IWETHEY v. 0.3.0 | TODO
1,095 registered users | 1 active user | 0 LpH | Statistics
Login | Create New User
IWETHEY Banner

Welcome to IWETHEY!

IWETHEY Home / IWETHEY Board / Theory and Practice of Programming Forum / Interesting....
Post #178,310
by Simon_Jester
10/7/04 5:58:41 PM
Reply
New Interesting....

my %norm =
{
"a" => (),
"b" => (),
"c" => ()
};


all of those ()s get expanded to nothing, Perl actually sees,

my %norm =
{
"a",
"b",
"c"


Just to refresh my memory, the key is that those () are effectively nothing. If he had used a reference to an empty array, that (part) would work, no?

so, one solution would be

@empty_array;

my %norm =
{
"a" => \\@empty_array;
"b" => \\@empty_array;
"c" => \\@empty_array;
}
Post #178,317
by ben_tilly
10/7/04 6:27:50 PM
Reply
New The solution would have its own problems
In particular, now all values are the same, and so updating one updates them all. You were looking for this:
\nmy %norm =\n  (\n  "a" => [],\n  "b" => [],\n  "c" => [],\n  );\n

(note that you still need to fix the () vs {} issue.)

But in any case the right solution is to assume that %norm needs no initialization at all. Perl will autovivify it to be The Right Thing.

In fact that's kind of the point of having all the syntax - Perl has enough clues to be able to figure out The Right Thing and do it for you. For instance you can write:
\n  push @{ my $norm{$key} }, $value;\n

and if need be it will realize, "There is no value in %norm for $key - I need to create one and clearly it should be an anonymous array." In, say, Ruby you'd have to write two lines - one to make sure that the array was there and another to push something onto it.

(Whether the syntax is worth that benefit I'll not debate. I'm trying to explain Perl, not advocate it.)

Cheers,
Ben
I have come to believe that idealism without discipline is a quick road to disaster, while discipline without idealism is pointless. -- Aaron Ward (my brother)
Post #178,370
by Simon_Jester
10/7/04 11:14:47 PM
Reply
New <bow> thank you.
     Perl frustrations - (tuberculosis) - (45) - Oct. 7, 2004, 03:55:53 PM EDT
         Simple solution - (ben_tilly) - (10) - Oct. 7, 2004, 05:31:13 PM EDT
             Bleh. - (admin) - (3) - Oct. 7, 2004, 04:42:49 PM EDT
                 That's because you approached it wrong - (ben_tilly) - (2) - Oct. 7, 2004, 05:22:57 PM EDT
                     I'm gradually warming up to it - (deSitter) - Oct. 7, 2004, 09:08:18 PM EDT
                     That should be in the man page. - (static) - Oct. 8, 2004, 02:17:42 AM EDT
             Is this a sort of typecasting? - (tuberculosis) - (2) - Oct. 7, 2004, 05:36:33 PM EDT
                 It's a reference -NT - (Simon_Jester) - Oct. 7, 2004, 06:00:50 PM EDT
                 Sort of - (ben_tilly) - Oct. 7, 2004, 06:10:14 PM EDT
             Interesting.... - (Simon_Jester) - (2) - Oct. 7, 2004, 05:58:41 PM EDT
                 The solution would have its own problems - (ben_tilly) - (1) - Oct. 7, 2004, 06:27:50 PM EDT
                     <bow> thank you. -NT - (Simon_Jester) - Oct. 7, 2004, 11:14:47 PM EDT
         THank you for stepping on this rake - (Arkadiy) - (33) - Oct. 7, 2004, 05:20:48 PM EDT
             He was overcomplicating it - (ben_tilly) - (32) - Oct. 7, 2004, 05:30:04 PM EDT
                 Well, maybe - (tuberculosis) - (17) - Oct. 7, 2004, 05:43:24 PM EDT
                     Agreed - (ben_tilly) - (16) - Oct. 7, 2004, 06:12:17 PM EDT
                         Agree its a mistake - (tuberculosis) - (15) - Oct. 7, 2004, 06:36:09 PM EDT
                             Same here. - (admin) - Oct. 7, 2004, 06:45:32 PM EDT
                             Right - (ben_tilly) - Oct. 7, 2004, 06:56:48 PM EDT
                             Shame about the inertia. Python's design is "least surprise" - (FuManChu) - (12) - Oct. 7, 2004, 10:02:43 PM EDT
                                 When I started with Perl... - (ben_tilly) - (11) - Oct. 7, 2004, 10:53:12 PM EDT
                                     On Perl 6 - (pwhysall) - (3) - Oct. 8, 2004, 01:53:55 AM EDT
                                         Re: On Perl 6 - (Yendor) - Oct. 8, 2004, 08:37:56 AM EDT
                                         I had a nice response to this typed up - (ben_tilly) - (1) - Oct. 8, 2004, 12:54:37 PM EDT
                                             Many thanks for that - (pwhysall) - Oct. 9, 2004, 02:53:04 AM EDT
                                     We're having a little brown bag on Ruby - (tuberculosis) - (4) - Oct. 8, 2004, 12:46:50 PM EDT
                                         Re: We're having a little brown bag on Ruby - (JimWeirich) - (3) - Oct. 11, 2004, 06:55:34 AM EDT
                                             Shhh! Anonymous Todd works at some other... - (CRConrad) - Oct. 11, 2004, 07:36:30 AM EDT
                                             I might have been there - (tuberculosis) - (1) - Oct. 11, 2004, 12:55:46 PM EDT
                                                 Re: I might have been there - (JimWeirich) - Oct. 11, 2004, 01:14:57 PM EDT
                                     Re: strict -- have you seen pychecker? - (FuManChu) - (1) - Oct. 8, 2004, 01:02:20 PM EDT
                                         No I hadn't, thanks - (ben_tilly) - Oct. 8, 2004, 01:08:02 PM EDT
                 BTW Arkadiy, I'm still waiting for a response - (ben_tilly) - (13) - Oct. 8, 2004, 12:57:52 PM EDT
                     I am not saying it's shorter in C... - (Arkadiy) - (10) - Oct. 8, 2004, 03:22:09 PM EDT
                         And now for my real comment - (ben_tilly) - (2) - Oct. 8, 2004, 03:34:09 PM EDT
                             May be it's a hindsight thing - (Arkadiy) - (1) - Oct. 8, 2004, 03:47:36 PM EDT
                                 It could be many things - (ben_tilly) - Oct. 8, 2004, 06:13:45 PM EDT
                         I don't know - multimap - (Simon_Jester) - (6) - Oct. 9, 2004, 12:22:32 AM EDT
                             Re: I don't know - multimap - (Arkadiy) - (5) - Oct. 9, 2004, 12:17:47 PM EDT
                                 Why the worry about efficiency? - (ben_tilly) - (4) - Oct. 9, 2004, 12:25:52 PM EDT
                                     I worry about efficiency - (Arkadiy) - (1) - Oct. 9, 2004, 12:30:44 PM EDT
                                         Ah - (ben_tilly) - Oct. 9, 2004, 12:59:07 PM EDT
                                     C vs Perl - efficiency - untrue for me - (broomberg) - (1) - Oct. 9, 2004, 12:41:13 PM EDT
                                         Point - but you may want to rebenchmark - (ben_tilly) - Oct. 9, 2004, 01:11:47 PM EDT
                     Here's a smalltalk version - (tuberculosis) - (1) - Oct. 8, 2004, 06:16:21 PM EDT
                         Yes, think of a reference as a pointer - (ben_tilly) - Oct. 8, 2004, 06:43:04 PM EDT

It is fine to have low opinions of people's friends when they are assholes.
86 ms