sub _mtime {
    my ($self, $filepath) = @_;
    my $options = $self->{options};

    return(undef) if ($options->{blind_cache});

    # make sure it still exists in the filesystem 
    (-r $filepath) or Carp::confess("HTML::Template : template file $f
+ilepath does not exist or is unreadable.");

    # get the modification time
    return (stat(_))[9];
}
[download]

As you can see, the _ construct is passed to stat following the -r (readable) test on the file, presumably given that this stat information is also cached in a similar manner to file test information.

Update

Following through on this at perlfunc:_X, it appears that these file tests obtain the information required to return a positive or negative test result via a system stat call. From perlfunc:_X ...

If any of the file tests (or either the stat() or lstat() operators) are given the special filehandle consisting of a solitary underline, then the stat structure of the previous file test (or stat operator) is used, saving a system call. (This doesn't work with -t, and you need to remember that lstat() and -l will leave values in the stat structure for the symbolic link, not the real file.)

print "Can do.\n" if -r $a || -w _ || -x _;

stat($filename);
print "Readable\n" if -r _;
print "Writable\n" if -w _;
print "Executable\n" if -x _;
print "Setuid\n" if -u _;
print "Setgid\n" if -g _;
print "Sticky\n" if -k _;
print "Text\n" if -T _;
print "Binary\n" if -B _;
[download]

Ooohhh, Rob no beer function well without!

The underscore pseudo-filehandle is actually one of the very few features that I can claim is entirely my own idea. I suggested to Larry during the Perl3 alpha testing phase that performing a separate stat operation for each component of -r $foo && -w $foo was wasteful and subject to phase errors. He came up with the underscore cache in response.

The other ideas I can claim directly are the list-slice, and the arrow syntax for dereferencing coderefs (introduced on a bet, but that's another story).

-- Randal L. Schwartz, Perl hacker

Other tricks? Hmm, well getting the hang of $" $, $\ $/ $; is worth your time, as well as learning the use and flexibility of perls command line options is also cool, I recommend having a look -l and -i as they are pretty useful for small scripts (thanks to various people, merlyn wog for pointing these out). Also for smaller scripts I have developed a fondness for using 1 argumnet opens as I mention here Re: auto-vivifying file handles. While it may not count as tricks, there is lots of stuff to know about perls extended regex syntax, watch for japhy posts as he seems to be an expert.

Anyways, one of the things that I like about perl is that whenever you think that you've learned all the perl tricks there are to learn some guru comes along and shows you five more that you never heard of....

:-)

Yves
--
You are not ready to use symrefs unless you already know why they are bad. -- tadmc (CLPM)

A2. I'll leave that one as an open question, though much perl wisdom is hidden under this link. ;-)

-Blake

Here is a simple script to illustrate the problem:

#! perl -w
use strict;
my $file = 'test.bat';
print "1. $file is a text file, executable\n" if -T $file && -x _;
print "2. $file is an executable text file\n" if -x $file && -T _;
[download]

2. test.bat is an executable text file
[download]

Rudif