Merge pull request #593 from jmaslak/joelle-24-2-1

Joelle's P5 solution for 24.2

1 files changed, 52 insertions, 0 deletions

diff --git a/challenge-024/joelle-maslak/perl5/ch-2.pl b/challenge-024/joelle-maslak/perl5/ch-2.pl
new file mode 100755
index 0000000000..93f8a076d6
--- /dev/null
+++ b/challenge-024/joelle-maslak/perl5/ch-2.pl
@@ -0,0 +1,52 @@
+#!/usr/bin/env perl
+
+use v5.16;  # for fc
+use strict;
+use warnings;
+
+use File::ByLine;
+use List::Util qw(uniq);
+
+# Using file by line means that a very long single file might be able to
+# be processed quicker (assuming IO can keep up).
+
+MAIN: {
+    my %docs;
+
+    my $fbl = File::ByLine->new();
+    $fbl->processes(10);
+
+    for my $fn (uniq @ARGV) {
+        # Read the file, seperate out the words (use File::ByLine to
+        # allow 10 threads to simultaniously read the file)
+        #
+        # F::BL->map applies the map to each line.
+        my %found;
+        $docs{$fn} = [
+            uniq $fbl->map(
+                sub {
+                    grep { $_ ne '' and exists $found{$_} ? undef : ($found{$_} = 1) }
+                        map   { fc }    # Case insensitive words
+                        split /\W+/, $_
+                },
+                $fn,
+            )
+        ];
+    }
+
+    # Build the index
+    my %index;
+    for my $fn (sort keys %docs) {
+        for my $word ($docs{$fn}->@*) {
+            $index{$word} = [] unless exists $index{$word};
+            push $index{$word}->@*, $fn;
+        }
+    }
+
+    # Output the index
+    for my $word (sort keys %index) {
+        say "$word: " . join(" ", $index{$word}->@*);
+    }
+}
+
+