/35) Oct 02 2007 Sometimes "broken" is difficult to define.
• Lars Ivar Igesund (10/33) Oct 02 2007 You are not directly mistaken, but the above point has resulted in a run...
• Lars Ivar Igesund (13/28) Oct 02 2007 You may not have experienced problems yet, but there are two main reason...
• Janice Caron (6/9) Oct 02 2007 Ah, but the whole point of this thread is that I don't /need/ to
• Bill Baxter (5/15) Oct 02 2007 Well an hour for the containers. io and net would probably take
• Lars Ivar Igesund (10/21) Oct 02 2007 And my original post could be translated into the following question; If...
• Bill Baxter (4/20) Oct 02 2007 If they *were* compatible then there would be less of a point. But no
• Alexander Panek (4/6) Oct 02 2007 It's been a month and some days since the conference - I don't think
• UB (4/4) Oct 02 2007 How about patching phobos source and build it locally?
• Alexander Panek (2/7) Oct 02 2007 It is legal, as you can see with Tango.
• UB (4/4) Oct 02 2007 So, in such way we can have one patched standard lib and
• Janice Caron (2/3) Oct 02 2007 You can have a patched standard lib, but not a standard patched lib :-)
• UB (3/3) Oct 02 2007 Yes, that's what I am saying. :) I would like to see patched Phobos with...
• Alexander Panek (6/11) Oct 02 2007 ..that's basically what Tango does, apart from the fact that it's not an...
• Charles D Hixson (4/16) Oct 03 2007 Except that installing Tango breaks Phobos.
• Robert Fraser (2/20) Oct 03 2007 I feel your pain, but D is still a relatively new language, so you proba...
• Derek Parnell (19/41) Oct 03 2007 It has nothing to do with D being "relatively" new. The problem is that
• Alexander Panek (6/7) Oct 03 2007 There's a difference between breaking and replacing. Just back up your
• Charles D Hixson (9/18) Oct 04 2007 When the flags become available, then the statement may well
• Lars Ivar Igesund (8/31) Oct 02 2007 It is not a "it it will happen", just a question of when as there are
• Bill Baxter (6/32) Oct 02 2007 Yeh, that's fine. I understand that. But I'm writing code that needs
• Alexander Panek (11/21) Oct 02 2007 Well it's up to you what you want, and how much effort you are willing

Bill Baxter <dnewsgroup billbaxter.com> writes:

I just decided to do a little experiment to see how hard it would be to 
make Tango's containers work with Phobos.

Answer: less than an hour's work.

Here's all it takes:
- Copy tango/util/collections/* to phango/util/collections/*.
- Copy tango/core/Exception.di to phango/core/Exception.d
- Replace all instances of 'tango' with 'phango'.
- Replace all instances of 'toUtf8' with 'toString'
- Comment out the IReader,IWriter imports in LinkMap.d
- Comment out all function prototypes in Exception.d.
- Make TracedException a simple subclass of Exception in Exception.d.
- Comment out anything else in Exception.d that gives compiler errors.

Then voila!  You have a set of container classes for Phobos.

I did some rudimentary testing by copying all the unittests from the 
containers to one big containers.d test program.  After changing Cout's 
to writefln's, all seems to work like a charm.

So this is a possible option for anyone who would like a set of 
Java-esque container classes to use with Phobos.  Certainly a heck of a 
lot easier than trying to port any implementation of C++ std::map to D.  :-)

--bb

Bruce Adams <tortoise_74 yahoo.co.uk> writes:

Bill Baxter Wrote:

 I just decided to do a little experiment to see how hard it would be to 
 make Tango's containers work with Phobos.
 
 Answer: less than an hour's work.
 
 Here's all it takes:
 - Copy tango/util/collections/* to phango/util/collections/*.
 - Copy tango/core/Exception.di to phango/core/Exception.d
 - Replace all instances of 'tango' with 'phango'.
 - Replace all instances of 'toUtf8' with 'toString'
 - Comment out the IReader,IWriter imports in LinkMap.d
 - Comment out all function prototypes in Exception.d.
 - Make TracedException a simple subclass of Exception in Exception.d.
 - Comment out anything else in Exception.d that gives compiler errors.
 
