package Courriel;
use 5.10.0;
use strict;
use warnings;
use namespace::autoclean;
our $VERSION = '0.42';
use Courriel::Headers;
use Courriel::Helpers qw( unique_boundary );
use Courriel::Part::Multipart;
use Courriel::Part::Single;
use Courriel::Types qw( ArrayRef Bool Headers Maybe Part Str StringRef );
use DateTime;
use DateTime::Format::Mail 0.403;
use DateTime::Format::Natural;
use Email::Address;
use Encode qw( encode );
use List::AllUtils qw( uniq );
use MooseX::Params::Validate 0.21 qw( validated_list );
use Moose;
use MooseX::StrictConstructor;
has top_level_part => (
is => 'rw',
writer => '_replace_top_level_part',
isa => Part,
init_arg => 'part',
required => 1,
handles => [
qw(
as_string
content_type
headers
is_multipart
stream_to
)
],
);
has subject => (
is => 'ro',
isa => Maybe [Str],
init_arg => undef,
lazy => 1,
builder => '_build_subject',
);
has datetime => (
is => 'ro',
isa => 'DateTime',
init_arg => undef,
lazy => 1,
builder => '_build_datetime',
);
has _to => (
traits => ['Array'],
isa => ArrayRef ['Email::Address'],
init_arg => undef,
lazy => 1,
builder => '_build_to',
handles => {
to => 'elements',
},
);
has _cc => (
traits => ['Array'],
isa => ArrayRef ['Email::Address'],
init_arg => undef,
lazy => 1,
builder => '_build_cc',
handles => {
cc => 'elements',
},
);
has from => (
is => 'ro',
isa => Maybe ['Email::Address'],
init_arg => undef,
lazy => 1,
builder => '_build_from',
);
has _participants => (
traits => ['Array'],
isa => ArrayRef ['Email::Address'],
init_arg => undef,
lazy => 1,
builder => '_build_participants',
handles => {
participants => 'elements',
},
);
has _recipients => (
traits => ['Array'],
isa => ArrayRef ['Email::Address'],
init_arg => undef,
lazy => 1,
builder => '_build_recipients',
handles => {
recipients => 'elements',
},
);
has plain_body_part => (
is => 'ro',
isa => Maybe ['Courriel::Part::Single'],
init_arg => undef,
lazy => 1,
builder => '_build_plain_body_part',
);
has html_body_part => (
is => 'ro',
isa => Maybe ['Courriel::Part::Single'],
init_arg => undef,
lazy => 1,
builder => '_build_html_body_part',
);
sub part_count {
my $self = shift;
return $self->is_multipart
? $self->top_level_part->part_count
: 1;
}
sub parts {
my $self = shift;
return $self->is_multipart
? $self->top_level_part->parts
: $self->top_level_part;
}
sub clone_without_attachments {
my $self = shift;
my $plain_body = $self->plain_body_part;
my $html_body = $self->html_body_part;
my $headers = $self->headers;
if ( $plain_body && $html_body ) {
my $ct = Courriel::Header::ContentType->new(
mime_type => 'multipart/alternative',
attributes => { boundary => unique_boundary },
);
return Courriel->new(
part => Courriel::Part::Multipart->new(
content_type => $ct,
headers => $headers,
parts => [ $plain_body, $html_body ],
)
);
}
elsif ($plain_body) {
return Courriel->new(
part => Courriel::Part::Single->new(
content_type => $plain_body->content_type,
headers => $headers,
encoding => $plain_body->encoding,
encoded_content => $plain_body->encoded_content,
)
);
}
elsif ($html_body) {
return Courriel->new(
part => Courriel::Part::Single->new(
content_type => $html_body->content_type,
headers => $headers,
encoding => $html_body->encoding,
encoded_content => $html_body->encoded_content,
)
);
}
die 'Cannot find a text or html body in this email!';
}
sub _build_subject {
my $self = shift;
my $subject = $self->headers->get('Subject');
return $subject ? $subject->value : undef;
}
{
my $mail_parser = DateTime::Format::Mail->new( loose => 1 );
my $natural_parser = DateTime::Format::Natural->new( time_zone => 'UTC' );
sub _build_datetime {
my $self = shift;
my @possible = (
( map { $_->value } $self->headers->get('Date') ),
