Mail::Cclient - Mailbox access via the c-client library API

    use Mail::Cclient;
    $c = Mail::Cclient->new(MAILBOX [, OPTION ...]);

    # Old API deprecated in next versions

    ($envelope, $body) = $c->fetchstructure(MSGNO [, FLAG ...] );
    $hdr = $c->fetchheader(MSGNO [, LINES [, FLAG ...]]);
    $text = $c->fetchtext(MSGNO [, FLAG ...]);
    $text = $c->fetchbody(MSGNO, SECTION [, FLAG ...]);

    $c->fetchflags(SEQUENCE);
    $c->fetchfast(SEQUENCE);

    # New API

    ($envelope, $body) = $c->fetch_structure(MSGNO [, FLAG ...] );
    $msg = $c->fetch_message(MSGNO [, FLAG ...]);
    $hdr = $c->fetch_header(MSGNO, SECTION [, LINES [, FLAG ...]]);
    $text = $c->fetch_text(MSGNO, SECTION [, FLAG ...]);
    $mime = $c->fetch_mime(MSGNO, SECTION [, FLAG ...]);
    $text = $c->fetch_body(MSGNO, SECTION [, FLAG ...]);

    $c->fetch_flags(SEQUENCE);
    $c->fetch_fast(SEQUENCE);

    $elt = $c->elt(MSGNO);

    $c->create(MAILBOX);
    $c->delete(MAILBOX);
    $c->rename(OLDNAME, NEWNAME);
    $c->open(MAILBOX);

    $nmsgs = $c->nmsgs;

    Mail::Cclient::set_callback KEY => CODE, ...
    $c->list(REF, PAT);                 # uses "list" callback
    $c->scan(REF, PAT, CONTENTS);
    $c->lsub(REF, PAT);
    $c->subscribe(MAILBOX);
    $c->unsubscribe(MAILBOX);
    $c->status(MAILBOX [, FLAG ...]);   # uses "status" callback

    $c->copy(SEQUENCE, MAILBOX [, FLAG ...]);
    $c->move(SEQUENCE, MAILBOX [, FLAG ...]);
    $c->append(MAILBOX, MESSAGE [, DATE [, FLAGS]);

    $arrayref = $c->sort(
        SORT      => ["crit_1", rev_1, ..., "crit_n", rev_n], # (n <= 7)
        CHARSET   => "MIME character",
        SEARCH    => "string",
        FLAG      => [flag_1, ..., flag_n] / "flag" );
    $c->search(                         # uses "searched" callback
        SEARCH    => "string",
        CHARSET   => "MIME character",
        FLAG      => ["flag_1", ..., "flag_n"] / "flag");
    $c->search_msg(MSGNO, CRITERIA [, CHARSET]);
    $arrayref = $c->thread(
        THREADING => "orderedsubject/references",
        CHARSET   => "MIME character",
        SEARCH    => "string",
        FLAG      => "flag");

    $c->ping;
    $c->check;                          # uses "log" callback
    $c->expunge;                        # uses "expunge" callback
    $uid = $c->uid(MSGNO);
    $msgno = $c->msgno(uid);
    $c->setflag(SEQUENCE, MAILFLAG [, FLAG ...]);
    $c->clearflag(SEQUENCE, MAILFLAG [, FLAG ...]);

    $c->gc( [FLAG, ...] );
    $c->debug;
    $c->nodebug;
    $c->set_sequence(SEQUENCE);
    $c->uid_set_sequence(SEQUENCE);
    $result = $c->parameters(PARAM);
    $c->parameters(PARAM1 => NEWVALUE1 [, PARAM2 => NEWVALUE2 ...]);

    $c->close;

    $smtp = Mail::Cclient::SMTP->new(
        hostlist => [HOSTS] / "HOST",
        service  => SERVICE,
        port     => PORT,
        options  => ["option_1", ..., "option_n"] / "option");

    $smtp->mail(
        transaction => MAILTYPE,
        defaulthost => HOST,
        filehandle  => FILEHANDLE,
        envelope    => { header => value, ... },
        body        => {HASHREF} );

    $smtp->debug();
    $smtp->nodebug();

    $smtp->close();

    Mail::Cclient::rfc822_output(HASHREF);

    $text = Mail::Cclient::rfc822_base64(SOURCE);
    $text = Mail::Cclient::rfc822_binary(SOURCE);
    $text = Mail::Cclient::rfc822_qprint(SOURCE);
    $text = Mail::Cclient::rfc822_8bit(SOURCE);

    $text = Mail::Cclient::utf8_mime2text(SOURCE);

    $str = Mail::Cclient::rfc822_date;

    $str = Mail::Cclient::rfc822_write_address(MAILBOX, HOST, PERSONAL);
    $str = Mail::Cclient::rfc822_parse_adrlist(ADDRESSES, HOST);

CWMail::Cclient gives access to mailboxes in many different formats (including remote IMAP folders) via the c-client API. The c-client library is the core library used by Pine and the University of Washington IMAP daemon (written by Mark Crispin).

