Ada not mentioned in GCC 4.3 release notes

Ada not mentioned in GCC 4.3 release notes

[Jerry]

I'm just wondering why "GCC 4.3 Release Series Changes, New Features,
and Fixes" at http://gcc.gnu.org/gcc-4.3/changes.html not mention Ada
anywhere? There is a section of changes to specific languages but not
Ada. I thought that 4.3 was the first gcc to handle Ada 2005, surely
worth mentioning.

Jerry

Re: Ada not mentioned in GCC 4.3 release notes

[anon]

Actual, any GCC after GCC-4.00 had Ada 2005.  

The problem may be that Adacore has not update the GPL license from 2 
to 3 and all GNU GCC type packages are to be under the GPL 3 license.

In <9cd86244-11e5-45a1-b77b-1fb13e99b9ee@s12g2000prg.googlegroups.com>, Jerry <lanceboyle@qwest.net> writes:
>I'm just wondering why "GCC 4.3 Release Series Changes, New Features,
>and Fixes" at http://gcc.gnu.org/gcc-4.3/changes.html not mention Ada
>anywhere? There is a section of changes to specific languages but not
>Ada. I thought that 4.3 was the first gcc to handle Ada 2005, surely
>worth mentioning.
>
>Jerry
>

Re: Ada not mentioned in GCC 4.3 release notes

[Tero Koskinen]

On Tue, 11 Mar 2008 06:35:21 GMT anon wrote:

> Actual, any GCC after GCC-4.00 had Ada 2005.  
> 
> The problem may be that Adacore has not update the GPL license from 2 
> to 3 and all GNU GCC type packages are to be under the GPL 3 license.

Ada components of GNU GCC are under GPLv3 (+linking exception):
http://gcc.gnu.org/viewvc?view=rev&revision=128332

> In <9cd86244-11e5-45a1-b77b-1fb13e99b9ee@s12g2000prg.googlegroups.com>, Jerry <lanceboyle@qwest.net> writes:
> >I'm just wondering why "GCC 4.3 Release Series Changes, New Features,
> >and Fixes" at http://gcc.gnu.org/gcc-4.3/changes.html not mention Ada
> >anywhere? There is a section of changes to specific languages but not
> >Ada. I thought that 4.3 was the first gcc to handle Ada 2005, surely
> >worth mentioning.
> >
> >Jerry

Ada maintainers probably either forgot to update the changes or they
are just not that interested in updating them. 4.2 changes was also
missing Ada section. 4.0 and 4.1 had some notes.

-- 
Tero Koskinen - http://iki.fi/tero.koskinen/

Re: Ada not mentioned in GCC 4.3 release notes

[Ludovic Brenta]

AdaCore ship release notes and lists of known problems in each release
of GNAT Pro or GNAT GPL (look in ${prefix}/share/doc/gnat).
Unfortunately, they omit to synchronise these release notes with GCC's
announcement page. This reduces the visibility of Ada to the casual
user and reinforces the impression that Ada is a second-class citizen
in GCC. Too bad.

If AdaCore would only put these files in the public source code
repository, outside contributors would be able to produce release
notes without having to read the entire detailed ChangeLog. Currently,
this is too much work so nobody stepped forward to do it.

--
Ludovic Brenta.

Re: Ada not mentioned in GCC 4.3 release notes

[Britt Snodgrass]

On Mar 11, 4:28 am, Ludovic Brenta <ludo...@ludovic-brenta.org> wrote:
> AdaCore ship release notes and lists of known problems in each release
> of GNAT Pro or GNAT GPL (look in ${prefix}/share/doc/gnat).
> Unfortunately, they omit to synchronise these release notes with GCC's
> announcement page. This reduces the visibility of Ada to the casual
> user and reinforces the impression that Ada is a second-class citizen
> in GCC. Too bad.
>

Yes, I agree that its looks bad.  I've raised the issue on the gcc
list before but nothing has happened.

However, I have GNAT Pro 6.1.1 (the latest release) at work and it
reports that it is using GCC 4.1.3, not 4.3, so maybe that's part of
the explanation.

