Return-Path: <samepst@cs.bu.edu> X-Spam-HitLevel: X-Spam-DCC: sonic.net: cs3.bu.edu 1117; Body=1 Fuz1=1 Fuz2=1 X-Spam-Checker-Version: SpamAssassin 3.1.0 (2005-09-13) on cs3.bu.edu X-Spam-Level: X-Spam-Status: No, score=-3.4 required=10.0 tests=ALL_TRUSTED,BAYES_00, NO_REAL_NAME autolearn=ham version=3.1.0 X-Spam-Pyzor: Received: from csa3.bu.edu (csa3 [128.197.12.5]) by cs3.bu.edu (8.13.6/8.13.6) with ESMTP id k9SI66IJ013326 for <cs520@cs.bu.edu>; Sat, 28 Oct 2006 14:06:06 -0400 Received: (from apache@localhost) by csa3.bu.edu (8.13.6/8.12.11/Submit) id k9SI667h002782; Sat, 28 Oct 2006 14:06:06 -0400 X-Authentication-Warning: csa3.bu.edu: apache set sender to samepst@cs.bu.edu using -f Received: from 71.192.166.44 (SquirrelMail authenticated user samepst) by cs-squirrelmail.bu.edu with HTTP; Sat, 28 Oct 2006 14:06:06 -0400 (EDT) Message-ID: <1811.71.192.166.44.1162058766.squirrel@cs-squirrelmail.bu.edu> In-Reply-To: <001501c6faae$566c00d0$6400a8c0@csnt.bu.edu> References: <1458.71.192.166.44.1162052282.squirrel@cs-squirrelmail.bu.edu> <001501c6faae$566c00d0$6400a8c0@csnt.bu.edu> Date: Sat, 28 Oct 2006 14:06:06 -0400 (EDT) Subject: RE: Implementation 3 Stuck terms From: samepst@cs.bu.edu To: cs520@cs.bu.edu User-Agent: SquirrelMail/1.4.6-4.fc2.1.legacy Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-Clamav-Status: No Status: RO X-Status: A Content-Length: 968 X-UID: 36 X-Keywords: NonJunk
Is it correct to assume that substituion does not recurse into raise terms?
For example, the following terms evaluates to true:
try (lam (Exit:int)=> raise Exit) 5 with Exit => true | 5 => false
Thank you,
Sam Epstein
>
> Yes. Those are cases you should take care of. No free lunch if you want to
> add some features to a language, right?
>
> Rui
>
>
> -----Original Message-----
> From: samepst@cs.bu.edu [mailto:samepst@cs.bu.edu]
> Sent: Saturday, October 28, 2006 12:18 PM
> To: cs520@cs.bu.edu
> Subject: Implementation 3 Stuck terms
>
> There are a couple cases which could lead to trouble. Which, if any,
> should
> we be responsible for creating an evaluation step?
>
> 1.raise in the test term of an if statement:
>
> if raise Exit then 1 else 0
>
>
> 2.raise t where t is not a value but types to exn:
>
> (lam (x:int)=>x) raise ((lam (x:exn)=> x) Exit)
>
>
> 3. raise as a parameter in an operation
>
> 5 + (raise Exit)
>
>
> Thank You,
> -Sam Epstein
>
This archive was generated by hypermail 2b29 : Thu Dec 14 2006 - 16:31:59 EST