Is 3.15p -still- the latest GNAT 'p' release?

Is 3.15p -still- the latest GNAT 'p' release?

[Dale Stanbrough]

I was looking at cs.nyu.edu the other day, and the latest Gnat version
still seems to be 3.15, which dates from 2002.

Is this -really- the latest release?

Dale

-- 
dstanbro@spam.o.matic.bigpond.net.au

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

Dale Stanbrough <MrNoSpam@bigpoop.net.au> writes:

> I was looking at cs.nyu.edu the other day, and the latest Gnat version
> still seems to be 3.15, which dates from 2002.
> 
> Is this -really- the latest release?

Yes, it is the latest official release, and is still recommended,
unless you want to participate in the development of GNAT.

There are 3 "branches" of GNAT: (1) the stable, official, "p" (public)
releases under GMGPL, (2) the "GNAT Pro" releases offered only to
paying customers, and (3) GCC.  The Ada front-end in GCC has been
available since GCC 3.1, but is not as stable as GNAT 3.15p.

The future is uncertain to me.  ACT has said they plan to make
additional "p" releases available, but there is no telling when.
Meanwhile, a lot of work has gone into the upcoming GCC 3.4.  It will
be more stable, offer some features proposed for Ada 2005, as well as
some additions to project files and the GNAT.* library.  But it does
not support tasking on powerpc-*-linux (a show stopper as far as I am
concerned), and lacks GLADE.

My recommendation is: stick with 3.15p.  If you want to be on the
bleeding edge, experiment with GCC 3.4 or, better yet, get involved in
the development of GCC 3.5.  Avoid GCC 3.1, 3.2 and 3.3.

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Hyman Rosen]

Ludovic Brenta wrote:
>(2) the "GNAT Pro" releases offered only to paying customers

You could always contact friends who are (or work for) paying
customers and get a Pro version from them.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Marin David Condic]

Even if 3.15p is a little old, it seems to me that there wouldn't really 
be much reason to complain about it. It works quite well and has a lot 
of features and, since we're talking about a compiler for a computer 
language, how much more does it really need to do? If the language 
changes - sure - there's cause to want a new release. If ACT increases 
their library of utilities, again, there's cause to want the new 
libraries - but that's almost a separate product in the sense that it 
would likely work with an older version of the compiler. You could 
probably freeze 3.15p and never make another release and still have a 
perfectly usable product. Have we been conditioned by Microsoft to think 
that unless we get a new release every quarter we are somehow missing 
out on something we absolutely have to have?

MDC

Ludovic Brenta wrote:
> 
> There are 3 "branches" of GNAT: (1) the stable, official, "p" (public)
> releases under GMGPL, (2) the "GNAT Pro" releases offered only to
> paying customers, and (3) GCC.  The Ada front-end in GCC has been
> available since GCC 3.1, but is not as stable as GNAT 3.15p.
> 


-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jsf.mil/NSFrames.htm

Send Replies To: m   o   d   c @ a   m   o   g
                    c   n   i       c   .   r

     "Face it ladies, its not the dress that makes you look fat.
     Its the FAT that makes you look fat."

         --  Al Bundy

======================================================================

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Preben Randhol]

On 2004-03-26, Marin David Condic <nobody@noplace.com> wrote:
> Even if 3.15p is a little old, it seems to me that there wouldn't really 
> be much reason to complain about it. It works quite well and has a lot 
> of features and, since we're talking about a compiler for a computer 
> language, how much more does it really need to do? 


I agree. Besides work *is* being done to get GNAT up to date with gcc 3.x
and this takes a lot of time (especially since gcc seems to be a moving
target as well). As I see it this is one of the biggest improvements one
can do to the compiler as one then hopefully can get better optimisation
(I have forgotten the name, but both Fortran and Ada can do better
optimising if the gcc backend is improved. I think this is something for
4.x.) 

So that Gnat 3.15p is from 2002 doesn't matter.


-- 
Preben Randhol -------- http://www.pvv.org/~randhol/

()  "Violence is the last refuge of the incompetent"
/\                                   - Isaac Asimov

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Mark H Johnson]

Marin David Condic wrote:
> You could 
> probably freeze 3.15p and never make another release and still have a 
> perfectly usable product.