- Britt

Re: Ada not mentioned in GCC 4.3 release notes

[Ludovic Brenta]

Britt Snodgrass wrote:
> On Mar 11, 4:28�am, Ludovic Brenta <ludo...@ludovic-brenta.org> wrote:
> > AdaCore ship release notes and lists of known problems in each release
> > of GNAT Pro or GNAT GPL (look in ${prefix}/share/doc/gnat).
> > Unfortunately, they omit to synchronise these release notes with GCC's
> > announcement page. This reduces the visibility of Ada to the casual
> > user and reinforces the impression that Ada is a second-class citizen
> > in GCC. Too bad.
> >
>
> Yes, I agree that its looks bad.  I've raised the issue on the gcc
> list before but nothing has happened.

Indeed. AdaCore spends their time supporting their paying customers,
not the "enthusiasts" looking at the public sources of GCC.

> However, I have GNAT Pro 6.1.1 (the latest release) at work and it
> reports that it is using GCC 4.1.3, not 4.3, so maybe that's part of
> the explanation.

Not really. The front-end and back-end have different life cycles.
Almost all of the changes in AdaCore's release notes are in the front-
end only. The front-end in GCC 4.3 is roughly equivalent to that of
GNAT GPL 2007 (GNAT Pro 6.1.0?) so the release notes for those
versions should apply, up to a point, to GCC 4.3. The date at which
each feature was implemented gives a clue as to which features are in
GCC 4.3. It should even be possible to correlate each item in the
release notes with a commit in the public GCC repository but that's
obviously a lot of work.

--
Ludovic Brenta.

Re: Ada not mentioned in GCC 4.3 release notes

[anon]

Download the MASTER version at Adacore.com!  It is still under Adacore 
modified GPL version 2. Also any changes that is may in GNAT Ada will 
be at Adacore web first, it may take 6 months before others become 
aware of those changes. And that because the chages will first appear 
in GNAT PRO first, which GNU has no control over, then Adacore will 
release the changes to the GPL version on their website. After which 
the GNU and other can obtain the changes at Adacore.

The copies that are at SF may or may not be complete!  Some of 
the features of Ada do not work between Archs as well as all 
Archs version are not archived. 

Plus, if you look at GNU website, you will see that they do not 
use SF either, they use their own web site archive. Which may or 
may not match the SF versions. And some do not match Adacore 
either which is a big problem with GNAT Ada.




In <20080311093324.7c3a0447.tero.koskinen@iki.fi>, Tero Koskinen <tero.koskinen@iki.fi> writes:
>On Tue, 11 Mar 2008 06:35:21 GMT anon wrote:
>
>> Actual, any GCC after GCC-4.00 had Ada 2005.  
>> 
>> The problem may be that Adacore has not update the GPL license from 2 
>> to 3 and all GNU GCC type packages are to be under the GPL 3 license.
>
>Ada components of GNU GCC are under GPLv3 (+linking exception):
>http://gcc.gnu.org/viewvc?view=rev&revision=128332
>
>> In <9cd86244-11e5-45a1-b77b-1fb13e99b9ee@s12g2000prg.googlegroups.com>, Jerry <lanceboyle@qwest.net> writes:
>> >I'm just wondering why "GCC 4.3 Release Series Changes, New Features,
>> >and Fixes" at http://gcc.gnu.org/gcc-4.3/changes.html not mention Ada
>> >anywhere? There is a section of changes to specific languages but not
>> >Ada. I thought that 4.3 was the first gcc to handle Ada 2005, surely
>> >worth mentioning.
>> >
>> >Jerry
>
>Ada maintainers probably either forgot to update the changes or they
>are just not that interested in updating them. 4.2 changes was also
>missing Ada section. 4.0 and 4.1 had some notes.
>
>-- 
>Tero Koskinen - http://iki.fi/tero.koskinen/

Re: Ada not mentioned in GCC 4.3 release notes

[Ivan Levashew]

Britt Snodgrass пишет:
> Yes, I agree that its looks bad.  I've raised the issue on the gcc
> list before but nothing has happened.


