|
Re: using s/// to remove file extensions
by polettix (Vicar) on Mar 30, 2005 at 13:23 UTC
|
Just delete the extension at the end of the name.
$approvedpdfs =~ s/\.pdf$//;
You could find File::Spec useful for this filename stuff, anyway.
Update: you should also take a look at perldoc perlretut, regexes aren't shell expansion rules :)
Flavio
Don't fool yourself.
| [reply]
[d/l]
[select] |
|
|
File::Basename might also be useful. Its got info on parsing out filenames (directories, basenames and extensions). I use the fileparse function to break apart the file name for renaming purposes.
| [reply] |
Re: using s/// to remove file extensions
by davis (Vicar) on Mar 30, 2005 at 13:32 UTC
|
You're confusing Perl's regular expressions with shell filename globbing rules. Using YAPE::Regex::Explain thusly:
#!/usr/bin/perl
use warnings;
use strict;
use YAPE::Regex::Explain;
my $re = qr/[A-Z]*.pdf/;
print YAPE::Regex::Explain->new($re)->explain;
gives
The regular expression:
(?-imsx:[A-Z]*.pdf)
matches as follows:
NODE EXPLANATION
----------------------------------------------------------------------
(?-imsx: group, but do not capture (case-sensitive)
(with ^ and $ matching normally) (with . not
matching \n) (matching whitespace and #
normally):
----------------------------------------------------------------------
[A-Z]* any character of: 'A' to 'Z' (0 or more
times (matching the most amount possible))
----------------------------------------------------------------------
. any character except \n
----------------------------------------------------------------------
pdf 'pdf'
----------------------------------------------------------------------
) end of grouping
----------------------------------------------------------------------
davis
Kids, you tried your hardest, and you failed miserably. The lesson is: Never try.
| [reply]
[d/l]
[select] |
Re: using s/// to remove file extensions
by cog (Parson) on Mar 30, 2005 at 13:32 UTC
|
You're confusing wildcards with regular expressions.
With wildcards, * is definitely "zero or more of any possible character including the space."
With regular expressions, * is "zero or more of the last token"
Your last token before the * is [A-Z]; hence, you'r [A-Z]*.pdf matches things like:
...but not something like file.pdf, for instance.
So regarding your regex, what you really wanted was [A-Z].*\.pdf... you see, a "." means (in Perl's regular expressions), "any single character except for the newline." (this is also why you have to escape the following dot, the one separating the filename from the extension)
Also, your right-hand side of the substitution won't work... that's going to name the file, literally, [A-Z]*
Instead, you have to capture the filename in a special variable (with brackets) and use that variable (in this case it's going to be $1, because it's going to be the first bracket counting from the left). You do it like this:
s/([A-Z].*)\.pdf/$1/;
There. Your filename, without the extension.
frodo72's solution is, however, better (and there are others, of course), but this is what you were trying to do, corrected. | [reply]
[d/l]
[select] |
Re: using s/// to remove file extensions
by ihb (Deacon) on Mar 30, 2005 at 13:30 UTC
|
s/// doesn't work like that. In s/PATTERN/REPLACEMENT/ the PATTERN in a regular expression, like the one you've given, but the REPLACEMENT part isn't. The replacement is just a string use like a regular double-quoted string. The /e makes the string evaluate, i.e. act like code instead of a piece of chars and [A-Z]* isn't valid Perl code. So remove the e, don't use a pattern in the replacement part, and ultimately fix your pattern. That might include capturing a part of the match matched by PATTERN, and you then interpolate that match in the replacement part (as the replacement is just a double-quoted string). What you really want to have, I think, is
$file =~ s/^([A-Z].*)\.pdf\z/$1/;
Now this introduce quite a few things, like ^, (), .*, \., \z, and $1. You can read more about them in perlretut and other documents that come with Perl. To learn about s/// see perlop.
Update: Didn't read closely enough; adjusted the pattern to fit the OP.
ihb
See perltoc if you don't know which perldoc to read!
| [reply]
[d/l]
[select] |
Re: using s/// to remove file extensions
by Plotinus (Sexton) on Mar 30, 2005 at 13:41 UTC
|
Scary people,
5 replies in 3 minutes. I'm having multiple 'doh!' moments whilst reading them. Thank you and I'll play some more.
Plotinus | [reply] |
|
|
| [reply] |
Re: using s/// to remove file extensions
by inman (Curate) on Mar 30, 2005 at 13:54 UTC
|
You may also wish to check out File::Find as this will help you to recursively scan an entire directory structure and do something with each file that you find. The code below should work OK for you.
use File::Find;
my @directories = @ARGV ? @ARGV : ('.');
find(\&wanted, @directories);
sub wanted
{
return unless /\.pdf$/i;
print "working with $File::Find::name\n";
}
| [reply]
[d/l] |
Re: using s/// to remove file extensions
by FitTrend (Pilgrim) on Mar 30, 2005 at 13:29 UTC
|
($fName, $extension) = split (/\./, $FullFileName);
print scalar $fName;
| [reply]
[d/l] |
|
|
| [reply] |