[JOB] Perl Software Developer and Database programmer
Peter Hickman
peter at semantico.com
Thu Feb 23 13:28:14 GMT 2006
Andy Armstrong wrote:
> I'm not completely opposed to that. Again it depends hugely on the
> context - but wrapping even a tiny bit of code like that in a
> descriptive name can - in some circumstances - have value as
> documentation. For example I've just written
>
> sub feed_name {
> return show_name(@_);
> }
>
> Which makes perfect sense to me - at the moment the name of a feed is
> the same as the name of a show - but that might change. This way, if
> it does change, I don't have to go looking at all the instances of
> show_name() to find out which of them should actually do something
> different to get the name of a feed.
>
> --Andy Armstrong, hexten.net
>
>
Depends, if you were using it as
my $name_of_feed = feed_name(@list_of_names);
then I would say that it was wrong but if you are using it to add some
semantic information to the code as in
validate(feed_name(@list_of_names))
then it would be right. The first tells you nothing that the receiving
variable name does not and the fact that you have written a subroutine
to do this might lead people to think that feed_name() is actually doing
something important. The second tells you something that is not readily
apparent, but of course a comment would also do the job.
How would you feel if you came across code littered with routines like
add_two_numbers(), append_item_to_list(), check_hash_key_is_defined()
and the like. Too much syntactic sugar for my liking.
More information about the london.pm
mailing list