 Then voila!  You have a set of container classes for Phobos.
 
 I did some rudimentary testing by copying all the unittests from the 
 containers to one big containers.d test program.  After changing Cout's 
 to writefln's, all seems to work like a charm.
 
 So this is a possible option for anyone who would like a set of 
 Java-esque container classes to use with Phobos.  Certainly a heck of a 
 lot easier than trying to port any implementation of C++ std::map to D.  :-)
 
 --bb

Good stuff. Have you posted the source anywhere?
  I see tango uses Red-black trees in some places so in those its STL
equivalent or better anyway. I guess we will get all this stuff as soon as the
tango / phobos incompatability is resolved. Out of interest, if you just throw
the all incompatibles out of tango what kind of phango are we left with? Does
it look like a worthwhile job?
I'm coming from the school of Phobos is standard, Tango is not (yet)
perspective.

Regards,

Bruce.

Bill Baxter <dnewsgroup billbaxter.com> writes:

Bruce Adams wrote:
 Bill Baxter Wrote:
 
 I just decided to do a little experiment to see how hard it would be to 
 make Tango's containers work with Phobos.

 Answer: less than an hour's work.

 Here's all it takes:
 - Copy tango/util/collections/* to phango/util/collections/*.
 - Copy tango/core/Exception.di to phango/core/Exception.d
 - Replace all instances of 'tango' with 'phango'.
 - Replace all instances of 'toUtf8' with 'toString'
 - Comment out the IReader,IWriter imports in LinkMap.d
 - Comment out all function prototypes in Exception.d.
 - Make TracedException a simple subclass of Exception in Exception.d.
 - Comment out anything else in Exception.d that gives compiler errors.

 Then voila!  You have a set of container classes for Phobos.

 I did some rudimentary testing by copying all the unittests from the 
 containers to one big containers.d test program.  After changing Cout's 
 to writefln's, all seems to work like a charm.

 So this is a possible option for anyone who would like a set of 
 Java-esque container classes to use with Phobos.  Certainly a heck of a 
 lot easier than trying to port any implementation of C++ std::map to D.  :-)

 --bb

 
 Good stuff. Have you posted the source anywhere?

No I mostly wanted to point out that the Tango containers are 95% 
independent of the Tango-specific runtime features, and easy to port to 
Phobos.  And I wanted to suggest the cool name "Phango, evil twin 
brother of Tangobos". :-)  Maybe "Phandango" would be a better name, 
though.

 I see tango uses Red-black trees in some places so in those its STL
 equivalent or better anyway. 

Yes, the TreeSet and TreeMap should have same asymptotic performance 
characteristics as STL std::set and std::map.

 I guess we will get all this stuff as
 soon as the tango / phobos incompatability is resolved. 

That's what I'm thinking too.  But I wanted to point out that anyone who 
is desperate for containers today can extract them from Tango with less 
than an hour's work of mostly mindless search and replace.

 Out of
 interest, if you just throw the all incompatibles out of tango what
 kind of phango are we left with? Does it look like a worthwhile job? 

I suspect you'd still have most of it.  Far easier than porting C++ code 
anyway.  Maybe the threading stuff is too dependent on tango's Object to 
port easily.  I'm not sure.

--bb

Bruce Adams <tortoise_74 yeah.who.co.uk> writes:

Bill Baxter Wrote:
 
 Good stuff. Have you posted the source anywhere?

 
 No I mostly wanted to point out that the Tango containers are 95% 
 independent of the Tango-specific runtime features, and easy to port to 
 Phobos.  And I wanted to suggest the cool name "Phango, evil twin 
 brother of Tangobos". :-)  Maybe "Phandango" would be a better name, 
 though.
 

