Re^4: Playing with non-localized $

<DATA>;
print $_;
__DATA__
This is a test.

__OUTPUT__
Use of uninitialized value in print at mytest.pl line 8, <DATA> line 1
+.
[download]

Hmm... You're right. There goes that theory. ;) I was erroneously sure that the while() loop's conditional had nothing to do with the fact that the diamond operator feeds $_. Thanks for the enlightenment. :D

Dave

Comment on Re^4: Playing with non-localized $_ in nested loops. Download Code

Replies are listed 'Best First'.
Re^5: Playing with non-localized $_ in nested loops. by ysth (Canon) on Aug 23, 2004 at 05:23 UTC
There are actually two pieces of dwimmery going on; first, if a readline or a glob is the only thing in a while condition (whether a statement modifier or while statement), it gets an implicit `$_ =`. Next, if a while condition is a scalar assignment from a readline, readdir, glob, or each, the whole condition gets an implicit defined() wrapped around it, so the loop continues on "0" or "".	[reply] [d/l]
Re^5: Playing with non-localized $_ in nested loops. by Aristotle (Chancellor) on Aug 23, 2004 at 10:06 UTC
That is why, in the node you referenced, I was talking about `while(<FH>)`. Invoking the DWIMmery really does require this entire combination of operations. Makeshifts last the longest.	[reply]