Actually on Linux, that is not quite true. I have a very large system 
implemented on GNAT / Linux (still 3.14a, but compatible with 3.17w) but 
someday, we have to step up beyond Red Hat 7.3 to something a little 
more modern. At that point, we have the pthread compatibility issue to 
deal with. We have the following options at that point:

  - step up to 3.16 or later to stay with the older backend
  - step up to 5.02 or later to use the new backend

Both of these official compiler releases have RH <9 and RH >=9 versions 
available.

I don't expect the first option to be a good long term strategy because 
I expect ACT to drop support for the 3.xx compilers within a year or so. 
They haven't said that directly but the following statement in the 5.02a 
release notification appears to imply that...
   "The 5.02a release completes the transition from GNAT3 to GNAT5 for
most of the GNAT Pro configurations."

The second option is also not quite as clean as we would like. The 
backend in 5.xx appears to do more aggressive optimization and we have 
some broken code to fix (it appears we need to add some more pragmas for 
aliasing or volatile). It is an annoyance at this point (since we 
obviously have working code / compiler) but time will be needed to make 
the fixes.

Yes - I am aware of the environment variable work around, but I don't 
see that as a good long term strategy either.
   --Mark

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Jano]

Marin David Condic dice...
> Even if 3.15p is a little old, it seems to me that there wouldn't really 
> be much reason to complain about it. 

I agree, it works very well. Though if you're planning to use it in 
windows with tasking or Gnat.Sockets there are some bugs you should 
avoid before they strike you in the head. Do a search in this group... 
They've been discussed here previously.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Craig Carey]

On Thu, 25 Mar 2004 22:40:36 GMT, Dale Stanbrough
<MrNoSpam@bigpoop.net.au> wrote:

>I was looking at cs.nyu.edu the other day, and the latest Gnat version
>still seems to be 3.15, which dates from 2002.
>
>Is this -really- the latest release?


Newer GNAT Ada files are in the FSF GCC CVS repository.

---

The MinGW Ada compiler would run if the environment variable, GNAT_ROOT,
is set to a correct pathname in the Win32 unix shell (Cygwin bash or
etc.)

---

I now report a success at getting WinCVS to download the FSF Gcc
compiler files. It got harder after about December 2003.

The GCC via CVS webpage: http://gcc.gnu.org/cvs.html

The GNATLIST mailing list is another place for this topic:
    http://homepage.sunrise.ch/mysunrise/gdm/gnatlist.htm

Instructions:

Get Cygwin (setup.exe at http://www.cygwin.com), and Openssh and Minires.

Also, WinCVS 1.3: http://www.wincvs.org/ (e.g. 1.3.14.1 beta 14 build 1).

Start WinCVS 1.3.

Select "Admin / Command Line"

Note: "Remote / Checkout module" and "Admin / Login" are not used.

This command line entered is:
  cvs -d :ext:anoncvs@savannah.gnu.org:/cvsroot/gcc -tz9 checkout -P gcc

( -P prunes directories)
( -t produces some debugging information.)
WinCVS might not run correctly if the command line has a trailing
line character/characters.

I didn't have success downloading Gcc when WinCVS entered part of the
 command line (in 1 test).

   Add global options       = [No] (Instead -P is manually entered)
   Force using CVSROOT (-d) = [No]
   Execute for Dir = [No]

General:
   Authentification = ssh
   Path = /cvsroot/gcc
   Username = ext:anoncvs
   CVSROOT: ext:anoncvs@savannah.gnu.org:/cvsroot/gcc
   Behind the settings button:
      RSA [tick]: V:\gcc\home\.ssh/identity
      SSH [tick]: V:\gcc\bin\ssh.exe
      Additional Ssh options: no

The -t can show why CVS uploads so much. Due to a bug, Ctrl-C would
be pressed, and then it lists the names of the files it is
unnecessarily uploading. During downloading, it can print out
"move aside", and so on.

WinCVS ought be able to update a single directory.




Craig Carey

Ada 95 bindings to GMP precise rational numbers C software:
http://www.ijs.co.nz/code/ada95_multiprecision_gmp_bindings.zip
GMP : http://www.swox.com/gmp/

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

Hyman Rosen <hyrosen@mail.com> writes:

> Ludovic Brenta wrote:
> >(2) the "GNAT Pro" releases offered only to paying customers
> 
> You could always contact friends who are (or work for) paying
> customers and get a Pro version from them.

