Trouble with UNIX IPC and Ada95

Trouble with UNIX IPC and Ada95

[berserker]

Hi,

I was wondering if anyone could give me some advice regarding programming
for UNIX IPC with Ada95. I don't know any equivalent Ada packages to C
functions such as 'pipe', 'close', 'popen'... etc. Does anyone have any
experience in this area, or have a URL which they could direct me to?

Thanks,

Luke B. Andrews.

Re: Trouble with UNIX IPC and Ada95

[Marc A. Criley]

Look at AdaSockets, available at http://www.infres.enst.fr/ANC/

berserker wrote:
>Hi,
>
>I was wondering if anyone could give me some advice regarding programming
>for UNIX IPC with Ada95. I don't know any equivalent Ada packages to C
>functions such as 'pipe', 'close', 'popen'... etc. Does anyone have any
>experience in this area, or have a URL which they could direct me to?
>
>Thanks,
>
>Luke B. Andrews.

Re: Trouble with UNIX IPC and Ada95

[wv12]

What is wrong with doing system programming in C on UNIX?
UNIX is written in C for crying out loud. If you don't trust
C, then don't use UNIX.
Wil
In article <euar5.4459$cr3.154535@ozemail.com.au>,
  "berserker" <ber@ser.ker> wrote:
> Hi,
>
> I was wondering if anyone could give me some advice regarding
programming
> for UNIX IPC with Ada95. I don't know any equivalent Ada packages to C
> functions such as 'pipe', 'close', 'popen'... etc. Does anyone have
any
> experience in this area, or have a URL which they could direct me to?
>
> Thanks,
>
> Luke B. Andrews.
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Trouble with UNIX IPC and Ada95

[Charles Hixson]

[-- Attachment #1: Type: text/plain, Size: 1876 bytes --]

"trust C"?  That's not exactly the question.  If the program is properly
designed in nearly any programming language, then it will work properly.
The problem is that it's more difficult to properly design a program in C
than in some other languages.  Since this is an Ada group, I'll mention
Eiffel and Java as other examples, with side thoughts for Smalltalk and
Haskell (well, I don't know much about Haskell, but what I've heard has been
pretty good).  (N.B.:  I'm not claiming that all of these languages are
suitable for system programming.  Merely that they reduce the work of
eliminating bugs.)

The main problem is pointers.  I suppose that the other problems (the need
to type cast frequently, etc.  The lack of reasonable type checking, etc.)
also make a big contribution, but the main problem is pointers and pointer
arithmetic.  This is made necessary by the lack of ADTs (struct-s just don't
make the cut), and by the lack of run-time allocation of array boundaries.

OTOH, gtk-- shows what can be done in C.  It's just much more work to do
things properly.

wv12@my-deja.com wrote:

> What is wrong with doing system programming in C on UNIX?
> UNIX is written in C for crying out loud. If you don't trust
> C, then don't use UNIX.
> Wil
> In article <euar5.4459$cr3.154535@ozemail.com.au>,
>   "berserker" <ber@ser.ker> wrote:
> > Hi,
> >
> > I was wondering if anyone could give me some advice regarding
> programming
> > for UNIX IPC with Ada95. I don't know any equivalent Ada packages to C
> > functions such as 'pipe', 'close', 'popen'... etc. Does anyone have
> any
> > experience in this area, or have a URL which they could direct me to?
> >
> > Thanks,
> >
> > Luke B. Andrews.
> >
> >
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.