I dunno. Sounds a bit grim to me :).

"Janice Caron" <caron800 googlemail.com> writes:

On 10/1/07, Bruce Adams <tortoise_74 yeah.who.co.uk> wrote:
 Maybe "Phandango" would be a better name,
 though.

 I dunno. Sounds a bit grim to me :).

Thunderbolt and lightning, very very frightning me!

Yes - Phandango! (PHobos AND tANGO) - Absolutely superb name!

Bruce Adams <tortoise_74 yeah.who.co.uk> writes:

Janice Caron Wrote:

 On 10/1/07, Bruce Adams <tortoise_74 yeah.who.co.uk> wrote:
 Maybe "Phandango" would be a better name,
 though.

 I dunno. Sounds a bit grim to me :).

 
 Thunderbolt and lightning, very very frightning me!
 
 Yes - Phandango! (PHobos AND tANGO) - Absolutely superb name!

Galileo?

Clay Smith <clayasaurus gmail.com> writes:

Bruce Adams wrote:
 Janice Caron Wrote:
 
 On 10/1/07, Bruce Adams <tortoise_74 yeah.who.co.uk> wrote:
 Maybe "Phandango" would be a better name,
 though.

 I dunno. Sounds a bit grim to me :).

 Thunderbolt and lightning, very very frightning me!

 Yes - Phandango! (PHobos AND tANGO) - Absolutely superb name!

 
 Galileo?

Photango?

Daniel Keep <daniel.keep.lists gmail.com> writes:

Clay Smith wrote:
 Bruce Adams wrote:
 Janice Caron Wrote:

 On 10/1/07, Bruce Adams <tortoise_74 yeah.who.co.uk> wrote:
 Maybe "Phandango" would be a better name,
 though.

 I dunno. Sounds a bit grim to me :).

 Thunderbolt and lightning, very very frightning me!

 Yes - Phandango! (PHobos AND tANGO) - Absolutely superb name!

 Galileo?

 
 Photango?

We are now the knights who say: ecky, ecky, ecky, PHTANGO zoop-boing!

Sorry; couldn't resist.

	-- Daniel

BLS <nanali nospam-wanadoo.fr> writes:

Bill Baxter schrieb:
 Bruce Adams wrote:
 Bill Baxter Wrote:

 I just decided to do a little experiment to see how hard it would be 
 to make Tango's containers work with Phobos.

 Answer: less than an hour's work.

 Here's all it takes:
 - Copy tango/util/collections/* to phango/util/collections/*.
 - Copy tango/core/Exception.di to phango/core/Exception.d
 - Replace all instances of 'tango' with 'phango'.
 - Replace all instances of 'toUtf8' with 'toString'
 - Comment out the IReader,IWriter imports in LinkMap.d
 - Comment out all function prototypes in Exception.d.
 - Make TracedException a simple subclass of Exception in Exception.d.
 - Comment out anything else in Exception.d that gives compiler errors.

 Then voila!  You have a set of container classes for Phobos.

 I did some rudimentary testing by copying all the unittests from the 
 containers to one big containers.d test program.  After changing 
 Cout's to writefln's, all seems to work like a charm.

 So this is a possible option for anyone who would like a set of 
 Java-esque container classes to use with Phobos.  Certainly a heck of 
 a lot easier than trying to port any implementation of C++ std::map 
 to D.  :-)

 --bb

 Good stuff. Have you posted the source anywhere?

 
 No I mostly wanted to point out that the Tango containers are 95% 
 independent of the Tango-specific runtime features, and easy to port to 
 Phobos.  And I wanted to suggest the cool name "Phango, evil twin 
 brother of Tangobos". :-)  Maybe "Phandango" would be a better name, 
 though.
 

Phandango is excellent reminds on a song from
PROCUL HARUM -
"A Brighter Shade Of Pale” :)

We skipped the light Phandango
Turned cartwheels 'cross the floor
I was feeling kind of seasick
But the crowd called out for more
Just kidding, Bjoern

 I see tango uses Red-black trees in some places so in those its STL
 equivalent or better anyway. 

