From: jcallen@Encore.COM (Jerry Callen)
Subject: Re: semantics of exception when exiting a procedure
Date: 15 Aug 90 17:08:25 GMT [thread overview]
Message-ID: <12489@encore.Encore.COM> (raw)
In-Reply-To: 1616@kuling.UUCP
In article <1616@kuling.UUCP> daniel@mizar.docs.uu.se (Daniel Wengelin) writes:
>Question:
>What are the semantics when a procedure with out parameters is
>exited by an exception. Can the out parameters be used, if they
>have been assigned prior to the exception ?
>
> [example deleted]
This whole area is described in something less than lucid detail in section
11.6 (Exceptions and Optimization). The example and comments in paragraphs
10 and 11 are helpful.
Basically, if the exception is raised in the body, you can depend only upon
initializations done in the declarations. In other words, the value of
out parameters is undefined.
-- Jerry Callen
jcallen@encore.com
next prev parent reply other threads:[~1990-08-15 17:08 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
1990-08-15 9:29 semantics of exception when exiting a procedure Daniel Wengelin
1990-08-15 17:01 ` Brian Hanafee
1990-08-15 17:08 ` Jerry Callen [this message]
1990-08-15 17:16 ` Fred Stluka
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox