Re^4: How can I group lines in a file?

[bichonfrise74]

oops, my bad. When I was writing my own version, I used "split" to get the values which is basically the regex thing that BrowserUk did. Also, my version is similar to what you just wrote... Read the file, get the values, push it in a hash of array.

But BrowserUk's version was done in a single line which I'm still perplexed. Why does it work? (regex thing "and" pushing it in a hash of array).

How does "and" work? How can it get the regex values and push it to HoA? Thanks.

[BrowserUk]

Why does this line work? /(\S+)\s+(\S+)/ and push @{ $HoA{ $1 } }, $2 while <$fh>;

The regex matches against $_ (which is set to each line in turn by the while), and sets $1 to the first whitespace delimited 'word', and $2 to the second.

The and checks that the line matches, and if it does pushes the second 'word' onto an array within the hash keyed by the first word.

You can also tighten the regex a tad and add some error checking:

#! perl -slw
use strict;
use Data::Dump qw[ pp ];

my %HoA;
open my $fh, '<', 'junk.dat' or die $!;

/(\S+)\s+(\d+)/
    and push @{ $HoA{ $1 } }, $2
    or warn "Bad data '$_' at line $.\n"
    while <$fh>;

close $fh;

pp \%HoA;

__END__
c:\test>junk
Bad data 'the quick brown fox
' at line 5
{ ernest => [38, 27], jim => [14, 34], john => [23, 44], matilda => [4
+3, 22] }