 
 Yes, the TreeSet and TreeMap should have same asymptotic performance 
 characteristics as STL std::set and std::map.
 
 I guess we will get all this stuff as
 soon as the tango / phobos incompatability is resolved. 

 
 That's what I'm thinking too.  But I wanted to point out that anyone who 
 is desperate for containers today can extract them from Tango with less 
 than an hour's work of mostly mindless search and replace.
 
 Out of
 interest, if you just throw the all incompatibles out of tango what
 kind of phango are we left with? Does it look like a worthwhile job? 

 
 I suspect you'd still have most of it.  Far easier than porting C++ code 
 anyway.  Maybe the threading stuff is too dependent on tango's Object to 
 port easily.  I'm not sure.
 
 --bb

0ffh <spam frankhirsch.net> writes:

Bill Baxter wrote:
 I just decided to do a little experiment to see how hard it would be to 
 make Tango's containers work with Phobos. [...]

Cool devices!
BTW I hope this is not too OT, but:
I have not much idea of the Phobos and Tango internals, but as I read it
it would mostly be the distinct object (and exception?) classes that
make the most trouble in making those libs compatible, or am I mistaken?
I suppose decoupling the gc's off the libs shouldn't be the main trouble?

This is just out of curiosity...

Rgeards, Frank

Bill Baxter <dnewsgroup billbaxter.com> writes:

0ffh wrote:
 Bill Baxter wrote:
 I just decided to do a little experiment to see how hard it would be 
 to make Tango's containers work with Phobos. [...]

 
 Cool devices!
 BTW I hope this is not too OT, but:
 I have not much idea of the Phobos and Tango internals, but as I read it
 it would mostly be the distinct object (and exception?) classes that
 make the most trouble in making those libs compatible, or am I mistaken?
 I suppose decoupling the gc's off the libs shouldn't be the main trouble?
 
 This is just out of curiosity...
 
 Rgeards, Frank

My guess is that the porting trouble spots all center around the modules 
that are supplied as .di headers in the tango distribution: 
tango/core/exception.di, tango/core/memory.di, and tango/core/thread.di.

--bb

"Janice Caron" <caron800 googlemail.com> writes:

On 10/2/07, 0ffh <spam frankhirsch.net> wrote:
 BTW I hope this is not too OT, but:
 I have not much idea of the Phobos and Tango internals, but as I read it
 it would mostly be the distinct object (and exception?) classes that
 make the most trouble in making those libs compatible, or am I mistaken?

I suspect there are a lot of people who are happy enough with Phobos's
objects and exceptions, but who wouldn't say no to a few extra snazzy
container and string classes.

Lars Ivar Igesund <larsivar igesund.net> writes:

Janice Caron wrote:

 On 10/2/07, 0ffh <spam frankhirsch.net> wrote:
 BTW I hope this is not too OT, but:
 I have not much idea of the Phobos and Tango internals, but as I read it
 it would mostly be the distinct object (and exception?) classes that
 make the most trouble in making those libs compatible, or am I mistaken?

 
 I suspect there are a lot of people who are happy enough with Phobos's
 objects and exceptions, but who wouldn't say no to a few extra snazzy
 container and string classes.

The differences in the runtime aren't really about the API's of those
(Phobos and Tango will become compatible in that respect). As it is, I
would expect most (if not all) to want Tango's runtime ahead of Phobos'
given compatible interfaces.

-- 
Lars Ivar Igesund
blog at http://larsivi.net
DSource, #d.tango & #D: larsivi
Dancing the Tango

"Janice Caron" <caron800 googlemail.com> writes:

On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 As it is, I
 would expect most (if not all) to want Tango's runtime ahead of Phobos'
 given compatible interfaces.

Well certainly not "all", because "all" includes me.

