Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!mcvax!enea!sommar
From: sommar@enea.UUCP (Erland Sommarskog)
Newsgroups: net.lang
Subject: Re: null statements
Message-ID: <1411@enea.UUCP>
Date: Wed, 30-Jul-86 15:24:55 EDT
Article-I.D.: enea.1411
Posted: Wed Jul 30 15:24:55 1986
Date-Received: Fri, 1-Aug-86 07:51:53 EDT
References: <484@valid.UUCP> <3188@utcsri.UUCP>
Reply-To: sommar@enea.UUCP (Erland Sommarskog)
Distribution: net
Organization: Enea Data, Sweden
Lines: 19
Summary: IF THEN ELSE statement is valid

In article <3188@utcsri.UUCP> greg@utcsri.UUCP (Gregory Smith) writes:
>Isn't there a null statement ""? I.e., can't you say
>
>	IF .... THEN ELSE BEGIN ... END
>
>??? ( i.e. just delete the ; on line 3. Sorry, can't find a reference ).
IF <expression> THEN ELSE <statement> 
is legal Pascal. Look e.g. in Wirth's report only book of Pascal that
contains the syntax graphs.
  You can express it that in Pascal the empty statement is written with
an empty string. 

>PS: a Pascal teaching compiler I once used could come up with the
>message 'semicolon is never legal before ELSE'. Neat, huh?
So does DEC's Pascal compiler for VMS:
"; ELSE is not valid Pascal"
But most compilers says    ELSE
                           ^6
And at the end of list explains: "Illegal symbol". It takes some until
you react normally and directly start to look for an illegal semicolon.