Questions about Grep

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

HI!

Sorry for this really newbie question, but I was just wondering if grep works the same in UNIX and perl.

Right now I can do this in grep:

grep pattern filename

And i'll know if that pattern occurs in the file. But from my reading so far, as I can tell, this function in perl only works in the sense that grep(pattern,list) but I would rather search a file then a list.

I have alternative ways to perform the action of grep. But I'd like to use the power of grep to do my dirty work for me.

Thanks!

Comment on Questions about Grep

Replies are listed 'Best First'.
Re: Questions about Grep by friedo (Prior) on May 29, 2008 at 19:02 UTC
If you want to grep a file in Perl, you can read it into an array first: `use strict; open my $fh, "file.txt" or die "Error opening file.txt: $!"; my @lines = <$fh>; my @matches = grep /pattern/, @lines;` [download] There's also the File::Grep module from CPAN which provides some easy-to-use file grepping functionality.	[reply] [d/l]
Re^2: Questions about Grep by pc88mxer (Vicar) on May 29, 2008 at 19:14 UTC
Don't know if this is any more efficient, but it's a few less keystrokes: `use strict; open my $fh, "file.txt" or die "Error opening file.txt: $!"; print grep /pattern/, <$fh>;` [download]	[reply] [d/l]
Re^2: Questions about Grep by Anonymous Monk on May 29, 2008 at 19:23 UTC
Thanks!	[reply]
Re: Questions about Grep by dwm042 (Priest) on May 29, 2008 at 20:06 UTC
As others have pointed out, grep in Perl and grep in various flavors of Unix are not the same. Further, there can be interesting variations in the kinds of patterns that different versions of (Unix) grep can handle (grep, egrep, fgrep, etc). Certain versions of Red Hat had a grep that accepted the -P flag, which then would allow the pattern to be a Perl style regular expression. I personally found it useful to be able to say things like: `grep -P "(199\|200)\d\w+" my.mail.log` [download]	[reply] [d/l]
Re^2: Questions about Grep by graff (Chancellor) on May 30, 2008 at 02:01 UTC
(shameless plug:) I seriously needed a command-line grep tool that could do real perl regexes, so I wrote one: grepp -- Perl version of grep -- it even handles encoding conversions and unicode expressions. (I know others have done similar exercises, but it's hard to resist doing it yourself...)	[reply]
Re: Questions about Grep by oko1 (Deacon) on May 29, 2008 at 20:28 UTC
If you take a look at 'man grep' and 'perldoc -f grep', you'll note a few differences. E.g. - commandline `grep` searches files, while Perl's `grep` searches lists. Commandline `grep` takes options; Perl's `grep` takes an expression or a subroutine. Commandline `grep` is limited to matching regular expressions; Perl's `grep` will return the matching elements for any test that returns 'true' for those elements. In short, the very basic premise of "global regular expression print" - matching and returning a specified item - is about the only thing that the two have in common, and you're much safer assuming that what you know about one does not apply to the other. -- Human history becomes more and more a race between education and catastrophe. -- HG Wells	[reply]
Re: Questions about Grep by planetscape (Chancellor) on May 29, 2008 at 21:45 UTC
For `grep` versions and related alternatives, please see grep , etc. HTH, planetscape	[reply] [d/l]


XP is just a number
	PerlMonks