The Perl API is mostly taken directly from the underlying c-client library with minor adjustments in places where Perl provides a more natural interface. The primary interface is an object oriented one via the CWMail::Cclient class but certain methods implicitly use callbacks set via the set_callback function.

The c-client library often provides a given piece of functionality by two different function names: one for simple usage under a name of the form CWmail_foo and one with an additional flags arguments under a name of the form CWmail_foo_full. The corresponding functionality is available from the Perl CWMail::Cclient class under the single name CWfoo.

Setting a flag bit such as CWFT_UID in the c-client API is done in Perl by appending an extra argument CW"uid" to the method call.

Arguments to c-client functions which are there only to pass or receive the length of a string argument or result are not present in the Perl API since Perl handles them automatically.

Some methods take arguments which refer to a message or messages in a mailbox. An argument denoted MSGNO is a number that refers to a single message. Message 1 refers to the first message in the mailbox, unless the uid option is passed as as additional argument in which case the number refers to the uid of the message. An argument denoted SEQUENCE refers to a list of messages and is a string such as '1,3,5-9,12'.

Creating a CWMail::Cclient object and associating a mailstream with it is done with the CW"new" constructor method (whereas the underlying c-client library uses the CWmail_open function).

new(MAILBOX [, OPTION ...])

Return a new C-Client object. The MAILBOX argument can be anything accepted by the underlying c-client library. This includes the following forms This is the driver-dependent INBOX folder.

an absolute filename

This specifies a mailbox in the default format (usually Berkeley format for most default library builds) These refer to an IMAP folder held on host. The folder part can be INBOX to reference the distinguished INBOX folder that the IMAP protocol defines. The username and password required for logging in to the IMAP server are obtained by using the login callback (q.v.). This refers to folder /bar/baz which is held in a non-default mailbox format handled by the c-client driver foo (e.g. mbx).

! The options which can be passed to the CWnew are as follows (taken almost verbatim from the c-client Internal.doc file):

debug

Log IMAP protocol telemetry through debug callback (q.v.).

readonly

Open mailbox read-only.

anonymous

Don't use or update a .newsrc file for news.

shortcache

Don't cache envelopes or body structures.

prototype

Return the prototype stream for the driver associated with this mailbox instead of opening the stream.

halfopen

For IMAP and NNTP names, open a connection to the server but don't open a mailbox.

expunge

Silently expunge the oldstream before recycling.

secure

Don't do non-secure authentication

tryssl

Try SSL first

! Read-only access to the fields of the underlying mailstream of a CWMail::Cclient object is supplied by the following methods:

nmsgs

Returns the number of messages in the mailbox.

mailbox

Returns the mailbox name.

rdonly

Stream is open read-only.

anonymous

Stream is open with anonymous access.

halfopen

Stream is half-open; it can be reopened or used for functions that don't need a open mailbox such as CW$c->create() but no message data can be fetched.

perm_seen, perm_deleted, perm_flagged, perm_answered, perm_draft

The relevant flag can be set permanently.

kwd_create

New user flags can be created by referencing them in setflag or clearflag method calls. This can change during a session (e.g. if there is a limit).

perm_user_flags

Returns a list of the user flags which can be set permanently.

recent

Number of recent messages in current mailbox.

uid_validity

The UID validity value.

uid_last

The highest currently assigned UID in the current mailbox.

! You can use the method

open(MAILBOX)

to get the mailstream object to open a different mailbox. The cclient library will try to reuse the same IMAP connection where possible in the case of IMAP mailboxes but the host part of the mailbox spec must be given exactly as in the original connection for this to work.

list(REF, PAT)

This method returns a list of mailboxes via the list() callback. REF should normally be just the server specification ({mailserver:110/imap}), and PAT (pattern) specifies where in the mailbox hierarchy to start searching. If you want all mailboxes, pass '*' for PAT. There are two special characters you can pass as part of the PAT: * and %. * means to return all mailboxes. If you pass PAT as *, you will get a list of the entire mailbox hierarchy. % means to return the current level only. % as the PAT parameter will return only the top level mailboxes; ~/mail/% on UW_IMAPD will return every mailbox in the ~/mail directory, but none in subfolders of that directory.

scan(REF, PAT, CONTENTS)

This method takes a string to search for in the text of the mailbox.

lsub(REF, PAT)

This method returns a list of subscribed mailboxes via the lsub() callback.

subscribe(MAILBOX)

This method adds the given name to the subscription list. If unsuccessful, an error message is returned via the log() callback.

unsubscribe(MAILBOX)

This method removes the given name from the subscription list. If unsuccessful, an error message is returned via the log() callback.

create(MAILBOX)

This method creates a mailbox with the given name. If unsuccessful, an error message is returned via the log() callback.

delete(MAILBOX)

This method deletes the named mailbox. If unsuccessful, an error message is returned via the log() callback.

rename(OLDNAME, NEWNAME)

This method renames the old mailbox to the new mailbox name. If unsuccessful, an error message is returned via the log() callback.

status(MAILBOX [, FLAG ...])

