Experiences/history of Aetech Ada compiler?

Experiences/history of Aetech Ada compiler?

[Dale Stanbrough]

Our department has the possibility of supplying students with 
the Aetech Ada compiler at a fairly low price (in conjuction with a book).

I was wondering if anyone has used this compiler and could tell me
of any weaknesses it may have.

So far it seems to handle things such as generics and tasks ok, but
it does not implement representation clauses at all.

Dale
-------------------------------------------------------------
Dale Stanbrough, RMIT, Melbourne, Australia, dale@rmit.edu.au
GNU Ada 94 (GNAT) => the best $0 you'll ever spend.
Available for DOS, Linux, OS/2, Sun Sparc, Sun Solaris, ...
Coming to a GNU supported platform near you soon...

Re: Experiences/history of Aetech Ada compiler? (more details!)

[Dale Stanbrough]

In article <358gnp$1lm@goanna.cs.rmit.oz.au> Dale Stanbrough,
dale@goanna.cs.rmit.edu.au writes:
>Our department has the possibility of supplying students with 
>the Aetech Ada compiler at a fairly low price (in conjuction with a book).

Whoops! Further details are...
IntegrAda, version 4.3.0, release 2, targetted for Dos computers.


Dale

Re: Experiences/history of Aetech Ada compiler?

[Jeff Creem]

I have used it a little (and people I work with have used it a lot).
There is no debugger, all tasks run at one priority, pragma inline
does not work. There is no package machine cod. If these things
don't matter then its not a bad little package. The code it
generates is not terrible (although there is no way to generate
annotated assembly listings)

Jeff Creem
member Team-Ada

Re: Experiences/history of Aetech Ada compiler?

[Robert C. Lokerson]

In Article <358gnp$1lm@goanna.cs.rmit.oz.au> "Dale Stanbrough <dale@goanna.cs.rmit.edu.au>" says:
> Our department has the possibility of supplying students with 
> the Aetech Ada compiler at a fairly low price (in conjuction with a book).
> 
> I was wondering if anyone has used this compiler and could tell me
> of any weaknesses it may have.
> 
> So far it seems to handle things such as generics and tasks ok, but
> it does not implement representation clauses at all.
> 
 
I got their "LearnAda" product about 15 months ago.  It is the only ADA
compiler I have any (real) experience with.  I learned a lot trying to use
it.  I wrote a few useful tools for other groups around here using it.  Its
been 7 months since I was able to justify using in in my work.
 
My personal style of programming has evolved into one which uses a lot of
arrays of pointers to constant strings of varying lengths.  I have not been
able to get this compiler to properly construct them unless they are created
at run time with the "access" type scheme.  This might be ok (and in the ADA
spirit) but I wish I could have "all" the initializations happen at
elaboration-time.  (If thay makes any sense... I don't really consider myself
an ADA expert... Much as I would like to be)
 
The editor that comes with LearnAda has a limited source file size, but one
can get around that by using another (less-helpful) editor when one
encounters that limit.  I have not found a meaningful limit to the Compiler
Source size.  (Hint... After getting around the size limit with another
editor... try going back to using the context-sensitive LearnAda Editor.)
  
 
 


  R.C. LOKERSON (BOB)      MAIL ADDRESS:      VOICE : 407.729.2257
  HARRIS MDSO/MIS/DCC/ATG     HARRIS E.S.S.   SMTP  : RLOKER01@HARRIS.COM
  BLDG 15/RM 838              P.O. BOX 37     CCMAIL: RLOKER01
  PALM BAY, FLORIDA           MELBOURNE FL    FCC   : WA3PKX
                              32902           HOME  : 407.725.8024

  (I WROTE IT, NOT MR. HARRIS)
 
"Mache'tes dont kill people, people kill people." NRA (Natl Rwandan Army)
D:\RRN

Re: Experiences/history of Aetech Ada compiler?

[Bob Duff]

In article <19940919111103.RLOKER01@rlokerson9.ess.harris.com>,
Robert C. Lokerson <RLOKER01@HARRIS.COM> wrote:
>My personal style of programming has evolved into one which uses a lot of
>arrays of pointers to constant strings of varying lengths.  I have not been
>able to get this compiler to properly construct them unless they are created
>at run time with the "access" type scheme.  This might be ok (and in the ADA
>spirit) but I wish I could have "all" the initializations happen at
>elaboration-time.  (If thay makes any sense... I don't really consider myself
>an ADA expert... Much as I would like to be)

In Ada 9X, you can do things like this:

    type String_Ptr is access constant String;
    type String_Table is array(Positive range <>) of String_Ptr;

    X: String_Table :=
        ( new String'("Hello, world."),
          new String'("Goodbye, cruel world."),
          new String'("Another message.") );

Since String_Ptr is an access-to-constant type, the compiler should
allocate the above stuff statically.  This works in part because
Unchecked_Deallocation is not allowed for access-to-constant types.
-- 
Bob Duff                                bobduff@inmet.com
Oak Tree Software, Inc.
Ada 9X Mapping/Revision Team (Intermetrics, Inc.)