How to handle files that do not contain a pattern? - regex

Question

I need help with my Perl program. The idea is to pass in a pattern and a file list from the command line. If the file name matches the pattern, print the file name. Then if the file name doesn't match, it should look for instances of the pattern in the text of the file and print filename : first line of text that contained occurrence.
However should the user add the -i option at the beginning the opposite should occur. If the filename does not match print it. Then print any files that do not contain any instances of the pattern in their text.
This last part is where I'm struggling I'm not exactly sure how to get files that don't have the pattern in their text. For example in my code
#!/usr/bin/perl -w
die("\n Usage: find.pl [-i] <perlRegexPattern> <listOfFiles>\n\n") if(#ARGV<2);
my (#array,$pattern,#filesmatch,#files);
#I can separate files based on name match
($pattern,#array) = ($ARGV[0] eq "-i") ? (#ARGV[1 .. $#ARGV]) : (#ARGV);
foreach(#array){
($_ =~ m/.*\/?$pattern/) ? (push #filesmatch,$_) : (push #files, $_);
}
#and I can get files that contain a pattern match in their text
if($ARGV[0] ne "-i"){
for my $matches(#filesmatch){ #remove path print just file name
$matches =~s/.*\///; #/
print "$matches\n";
}
for my $file(#files){
open(FILE,'<',$file) or die("\nCould not open file $file\n\n");
while(my $line = <FILE>){
if($line =~ m/$pattern/){
$file =~ s/.*\///; #/ remove path print just file name
print "$file: $line";
next;
}
}
}
}
#however I'm not sure how to say this file dosen't have any matches so print it
else{
for my $matches(#files){ #remove path print just file name
$matches =~ s/.*\///;
print "$matches\n";
}
for my $file(#filesmatch){
open(FILE,'<',$file) or die("\nCould not open file $file\n\n");;
while(my $line = <FILE>){...
I'm not sure if something like grep could be used to do this but I'm having a hard time working with Perl's grep.

Answer 1

In order to decide whether to print or not a file based on its content you have to first read the file. With your criterion -- that a phrase does not exist -- you have to check the whole file.
A standard way is to use a separate variable ("flag") to record the condition then go back to print
my $has_match;
while (<$fh>) {
if (/$pattern/) {
$has_match = 1;
last;
}
}
if (not $has_match) {
seek $fh, 0, 0; # rewind to the beginning
print while <$fh>;
}
This can be simplified by reading the file into a variable first, and by using labels (also see perlsyn)
FILE: foreach my $file (#filesmatch) {
open my $fh, '<', $file or die "Can't open $file: $!";
my #lines = <$fh>;
for (#lines) {
next FILE if /$pattern/;
}
print for #lines;
}
Note that skipping an iteration in the middle of a loop isn't the cleanest way since one has to always keep in mind that the rest of the loop may not run.
Each file is read first so that we don't read it twice, but don't do that if any of the files can be huge.
If there is any command line processing it is better to use a module; Getopt::Long is nice.
use Getopt::Long;
my ($inverse, $pattern);
GetOptions('inverse|i' => \$inverse, 'pattern=s' => \$pattern)
or usage(), exit;
usage(), exit if not $pattern or not #ARGV;
sub usage { say STDERR "Usage: $0 ... " }
Call the program as progname [-i] --patern PATTERN files. The module provides a lot, please see docs. For example, in this case you can also just use -p PATTERN.
As GetOptions parses the command line the submitted options are removed from #ARGV and what remains in it are file names. And you have the $inverse variable to nicely make decisions.
Please have use warnings; (not -w) and use strict; at the top of every program.