This is pirating.  I will never encourage such a disrespectful
attitude.  ACT are doing an outstanding job.  As a professional
software developer, I happen to know just how much effort goes into
testing, verification and certification.  I pray everyone on this
newsgroup not to listen to Hyman Rosen.  If you require GNAT Pro, pay.

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

Marin David Condic <nobody@noplace.com> writes:

> Even if 3.15p is a little old, it seems to me that there wouldn't
> really be much reason to complain about it. It works quite well and
> has a lot of features and, since we're talking about a compiler for a
> computer language, how much more does it really need to do? If the
> language changes - sure - there's cause to want a new release. If ACT
> increases their library of utilities, again, there's cause to want the
> new libraries - but that's almost a separate product in the sense that
> it would likely work with an older version of the compiler. You could
> probably freeze 3.15p and never make another release and still have a
> perfectly usable product. Have we been conditioned by Microsoft to
> think that unless we get a new release every quarter we are somehow
> missing out on something we absolutely have to have?

Heh, I was refraining from saying just that for fear of looking like
an emotional Ada bigot and Microsoft basher.  Now there's two of us :)

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[tmoran]

> > You could always contact friends who are (or work for) paying
> > customers and get a Pro version from them.
>
> This is pirating.  I will never encourage such a disrespectful
    I thought Gnat was Open Source and was developed with taxpayer funds
on that basis.  No?

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

Mark H Johnson <Mark_H_Johnson@raytheon.com> writes:

> Marin David Condic wrote:
> > You could probably freeze 3.15p and never make another release and
> > still have a perfectly usable product.
> 
> Actually on Linux, that is not quite true. I have a very large system
> implemented on GNAT / Linux (still 3.14a, but compatible with 3.17w)
> but someday, we have to step up beyond Red Hat 7.3 to something a
> little more modern. At that point, we have the pthread compatibility
> issue to deal with. We have the following options at that point:
> 
>   - step up to 3.16 or later to stay with the older backend
>   - step up to 5.02 or later to use the new backend
> 
> Both of these official compiler releases have RH <9 and RH >=9
> versions available.
> 
> I don't expect the first option to be a good long term strategy
> because I expect ACT to drop support for the 3.xx compilers within a
> year or so. They haven't said that directly but the following
> statement in the 5.02a release notification appears to imply that...
>    "The 5.02a release completes the transition from GNAT3 to GNAT5 for
> most of the GNAT Pro configurations."

So, you seem to be a paying customer of ACT's.  Good.  This means that
ACT does all the work of validating the compiler against several
platforms for you.
 
> The second option is also not quite as clean as we would like. The
> backend in 5.xx appears to do more aggressive optimization and we have
> some broken code to fix (it appears we need to add some more pragmas
> for aliasing or volatile). It is an annoyance at this point (since we
> obviously have working code / compiler) but time will be needed to
> make the fixes.
> 
> Yes - I am aware of the environment variable work around, but I don't
> see that as a good long term strategy either.
>    --Mark

From what I understand, the problem, then, is neither Ada, nor ACT,
nor GNAT, nor the pthreads library, but only the fact that "someday,
[you] will have to step up beyond Red Hat 7.3 to something a little
more modern".

If you have strong validation requirements, you will have to explain
why you absolutely have to move away from Red Hat 7.3 (as a side note,
I still consider that particular release the best ever from Red Hat;
the ones after that were, IME, of lesser quality).  If you can justify
the effort of re-validating your application on a newer Red Hat, then
surely the effort of switching compilers will be minimal compared to
it, given that ACT already does most of the work for you.

If you are looking for a long-term strategy, then your first task is
to define long-term.  Then, try to find a platform, the long-term
strategy of which matches yours as closely as possible.  For example,
Red Hat has 6-month release cycles; Red Hat Advanced Server has 3-year
release cycles; and Debian has a "when it's ready" release cycle.

Use Debian :)

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Georg Bauhaus]

tmoran@acm.org wrote:
:> > You could always contact friends who are (or work for) paying
:> > customers and get a Pro version from them.
:>
:> This is pirating.  I will never encourage such a disrespectful
:    I thought Gnat was Open Source and was developed with taxpayer funds
: on that basis.  No?

Isn't the compiler funded this way available as Open Source?

Would CLAW profit from Pro versions?

