Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uunet!yale!mfci!karzes
From: karzes@mfci.UUCP (Tom Karzes)
Newsgroups: comp.lang.c
Subject: Re: syntax for unary assignment operators (was Re: C history question)
Keywords: C design, XOR
Message-ID: <1036@m3.mfci.UUCP>
Date: 19 Sep 89 16:22:14 GMT
References: <575@calmasd.Prime.COM> <1687@sunset.MATH.UCLA.EDU> <174@cpsolv.UUCP> <1989Sep17.150504.16643@jarvis.csri.toronto.edu> <1032@m3.mfci.UUCP> <2562@jhunix.HCF.JHU.EDU>
Sender: karzes@mfci.UUCP
Reply-To: karzes@mfci.UUCP (Tom Karzes)
Organization: Multiflow Computer Inc., Branford Ct. 06405
Lines: 24

In article <2562@jhunix.HCF.JHU.EDU> ins_akaa@jhunix.UUCP (Ta06) writes:
->>A somewhat consistent but fairly bizarre syntax would be
->>	x -=;
->The problem with this is that you would like it to have the same precedence
->as ++ and --.  ...
-
-Why?  We don't expect the regular += to have the same precedence as +...

Because all unary operators in C have the same precedence, which is higher
than the precedence of any binary or ternary operator (other than the primary
expression operators).  Introducing a unary operator with lower precedence
than assignment would be a disgusting wart.  Furthermore, it would make C
agonizing to parse.  What would the rules be for deciding when to reduce
x-= as opposed to looking for some expression following it?  Since you
naively think that binary -= should have higher precedence than unary -=,
it means that you'd always have to look beyond the -= to see if there's
something there that you can use as a second operand.  So you wouldn't be
able to write x -= * 3 because this would be parsed as x -= (*3), but you
might be able to write x -= / 3 since / isn't a valid unary operator.  The
whole thing it utterly absurd.

As for += not having the same precedence as +, what does that have to do with
anything?  All binary assignment operators in C have the same precedence,
which is lower than the precedence of +.