Re: chaining method calls

I also find methods that return the object extremely annoying. Of course it is okay to chain method calls, but imNSho, only if the object returned if different from the one used to call the method. That makes sense and is useful, even though that too is not ideal when debugging. I don't mind doing

my $page = LWP::UserAgent->new->request(...)->content;
[download]

my ($name) = DBIx::Simple
    ->connect('dbi:SQLite:people')
    ->query('SELECT name FROM people WHERE email=? LIMIT 1', $email)
    ->list;
[download]

, but I do dislike when a method returns its $_[0].

The argument that I hear the most is that chaining methods like this reduces typing. While that is true, I think it's a very bad reason for actually doing so. Using only globals and no strict is another way to save three characters per first use. Not checking open's return value is another way of typing less. It seems that length reduction and bad style often go together.

Juerd # { site => 'juerd.nl', plp_site => 'plp.juerd.nl', do_not_use => 'spamtrap' }

Comment on Re: chaining method calls Select or Download Code

Replies are listed 'Best First'.

Re^2: chaining method calls
by adrianh (Chancellor) on Jun 12, 2003 at 09:59 UTC

The argument that I hear the most is that chaining methods like this reduces typing. While that is true, I think it's a very bad reason for actually doing so. Using only globals and no strict is another way to save three characters per first use. Not checking open's return value is another way of typing less. It seems that length reduction and bad style often go together.

Using subroutines is a way to save characters. So are foreach loops. You can argue it both ways.

In many cases I'd agree with Paul Graham that Succinctness is Power.

I think this is going to be one of those issues that people can agree to disagree over. The idiom has always seemed very natural and clear to me.

[reply]


Clear questions and runnable code get the best and fastest answer
	PerlMonks