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Steve]

"Ludovic Brenta" <ludovic.brenta@insalien.org> wrote in message
news:87isgryxkz.fsf@insalien.org...
> Hyman Rosen <hyrosen@mail.com> writes:
>
> > Ludovic Brenta wrote:
> > >(2) the "GNAT Pro" releases offered only to paying customers
> >
> > You could always contact friends who are (or work for) paying
> > customers and get a Pro version from them.
>
> This is pirating.  I will never encourage such a disrespectful
> attitude.  ACT are doing an outstanding job.  As a professional
> software developer, I happen to know just how much effort goes into
> testing, verification and certification.  I pray everyone on this
> newsgroup not to listen to Hyman Rosen.  If you require GNAT Pro, pay.
>

In my vocabulary I reserve the term "pirating" for use that violates
copyrights.
This subject has come up before, back in the days when representatives from
ACT (unofficially) frequented the list, and they never voiced any objections
to anyone sharing an official distribution.  There was some indication that
they had never heard of it happening.

When you purchase GNAT Pro, you're not just purchasing a compiler, you're
contracting for support.  High quality support.  If you plan to use GNAT for
professional work, you really should buy support.  Not because you have to,
but because it makes good buisness sense.

Steve
(The Duck)

> -- 
> Ludovic Brenta.
>

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Alexandre E. Kopilovitch]

Ludovic Brenta write:

> Hyman Rosen <hyrosen@mail.com> writes:
>
> > Ludovic Brenta wrote:
> > >(2) the "GNAT Pro" releases offered only to paying customers
> > 
> > You could always contact friends who are (or work for) paying
> > customers and get a Pro version from them.
>
> This is pirating.

You made me smiling, almost laughing, thanks -:) . What do you know about
so-called "piracy"? I guess - from that your sentence - that you know very
little, too little about this controversial issue.

I think that taking GNAT Pro from a friend (if it appears possible) can't
be "pirating" in any sense (and if fact, if someone managed to "pirate" ACT,
that is, Robert Dewar, it would be big and admirable achievement, and true
contribution into software business art and experience -:) .

There is little sense to take GNAT Pro from a friend unless the "pirate" has
the same profile (as a developer) as that friend, which is unlikely. If fact,
it may be only reasonable for evaluating either GNAT Pro itself or some new
language or library features provided there - but this hardly can make any
damage to ACT.

>  I will never encourage such a disrespectful attitude.

I can't see why you call this attitude "disrespectful". On the contrary,
taking some product for use shows respect for that product and for its vendor.
Note that it is unimaginable that a company that is able to pay for GNAT Pro
will choose to take it "from a friend". Only those who can't pay and do not
care much about legality of *their* own software products may go this way.

>  ACT are doing an outstanding job.  As a professional
> software developer, I happen to know just how much effort goes into
> testing, verification and certification.

Very good, but so what? GNAT Pro is not a computer game or a song. Note that
Robert Dewar repeated many times that ACT sells *support*, and GNAT Pro is
just supported version of GNAT. And you can't easily take that support from
a friend, because the customer's profile is stated in the support contract.

> I pray everyone on this newsgroup not to listen to Hyman Rosen.

It sounds like you accused him in heresy -:)

>  If you require GNAT Pro, pay.

I think that more proper formulation will be: "If you can easily pay for GNAT
Pro then, perhaps, you need it".




Alexander Kopilovitch                      aek@vib.usr.pu.ru
Saint-Petersburg
Russia

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

tmoran@acm.org writes:

> > > You could always contact friends who are (or work for) paying
> > > customers and get a Pro version from them.
> >
> > This is pirating.  I will never encourage such a disrespectful
>     I thought Gnat was Open Source and was developed with taxpayer funds
> on that basis.  No?

This is true for the "p" releases and for GCC.  Not GNAT Pro.

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Jeff C,]

"Ludovic Brenta" <ludovic.brenta@insalien.org> wrote in message
news:87ekrey757.fsf@insalien.org...
> tmoran@acm.org writes:
>
> > > > You could always contact friends who are (or work for) paying
> > > > customers and get a Pro version from them.
> > >
> > > This is pirating.  I will never encourage such a disrespectful
> >     I thought Gnat was Open Source and was developed with taxpayer funds
> > on that basis.  No?
>
> This is true for the "p" releases and for GCC.  Not GNAT Pro.
>
> -- 
> Ludovic Brenta.

No.... The original poster is somewhat wrong and you are even more wrong.

Originally...many many years back, GNAT was developed partially using a
grant of federal money.

It was a GPL project then.
It is a GPL project now.
The p versions are GPL.
The version in the FSF tree is GPL.
The GNAT Pro versions ARE GPL.

