I found this advice in a thread in a Linux newsgroup. The original question was typical newbie confusion about use strict forcing them to declare variables. Amongst all the advice (of variable quality!) on how to declare variables, I found someone saying that you could declare variables using syntax like:
use strict qw($foo);
The confusing thing is that it seems to work but, of course, it's not doing what the poster thinks at all. In fact, it has the potentially disasterous effect of turning "strict" off completely.
Perhaps it would be better if use strict complained if it got parameters that it didn't understand.
'Not turning "strict" on"' even.it has the potentially disasterous effect of turning "strict" off completely.
Oh, can I get a +Informative if I link to the thread on London.pm discussing this
Re:Pedantic comment ahoy...
davorg on 2002-11-14T12:39:38
can I get a +Informative if I link to the thread on London.pm discussing this;-) No, because then everyone will see me being really stupid and not working out what was going on until it was explaned to me in very simple language
:) Re:Pedantic comment ahoy...
mir on 2002-11-14T12:48:50
Well worth a +Funny then...
Perhaps it would be better if use strict complained if it got parameters that it didn't understand.
I'm shocked that a co-provoker of Acme::USIG would suggest something that would break this delightful module. USIG passes to use strict the URL of the original, historical and theatrical use strict is gay bulletin board exchange between BK, P.Cawley and D.Cross. Maybe I should file a bug report because this URL is long gone and should be replaced with the new and improved wayback machine one.
Moreover, having use strict complain about unknown strictures means code that uses a new stricture will break when run on older versions of Perl that do not understand it.
Re:But this would break davorg-inspired Acme::USIG
richardc on 2002-11-15T16:39:42
Maybe I should file a bug report because this URL is long gone and should be replaced with the new and improved wayback machine one.Or, you could just upgrade to Acme::USIG 1.02 which had a patch from Kake for this very issue applied about a month ago.