(
reverse
map { $self->_find_date_received( $_->value ) }
$self->headers->get('Received')
),
( map { $_->value } $self->headers->get('Resent-Date') ),
);
# Stolen from Email::Date and then modified
for my $possible (@possible) {
next unless defined $possible && length $possible;
my $dt = eval { $mail_parser->parse_datetime($possible) };
unless ($dt) {
$dt = $natural_parser->parse_datetime($possible);
next unless $natural_parser->success;
}
$dt->set_time_zone('UTC');
return $dt;
}
return DateTime->now( time_zone => 'UTC' );
}
}
# Stolen from Email::Date and modified
sub _find_date_received {
shift;
my $received = shift;
return unless defined $received && length $received;
$received =~ s/.+;//;
return $received;
}
sub _build_to {
my $self = shift;
my @addresses = map { Email::Address->parse( $_->value ) }
$self->headers->get('To');
return $self->_unique_addresses( \@addresses );
}
sub _build_cc {
my $self = shift;
my @addresses = map { Email::Address->parse( $_->value ) }
$self->headers->get('CC');
return $self->_unique_addresses( \@addresses );
}
sub _build_from {
my $self = shift;
my @addresses = Email::Address->parse( map { $_->value }
$self->headers->get('From') );
return $addresses[0];
}
sub _build_recipients {
my $self = shift;
my @addresses = ( $self->to, $self->cc );
return $self->_unique_addresses( \@addresses );
}
sub _build_participants {
my $self = shift;
my @addresses = grep {defined} ( $self->from, $self->to, $self->cc );
return $self->_unique_addresses( \@addresses );
}
sub _unique_addresses {
my $self = shift;
my $addresses = shift;
my %seen;
return [ grep { !$seen{ $_->original }++ } @{$addresses} ];
}
sub _build_plain_body_part {
my $self = shift;
return $self->first_part_matching(
sub {
$_[0]->mime_type eq 'text/plain'
&& $_[0]->is_inline;
}
);
}
sub _build_html_body_part {
my $self = shift;
return $self->first_part_matching(
sub {
$_[0]->mime_type eq 'text/html'
&& $_[0]->is_inline;
}
);
}
sub first_part_matching {
my $self = shift;
my $match = shift;
my @parts = $self->top_level_part;
## no critic (ControlStructures::ProhibitCStyleForLoops)
for ( my $part = shift @parts; $part; $part = shift @parts ) {
return $part if $match->($part);
push @parts, $part->parts if $part->is_multipart;
}
}
sub all_parts_matching {
my $self = shift;
my $match = shift;
my @parts = $self->top_level_part;
my @match;
## no critic (ControlStructures::ProhibitCStyleForLoops)
for ( my $part = shift @parts; $part; $part = shift @parts ) {
push @match, $part if $match->($part);
push @parts, $part->parts if $part->is_multipart;
}
return @match;
}
{
my @spec = ( text => { isa => StringRef, coerce => 1 } );
# This is needed for Email::Abstract compatibility but it's a godawful
# idea, and even Email::Abstract says not to do this.
#
# It's much safer to just make a new Courriel object from scratch.
sub replace_body {
my $self = shift;
my ($text) = validated_list(
\@_,
@spec,
);
my $part = Courriel::Part::Single->new(
headers => $self->headers,
encoded_content => $text,
);
$self->_replace_top_level_part($part);
return;
}
}
{
my @spec = (
text => { isa => StringRef, coerce => 1 },
is_character => { isa => Bool, default => 0 },
);
sub parse {
my $class = shift;
my ( $text, $is_character ) = validated_list(
\@_,
@spec,
);
if ($is_character) {
${$text} = encode( 'UTF-8', ${$text} );
}
return $class->new( part => $class->_parse($text) );
}
}
sub _parse {
my $class = shift;
my $text = shift;
my ( $sep_idx, $headers ) = $class->_parse_headers($text);
substr( ${$text}, 0, $sep_idx, q{} );
return $class->_parse_parts( $text, $headers );
}
sub _parse_headers {
my $class = shift;
my $text = shift;
my $header_text;
my $sep_idx;
