Re^2: Make $^V and "my" implicit

"my" or it's equivalent should only be necessary in the not-so-common case where you have 2 variables with the same name and type ($@%). Why is "my" even necessary for the global declaration of $user in your example? "my" should only be necessary for the function-level $user variable in your case and, as a general rule, only necessary when distinguishing variables which have the same name and type ($@%).

Perl shoud apply rules such as these in a more fine-grained manner instead of peppering code with redundant scoping declarations. However, I am not arguing against the principle of "my" in general.

Comment on Re^2: Make $^V and "my" implicit

Replies are listed 'Best First'.
Re^3: Make $^V and "my" implicit by Corion (Patriarch) on Feb 03, 2014 at 17:52 UTC
But if that case is not-so-common, then you don't need `my` at all. `my` forces you to predeclare variables. This is highly important, as Perl can then tell you if you misspell a variable. It has the additional advantage of allowing you to explicitly declare the scope of a variable, to tell Perl whether you meant to create a new, lexically scoped variable or whether you meant to use the global variable. I think most of your complaints can be done away with by not using strict. Then you don't need to predeclare your variables but still can use lexical variables with `my`. Of course, recursion is a harsh mistress if you're using global variables. Typos also are a nasty problem. But you don't seem to have these.	[reply] [d/l] [select]
Re^4: Make $^V and "my" implicit by gunzip (Pilgrim) on Feb 03, 2014 at 18:02 UTC
I'm aware you do need a way of distinguishing variables of the same type and name used in different scopes but "my" just seems to over-generalise the issue to the point that it becomes boilerplate. Depends on your view at the end of the day, I suppose.	[reply]
A reply falls below the community's threshold of quality. You may see it by logging in.