GNAT Pro has now (for many many years) been mostly maintained by ACT using
private money from
selling support services.

GNAT Pro continues to be distributed under the terms of the GPL.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

"Jeff C," <jcreem@yahoo.com> writes:

> GNAT Pro continues to be distributed under the terms of the GPL.

OK, I stand corrected.  Thanks for the info.

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Marin David Condic]

Jano wrote:
> 
> I agree, it works very well. Though if you're planning to use it in 
> windows with tasking or Gnat.Sockets there are some bugs you should 
> avoid before they strike you in the head. Do a search in this group... 
> They've been discussed here previously.

I have encountered bugs with tasking. I believe I've even raised them in 
this forum. I don't know about the sockets problem. Still, I've never 
used a compiler that was perfect and always succeeded in finding a way 
to get the job done anyway. When you get a new compiler release, you may 
or may not get these fixed and I'd bet you'll get some new ones to deal 
with anyway. So I tend to take the attitude that if ACT hasn't made a 
new public release in a while, I could still utilize what they have 
released quite well. I think we get a little too acustomed to the notion 
that if software isn't released quarterly there must be something wrong. 
An old program that works predictably with a reliable set of features is 
often a blessing if you can still use it to get your job done.

MDC
-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jsf.mil/NSFrames.htm

Send Replies To: m   o   d   c @ a   m   o   g
                    c   n   i       c   .   r

     "Face it ladies, its not the dress that makes you look fat.
     Its the FAT that makes you look fat."

         --  Al Bundy

======================================================================

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Marin David Condic]

Well, we might discover that Microsoft actually *agrees* with that 
assessment in the sense that they can see their own customers resisting 
purchasing new versions of software because the old ones still work & 
get the job done fine. Hence, they have to keep coming up with schemes 
to either tempt you or force you to buy the next generation product. 
That's one reason they feel so threatened by things like Linux and Open 
Office - sooner or later those products stabilize and provide 99.9% of 
all the features anyone really needs and so why does someone need to buy 
a new generation of word processor or spreadsheet?

Ada compiler vendors OTOH can likely count on ongoing language changes, 
new computer architectures to target and continual improvement in the 
efficiency and quality of the code they generate to keep demand up for 
new releases of a compiler. But still, I've had projects freeze the 
compiler version in order to achieve stability and were still able to 
get the job done. Old software that works is often "Good Enough".

MDC

Ludovic Brenta wrote:
> 
> Heh, I was refraining from saying just that for fear of looking like
> an emotional Ada bigot and Microsoft basher.  Now there's two of us :)
> 


-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jsf.mil/NSFrames.htm

Send Replies To: m   o   d   c @ a   m   o   g
                    c   n   i       c   .   r

     "Face it ladies, its not the dress that makes you look fat.
     Its the FAT that makes you look fat."

         --  Al Bundy

======================================================================

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

Marin David Condic <nobody@noplace.com> writes:

> Jano wrote:
> > I agree, it works very well. Though if you're planning to use it in
> > windows with tasking or Gnat.Sockets there are some bugs you should
> > avoid before they strike you in the head. Do a search in this
> > group... They've been discussed here previously.
> 
> I have encountered bugs with tasking. I believe I've even raised them
> in this forum. I don't know about the sockets problem. Still, I've
> never used a compiler that was perfect and always succeeded in finding
> a way to get the job done anyway. When you get a new compiler release,
> you may or may not get these fixed and I'd bet you'll get some new
> ones to deal with anyway. So I tend to take the attitude that if ACT
> hasn't made a new public release in a while, I could still utilize
> what they have released quite well. I think we get a little too
> acustomed to the notion that if software isn't released quarterly
> there must be something wrong. An old program that works predictably
> with a reliable set of features is often a blessing if you can still
> use it to get your job done.

I agree wholeheartedly.  People who use free and open-source software
are especially prone to this thinking.  As ESR said: "release early,
release often".  This is good for people who program in their spare
time and cannot or will not test their software thoroughly, and so
want to benefit from the "many eyeballs" paradigm that goes with
open-source.  But with GNAT, this need is reduced by the fact that ACT
has already tested the compiler before releasing it to the public, and
by the fact that with Ada, there are fewer bugs to begin with.