You have to realise, a lot of people, myself included, adopt the
philosophy "If an engine ain't broke, don't fix it". If there is /no
need/ to replace the runtime (which there isn't), then I'm not going
to do it, because what I have right now works just fine, so I'm not
going to risk breaking anything by changing it.

On the other hand, I'd have no problems with just importing a module
to get me lots of cool new classes.

Regan Heath <regan netmail.co.nz> writes:

Janice Caron wrote:
 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 As it is, I
 would expect most (if not all) to want Tango's runtime ahead of Phobos'
 given compatible interfaces.

 
 Well certainly not "all", because "all" includes me.
 
 You have to realise, a lot of people, myself included, adopt the
 philosophy "If an engine ain't broke, don't fix it". If there is /no
 need/ to replace the runtime (which there isn't), then I'm not going
 to do it, because what I have right now works just fine, so I'm not
 going to risk breaking anything by changing it.
 
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

Sometimes "broken" is difficult to define.

I have the impression (perhaps mistaken) that the main point of 
difference between the phobos and tango runtime is that tango's is 
better organised, has less coupling, and allows for more advanced 
features to be built upon it (some of which are already present in Tango).

Now, I could be wrong, and if so perhaps Sean or Kris can fill me (and I 
suspect some of the rest of us) in on what exactly the main differences 
are in the core runtime.

If I'm not too far off the mark, and if you take the view that those 
things are important and/or requirements then in that sense phobos is 
broken.

I'm not knocking phobos or Walter, heck I'm not even using tango at this 
point, but then, I'm not working on any projects in D of any significant 
size at the moment.

Assuming I am not delusional then it is my hope that the tango runtime 
will either replace or be a model for change to the phobos one.

I also have complete faith in both Walter and the tango guys, in that I 
simply assume they will both do whatever is best for D, and by extension me.

Regan

Lars Ivar Igesund <larsivar igesund.net> writes:

Regan Heath wrote:

 Janice Caron wrote:
 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 As it is, I
 would expect most (if not all) to want Tango's runtime ahead of Phobos'
 given compatible interfaces.

 
 Well certainly not "all", because "all" includes me.
 
 You have to realise, a lot of people, myself included, adopt the
 philosophy "If an engine ain't broke, don't fix it". If there is /no
 need/ to replace the runtime (which there isn't), then I'm not going
 to do it, because what I have right now works just fine, so I'm not
 going to risk breaking anything by changing it.
 
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 
 Sometimes "broken" is difficult to define.
 
 I have the impression (perhaps mistaken) that the main point of
 difference between the phobos and tango runtime is that tango's is
 better organised, has less coupling, and allows for more advanced
 features to be built upon it (some of which are already present in Tango).

You are not directly mistaken, but the above point has resulted in a runtime
that is easier to mantain, easier to fix bugs in and has thus become
something of higher quality. There are also a few features not in the
Phobos runtime.

-- 
Lars Ivar Igesund
blog at http://larsivi.net
DSource, #d.tango & #D: larsivi
Dancing the Tango

Lars Ivar Igesund <larsivar igesund.net> writes:

Janice Caron wrote:

 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 As it is, I
 would expect most (if not all) to want Tango's runtime ahead of Phobos'
 given compatible interfaces.

 
 Well certainly not "all", because "all" includes me.
 
 You have to realise, a lot of people, myself included, adopt the
 philosophy "If an engine ain't broke, don't fix it". If there is /no
 need/ to replace the runtime (which there isn't), then I'm not going
 to do it, because what I have right now works just fine, so I'm not
 going to risk breaking anything by changing it.

You may not have experienced problems yet, but there are two main reasons
cited when someone switch from Phobos to Tango, where one is issues with
Phobos' runtime (typically threading and the GC). In the cases where the
same issue is identified in Tango, it is usually fixed by the next release
(and often within days in our repository). This may sound like bragging,
but it is meant as a testament to what is our main focus, the library.

 
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

This is the other reason for people switching.

