Ada serial Communication packages

Ada serial Communication packages

[AdaNewbie]

Greetings,

I'm keen to view an implementation of a sophisticated(timeout support
etc) rs-232(Serial Comms) package, if availiable in the public domain.
Could anyone  point me in the correct direction, please.


Regards

Newbie

Re: Ada serial Communication packages

[Stephen Leake]

robin-hall@utvinternet.com (AdaNewbie) writes:

> Greetings,
> 
> I'm keen to view an implementation of a sophisticated(timeout support
> etc) rs-232(Serial Comms) package, 

My package does time outs.

http://www.toadmail.com/~ada_wizard/

See the "com ports" link.

> if availiable in the public domain. 

As a newbie, one of the things you need to learn about is licensing
terms. "Public domain" is defined in the United States to mean "no
copyright owner". That means anyone can take it and claim it as their
own.

There are other licenses that allow you to use the software, but
retain control for the owner. One of the most common used with Ada
code is the GNAT Modified Gnu Public License (GMGPL). It allows you to
use the code in Open Source or proprietary projects. My com ports
package should be released under that license, but I wrote it before I
learned about this stuff myself :). I guess it's time to update it.

-- 
-- Stephe

Re: Ada serial Communication packages

[AdaNewbie]

Stephen,

Thanks very much for the reply and the low down on licensing issues.
Your code is very interesting but based on an OS. I had hope to see a
pure Ada implementation,any ideas?

Best regards

Rob



Stephen Leake <Stephe.Leake@nasa.gov> wrote in message news:<ud6iazq7g.fsf@nasa.gov>...
> robin-hall@utvinternet.com (AdaNewbie) writes:
> 
> > Greetings,
> > 
> > I'm keen to view an implementation of a sophisticated(timeout support
> > etc) rs-232(Serial Comms) package, 
> 
> My package does time outs.
> 
> http://www.toadmail.com/~ada_wizard/
> 
> See the "com ports" link.
> 
> > if availiable in the public domain. 
> 
> As a newbie, one of the things you need to learn about is licensing
> terms. "Public domain" is defined in the United States to mean "no
> copyright owner". That means anyone can take it and claim it as their
> own.
> 
> There are other licenses that allow you to use the software, but
> retain control for the owner. One of the most common used with Ada
> code is the GNAT Modified Gnu Public License (GMGPL). It allows you to
> use the code in Open Source or proprietary projects. My com ports
> package should be released under that license, but I wrote it before I
> learned about this stuff myself :). I guess it's time to update it.

Re: Ada serial Communication packages

[Stephen Leake]

robin-hall@utvinternet.com (AdaNewbie) writes:

> Stephen,
> 
> Thanks very much for the reply and the low down on licensing issues.
> Your code is very interesting but based on an OS. I had hope to see a
> pure Ada implementation,any ideas?

Since the phrase "Serial port" is not in the Ada Reference Manual, it
is by definition not Pure Ada, so I don't understand what you want :).

To be less pedantic:

Serial ports are hardware; on most computers, the only way to talk to
hardware is thru an operating system.  

Perhaps there is some standard that defines an operating system API
for serial ports. If that is true, then an Ada binding for that
standard would be useful, and might be what you are looking for. There
is an Ada binding to Posix; I don't know if it covers serial ports.

It would also be possible to treat my Com_Ports Ada spec as a
standard, and provide a different body for a different operating
system. So far, I haven't had to do that.

On a hard-core embedded system, with no operating system, a package
that dealt with serial ports would do direct port IO. That could be
done with machine-code insertions in Ada; it would be CPU and serial
port controller chip specific (which is part of the reason operating
systems were invented :).

Hope this helps,

-- 
-- Stephe

Re: Ada serial Communication packages

[Martin Dowie]

"Stephen Leake" <Stephe.Leake@nasa.gov> wrote in message
news:uznl4mlpm.fsf@nasa.gov...
> On a hard-core embedded system, with no operating system, a package
> that dealt with serial ports would do direct port IO. That could be
> done with machine-code insertions in Ada; it would be CPU and serial

It can be done in Ada too! There's not much you can't do in Ada with
address clauses and modular types :-)

Re: Ada serial Communication packages

[David C. Hoos]

"Stephen Leake" <Stephe.Leake@nasa.gov> wrote in message
news:uznl4mlpm.fsf@nasa.gov...
> robin-hall@utvinternet.com (AdaNewbie) writes:
>
> > Stephen,
> >
> > Thanks very much for the reply and the low down on licensing issues.
> > Your code is very interesting but based on an OS. I had hope to see a
> > pure Ada implementation,any ideas?
>
> Since the phrase "Serial port" is not in the Ada Reference Manual, it
> is by definition not Pure Ada, so I don't understand what you want :).
>
> To be less pedantic:
>
> Serial ports are hardware; on most computers, the only way to talk to
> hardware is thru an operating system.
>
> Perhaps there is some standard that defines an operating system API
> for serial ports. If that is true, then an Ada binding for that
> standard would be useful, and might be what you are looking for. There
> is an Ada binding to Posix; I don't know if it covers serial ports.

FWIW, the Posix/Ada binding does cover Serial Ports, but it's limited
to 38,400 baud, so hasn't kept up with reality.
Also, it's been several years since I wrote serial port code using the
Posix/Ada bindings, but if I remember correctly, there was some problem
with setting port modes -- e.g., Canoinical, Raw, Cooked, etc.

>
> It would also be possible to treat my Com_Ports Ada spec as a
> standard, and provide a different body for a different operating
> system. So far, I haven't had to do that.
>
> On a hard-core embedded system, with no operating system, a package
> that dealt with serial ports would do direct port IO. That could be
> done with machine-code insertions in Ada; it would be CPU and serial
> port controller chip specific (which is part of the reason operating
> systems were invented :).
>
> Hope this helps,
>
> -- 
> -- Stephe
> _______________________________________________
> comp.lang.ada mailing list
> comp.lang.ada@ada.eu.org
> http://ada.eu.org/mailman/listinfo/comp.lang.ada
>