|
Re: Easy way to search files?
by adrianh (Chancellor) on Jan 06, 2003 at 13:00 UTC
|
use File::Find::Rule;
+
# find all the subdirectories of a given directory
+
my @subdirs = File::Find::Rule->directory->in( $directory );
+
+
# find all the .pm files in @INC
+
my @files = File::Find::Rule->file()
+
->name( '*.pm' )
+
->in( @INC );
Once you have found the files you are interested, just open them and use m// (or whatever) to search for what you're interested in. | [reply]
[d/l]
[select] |
|
|
I wonder why people choose the OO interface for File::Find::Rule.
Personally, I prefer its functional interface. To demonstrate how clean it is, I'm rewriting the code you pasted from the documentation:
use File::Find::Rule qw(find);
my @subdirs = find 'directory', in => $directory;
my @files = find 'file', name => '*.pm', in => \@INC;
Would you choose the OO interface? Why?
Another nice syntax for exactly the same is find file => (name => '*.pm', in => \@INC).
- Yes, I reinvent wheels.
- Spam: Visit eurotraQ.
| [reply]
[d/l]
[select] |
|
|
Would you choose the OO interface?
Yes :-)
Why?
Because:
- I find the OO style more readable.
- Most of the code I write is OO. Using the OO API to File::Find::Rule saves me switching mental models.
- Sometimes the intermediate objects are useful. You can use it to generate rules at runtime based on other conditions. This can be harder using the functional style.
| [reply] |
|
|
my @files = File::Find::Rule
->file()
->grep(qr/\Q$word\E/)
->in(@directory);
Makeshifts last the longest. | [reply]
[d/l] |
|
|
On the topic of File::Find::Rule I'd like to mention that this module was featured in the Perl Advent Calender here (off-site link).
-- Hofmator
| [reply] |
Re: Easy way to search files?
by Zaxo (Archbishop) on Jan 06, 2003 at 13:15 UTC
|
File::Find will handle recursive directory spanning.
Here's a simple sub which takes a filepath and a string, returning true if the file contains the string. Slurp is used, so files should be smallish.
sub contains {
my ($file, $string) = @_;
my $content;
{
local $/;
open my $fh, '<', $file or die $!;
$content = <$fh>;
close $fh or die $!;
}
index( $content, $string) == -1
? 0
: 1;
}
That should be fairly quick.
Update: Any sub returns the value of its last expression unless an explicit return comes before. In sub contains, the trinary operator ?: tests whether index returned -1, and evaluates to zero if so, otherwise evaluates to 1. That could as well have been written
return not index( $content, $string) == -1
or
return index( $content, $string) != -1
where the return keyword is optional.
After Compline,
Zaxo | [reply]
[d/l]
[select] |
|
|
That looks like what I need. I have a basic question though. I know how to return variables from a subroutine. I'm not, however, sure how this subroutine returns true. Or should I say that I am not sure how to return the fact that this subroutine returns true for each file. Can i just code to return a variable called $file_hit or something? That if indeed the file has been found to be true. I think that this post might be a bit confusing. I'm posting it anyway just in case people understand what is going through my mind.<\p>
| [reply] |
|
|
Hi ,
The file content or pointer could be returned .
| [reply] |
Re: Easy way to search files?
by Anonymous Monk on Jan 07, 2003 at 04:56 UTC
|
Not to be anti-perl, but what you need is "grep". There are versions for win and obviously unix. This program is optimized to do exactly what you want. But, if your doing this to stimulate your perl brain cells, GO for it. | [reply] |
Re: Easy way to search files?
by Anonymous Monk on Jan 07, 2003 at 04:56 UTC
|
The easiest way would be to use the find command.
$ find . -exec grep -l TheWordIAmLookingFor {} \;
This assumes a Unix OS or that you have cygwin installed
| [reply]
[d/l] |
|
|
This assumes a Unix OS or that you have cygwin installed
No need to go that far-
http://gnuwin32.sourceforge.net/
| [reply] |
|
|
| [reply]
[d/l] |