-- 
Lars Ivar Igesund
blog at http://larsivi.net
DSource, #d.tango & #D: larsivi
Dancing the Tango

"Janice Caron" <caron800 googlemail.com> writes:

On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 This is the other reason for people switching.

Ah, but the whole point of this thread is that I don't /need/ to
switch, because now we know that with about an hour's effort we can
make those classes work with Phobos. That's what "Phango" (or
"Phandango") means. I don't /want/ to have to rip out the innards of D
just to make a string class work.

Bill Baxter <dnewsgroup billbaxter.com> writes:

Janice Caron wrote:
 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 This is the other reason for people switching.

 
 Ah, but the whole point of this thread is that I don't /need/ to
 switch, because now we know that with about an hour's effort we can
 make those classes work with Phobos. That's what "Phango" (or
 "Phandango") means. I don't /want/ to have to rip out the innards of D
 just to make a string class work.

Well an hour for the containers.  io and net would probably take 
significantly longer.  math would probably take less (it still has 
'writefln's commented out in the source :-))

--bb

Lars Ivar Igesund <larsivar igesund.net> writes:

Janice Caron wrote:

 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 This is the other reason for people switching.

 
 Ah, but the whole point of this thread is that I don't /need/ to
 switch, because now we know that with about an hour's effort we can
 make those classes work with Phobos. That's what "Phango" (or
 "Phandango") means. I don't /want/ to have to rip out the innards of D
 just to make a string class work.

And my original post could be translated into the following question; If the
runtimes are compatible, why spend time on porting the utility classes when
it would be much faster to install a different runtime (it is a question of
replacing one file) that usually is considered to be better?

-- 
Lars Ivar Igesund
blog at http://larsivi.net
DSource, #d.tango & #D: larsivi
Dancing the Tango

Bill Baxter <dnewsgroup billbaxter.com> writes:

Lars Ivar Igesund wrote:
 Janice Caron wrote:
 
 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 This is the other reason for people switching.

 Ah, but the whole point of this thread is that I don't /need/ to
 switch, because now we know that with about an hour's effort we can
 make those classes work with Phobos. That's what "Phango" (or
 "Phandango") means. I don't /want/ to have to rip out the innards of D
 just to make a string class work.

 
 And my original post could be translated into the following question; If the
 runtimes are compatible, why spend time on porting the utility classes when
 it would be much faster to install a different runtime (it is a question of
 replacing one file) that usually is considered to be better?

If they *were* compatible then there would be less of a point.  But no 
one really knows when or if this compatibility is going to happen.

--bb

Alexander Panek <alexander.panek brainsware.org> writes:

Bill Baxter wrote:
 If they *were* compatible then there would be less of a point.  But no 
 one really knows when or if this compatibility is going to happen.

It's been a month and some days since the conference - I don't think 
everything is sorted out so fast. But I am certain it /will/ be sorted 
out. Just have patience.

UB <me mymail.com> writes:

How about patching phobos source and build it locally?
Is it legal?
Then there will be only one standard lib.

ub

Alexander Panek <alexander.panek brainsware.org> writes:

UB wrote:
 How about patching phobos source and build it locally?
 Is it legal?
 Then there will be only one standard lib.
 
 ub

It is legal, as you can see with Tango.

UB <me mymail.com> writes:

So, in such way we can have one patched standard lib and
new/additional code can go in phobos_ext.lib
what do you think about it?

ub

"Janice Caron" <caron800 googlemail.com> writes:

On 10/2/07, UB <me mymail.com> wrote:
 So, in such way we can have one patched standard lib and

You can have a patched standard lib, but not a standard patched lib :-)

UB <me mymail.com> writes:

Yes, that's what I am saying. :) I would like to see patched Phobos with Tango
extensions.

ub

Alexander Panek <alexander.panek brainsware.org> writes:

UB wrote:
 So, in such way we can have one patched standard lib and
 new/additional code can go in phobos_ext.lib
 what do you think about it?
 
 ub

..that's basically what Tango does, apart from the fact that it's not an 
extension, but rather a replacement. The problem of having two 
incompatible runtime libraries (the core runtime, that is) still 
persists - it is addressed, though........I think I just repeated it the 
95734897th time now.