That said, if you find a bug in GNAT 3.15p for which you have a patch
or a workaround, please do send the info to me and I'll include it in
Debian.  The GNAT in Debian already has a couple of such patches
applied.

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Ludovic Brenta]

Marin David Condic <nobody@noplace.com> writes:

> Well, we might discover that Microsoft actually *agrees* with that
> assessment in the sense that they can see their own customers
> resisting purchasing new versions of software because the old ones
> still work & get the job done fine. Hence, they have to keep coming up
> with schemes to either tempt you or force you to buy the next
> generation product. That's one reason they feel so threatened by
> things like Linux and Open Office - sooner or later those products
> stabilize and provide 99.9% of all the features anyone really needs
> and so why does someone need to buy a new generation of word processor
> or spreadsheet?

Yes, this is why they keep changing the file formats for no apparent
reason.  And that is why I systematically complain when people send me
data in an opaque data format such as Microsoft Word or Excel, even
though I can read them with OpenOffice.

The other trick they use is to change the architecture every few
years, so that the development tools sold by their competitors
(Borland, IBM, Watcom) are obsolete.  As for the third-party
applications, they have to be rewritten using Microsoft development
tools.  Microsoft don't mind that they, too, have to rewrite their
apps, because it allows them to sell new versions that are "up to
date" and "integrated with the latest Windows".  They are working on
this right now; the next version of Windows ("Longhorn") will
deprecate the Win32 API, COM, and ActiveX in favour of .Net.  I expect
that new "up-to-date" versions of Microsoft Office, SQL Server,
etc. will follow.

> Ada compiler vendors OTOH can likely count on ongoing language
> changes, new computer architectures to target and continual
> improvement in the efficiency and quality of the code they generate to
> keep demand up for new releases of a compiler. But still, I've had
> projects freeze the compiler version in order to achieve stability and
> were still able to get the job done. Old software that works is often
> "Good Enough".

Yes.  At Barco we use a compiler released in 1999.  It is good enough.

-- 
Ludovic Brenta.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[tmoran]

>Still, I've never used a compiler that was perfect and always succeeded in
>finding a way to get the job done anyway.  When you get a new compiler
>release, you may or may not get these fixed and I'd bet you'll get some
>new ones to deal with anyway.
   When you have code that needs to be portable across multiple compilers,
having to maintain a special "working around" version for one of those
compilers make life significantly harder.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Robert I. Eachus]

tmoran@acm.org wrote:
>>>You could always contact friends who are (or work for) paying
>>>customers and get a Pro version from them.
>>
>>This is pirating.  I will never encourage such a disrespectful
> 
>     I thought Gnat was Open Source and was developed with taxpayer funds
> on that basis.  No?

The situation is much more complex than that.  First, having a 
non-public release of GNAT is not piracy, whether or not you paid for 
it.  Paying ACT customers are encouraged not to redestribute the 
non-public releases in this way, but there is nothing that legally 
prevents them--or ACT from doing so.  In fact, there have been times 
when I have been given access to such versions--by ACT--either to help 
decide whether some behavior was a bug, or to help fix a bug.  In those 
cases, I have disposed of the wavefront (or in one case, I think, 3.12a 
release) once the problem was solved.  I didn't do this for legal or 
moral reasons, but because I wanted any code I publically distributed to 
compile and run on the current GNAT public release.

On the other hand, if you as a GNAT licensee give a copy of the compiler 
to someone, then report their bugs and problems to ACT as your own, that 
is fraud.

Which brings us full-circle to the issue of public releases by ACT. 
First, AFAIK, any contracts involved in the creation of GNAT were 
between the government and New York University.  NYU is, I think, a part 
owner of ACT.  But the public releases of GNAT are intended as a public 
service, and if you check, you should download them from NYU, not from 
ACT.  ACT is involved in deciding when a version of GNAT is stable 
enough to warrant becomming a public release.

If you are familiar with the concept of the Cathedral and the Bazaar, 
for some software the cathedral approach is much more appropriate. 
Public releases of compilers are one such case.  If you want to put 
together several publically available software packages into a project, 
you are much better served if all of the developers of those packages 
used the same compiler.  If each used different versions of GNAT, or 
required different versions of the same libraries, you end up having to 
do a lot more work.

So as far as I am concerned, it will be nice when there is a stable GNAT 
release that uses the new GCC backend.  But I am quite content to use 
3.15p until such a version is available.  (Well, until the issue of Ada 
2005 compatibility starts to become important.  But right now, even the 
2005 in that is just a guess.)