# We want to ignore mbox message separators - this is a pretty lax parser,
# but we may find broken lines. The key is that it starts with From
# followed by space, not a colon.
${$text} =~ s/^From\s+.+$Courriel::Helpers::LINE_SEP_RE//;
# Some broken emails may split the From line in an arbitrary spot
${$text} =~ s/^[^:]+$Courriel::Helpers::LINE_SEP_RE//g;
if ( ${$text} =~ /^(.+?)($Courriel::Helpers::LINE_SEP_RE)\g{2}/s ) {
$header_text = $1 . $2;
$sep_idx = ( length $header_text ) + ( length $2 );
}
else {
return ( 0, Courriel::Headers::->new );
}
my $headers = Courriel::Headers::->parse(
text => \$header_text,
);
return ( $sep_idx, $headers );
}
{
my $fake_ct = Courriel::Header::ContentType->new_from_value(
name => 'Content-Type',
value => 'text/plain'
);
sub _parse_parts {
my $class = shift;
my $text = shift;
my $headers = shift;
my @ct = $headers->get('Content-Type');
if ( @ct > 1 ) {
die 'This email defines more than one Content-Type header.';
}
my $ct = $ct[0] // $fake_ct;
if ( $ct->mime_type !~ /^multipart/i ) {
return Courriel::Part::Single->new(
headers => $headers,
encoded_content => $text,
);
}
return $class->_parse_multipart( $text, $headers, $ct );
}
}
sub _parse_multipart {
my $class = shift;
my $text = shift;
my $headers = shift;
my $ct = shift;
my $boundary = $ct->attribute_value('boundary');
die q{The message's mime type claims this is a multipart message (}
. $ct->mime_type
. q{) but it does not specify a boundary.}
unless defined $boundary && length $boundary;
my ( $preamble, $all_parts, $epilogue ) = ${$text} =~ /
(.*?) # preamble
^--\Q$boundary\E\s*
(.+) # all parts
^--\Q$boundary\E--\s*
(.*) # epilogue
/smx;
my @part_text;
if ( defined $all_parts ) {
@part_text = split /^--\Q$boundary\E\s*/m, $all_parts;
}
unless (@part_text) {
${$text} =~ s/^--\Q$boundary\E\s*//m;
push @part_text, ${$text};
}
return Courriel::Part::Multipart->new(
headers => $headers,
(
defined $preamble
&& length $preamble
&& $preamble =~ /\S/ ? ( preamble => $preamble ) : ()
),
(
defined $epilogue
&& length $epilogue
&& $epilogue =~ /\S/ ? ( epilogue => $epilogue ) : ()
),
boundary => $boundary,
parts => [ map { $class->_parse( \$_ ) } @part_text ],
);
}
__PACKAGE__->meta->make_immutable;
1;
# ABSTRACT: High level email parsing and manipulation
__END__
=pod
=encoding utf-8
=head1 NAME
Courriel - High level email parsing and manipulation
=head1 VERSION
version 0.42
=head1 SYNOPSIS
my $email = Courriel->parse( text => $raw_email );
print $email->subject;
print $_->address for $email->participants;
print $email->datetime->year;
if ( my $part = $email->plain_body_part ) {
print $part->content;
}
=head1 DESCRIPTION
This class exists to provide a high level API for working with emails,
particular for processing incoming email. It is primarily a wrapper around the
other classes in the Courriel distro, especially L<Courriel::Headers>,
L<Courriel::Part::Single>, and L<Courriel::Part::Multipart>. If you need lower
level information about an email, it should be available from one of these
classes.
=head1 API
This class provides the following methods:
=head2 Courriel->parse( text => $raw_email, is_character => 0|1 )
This parses the given text and returns a new Courriel object. The text can be
provided as a string or a reference to a string.
If you pass a reference, then the scalar underlying the reference I<will> be
modified, so don't pass in something you don't want modified.
By default, Courriel expects that content passed in text is binary data. This
means that it has not been decoded into utf-8 with C<Encode::decode()> or by
using a C<:encoding(UTF-8)> IO layer.
In practice, this doesn't matter for most emails, since they either contain
only ASCII data or they actually do contain binary (non-character)
data. However, if an email is using the 8bit Content-Transfer-Encoding, then
this does matter.
If the email has already been decoded, you must set C<is_character> to a true
value.
It's safest to simply pass binary data to Courriel and let it handle decoding
internally.
=head2 $email->parts()
Returns an array (not a reference) of the parts this email contains.
=head2 $email->part_count()
Returns the number of parts this email contains.
=head2 $email->is_multipart()
Returns true if the top-level part is a multipart part, false otherwise.
=head2 $email->top_level_part()
Returns the actual top level part for the object. You're probably better off
just calling C<< $email->parts() >> most of the time, since when the email is
multipart, the top level part is just a container.
=head2 $email->subject()
Returns the email's Subject header value, or C<undef> if it doesn't have one.
=head2 $email->datetime()
Returns a L<DateTime> object for the email. The DateTime object is always in
the "UTC" time zone.
This uses the Date header by default one. Otherwise it looks at the date in
each Received header, and then it looks for a Resent-Date header. If none of
these exists, it just returns C<< DateTime->now() >>.
=head2 $email->from()
This returns a single L<Email::Address> object based on the From header of the
email. If the email has no From header or if the From header is broken, it
returns C<undef>.
=head2 $email->participants()
This returns a list of L<Email::Address> objects, one for each unique
participant in the email. This includes any address in the From, To, or CC
headers.
Just like with the From header, broken addresses will not be included.
=head2 $email->recipients()
This returns a list of L<Email::Address> objects, one for each unique
recipient in the email. This includes any address in the To or CC headers.
Just like with the From header, broken addresses will not be included.
=head2 $email->to()
This returns a list of L<Email::Address> objects, one for each unique
address in the To header.
Just like with the From header, broken addresses will not be included.
=head2 $email->cc()
This returns a list of L<Email::Address> objects, one for each unique
address in the CC header.
Just like with the From header, broken addresses will not be included.
=head2 $email->plain_body_part()
This returns the first L<Courriel::Part::Single> object in the email with a
mime type of "text/plain" and an inline disposition, if one exists.
=head2 $email->html_body_part()
This returns the first L<Courriel::Part::Single> object in the email with a
mime type of "text/html" and an inline disposition, if one exists.
=head2 $email->clone_without_attachments()
Returns a new Courriel object that only contains inline parts from the
original email, effectively removing all attachments.
=head2 $email->first_part_matching( sub { ... } )
Given a subroutine reference, this method calls that subroutine for each part
in the email, in a depth-first search.
The subroutine receives the part as its only argument. If it returns true,
this method returns that part.
=head2 $email->all_parts_matching( sub { ... } )
Given a subroutine reference, this method calls that subroutine for each part
in the email, in a depth-first search.
The subroutine receives the part as its only argument. If it returns true,
this method includes that part.
This method returns all of the parts that match the subroutine.
=head2 $email->content_type()
Returns the L<Courriel::Header::ContentType> object associated with the email.
=head2 $email->headers()
Returns the L<Courriel::Headers> object for this email.
=head2 $email->stream_to( output => $output )
This method will send the stringified email to the specified output. The
output can be a subroutine reference, a filehandle, or an object with a
C<print()> method. The output may be sent as a single string, as a list of
strings, or via multiple calls to the output.
For large emails, streaming can be much more memory efficient than generating
a single string in memory.
=head2 $email->as_string()
Returns the email as a string, along with its headers. Lines will be
terminated with "\r\n".
=head1 ROBUSTNESS PRINCIPLE
Courriel aims to respect the common Internet robustness principle (aka
Postel's law). Courriel is conservative in the output it generates, and
liberal in what it accepts.
When parsing, the goal is to never die and always return as much information
as possible. Any input that causes the C<< Courriel->parse() >> to die means
there's a bug in the parser. Please report these bugs.
Conversely, Courriel aims to respect all relevant RFCs in its output, except
when it preserves the original data in a parsed email. If you're using
L<Courriel::Builder> to create emails from scratch, any output that isn't
RFC-compliant is a bug.
=head1 FUTURE PLANS
This release is still rough, and I have some plans for additional features:
=head2 More methods for walking all parts
Some more methods for walking/collecting multiple parts would be useful.
=head2 More?
Stay tuned for details.
=head1 WHY DID I WRITE THIS MODULE?
There a lot of email modules/distros on CPAN. Why didn't I use/fix one of them?
=over 4
=item * L<Mail::Box>
This one probably does everything this module does and more, but it's really,
really big and complicated, forcing the end user to make a lot of choices just
to get started. If you need it, it's great, but I generally find it to be too
much module for me.
=item * L<Email::Simple> and L<Email::MIME>
These are surprisingly B<not> simple. They suffer from a problematic API (too
high level in some spots, too low in others), and a poor separation of
concerns. I've hacked on these enough to know that I can never make them do
what I want.
=item * Everything Else
There's a lot of other email modules on CPAN, but none of them really seem any
better than the ones mentioned above.
=back
=head1 CREDITS
This module rips some chunks of code from a few other places, notably several
of the Email suite modules.
=head1 DONATIONS
If you'd like to thank me for the work I've done on this module, please
consider making a "donation" to me via PayPal. I spend a lot of free time
creating free software, and would appreciate any support you'd care to offer.
Please note that B<I am not suggesting that you must do this> in order for me
to continue working on this particular software. I will continue to do so,
inasmuch as I have in the past, for as long as it interests me.
Similarly, a donation made in this way will probably not make me work on this
software much more, unless I get so many donations that I can consider working
on free software full time, which seems unlikely at best.
To donate, log into PayPal and send money to autarch@urth.org or use the
button on this page: L<http://www.urth.org/~autarch/fs-donation.html>
=head1 BUGS
Please report any bugs or feature requests to C<bug-courriel@rt.cpan.org>, or
through the web interface at L<http://rt.cpan.org>. I will be notified, and
then you'll automatically be notified of progress on your bug as I make
changes.
Bugs may be submitted through L<the RT bug tracker|http://rt.cpan.org/Public/Dist/Display.html?Name=Courriel>
(or L<bug-courriel@rt.cpan.org|mailto:bug-courriel@rt.cpan.org>).
I am also usually active on IRC as 'drolsky' on C<irc://irc.perl.org>.
=head1 DONATIONS
If you'd like to thank me for the work I've done on this module, please
consider making a "donation" to me via PayPal. I spend a lot of free time
creating free software, and would appreciate any support you'd care to offer.
Please note that B<I am not suggesting that you must do this> in order for me
to continue working on this particular software. I will continue to do so,
inasmuch as I have in the past, for as long as it interests me.
Similarly, a donation made in this way will probably not make me work on this
software much more, unless I get so many donations that I can consider working
on free software full time (let's all have a chuckle at that together).
To donate, log into PayPal and send money to autarch@urth.org, or use the
button at L<http://www.urth.org/~autarch/fs-donation.html>.
=head1 AUTHOR
Dave Rolsky <autarch@urth.org>
=head1 CONTRIBUTORS
=for stopwords Gregory Oschwald Ricardo Signes Zbigniew Łukasiak
=over 4
=item *
Gregory Oschwald <goschwald@maxmind.com>
=item *
Ricardo Signes <rjbs@users.noreply.github.com>
=item *
Zbigniew Łukasiak <zzbbyy@gmail.com>
=back
=head1 COPYRIGHT AND LICENCE
This software is Copyright (c) 2016 by Dave Rolsky.
This is free software, licensed under:
The Artistic License 2.0 (GPL Compatible)
=cut