-- (c) Charles Hixson
--  Addition of advertisements or hyperlinks to products specifically
prohibited


[-- Attachment #2: Card for Charles Hixson --]
[-- Type: text/x-vcard, Size: 145 bytes --]

begin:vcard 
n:Hixson;Charles
x-mozilla-html:FALSE
adr:;;;;;;
version:2.1
email;internet:charleshixson@earthling.net
fn:Charles Hixson
end:vcard

Re: Trouble with UNIX IPC and Ada95

[wv12]

>
> "trust C"?  That's not exactly the question.  If the program is
properly
> designed in nearly any programming language, then it will work
properly.
> The problem is that it's more difficult to properly design a program
in C
> than in some other languages.
Apparently this is not the case. Twenty years after IBM launches the
IBM PC, you see a myriads of operating systems written in C: MINIX,
XINU, Linux, XENIX, FreeBSD, etc... I haven't seen anything in Ada
on any PC that provides half the functionality of DOS. Maybe there
is a language that makes OS design become a walk in the park. That
language is not Ada.



Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Trouble with UNIX IPC and Ada95

[David Starner]

On Thu, 31 Aug 2000 17:50:57 GMT, wv12@my-deja.com <wv12@my-deja.com> wrote:
>Apparently this is not the case. Twenty years after IBM launches the
>IBM PC, you see a myriads of operating systems written in C: MINIX,
>XINU, Linux, XENIX, FreeBSD, etc... I haven't seen anything in Ada
>on any PC that provides half the functionality of DOS. Maybe there
>is a language that makes OS design become a walk in the park. That
>language is not Ada.

Your logic is wrong: No major OS for the PC is written in Ada does
not imply anything about Ada's use for writting OS's. It just means
that Linus Torvald, UC-Berkeley and Microsoft chose to write their
operating systems in C and C++. Two main reasons are probably cultural
(Unix is traditionally written in C, and C was more popular) and 
resources (a free OS needs a free compiler, but GNAT didn't appear
until '94-'95).

Of course Ada doesn't make OS design a walk in the park. No language
that exists does, and any language that did would be like Lex -
a very narrow, specialized tool. OS design is naturally a highly
complex job, not a walk in the park.

-- 
David Starner - dstarner98@aasaa.ofe.org
http/ftp: dvdeug.net.dhis.org
It was starting to rain on the night that they cried forever,
It was blinding with snow on the night that they screamed goodbye.
	- Dio, "Rock and Roll Children"

Re: Trouble with UNIX IPC and Ada95

[mjsilva]

In article <8om5tn$q0$1@nnrp1.deja.com>,
  wv12@my-deja.com wrote:
>
>
> >
> > "trust C"?  That's not exactly the question.  If the program is
> properly
> > designed in nearly any programming language, then it will work
> properly.
> > The problem is that it's more difficult to properly design a program
> in C
> > than in some other languages.
> Apparently this is not the case. Twenty years after IBM launches the
> IBM PC, you see a myriads of operating systems written in C: MINIX,
> XINU, Linux, XENIX, FreeBSD, etc... I haven't seen anything in Ada
> on any PC that provides half the functionality of DOS. Maybe there
> is a language that makes OS design become a walk in the park. That
> language is not Ada.

You have quite a bee in your bonnet regarding Ada, haven't you?  Your
reply totally avoids responding to the assertion.  If you claim the
opposite, that it is *no* more difficult to properly design a program
in C than in *any* other language, then you've staked out a very lonely
and unsupported position.  Are you aware that "When Bell Labs were
invited to evaluate C against the DoD requirements, they said that
there was no chance of C meeting the requirements of readability,
safety, etc., for which we were striving, and that it should not even
be on the list of evaluated languages"? (from "Ada - The Project, The
DoD High Order Language Working Group")

Maybe a few years down the road you'll finally get tired of chasing
down the C bugs that other languages (including Ada) would have caught,
and stop looking at C through rose-colored glasses.

Mike


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Trouble with UNIX IPC and Ada95

[Larry Kilgallen]

In article <8om5tn$q0$1@nnrp1.deja.com>, wv12@my-deja.com writes:
> 
> 
>>
>> "trust C"?  That's not exactly the question.  If the program is
> properly
>> designed in nearly any programming language, then it will work
> properly.
>> The problem is that it's more difficult to properly design a program
> in C
>> than in some other languages.
> Apparently this is not the case. Twenty years after IBM launches the
> IBM PC, you see a myriads of operating systems written in C: MINIX,
> XINU, Linux, XENIX, FreeBSD, etc... I haven't seen anything in Ada
> on any PC that provides half the functionality of DOS. Maybe there
> is a language that makes OS design become a walk in the park. That
> language is not Ada.