-- 

                                           Robert I. Eachus

"The terrorist enemy holds no territory, defends no population, is 
unconstrained by rules of warfare, and respects no law of morality. Such 
an enemy cannot be deterred, contained, appeased or negotiated with. It 
can only be destroyed--and that, ladies and gentlemen, is the business 
at hand."  -- Dick Cheney

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Jano]

Marin David Condic dice...
> Jano wrote:
> > 
> > I agree, it works very well. Though if you're planning to use it in 
> > windows with tasking or Gnat.Sockets there are some bugs you should 
> > avoid before they strike you in the head. Do a search in this group... 
> > They've been discussed here previously.
> 
> I don't know about the sockets problem. 

It is in the select call. For the maximum read-enabled socket, the call 
uses the max from the write sockets, or vice versa. Something like that.

Oh, and every time you get a stream from a socket you're causing an 
allocation you should explicitly free to avoid a memory leak. This is 
more a detail than a bug, I suppose. Though is something that doesn't 
happen for the streams in Stream_IO

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Marin David Condic]

Sure. I understand that. Too bad that Ada doesn't have some kind of 
conditional compile directive that lets one make a "work around" for 
compiler weaknesses so it would be easier to make code that would work 
on several different compilers. :-)

MDC


tmoran@acm.org wrote:
> 
>    When you have code that needs to be portable across multiple compilers,
> having to maintain a special "working around" version for one of those
> compilers make life significantly harder.


-- 
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jsf.mil/NSFrames.htm

Send Replies To: m   o   d   c @ a   m   o   g
                    c   n   i       c   .   r

     "Face it ladies, its not the dress that makes you look fat.
     Its the FAT that makes you look fat."

         --  Al Bundy

======================================================================

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Hyman Rosen]

Ludovic Brenta wrote:
> Hyman Rosen <hyrosen@mail.com> writes:
>>Ludovic Brenta wrote:
>>>(2) the "GNAT Pro" releases offered only to paying customers
>>You could always contact friends who are (or work for) paying
>>customers and get a Pro version from them.
> This is pirating.  I will never encourage such a disrespectful
> attitude.  ACT are doing an outstanding job.  As a professional
> software developer, I happen to know just how much effort goes into
> testing, verification and certification.  I pray everyone on this
> newsgroup not to listen to Hyman Rosen.  If you require GNAT Pro, pay.

Excuse me, but this is most assuredly *not* "pirating". All versions
of GNAT, including GNAT pro, are released under the GPL. Indeed it
could hardly be otherwise, since it is a derived work of GCC. It is
completely legal, fair, and honest to distribute GNAT Pro to whomever
you like, and ACT would be in breach of the GPL if they acted to
disallow such a thing (not that they would, of course).

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Marius Amado Alves]

> Excuse me, but this is most assuredly *not* "pirating". All versions
> of GNAT, including GNAT pro, are released under the GPL. Indeed it
> could hardly be otherwise, since it is a derived work of GCC. It is
> completely legal, fair, and honest to distribute GNAT Pro to whomever
> you like, and ACT would be in breach of the GPL if they acted to
> disallow such a thing (not that they would, of course).

You may be right, but strictly not for the reason you give. If ACT owns the
copyright to GNAT Pro (which I think they do), they are perfectly entitled
to distribute GNAT Pro under a different license than GPL (which I think
they don't), including one that makes it piracy to give copies to friends.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Hyman Rosen]

Marius Amado Alves wrote:
> You may be right, but strictly not for the reason you give. If ACT owns the
> copyright to GNAT Pro (which I think they do), they are perfectly entitled
> to distribute GNAT Pro under a different license than GPL (which I think
> they don't), including one that makes it piracy to give copies to friends.

No. GNAT is an Ada compiler built upon existing GCC technology,
and is absolutely a derived work of pre-existing GPLed software.

Re: Is 3.15p -still- the latest GNAT 'p' release?

[Jacob Sparre Andersen]

Ludovic Brenta wrote:
> Hyman Rosen wrote:

> > You could always contact friends who are (or work for) paying
> > customers and get a Pro version from them.
> 
> This is pirating.

How could it possibly be that?  Also the "Pro" version of GNAT should
be distributed under the GNU General Public License, which allows
people to redistribute the binaries as long as they also redistribute
the source.

Jacob
-- 
"It ain't rocket science!"