day_one_splitter.pl

#! /usr/bin/env perl
use warnings;
use strict;

use File::Spec::Functions qw( catfile );
use Getopt::Long;
use Pod::Usage;
use Time::Piece;

# Default options
my $output_dir      = 'split';
my $suffix          = '.md';
my $filename_format = '%F-%H%M';
my ($help, $man);

GetOptions(
    'dir=s'        => \$output_dir,
    'suffix=s'     => \$suffix,
    'fmt|format=s' => \$filename_format,
    'help'         => \$help,
    'man'          => \$man
) or pod2usage(2);

pod2usage(1) if $help;
pod2usage(-exitval => 0, -verbose => 2) if $man;

if ( ! -d $output_dir) {
    pod2usage(-message => "Destination directory $output_dir doesn't exist");
}

my $input = shift @ARGV;

pod2usage(-message => "No input file specified.") unless $input;
pod2usage(-message => "Input file $input doesn't exist") unless -r $input;

my $date_string;
my $fh;
my $filename;
my $in_header = 0;

open(my $input_fh, '<', $input);

while (<$input_fh>) {
    # There's indentation in Day One's header blocks
    # This is handy to help distinguish them from HTTP responses!
    if (/^\tDate:\s*(.*)\s+(GMT|BST)$/) {

        $date_string = $1;

        # At the start of a header block we have a new entry, so close any open files
        close($fh) if $fh;

        # Now convert the date string to something we can use in a filename
        $filename = Time::Piece->strptime($date_string, "%d %B %Y at %T")->strftime($filename_format);

        my $output_filename = catfile($output_dir, $filename) . $suffix;
        if (-e $output_filename) {
            warn "Output filename $output_filename exists. Refusing to overwrite it and quitting instead!\n";
            last;
        }

        open($fh, ">", $output_filename);

        print { $fh } "---\n";
    }

    # If we're no longer in a header block, we need to close it
    if ($in_header && $_ !~ /^\t/) {
        $in_header = 0;
        $_ .= "---\n\n";
    }

    # Clean up Day One's front matter
    if (/^\t(Location|Date|Weather)/) {
        $_ =~ s{^\s+}{};
        $_ =~ s{\t}{ };
        $_ = lcfirst $_;
        $in_header++;
    }

    print { $fh } $_;
}

close($input_fh);

__END__

=head1 NAME 

day_one_splitter - split Day One exports

=head1 SYNOPSIS 

day_one_splitter [options] filename

  Options:
      --dir		Output directory (must exist)
      --suffix		Output file suffix
      --format		Output filename format (strftime format)
      --help	        Show help.

=head1 OPTIONS

=over 4

=item B<--dir>

Specify an output directory. This must exist, and defaults to S<split>.

=item B<--suffix>

Specify the filename suffix to use when writing output files.
Defaults to S<.md>.

=item B<--format>

Specify the format to use for output filenames. Defaults to
S<%F-%H%M>, which leads to filenames that look like: 
S<2016-12-31-1410.md> with the default S<.md> suffix.

=item B<--help>

Print this help.

=item B<--man>

Print more help, and page it.

=back

=head1 ABOUT

This script takes a Day One output file and splits it into one file per post
(ignoring images). Use it to split exports up for easy import into other
software, or use with flat file based journalling systems.

This was written by Matt Foster S<mpf@hackerific.net>

=cut