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 autolearn=ham autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,ce1e7170ab2bc91c X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 2001-03-27 15:36:05 PST Path: supernews.google.com!sn-xit-03!supernews.com!freenix!sunqbc.risq.qc.ca!cpk-news-hub1.bbnplanet.com!bstnma1-snf1.gtei.net!news.gtei.net!USEAST.RATIONAL.com NNTP-Posting-Host: 172.20.21.30 Newsgroups: comp.lang.ada Date: Tue, 27 Mar 2001 18:17:01 -0500 Message-ID: From: "Corey Ashford" References: <_pNv6.15345$ue1.1278082@newsread2.prod.itd.earthlink.net> <3ABFB3AD.DF3AA617@lmco.com> Subject: Re: Why do so many companies use Apex? X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2919.6700 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6700 Xref: supernews.google.com comp.lang.ada:6144 Date: 2001-03-27T18:17:01-05:00 List-Id: "Keith Thompson" wrote in message news:yecpuf2c1hf.fsf@king.cts.com... > "Corey Ashford" writes: > [...] > > Apex's support for "fast path" recompilation is far more > > extensive. For example, let's say you change the definition > > of a record in a spec. In Apex, only those units which > > use the record definition will be recompiled, whereas in most > > other compilers (and I assume including GNAT), the compilation > > system will cause all units that "with" the spec of that package > > containing the record definition to be recompiled because > > more than just comments or white space has changed. > > It would be interesting to see a cost-benefit analysis for this kind > of thing -- in particular, whether the time saved by avoiding major > rebuilds outweighs the time spent in bookkeeping. > > Note that 100 6-second delays might well be more annoying to the user > than a single 600-second delay. The former could give a general > impression that the system is slow; the latter is understandable if > it's doing a lot of work, and it gives the user a chance to grab a cup > of coffee. > > I really have no idea what the outcome of such a study would be, and > I've never used Apex. Just randomly firing neurons. > Yes, that would be very interesting. I think it would depend a great deal upon how often changes were made and how deep and complex the dependency graph is. I can say from my use of it that it is very nice, and personally I've never run into a problem with it not recompiling something that should've recompiled. I used to use Verdix VADS and it had the attribute that you'd avoid making cosmetic changes to units to avoid recompilation. Cosmetic changes aren't always just white space or comments - they can be things like variable names, function names, etc. which would cause recompilation in most other systems. With Apex, it's hardly a consideration. So I think fast path recompilation adds to the maintainability of code. - Corey