Re: Very elementry query

[Marin David Condic <condicma@bogon.pwfl.com>]

rishikeshjha@my-deja.com wrote:

> I am an absolute newbie to ADA. I have been assigned the task of sizing
> a project wherein a ADA app. has to be ported from Solaris to NT. Can
> someone tell me how complicated such efforts are.
>

Please note that the language is "Ada" - named after Ada Byron and not
"ADA" (Americans with Disabilities Act? American Dental Association?). For
some reason we have a tendancy to be touchy about this. :-)

I don't think your problems are going to be Ada specific. My experience in
porting Ada between PC's and Unix has been very positive. But then, the
applications were simple command-line-interface driven and made little to
no system calls. Given that GNAT exists on both platforms, I found that
compilation had zero problems and things ran mostly identically. (Some
problems with byte sex differences in external, custom built data files,
but this may be avoided if the Ada code is structured to detect this
possibility.) With a large number of applications I've built/ported, the
effort was near zero - just recompile and go.

Your real problem is going to be in the area of OS dependencies. You will
want to assess how much of the application performs GUI functions, asks for
OS services, manipulates files with unusual design, etc. That's the code
you will need to worry about and the level of porting effort will depend on
the skill of the programmers, the portability/isolation of the original
design and so on.

If it were me, I'd assess what portion of the app has such sensitive
features and count up the SLOCs. Anything that does more or less general
computational stuff ought to port very easily. Since our organization has
several years of productivity data (counting SLOCs per fortnight) I could
take a SWAG at how many hours would be needed to review/rewrite that
portion of code which may be bound to the OS. If you have similar data for
your organization, this would be a major help. (There is no way that one
organization's productivity measures will have anything to do with another
organization's behavior, so it wouldn't be much use for me to share that
data)

If you have more information about the nature of the app, and you are free
to post it here, its possible you may get more helpful answers.

MDC
--
Marin David Condic
If you act quick, you may still find me at.......
Real Time & Embedded Systems, Propulsion Systems Analysis
United Technologies, Pratt & Whitney, Large Military Engines
M/S 731-95, P.O.B. 109600, West Palm Beach, FL, 33410-9600
***To reply, remove "bogon" from the domain name.***

Visit my web page at: http://www.mcondic.com/