I think there are better ways to raise popularity of Ada. Those who are 
heading FOSS development seem to be completely unaware of Ada. For 
example, A. S. Tanenbaum in his "Modern Operating Systems", chapter 2 
"Processes and threads" used Pidgin Pascal he just invented to 
demonstrate Hoare monitors:

monitor ProducerConsumer
         condition full, empty;
         integer count;

         procedure insert(item: integer);
         begin
                 if count = N then wait(full);
                 insert_item(item);
                 count := count+1;
                 if count = 1 then signal(empty)
         end;
         function remove: integer;
         begin
                 if count = 0 then wait(empty);
                 remove = remove_item;
                 count := count+1;
                 if count = 1 then signal(full)
         end;
         count := 0;
end monitor;

procedure producer;
begin
         while true do
         begin
                 item = produce_item;
                 ProducerConsumer.insert(item)
         end
end;

procedure consumer;
begin
         while true do
         begin
                 item = ProducerConsumer.remove;
                 consume_item(item)
         end
end;

It's not about preferences, it's what complete unawareness is! How many 
people are reinventing Ada?

Tanenbaum's Minix was foundation to Linux, and Tanenbaum's book is very 
popular in systems programming community.

Systems programmers are completely unaware of Ada.
Systems programmers read Tanenbaum's book.
Systems programmers remain completely unaware of Ada.

Another point:

Ada bindings are aliens in the FOSS world. Consider 2 examples:
1. gettext
gettext comes with Java, Lisp, *C#* bindings, but no Ada. If you need 
gettext in Ada, you must use separate package GtkAda (BTW, is there a 
standalone version of gettext/Ada?)

Note C# here.

http://www.roughlydrafted.com/RD/RDM.Tech.Q2.07/878F362F-2BF5-4C86-84E7-9C976F7BDDD4.html

----------quote------------
Does this mean that Microsoft is now aligned with open source developers 
and working to push open, interoperable implementations of its software? 
Is the old triangle of contention between Microsoft, Linux and Apple 
dissolving into a free and open love circle?

Ha Ha, No.
Microsoft is not trying to usher in a new OpenStep with .Net. It is 
working to usher in a new Win32: another decade of dependance upon 
Microsoft software that can only work on Windows. Why the subterfuge on 
submitting portions of .Net to standards bodies? Three guesses, and the 
first two don't count!

The best way to keep opponents busy is to give them false directions 
that lead into traps. This will distract them from blazing their own 
successful, competing trail, and will lead them directly into 
containment with the least mess and inconvenience.

Microsoft is leading Mono users and developers into a pleasant feeling 
trap. Along the way, they gain appreciation for Microsoft's development 
tools as they struggle to make their own open source copies. They will 
grow increasingly familiar with Microsoft's directions, up to the point 
where they are hopelessly brainwashed into thinking that Microsoft is 
leading technology into a paradise of openness.
----------quote------------

That's why C# must be moved off from gettext to the Gtk#, and Ada 
binding be moved from GtkAda to the gettext instead.

2. SWIG
SWIG support many languages out-of-box, but guess what? GNAT-SWIG is 
still a separate project. It would be natural to aquire GNAT-SWIG into 
the SWIG. Even if it's not tested as good as another bindings. Mark Ada 
support "experimental" and bundle GNAT-SWIG with SWIG.

There must be Ada FOSS wins other than just GCC.

Re: Ada not mentioned in GCC 4.3 release notes

[Georg Bauhaus]

On Thu, 2008-03-13 at 22:29 +0600, Ivan Levashew wrote:
> Britt Snodgrass пишет:
> > Yes, I agree that its looks bad.  I've raised the issue on the gcc
> > list before but nothing has happened.
> 
> 
> I think there are better ways to raise popularity of Ada. Those who are 
> heading FOSS development seem to be completely unaware of Ada. For 
> example, A. S. Tanenbaum in his "Modern Operating Systems", chapter 2 
> "Processes and threads" used Pidgin Pascal he just invented to 
> demonstrate Hoare monitors:
> 
> monitor ProducerConsumer
>          condition full, empty;
...
> It's not about preferences, it's what complete unawareness is! How many 
> people are reinventing Ada?

Without reading the red-top press of computer society 
I can't tell who publicly announces awareness of what or
whom for what reasons etc..
 But for one thing, is seems a certain truth that Tanenbaum has
heard of Ada. (And I hate to judge the ears of other people.)
I think it is one of the conclusions one may draw from an
article comparing concurrency features of the Orca
programming language and the then new protected types of Ada.

Even Ada instructors seem to have reason to use a language besides
Ada. I see that in the Monitors section (3.9) of Burns/Wellings.
They show a non-Ada source including the keyword "monitor"
and the type "CONDITION_VARIABLE". Compare this to the example
you found ;-)