From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,FREEMAIL_FROM autolearn=unavailable autolearn_force=no version=3.4.4 X-Received: by 10.107.19.144 with SMTP id 16mr9490910iot.0.1466181551450; Fri, 17 Jun 2016 09:39:11 -0700 (PDT) X-Received: by 10.157.45.136 with SMTP id g8mr115845otb.3.1466181551295; Fri, 17 Jun 2016 09:39:11 -0700 (PDT) Path: eternal-september.org!reader01.eternal-september.org!reader02.eternal-september.org!news.eternal-september.org!mx02.eternal-september.org!feeder.eternal-september.org!usenet.blueworldhosting.com!feeder01.blueworldhosting.com!border2.nntp.dca1.giganews.com!nntp.giganews.com!v48no3286229qgd.0!news-out.google.com!v8ni564qgv.0!nntp.google.com!31no1639395qgh.1!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail Newsgroups: comp.lang.ada Date: Fri, 17 Jun 2016 09:39:11 -0700 (PDT) In-Reply-To: <5f9e49b3-db28-4a4d-b9ed-887391ec6ff1@googlegroups.com> Complaints-To: groups-abuse@google.com Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=2601:191:8202:8510:5985:2c17:9409:aa9c; posting-account=fdRd8woAAADTIlxCu9FgvDrUK4wPzvy3 NNTP-Posting-Host: 2601:191:8202:8510:5985:2c17:9409:aa9c References: <5f9e49b3-db28-4a4d-b9ed-887391ec6ff1@googlegroups.com> User-Agent: G2/1.0 MIME-Version: 1.0 Message-ID: <209d8955-7377-4747-a0f5-31cfc96425ac@googlegroups.com> Subject: Re: Help Needed - Telesoft Telegen 3.2.5 expert VAX/VMS From: rieachus@comcast.net Injection-Date: Fri, 17 Jun 2016 16:39:11 +0000 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Xref: news.eternal-september.org comp.lang.ada:30770 Date: 2016-06-17T09:39:11-07:00 List-Id: On Wednesday, June 15, 2016 at 5:12:56 PM UTC-4, Rick Cottle wrote: > We need to touch an old project and of course there is very little docume= ntation available (or readily found) and previous contributors move on a lo= ng time ago. >=20 > In short, the project is an embedded system built with Mil-Std-1750 built= with ADA on a Vax. We have recreated the project to the point of trying t= o link but do not have a linker map to go the final step. If you take the source and feed it to gnatchop for any version of GNAT, you= will get a set of files properly named for GNAT compilation. If you then = compile the main program, it will (recursively if necessary) compile all of= the source files, and create a make file for that main program. There may be more than one main program. If so you can compile them all in= a single project and get a make that will create each one. (Compilation u= nits used by multiple main programs will be compiled just once.) Can't fig= ure out from the documentation which are the main programs? Just look for = procedure bodies as files, probably without a corresponding specification. If all of this is Greek to you? Contact AdaCore (www.adacore.com) and hire= them to do it for you. Personally, I'm retired. I assume you have access= to the Telesoft compiler and a machine to run it on. Note that if the sou= rce is classified that may be the major cost and schedule driver in getting= the work done. In other words, it might be much cheaper to educate one of= your own software engineers to do the work. If there are only a dozen or so source files, it will be an easy task. If t= here are hundreds? One engineer can do it. If there are thousands? You w= ill need to partition the work among multiple engineers--assuming you need = to do more that retarget to a different CPU or operating system. Why? You= need to recover the information about how the software works that was in t= he brains of the original project engineers to know how changes will affect= the system. As an anecdote to warn you about the effort involved. I and another softwa= re engineer, both of us very familiar with the software, spent about a mont= h on one change proposal. After an intense review which lasted all afterno= on, we made the software change--changing an "and" to an "or," and ran the = test suite. In other words, learning the software may be the biggest task = in any change, and the time to do that will be proportional to the source c= ode size, not the magnitude of the change. If the source code contains useful comments it will help, even if you can't= find the other project documentation. Again though, it will take time to r= ead that documentation.