Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!usc!apple!sun-barr!newstop!texsun!convex!convex.COM
From: tchrist@convex.COM (Tom Christiansen)
Newsgroups: comp.lang.perl
Subject: Re: inlining your sort subroutine
Message-ID: <109359@convex.convex.com>
Date: 28 Nov 90 01:39:33 GMT
References: <109155@convex.convex.com> <10521@jpl-devvax.JPL.NASA.GOV>
Sender: usenet@convex.com
Reply-To: tchrist@convex.COM (Tom Christiansen)
Organization: CONVEX Software Development, Richardson, TX
Lines: 39

In article <10521@jpl-devvax.JPL.NASA.GOV> lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall) writes:
>In article <109155@convex.convex.com> I wrote:
>: wouldn't this be nice:
>: 
>:     @out = sort { $a <=> $b; } @in;

>It would be possible, but apart from the problem you mentioned of making it
>hard to find the list, I also think people should occasionally be encouraged
>to name their abstractions.

By all means.  It's just that the quoted line above seems to be more
readable than naming a subroutine for the sort; it would seem easier
on both the writer and the reader of the code.  I know in teaching 
perl, students have more than once asked me whether this was possible
as soon as I show them how sort takes an optional subroutine.  My 
response generally begins with "Alas, no; ...". 

>Of course, the fact that it would be hard to implement has nothing to
>do with it...  :-)

I see we come at last to the meat of the matter. :-)

In the meantime (and perhaps forever), you can always write
a sort() function like this if you really want to:

    @out = &sort('$a <=> $b', @in);

and a byreference version like this:

    @out = &sortr('$a <=> $b', *in);

so that you don't try this:

    @out = &sortr('$a <=> $b', 3, 4, 2, 1, 4, 98);

Naw, that's too complex.  I'll go back to naming subroutines
for a while.

--tom