Charles D Hixson <charleshixsn earthlink.net> writes:

Alexander Panek wrote:
 UB wrote:
 So, in such way we can have one patched standard lib and
 new/additional code can go in phobos_ext.lib
 what do you think about it?

 ub

 
 ...that's basically what Tango does, apart from the fact that it's not 
 an extension, but rather a replacement. The problem of having two 
 incompatible runtime libraries (the core runtime, that is) still 
 persists - it is addressed, though........I think I just repeated it the 
 95734897th time now.

Except that installing Tango breaks Phobos.

(I know, patience.  While I'm being patient, I end up 
programming in a different language.)

Robert Fraser <fraserofthenight gmail.com> writes:

Charles D Hixson Wrote:

 Alexander Panek wrote:
 UB wrote:
 So, in such way we can have one patched standard lib and
 new/additional code can go in phobos_ext.lib
 what do you think about it?

 ub

 
 ...that's basically what Tango does, apart from the fact that it's not 
 an extension, but rather a replacement. The problem of having two 
 incompatible runtime libraries (the core runtime, that is) still 
 persists - it is addressed, though........I think I just repeated it the 
 95734897th time now.

 
 Except that installing Tango breaks Phobos.
 
 (I know, patience.  While I'm being patient, I end up 
 programming in a different language.)

I feel your pain, but D is still a relatively new language, so you probably
shouldn't be using it for production code, yet, anyway. By programming in D,
you're accepting that the language, as powerful as it is, is definitely not
"everything it will be."

Derek Parnell <derek psych.ward> writes:

On Wed, 03 Oct 2007 18:42:17 -0400, Robert Fraser wrote:

 Charles D Hixson Wrote:
 
 Alexander Panek wrote:
 UB wrote:
 So, in such way we can have one patched standard lib and
 new/additional code can go in phobos_ext.lib
 what do you think about it?

 ub

 
 ...that's basically what Tango does, apart from the fact that it's not 
 an extension, but rather a replacement. The problem of having two 
 incompatible runtime libraries (the core runtime, that is) still 
 persists - it is addressed, though........I think I just repeated it the 
 95734897th time now.

 
 Except that installing Tango breaks Phobos.
 
 (I know, patience.  While I'm being patient, I end up 
 programming in a different language.)

 
 I feel your pain, but D is still a relatively new language

It has nothing to do with D being "relatively" new. The problem is that
there are far too few people who a working on D and its library on a
full-time basis. Thus we get a dribble of output, which enforces the need
for patience. I know there is a such a thing as balance between too many
and too few, but I'm pretty sure we are still on the "too few" side of the
see-saw.

The bottleneck is Walter's ability to encorporate change, whether change
comes from himself or others. This has a direct effect on GDC too as it
can't "outstrip" Walter's output because it doesn't know where D is going
until Walter implements stuff.

From my uninformed position, it appears that Tango and Phobos can never
co-exist until Walter agrees to replace Phobos' runtime paradigm with the
superior Tango one. And that's is not going to be a simple thing for Walter
to agree to.

-- 
Derek Parnell
Melbourne, Australia
skype: derek.j.parnell

Alexander Panek <a.panek brainsware.org> writes:

Charles D Hixson wrote:
 Except that installing Tango breaks Phobos.

There's a difference between breaking and replacing. Just back up your 
libphobos.a and there you go.

Apart from that, there's soon to come support for the new compiler flags 
(-defaultlib, -debuglib) in Tango, which should render your statement 
obsolete.

Charles D Hixson <charleshixsn earthlink.net> writes:

Alexander Panek wrote:
 Charles D Hixson wrote:
 Except that installing Tango breaks Phobos.

