Newsgroups: comp.lang.smalltalk
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!noc.near.net!MathWorks.Com!news.alpha.net!uwm.edu!math.ohio-state.edu!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!scsing.switch.ch!ubilab!svelte.flur.zuerich.ubs.ch!wsmango!zhka8
From: Paul.z.h.k.a.8.Kennedy@zhflur.ubs.ubs.ch (Paul Kennedy)
Subject: Running ST Code Stored In RDBMS's
Message-ID: <1994Oct14.152716.28346@zh014.ubs.ubs.ch>
Sender: news@zh014.ubs.ubs.ch
Nntp-Posting-Host: wsmango.flur.zuerich.ubs.ch
Reply-To: Paul.z.h.k.a.8.Kennedy@zhflur.ubs.ubs.ch
Organization: Union Bank Switzerland, CH
Date: Fri, 14 Oct 1994 15:27:16 GMT
Lines: 24

I think there was a thread in comp.lang.smalltalk a little while back
concerning storing ST methods in an external RDBMS and then selecting
and executing the required routine at runtime. I'm afraid at the time
I didn't follow this thread too closely, however it now looks as though
this sort of approach might be useful for my current project, so I was
wondering if people could (re)post their thoughts/opinions/experiences
on this subject.

As a simple example, it might be necessary for a user to select a subset
of objects from a list, where the criteria for this selection are stored
in the database records as ST blocks, each of which takes the object as
its argument and returns true or false.

If nothing else, I'm sure our system test people will pull their hair out
when they find out that the code isn't nicely seperated from the data
anymore :-)

Thanks in advance for any pointers.

-- Paul --

EMail: Paul.z.h.k.a.8.Kennedy@zhflur.ubs.ubs.ch


