#!/usr/bin/perl

# Copyright (c) 2009 Jonathan Kamens <jik@alum.mit.edu>.

# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.

# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
# General Public License for more details.

# To read the full GNU General Public License, see
# <http://www.gnu.org/licenses/>.

# Converts a CSV generated by Outlook 2007 contacts export into a CSV
# that can more easily be imported by Thunderbird 3.
#
# Specify the name of the Outlook-generated CSV file on the command
# line, or feed it via stdin.
#
# Thunderbird-compatible CSV is generated to stdout.
#
# After generating the Thunderbird-compatible CSV and saving it to a
# *.csv file, run Tools > Import... in Thunderbird, tell it you want
# to import Address Books, then tell it you want to import a CSV file,
# then open the CSV file you generated.
#
# Then check the column mappings.  If they're wrong, it's probably
# because the Thunderbird you're using has different address book
# fields from mine, or puts them in a different order.  Adjust the
# @columns array below to compensate for this, generate a new CSV, and
# try again.

use strict;
use warnings;

use File::Basename;
use Getopt::Long;
use POSIX qw(strftime);
use Text::CSV;

my $whoami = basename $0;

# -2 means Thunderbird 2 compatibility.
# -3 means Thunderbird 3 compatibility.
my $usage = "Usage: $whoami [-2|-3]\n";
my($t2, $t3);

die $usage if (! GetOptions("2" => \$t2,
			    "3" => \$t3,
			    ));
die "Don't specify both -2 and -3\n$usage" if ($t2 && $t3);

# First item in each array is the Thunderbird field name.  Remainder
# of columns are Outlook fields to look in, or subroutine references
# to call.  First Outlook field that has a value, or the first
# subroutine to return a value, is the one that gets used.  The
# subroutine gets passed a hashref containing the Outlook fields.

my(@columns) = (
    [ 'First Name',		'First Name' ],
    [ 'Last Name',		'Last Name' ],
    [ 'Display Name',		\&display_name ],
    [ 'Nickname' ],
    [ 'Primary Email',		\&email1 ],
    [ 'Secondary Email',	\&email2 ],
    [ 'Screen Name' ],
    [ 'Work Phone',		'Business Phone' ],
    [ 'Home Phone',		'Home Phone' ],
    [ 'Fax Number',		'Business Fax', 'Home Fax', 'Other Fax' ],
    [ 'Pager Number',		'Pager' ],
    [ 'Mobile Number',		'Mobile Phone' ],
    [ 'Home Address',		'Home Street' ],
    [ 'Home Address 2',		'Home Street 2' ],
    [ 'Home City',		'Home City' ],
    [ 'Home State',		'Home State' ],
    [ 'Home ZipCode',		'Home Postal Code' ],
    [ 'Home Country',		'Home Country/Region' ],
    [ 'Work Address',		'Business Street' ],
    [ 'Work Address 2',		'Business Street 2' ],
    [ 'Work City',		'Business City' ],
    [ 'Work State',		'Business State' ],
    [ 'Work ZipCode',		'Business Postal Code' ],
    [ 'Work Country',		'Business Country/Region' ],
    [ 'Job Title',		'Job Title' ],
    [ 'Department',		'Department' ],
    [ 'Organization',		'Company' ],
    [ 'Web Page 1',		'Web Page' ],
    [ 'Web Page 2' ],
    [ 'Birth Year',		\&birth_year ],
    [ 'Birth Month',		\&birth_month ],
    [ 'Birth Day',		\&birth_day ],
    [ 'Custom 1',		'User 1' ],
    [ 'Custom 2',		'User 2' ],
    [ 'Custom 3',		'User 3' ],
    [ 'Custom 4',		'User 4' ],
    [ 'Notes',			'Notes' ],
    );

if ($t2) {
    # Thunderbird 2 doesn't have 'Screen Name' field
    @columns = grep($_->[0] ne 'Screen Name', @columns);
}
elsif ($t3) {
    # nothing to do here
}

my $csv = Text::CSV->new({binary => 1});
if (@ARGV) {
    open(STDIN, "<", $ARGV[0]) or die "open($ARGV[0]): $!\n";
    shift @ARGV;
    die "Extra arguments: @ARGV\n" if (@ARGV);
}

{
    my $header_ref = $csv->getline(\*STDIN) or die;
    $csv->column_names(@$header_ref);
}

$csv->combine(map($_->[0], @columns));
print $csv->string(), "\n";

while (my $row = $csv->getline_hr(\*STDIN)) {
    my @values;
    foreach my $colref (@columns) {
	my($thunderbird, @outlook) = @$colref;
	my $val = '';
	foreach my $outlook (@outlook) {
	    if (ref $outlook eq 'CODE') {
		$val = &$outlook($row);
	    }
	    else {
		$val = $row->{$outlook};
	    }
	    last if $val;
	}
	push(@values, $val);
    }
    $csv->combine(@values);
    print $csv->string(), "\n";
}

sub birth_parts {
    my($row) = @_;
    my($birthday) = $row->{Birthday};
    return('', '', '') if (! $birthday);
    return('', '', '') if ($birthday =~ m,^[0/]+$,);
    if ($birthday !~ m,^(\d+)/(\d+)/(\d+)$,) {
	warn "Invalid birthday: $birthday\n";
	return('', '', '');
    }
    my($month, $day, $year) = ($1, $2, $3);

    # Ugh!
    if ($year < 100) {
	my($this_year) = strftime("%Y", localtime());
	my($century) = $this_year - ($this_year % 100);
	$year += $century;
	$year -= 100 while ($year > $this_year);
    }
    return($year, $month, $day);
}

sub birth_year {
    (&birth_parts(@_))[0];
}

sub birth_month {
    (&birth_parts(@_))[1];
}

sub birth_day {
    (&birth_parts(@_))[2];
}

sub display_name {
    my($row) = @_;
    $row->{'First Name'} . ' ' . $row->{'Last Name'};
}

sub emails {
    my($row) = @_;
    # Return all email address, ignoring bogus Exchange addresses
    # without any '@' signs in them.
    grep(/\@/, map($row->{$_}, 'E-mail Address', 'E-mail 2 Address',
		   'E-mail 3 Address'));
}

sub email1 {
    (&emails(@_))[0] || '';
}

sub email2 {
    (&emails(@_))[1] || '';
}

# $Id: outlook-contacts-to-thunderbird.pl,v 1.6 2009/09/11 14:30:58 jik Exp $