Perl Weekly Challenge 166: files, directories and hex!

It is sad that, after more than two years of me doing Raku, I still don’t have any production code project to work on. Therefore, in order to keep my coding and Raku-ing (is that a term?) knowdledge, I try to solve every Perl Weekly Challenge tasks.

In the following, the assigned tasks for Challenge 166.

• Task 1
• Task 2

and for the sake of some Perl 5, let’s do some stuff also in PostgreSQL Pl/Perl:

• Task 1 in PostgreSQL Pl/Perl
• Task 2 in PostgreSQL Pl/Perl

Last, the solutions in PostgreSQL PL/PgSQL:

• Task 1 in PostgreSQL Pl/PgSQL
• Task 2 in PostgreSQL Pl/PgSQL

PWC 166 - Task 1

Given a dictionary, a text file with a word per-line, find out all the words that can be expressed as an hexadecimal string substituting some letters with numbers. Each word must be between two and eight chars. This is my Raku implementation:

sub MAIN( Str $filename = '../../data/dictionary.txt', Int $limit = 100 ) {
    my %substitutions = 'o' => 0, 'l' => 1, 'i' => 1, 's' => 5, 't' => 7;

    my @results = lazy gather {
        for $filename.IO.lines -> $word is copy {
            next if $word.chars > 8 || $word.chars < 2;
            next if $word.lc !~~ / ^ <[a..folist]>+ $ /;

            $word .= lc;
            my $src-word = $word;

            #$word .= subst( $_, %substitutions{ $_ } ) for %substitutions.keys;
            for %substitutions.kv -> $k, $v {
                $word ~~ s:g/$k/$v/;
            }

            take [ $src-word.uc, $word.uc ];
        }
    };


    @results[0 .. $limit].map( { "The word $_[0] is translated to $_[1]" } ).join( "\n" ).say;
}

I use a lazy gather to provide every new word. First, I skip any $word that has not hexadceimal letters (i.e, not in the range ‘a’..’f’) and that do not contain substituble letters.
Then I do an iteration over all possible substitutions and do a regexp global substitution (that is the same as the subst commented out line). Last, I return an array with the original word and the substituted one.
The last line does only a pretty printing of the result.

PWC 166 - Task 2

Given three directories, find out which file are not simultaneously available in all of them.

sub MAIN( Str $dir-a, Str $dir-b, Str $dir-c ) {


    my @dirs = $dir-a, $dir-b, $dir-c;
    my %files;

    # build the directory content
    for @dirs -> $current-dir {
        %files{ $_.basename  ~ ( $_.d ?? '/' !! '' ) }.push: $current-dir for $current-dir.IO.dir( test => { $current-dir.IO.d || $current-dir.IO.f } );
    }


    # print the stuff
    my $header = False;
    for %files.kv -> $file-name, $dir-names {
        # skip all entries that are there for all the directories!
        next if %files{ $file-name }.elems == @dirs.elems;

        "|%-20s|%-20s|%-20s|\n|%s|%s|%s|".sprintf( @dirs, ( '-' x 20 ) xx @dirs.elems ).say and $header = True if ! $header;
        printf "|%-20s", $dir-names.grep( $_ ) ?? $file-name !! ' '         for @dirs;
        print "|";

        say "";
    }

}

First of all, I scan every directory and place the files as keys in an %files hash. Each entry in the hash contains an array with the directories where the file is present. Therefore, given every file, the entries indicates in which directories the file is present.
Next I iterate over all the found files, and the number of directories where the file is found is the same as the number of directories, then I can skip its printing (i.e., the file is everywhere). On the other hand, I print a formatted string with the directory name or an empty string if the file is not within a directory.

PWC 166 - Task 1 in PostgreSQL PL/Perl

Same implementation as Raku, but using plperlu since there is the need to access the filesystem.



I miss auto-chomping! In fact, in the first implementation I was unable to get a word-per-line.

PWC 166 - Task 2 in PostgreSQL PL/Perl

Similar to the Raku implementation:



In this implementation I use opendir and readdir functions to get the listing (i.e., content) as filenames. One difference between the Raku implementation is that in Perl 5 readdir returns a relative filename.
Once the hash with the files is built, I iterate over all the files skipping those that are present ion the same number of occurrencies as the number of the directories, and then prefill an $hash with the values.
Since we need to keep the list of columns constant, I remap the directories into the $files hash with a trick like "dir_" . ( 'a' .. 'c' )[ $index ] that maps any directory name into dir_a thru dir_c.

PWC 166 - Task 1 in PostgreSQL PL/PgSQL

Just a nested query, no need to use PL/PgSQL:



The idea is to populate the dictionary table with the content of the dictionary.txt file (via COPY). Then the query selects all words within two and eight chars that match a case-insensitive regular expression with the hex letters (via ~*).
Then, I nest regexp_replace() calls to change every single letter globally.

PWC 166 - Task 2 in PostgreSQL PL/PgSQL

Another “simple” query:



The three CTEs provide a list of files within each directory using pg_ls_dir(). Then I do a full outer join on all the materialzed tables where any of the elements is NULL.
This implementation has a little drawback: it cannot discriminate between a file and a directory with the same name.