I think you will find any Ada work concentrated in embedded systems.

Do you have any evidence the language choice for those other operating
systems were based on features of the language ?  It seems to me more
likely based on habit.

Re: Trouble with UNIX IPC and Ada95

[Tarjei T. Jensen]

wv12@my-deja.com wrote
>Apparently this is not the case. Twenty years after IBM launches the
>IBM PC, you see a myriads of operating systems written in C: MINIX,
>XINU, Linux, XENIX, FreeBSD, etc... I haven't seen anything in Ada
>on any PC that provides half the functionality of DOS. Maybe there
>is a language that makes OS design become a walk in the park. That
>language is not Ada.

Operating systems are written in other languages than C. OS/400 is supposed to
be written in Modula-2. VMS is mostly written in bliss. Other languages are
also used including C.


Originally plan9 should have been written in modula-2, but Rob Pike decided for
an extended C since Ken Thompson was good at writing C compilers. I think he
made the wrong trade off. Reducing the work load for Ken Thomson is silly if
you can reduce total time used for the entire project.

I believe that the popularity of C++ and java reflects the problems real world
people have with C. The doctrine says that C++ must be easier than C. Reality
is that C++ adds complexity and time. Java is a time saver compared to C++ and
possibly C and its popularity reflects that people just do not hack it with C
and C++.

BTW. Programming in C is fun. It has a absolutely huge tinkering apeal. Because
it is fun one tend not to notice that time flies. As a professional I cannot
let "fun" decide engineering choices when that choice will incure a large
time/cost penalty. If I choose to use C, I do it knowing the problems that
follows that choice.

The reality of life is that both the C and Unix community is plagued with bad
engineering decisions. We all pay for it.

Greetings,

Re: Trouble with UNIX IPC and Ada95

[gdemont]

> > "trust C"?  That's not exactly the question.  If the program is
> properly
> > designed in nearly any programming language, then it will work
> properly.
> > The problem is that it's more difficult to properly design a program
> in C
> > than in some other languages.

wv12@my-deja.com:

> Apparently this is not the case. Twenty years after IBM launches the
> IBM PC, you see a myriads of operating systems written in C: MINIX,
> XINU, Linux, XENIX, FreeBSD, etc... I haven't seen anything in Ada
> on any PC that provides half the functionality of DOS. Maybe there
> is a language that makes OS design become a walk in the park. That
> language is not Ada.

It is not C either: "twenty years after IBM launches the IBM PC",
there are still bugs from the first DOS at in Windows 9x shell...
The OS business is not a good example: the vendors can have a large
number of programmers, and still more testers; the buyers accept to
buy the OS without warranty and no real support, survive with bugs
for years and even accept to re-buy updates with the same and more
bugs!

G-)


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Trouble with UNIX IPC and Ada95

[Marin David Condic]

Nick Roberts wrote:

> There is, at least, a project to build such an OS ;-)
>
> http://www.AdaOS.org
>

Is this project making any headway? Is there anything that might be at a
"release" stage (perhaps the kernel?) that might be reviewed/used? While
I'm aware that this is a "labor of love" and y'all have "real jobs" to
attend to, I was hoping to see something concrete. Any expected time
frame in which there might be something to look over?