 
 There's a difference between breaking and replacing. Just back up your 
 libphobos.a and there you go.
 
 Apart from that, there's soon to come support for the new compiler flags 
 (-defaultlib, -debuglib) in Tango, which should render your statement 
 obsolete.

When the flags become available, then the statement may well 
be obsolete.  At the moment...being able to restore phobos 
doesn't mean that it isn't currently broken.

Having them both installed and neither broken implies that I 
can have programs that use a mix of phobos and tango 
calls...possibly in different object files.  Until that's 
true, the D code base is fragmented and at least one of the 
libraries is either broken or not installed.

Lars Ivar Igesund <larsivar igesund.net> writes:

Bill Baxter wrote:

 Lars Ivar Igesund wrote:
 Janice Caron wrote:
 
 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 This is the other reason for people switching.

 Ah, but the whole point of this thread is that I don't /need/ to
 switch, because now we know that with about an hour's effort we can
 make those classes work with Phobos. That's what "Phango" (or
 "Phandango") means. I don't /want/ to have to rip out the innards of D
 just to make a string class work.

 
 And my original post could be translated into the following question; If
 the runtimes are compatible, why spend time on porting the utility
 classes when it would be much faster to install a different runtime (it
 is a question of replacing one file) that usually is considered to be
 better?

 
 If they *were* compatible then there would be less of a point.  But no
 one really knows when or if this compatibility is going to happen.
 
 --bb

It is not a "it it will happen", just a question of when as there are
bandwidth restrictions :) We want to resolve this ASAP.

-- 
Lars Ivar Igesund
blog at http://larsivi.net
DSource, #d.tango & #D: larsivi
Dancing the Tango

Bill Baxter <dnewsgroup billbaxter.com> writes:

Lars Ivar Igesund wrote:
 Bill Baxter wrote:
 
 Lars Ivar Igesund wrote:
 Janice Caron wrote:

 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 This is the other reason for people switching.

 Ah, but the whole point of this thread is that I don't /need/ to
 switch, because now we know that with about an hour's effort we can
 make those classes work with Phobos. That's what "Phango" (or
 "Phandango") means. I don't /want/ to have to rip out the innards of D
 just to make a string class work.

 And my original post could be translated into the following question; If
 the runtimes are compatible, why spend time on porting the utility
 classes when it would be much faster to install a different runtime (it
 is a question of replacing one file) that usually is considered to be
 better?

 If they *were* compatible then there would be less of a point.  But no
 one really knows when or if this compatibility is going to happen.

 --bb

 
 It is not a "it it will happen", just a question of when as there are
 bandwidth restrictions :) We want to resolve this ASAP.

Yeh, that's fine.  I understand that.  But I'm writing code that needs 
container classes *today*.  So I just thought I'd give porting to phobos 
a try.  And hey! It wasn't that hard and now I have some container 
classes I can use until the schism is mended.   That's all I'm saying.

--bb

Alexander Panek <alexander.panek brainsware.org> writes:

Janice Caron wrote:
 On 10/2/07, Lars Ivar Igesund <larsivar igesund.net> wrote:
 On the other hand, I'd have no problems with just importing a module
 to get me lots of cool new classes.

 This is the other reason for people switching.

 
 Ah, but the whole point of this thread is that I don't /need/ to
 switch, because now we know that with about an hour's effort we can
 make those classes work with Phobos. That's what "Phango" (or
 "Phandango") means. I don't /want/ to have to rip out the innards of D
 just to make a string class work.

Well it's up to you what you want, and how much effort you are willing 
to put into /porting/ the parts you try to use - as opposed to just 
using them inside the complete Tango environment.

..but before you get started: don't. I have read enough to see your 
point: "Don't swim upstream, ffs! What are you doing! That's not 
standard!". No offense intended, but you seem to be ridiculously 
reluctant to changes made by other than the original author of an 
environment (which is Walter in case of D and Phobos). You also seem to 
forget, that we're talking all open source and partly even free software 
(F/OSS).