Check whether a field from a line of text line matches a value - perl

Question

I have been using the following Perl code to extract text from multiple text files. It works fine.
Example of a couple of lines in one of the input files:
Fa0/19 CUTExyz notconnect 129 half 100 10/100BaseTX
Fa0/22 xyz MLS notconnect 1293 half 10 10/100BaseTX
What I need is to match the numbers in each line exactly (i.e. 129 is not matched by 1293) and print the corresponding lines.
It would also be nice to match a range of numbers leaving specific numbers out i.e. match 2 through 10 but not 11 the 12 through 20
#!/perl/bin/perl
use warnings;
my #files = <c:/perl64/files/*>;
foreach $file ( #files ) {
open( FILE, "$file" );
while ( $line = <FILE> ) {
print "$file $line" if $line =~ /123/n;
}
close FILE;
}
Thank you for the suggestions, but can it can be done using the code structure above?

Answer 1

I suggest that you take a look at perldoc perlre.
You need to anchor your regex pattern. The easiest way is probably using \b which is a zero-width boundary between alphanumerics and non-alphanumerics.
#!/perl/bin/perl
use warnings;
use strict;
foreach my $file ( glob "c:/perl64/files/*" ) {
open( my $input, '<', $file ) or die $!;
while (<$input>) {
print "$file $_" if m/\b123\b/;
}
close $input;
}
Note - you should use three-argument open with lexical file handles as above, because it is better practice.
I've also removed the n pattern modifier, as it appears redundant.
Following your edit though, to give us some source data. I'd suggest the solution is not to use a regex - your source data looks space delimited. (Maybe those are tabs?).
So I'd suggest you're better off using split and selecting the field you want, and testing it numerically, because you mention matching ranges. This is not a good fit for regexes because they don't understand the numeric content.
Instead:
while ( <$input> ) {
print if (split)[-4] == 129;
}
Note - I use -4 in the split, which indexes from the end of the list.
This is because column 3 contains spaces, so splitting on whitespace is going to produce the wrong result unless we count down from the end of the array. Using a negative index we get the right field each time.
If your data is tab separated then you could use chomp and split /\t/. Or potentially split on /\s{2,}/ to split on 2-or-more spaces
But by selecting the field, you can do numeric tests on it, like
if $fields[-4] > 100 and $fields[-4] < 200
etc.

Answer 2

I hope you don't get the answers you're asking for, which discard best practice because of your unfamiliarity with Perl. It is inappropriate to ask how to write an ugly solution because proper Perl is beyond your reach
As has been said repeatedly on this site, if you don't know how to do a job then you should hire someone who does know and pay them for their work. No other profession that I know has the expectation of getting quality work done for free
Here's a few notes on your code. Wherever you have learned your techniques, you have been looking at a very outdated resource
Do you really have a root directory perl, so that your compiler is /perl/bin/perl? That's very unusual, and there is no need to use a shebang line in Windows
You must always add use strict and use warnings 'all' at the top of every Perl program you write, and declare all of your variables using my as close as possible to their first point of use. For some reason you do this with #files but not with $file
It is better to replace <c:/perl64/files/*> with glob 'C:/perl64/files/*'. Otherwise the code is less clear because Perl overloads the <> operator
Don't put variable names inside double quotes. It is unnecessary at best, and may cause bugs. So "$file" should be $file
Always use the three-parameter version of open, so that the second parameter is the open mode
Don't use global file handles. And always test whether the file has been opened correctly, dying with a message including $!—the reason for the failure—if the open fails
open( FILE, "$file" )
should be something like
open my $fh, '<', $file or die qq{Unable to open "$file" for input: $!}
Don't rely on regex patterns for everything. In this case it looks like split would be a better option, or perhaps unpack if your records have fixed-width fields. In my solution below I have used split on "more than one space", but if your real data is different from what you have shown (tab-delimited?) then this is not going to work
Note that Fa0/129 will also be matched by your current approach
This Perl program filters your data, printing lines where the fourth field $lines[3] (delineated by more than one whitespace character) is numerically equal to 129
The output shown is produced when the input is the single file splitn.txt, containing the data shown in your question
use strict;
use warnings 'all';
for my $file ( glob 'C:/perl64/files/*' ) {
open my $fh, '<', $file or die qq{Unable to open "$file" for input: $!};
while ( my $line = <$fh> ) {
chomp;
my #fields = split /\s\s+/, $line;
print "$file $line" if $fields[3] == 129;
}
}
output
splitn.txt Fa0/19 CUTExyz notconnect 129 half 100 10/100BaseTX

Answer 3

Your question is unclear. When you say:
What I need is to match numbers in the on each line exactly
That could mean a couple of things. It could mean that each line contains nothing but a single number which you want to match. In that case, using == is probably better than using a regular expression. Or it could mean that you have lots of text on a line and you only want to match complete numbers. In that case you should use \b (the "word boundary" anchor) - /\b123\b/.
If you're clearer in your questions (perhaps by giving us sample input) then people won't have to guess at your meaning.
A few more points on your code:
Always include both use strict and use warnings.
Always check the return value from open() and take appropriate action on failure.
Use lexical filehandles and 3-arg version of open().
No need to quote $file in your open() call.
Using $_ can simplify your code.
/n on the match operator has no effect unless your regex contains parentheses.
Putting that all together (and assuming my second interpretation of your question is correct), your code could look like this:
#!/perl/bin/perl
use strict;
use warnings;
my #files = <c:/perl64/files/*>;
foreach my $file (#files) {
open my $file_h, '<', $file
or die "Can't open $file: $!";
while (<$file_h>) {
print "$file $_\n" if /\b123\b/;
}
# No need to close $file_h as it is closed
# automatically when the variable goes out
# of scope.
}