Just asking. I don't mean to be a nudzh. :-)

MDC
--
======================================================================
Marin David Condic - Quadrus Corporation - http://www.quadruscorp.com/
Send Replies To: m c o n d i c @ q u a d r u s c o r p . c o m
Visit my web site at:  http://www.mcondic.com/

    "Take away the punchbowl just when the party gets going"

        --  William McChesney Martin, Former Fed chairman, explaining
            what a sound central bank must always do.
======================================================================

AdaOS (was: Trouble with UNIX IPC and Ada95)

[Nick Roberts]

It's making headway, but it's doing the breast stroke rather than the front
crawl at the moment. We are still way off producing any kind of working
code, and this will not appear for a long, long time yet (sorry). However,
some interesting things are (hopefully) not too far off, e.g. a detailed
design for the microkernel.

--
Nick Roberts
http://www.AdaOS.org

"Marin David Condic" <mcondic.nospam@acm.org> wrote in message
news:39B0FF69.B3AB90A1@acm.org...
> Nick Roberts wrote:
>
> > There is, at least, a project to build such an OS ;-)
>
> Is this project making any headway? Is there anything that might be at a
> "release" stage (perhaps the kernel?) that might be reviewed/used? While
> I'm aware that this is a "labor of love" and y'all have "real jobs" to
> attend to, I was hoping to see something concrete. Any expected time
> frame in which there might be something to look over?
>
> Just asking. I don't mean to be a nudzh. :-)

Re: Trouble with UNIX IPC and Ada95

[John McCabe]

Charles Hixson <charleshixsn@earthlink.net> wrote:

>"trust C"?  That's not exactly the question. 

>The main problem is pointers.

Refer to the package Interfaces.C.Pointers.


Best Regards
John McCabe <john@assen.demon.co.uk>

Re: Trouble with UNIX IPC and Ada95

[David Starner]

On Thu, 31 Aug 2000 14:36:32 GMT, wv12@my-deja.com <wv12@my-deja.com> wrote:
>What is wrong with doing system programming in C on UNIX?
>UNIX is written in C for crying out loud. If you don't trust
>C, then don't use UNIX.

Obviously C isn't his language of choice. What is wrong with
perferring one language over another?

-- 
David Starner - dstarner98@aasaa.ofe.org
http/ftp: dvdeug.net.dhis.org
It was starting to rain on the night that they cried forever,
It was blinding with snow on the night that they screamed goodbye.
	- Dio, "Rock and Roll Children"

Re: Trouble with UNIX IPC and Ada95

[Robert Dewar]

In article <8olqhe$hsv$1@nnrp1.deja.com>,
  wv12@my-deja.com wrote:

> What is wrong with doing system programming in C on UNIX?
> UNIX is written in C for crying out loud. If you don't trust
> C, then don't use UNIX.

It is hard to imagine a more transparent troll on comp.lang.ada,
but it still seems to be working pretty well and dragging in
expected replies. I expect wv12 is getting some amusement
from this thread :-)


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Trouble with UNIX IPC and Ada95

[Luke B. Andrews]

> What is wrong with doing system programming in C on UNIX?
> UNIX is written in C for crying out loud. If you don't trust
> C, then don't use UNIX.

Perhaps I didn't explain myself clearly enough. My final year CS project
*MUST* be written in Ada95 on a Sun Solaris system running UNIX. Now you may
see my dilemma :)

P.S. Thanks Marc for the URL, much appreictaed.

Re: Trouble with UNIX IPC and Ada95

[Anders Gidenstam]

In article <euar5.4459$cr3.154535@ozemail.com.au>,
	"berserker" <ber@ser.ker> writes:
>Hi,
>
>I was wondering if anyone could give me some advice regarding programming
>for UNIX IPC with Ada95. I don't know any equivalent Ada packages to C
>functions such as 'pipe', 'close', 'popen'... etc. Does anyone have any
>experience in this area, or have a URL which they could direct me to?

Hi,

Have a look at Florist, a POSIX Ada Binding, it should work on solaris.
I got the precompiled Linux version at http://www.gnuada.org/alt.html, and
I think they have a link back to the Florist development site.

Regards

Anders
-- 
"A well-written program is its own heaven; 
 a poorly-written program is its own hell."
  - The Tao of Programming