comp.lang.ada
 help / color / mirror / Atom feed
From: "Hibou57 (Yannick Duchêne)" <yannick_duchene@yahoo.fr>
Subject: Re: Specifying the order of ops on an ADT with aspects
Date: Fri, 5 Feb 2010 19:35:58 -0800 (PST)
Date: 2010-02-05T19:35:58-08:00	[thread overview]
Message-ID: <f5252d62-d4bf-440d-b5ce-4b2e39d9cc01@u9g2000yqb.googlegroups.com> (raw)
In-Reply-To: a5c90484-a8c0-4ed4-a90a-9ba6fedbcb42@f15g2000yqe.googlegroups.com

Side comment.

There is another good side effect of pre-post-conditions, apart their
abilities to better express contracts and to check usages and
implementations conformance : as you define pre-post-conditions, you
see what's needed to define these, so it enforce completeness of ADTs.
With this kind of design, one is less likely to miss a required
function to test an instance of an ADT. Last and although indirectly,
these functions, in turn sometimes makes the need of some associated
procedures more clear.

This is an aid in setting up specifications.



      reply	other threads:[~2010-02-06  3:35 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-02-04 11:26 Specifying the order of ops on an ADT with aspects Georg Bauhaus
2010-02-04 18:07 ` Hibou57 (Yannick Duchêne)
2010-02-05 16:44   ` Colin Paul Gloster
2010-02-05 16:55     ` Hibou57 (Yannick Duchêne)
2010-02-05 18:34       ` Dmitry A. Kazakov
2010-02-08 10:41         ` Colin Paul Gloster
2010-02-05 22:15   ` Randy Brukardt
2010-02-05 22:45     ` Hibou57 (Yannick Duchêne)
2010-02-06  3:35       ` Hibou57 (Yannick Duchêne) [this message]
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox