man Mail::MboxParser::Mail () - Provide mail-objects and methods upon


Mail::MboxParser::Mail - Provide mail-objects and methods upon


See Mail::MboxParser for an outline on usage. Examples however are also provided in this manpage further below.


Mail::MboxParser::Mail objects are usually not created directly though, in theory, they could be. A description of the provided methods can be found in Mail::MboxParser.

However, go on reading if you want to use methods from MIME::Entity and learn about overloading.


new(header, body)
This is usually not called directly but instead by CWget_messages(). You could however create a mail-object manually providing the header and body each as either one string or as an array-ref representing the lines. Here is a common scenario: Retrieving mails from a remote POP-server using Mail::POP3Client and directly feeding each mail to CWMail::MboxParser::Mail->new:
    use Mail::POP3Client;
    use Mail::MboxParser::Mail;
    my $pop = new Mail::POP3Client (...);
    for my $i (1 .. $pop->Count) {
        my $msg = Mail::MboxParser::Mail->new( [ $pop->Head($i) ],
                                               [ $pop->Body($i) ] );
        $msg->store_all_attachments( path => '/home/user/dump' );
The above effectively behaves like an attachment-only retriever.
Returns the mail-header as a hash-ref with header-fields as keys. All keys are turned to lower-case, so CW$header{Subject} has to be written as CW$header{subject}. If a header-field occurs more than once in the header, the value of the key is an array_ref. Example:
    my $field = $msg->header->{field};
    print $field->[0]; # first occurance of 'field'
    print $field->[1]; # second one
Returns the From -line of the message.
This method returns the Received: -lines of the message as a list.
Returns a Mail::MboxParser::Mail::Body object. For methods upon that see further below. When called with the argument n, the n-th body of the message is retrieved. That is, the body of the n-th entity. Sets CW$mail->error if something went wrong.
This will return an index number that represents what Mail::MboxParser::Mail considers to be the actual (main)-body of an email. This is useful if you don't know about the structure of a message but want to retrieve the message's signature for instance:
        $signature = $msg->body($msg->find_body)->signature;
Changes are good that find_body does what it is supposed to do.
Returns a Mail::MboxParser::Mail::Convertable object. For details on what you can do with it, read Mail::MboxParser::Mail::Convertable.
Returns the specified raw field from the message header, that is: the fieldname is not stripped off nor is any decoding done. Returns multiple lines as needed if the field is Received or another multi-line field. Not case sensitive. CWget_field() always returns one string regardless of how many times the field occured in the header. Multiple occurances are separated by a newline and multiple whitespaces squeezed to one. That means you can process each occurance of the field thusly:
    for my $field ( split /\n/, $msg->get_field('received') ) {
        # do something with $field
Sets CW$mail->error if the field was not found in which case CWget_field() returns CWundef.
Returns a hash-ref with the two fields 'name' and 'email'. Returns CWundef if empty. The name-field does not necessarily contain a value either. Example:
        print $mail->from->{email};
On behalf of suggestions I received from users, from() tries to be smart when 'name'is empty and 'email' has the form ''. In this case, 'name' is set to First Name.
Returns an array of hash-references of all to-fields in the mail-header. Fields are the same as those of CW$mail->from. Example:
        for my $recipient ($mail->to) {
                print $recipient->{name} || "<no name>", "\n";
                print $recipient->{email};
The same 'name'-smartness applies here as described under CWfrom().
Identical with to() but returning the hash-refed Cc: -line. The same 'name'-smartness applies here as described under CWfrom().
Returns the message-id of a message cutting off the leading and trailing '<' and '>' respectively.
Returns the number of MIME-entities. That is, the number of sub-entitities actually. If 0 is returned and you think this is wrong, check CW$mail->log.
Either returns an array of all MIME::Entity objects or one particular if called with a number. If no entity whatsoever could be found, an empty list is returned. CW$mail->log instantly called after get_entities will give you some information of what internally may have failed. If set, this will be an error raised by MIME::Entity but you don't need to worry about it at all. It's just for the record.
Returns the body of the n-th MIME::Entity as a single string, undef otherwise in which case you could check CW$mail->error.
store_entity_body(n, handle => FILEHANDLE)
Stores the stringified body of n-th entity to the specified filehandle. That's basically the same as:
 my $body = $mail->get_entity_body(0);
 print FILEHANDLE $body;
and could be shortened to this:
 $mail->store_entity_body(0, handle => \*FILEHANDLE);
It returns a true value on success and undef on failure. In this case, examine the value of CW$mail->error since the entity you specified with 'n' might not exist.
store_attachment(n, options)
It is really just a call to store_entity_body but it will take care that the n-th entity really is a saveable attachment. That is, it wont save anything with a MIME-type of, say, text/html or so. Unless further 'options' have been given, an attachment (if found) is stored into the current directory under the recommended filename given in the MIME-header. 'options' are specified in key/value pairs:
    key:       | value:        | description:
    path       | relative or    | directory to store attachment
    (".")      | absolute       |
               | path           |
    encode     | encoding       | Some platforms store files 
               | suitable for   | in e.g. UTF-8. Specify the
               | Encode::encode | appropriate encoding here and
               |                | and the filename will be en-
               |                | coded accordingly.
    store_only | a compiled     | store only files whose file
               | regex-pattern  | names match this pattern
    code       | an anonym      | first argument will be the 
               | subroutine     | $msg-object, second one the 
               |                | index-number of the current
               |                | MIME-part
               |                | should return a filename for
               |                | the attachment
    args       | additional     | this array-ref will be passed  
               | arguments as   | on to the 'code' subroutine
               | array-ref      | as a dereferenced array
                            path => "/home/ethan/", 
                            code => sub {
                                        my ($msg, $n, @args) = @_;
                                        return $msg->id."+$n";
                            args => [ "Foo", "Bar" ]);
This will save the attachment found in the second entity under the name that consists of the message-ID and the appendix +1 since the above code works on the second entity (that is, with index = 1). 'args' isn't used in this example but should demonstrate how to pass additional arguments. Inside the 'code' sub, CW@args equals (Foo, Bar). If 'path' does not exist, it will try to create the directory for you. You can specify to save only files matching a certain pattern. To do that, use the store-only switch:
    $msg->store_attachment(1, path       => "/home/ethan/", 
                              store_only => qr/\.jpg$/i);
The above will only save files that end on '.jpg', not case-sensitive. You could also use a non-compiled pattern if you want, but that would make for instance case-insensitive matching a little cumbersome:
    store_only => '(?i)\.jpg$'
If you are working on a platform that requires a certain encoding for filenames on disk, you can use the 'encode' option. This becomes necessary for instance on Mac OS X which internally is UTF-8 based. If the filename contains 8bit characters (like the German umlauts or French accented characters as in 'é'), storing the attachment under a non-encoded name will most likely fail. In this case, use something like this:
    $msg->store_attachment(1, path => '/tmp', encode => 'utf-8');
See Encode::Supported for a list of encodings that you may use. Returns the filename under which the attachment has been saved. undef is returned in case the entity did not contain a saveable attachement, there was no such entity at all or there was something wrong with the 'path' you specified. Check CW$mail->error to find out which of these possibilities apply.
Walks through an entire mail and stores all apparent attachments. 'options' are exactly the same as in CWstore_attachement() with the same behaviour if no options are given. Returns a list of files that have been succesfully saved and an empty list if no attachment could be extracted. CW$mail->error will tell you possible failures and a possible explanation for that.
This method returns a mapping from attachment-names (if those are savable) to index-numbers of the MIME-part that represents this attachment. It returns a hash-reference, the file-names being the key and the index the value:
    my $mapping = $msg->get_attachments;
    for my $filename (keys %$mapping) {
        print "$filename => $mapping->{$filename}\n";
If called with a string as argument, it tries to look up this filename. If it can't be found, undef is returned. In this case you also should have an error-message patiently awaiting you in the return value of CW$mail->error. Even though it looks tempting, don't do the following:
    # BAD!
    for my $file (qw/file1.ext file2.ext file3.ext file4.ext/) {
        print "$file is in message ", $msg->id, "\n"  
            if defined $msg->get_attachments($file);
The reason is that CWget_attachments() is currently not optimized to cache the filename mapping. So, each time you call it on (even the same) message, it will scan it from beginning to end. Better would be:
    # GOOD!
    my $mapping = $msg->get_attachments;
    for my $file (qw/file1.ext file2.ext file3.ext file4.ext/) {
        print "$file is in message ", $msg->id, "\n" 
            if exists $mapping->{$file};
Returns the message as one string. This is the method that string overloading depends on, so these two are the same:
    print $msg;
    print $msg->as_string;


Mail::MboxParser::Mail implements an autoloader that will do the appropriate type-casts for you if you invoke methods from external modules. This, however, currently only works with MIME::Entity. Support for other modules will follow. Example:

        my $mb = Mail::MboxParser->new("/home/user/Mail/received");
        for my $msg ($mb->get_messages) {
                print $msg->effective_type, "\n";

CWeffective_type() is not implemented by Mail::MboxParser::Mail and thus the corresponding method of MIME::Entity is automatically called.

To learn about what methods might be useful for you, you should read the Access-part of the section PUBLIC INTERFACE in the MIME::Entity manpage. It may become handy if you have mails with a lot of MIME-parts and you not just want to handle binary-attachments but any kind of MIME-data.


Mail::MboxParser::Mail overloads the operator. Overloading operators is a fancy feature of Perl and some other languages ( for instance) which will change the behaviour of an object when one of those overloaded operators is applied onto it. Here you get the stringified mail when you write CW$mail while otherwise you'd get the stringified reference: CWMail::MboxParser::Mail=HASH(...).


This is version 0.54.


Tassilo von Parseval <>

Copyright (c) 2001-2005 Tassilo von Parseval. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.



Mail::MboxParser, Mail::MboxParser::Mail::Body, Mail::MboxParser::Mail::Convertable