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=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: fac41,2c6139ce13be9980 X-Google-Attributes: gidfac41,public X-Google-Thread: 103376,3d3f20d31be1c33a X-Google-Attributes: gid103376,public X-Google-Thread: 1108a1,2c6139ce13be9980 X-Google-Attributes: gid1108a1,public X-Google-Thread: f43e6,2c6139ce13be9980 X-Google-Attributes: gidf43e6,public From: fjh@mundook.cs.mu.OZ.AU (Fergus Henderson) Subject: Re: Separation of IF and Imp: process issue? Date: 1997/09/09 Message-ID: <5v4093$of7@mulga.cs.mu.OZ.AU>#1/1 X-Deja-AN: 271056019 References: <33E9ADE9.4709@flash.net> <5upe9k$7he@newshub.atmnet.net> <5utag9$o6s@newshub.atmnet.net> <5v1gua$fkk@newshub.atmnet.net> Organization: Comp Sci, University of Melbourne Newsgroups: comp.object,comp.software-eng,comp.lang.ada,comp.lang.eiffel Date: 1997-09-09T00:00:00+00:00 List-Id: dnew@zloty.fv.com (Darren New) writes: >>There's the rub. My point was, why do you think traditional >>configuration management tools can't generate a short form from the >>Eiffel class text and disallow checkins that have changed the spec in >>incompatible ways? Often you don't want to actually _disallow_ such checkins. It is just important that they be given more visibility. For example, there may be a guideline such as "any change to the library interface should be accompanied by a change to the NEWS file". This is a guideline, not a rule, and it should not be enforced by the configuration management tool (at least not until configuration management tools get a lot smarter! ;-). But if it is not _immediately obvious_ to those reviewing a change which parts of the change are interface changes, then they may not realize the significance of those parts, and they may easily >Actually (following up on my own post) how does a traditional CM tool >make sure the Ada body matches the interface file? Does it actually >run some external tool at checkin time to make sure the equivalence >hasn't been broken? I mean, if I edit the return type of a function >in the body and not the spec in Ada, how does the CM tool know it won't >compile next time someone checks it out? Sometimes such checks are run at checkin time; for example, Robert Dewar has explained how ACT have a checkin server that receives checkins, builds the software, and runs their test suite; only if these tests pass will the checkin be accepted. Another common mode of operation is to run nightly tests via cron; with this scheme, violations might not be caught at checking time, but they would be caught overnight. >It seems to me that if you've solved this problem, you can use exactly >the same approach to solve the problem for Eiffel. I don't think it's as easy as that, because of the humans in the loop. One approach that might work would be for the CM tool to require the log message accompanying a change to specify which interfaces had been modified, and for the tool to verify that only those interfaces had changed. But I'll bet most Eiffel developers don't use a CM tool with that sort of support. -- Fergus Henderson | "I have always known that the pursuit WWW: | of excellence is a lethal habit" PGP: finger fjh@128.250.37.3 | -- the last words of T. S. Garp.