This method provides status information about MAILBOX. The information calculated is limited to those mentioned in FLAG arguments and is returned via the status callback (q.v.). The FLAG arguments possible are precisely those mentioned in the documentation below for the status callback.

close()

This method closes the MAIL stream.

! The following methods provide access to messages themselves: their headers, structures and the text of their bodies (or parts thereof).

fetch_fast(SEQUENCE)

This method causes a cache load of all the fast information (internal date, RFC 822 size, and flags) for the given sequence.

fetch_flags(SEQUENCE)

This method causes a fetch of the flags for the given sequence.

fetch_structure(MSGNO [, FLAG ...] )

This returns the equivalent of what c-client calls an ENVELOPE structure for message MSGNO. If called in an array context then the equivalent of a BODY structure is passed as a second return value. The ENVELOPE structure is in the form of a Perl object of class CWMail::Cclient::Envelope. The BODY structure is in the form of a Perl object of class CWMail::Cclient::Body. See later on for the description of these objects. The flag "uid" can be passed which makes the MSGNO argument be interpreted as a message uid.

fetch_message(MSGNO [, FLAG ...])

This fetch a complete message (headers + body) from an IMAP server (a message uid if the flag "uid" is included).

fetch_header(MSGNO, SECTION [, LINES [, FLAG ...]])

This returns the message header in RFC822 format (as a single string) of message MSGNO (which is interpreted as a message uid if the flag "uid is included). The SECTION specification is a string of integers delimited by period (#.#.#...#) and if is specifed then return only the header section. With no LINES argument, all headers are put into the string. If an array ref argument is passed then it is taken to be a reference to a list of header names. Those headers are the ones that are included in the result, unless the flag not is passed in which case all headers are included except those in the list. The flag internal can be passed to avoid canonicalising the header texts. The flag prefetchtext" can be passed to pre-fetch the RFC822.TEXT part of the message at the same time.

fetch_mime(MSGNO, SECTION [, FLAG ...])

This returns the message text of message MSGNO (which is interpreted as a message uid if the flag "uid is included) or a message text section if a valid SECTION is provided. The body is returned as a MIME single string. The SECTION specification is a string of integers delimited by period (#.#.#...#). The flag internal can be passed to avoid canonicalising of internal" format to CRLF newlines.

fetch_text(MSGNO, SECTION [, FLAG ...])

This returns the body of message MSGNO (a message uid if the flag "uid is included) or a message text section if a valid SECTION is provided. The whole body is returned as a single string with no MIME processing done. Line endings are canonicalised to CRLF unless the internal flag is included. If the peek" flag is included then the \Seen flag is not actively set (though it may already have been set previously, of course).

fetch_body(MSGNO, SECTION, [, FLAG ...])

This returns a single (MIME) section of message MSGNO (a message uid if the flag "uid is included). The SECTION argument determines which section is returned and is a string in the form of a dot-separated list of numbers. See the IMAP specification for details. As an example, a multipart/mixed MIME message has sections 1, 2, 3 and so on. If section 3 is multipart/mixed itself, then it has subsections 3.1, 3.2 and so on. The peek and internal" flags may also be passed and have the same effect as in CWfetch_text documented above.

uid(MSGNO)

Returns the uid associated with message MSGNO.

elt(MSGNO)

This returns the MESSAGECACHE (commonly known as elt) information associated with message MSGNO as an object in class Mail::Cclient::Elt. See below for what such an object contains. Important note: for this method to be valid, a previous CWfetch_structure or CWfetch_flags must have been called on this message. Otherwise, you are into the realms of undefined behaviour and at the mercy of the underlying c-client library.

setflag(SEQUENCE, MAILFLAG [, FLAG ...])

Sets flag MAILFLAG on each message in SEQUENCE (taken to be a sequence of message uids if the "uid flag is passed). The silent" flag causes the local cache not to be updated.

clearflag(SEQUENCE, MAILFLAG [, FLAG ...])

Clears flag MAILFLAG from each message in SEQUENCE (taken to be a sequence of message uids if the "uid flag is passed). The silent" flag causes the local cache not to be updated.

search(KEY => VALUE [, KEY1 => VALUE1 ...])

Search for messages satisfying SEARCH keyword value. The searched callback (q.v.) is called for each matching message. The SEARCH keyword value is a string containing a search specification as defined in item SEARCH Command (6.4.4.) of RFC2060 (imap2000/docs/rfc/rfc2060.txt). The SEARCH keyword value is like 'ANSWERED TO malcolm' or 'FLAGGED SINCE 1-Feb-1994 NOT FROM Smith' (See the searching criteria section for details). The CHARSET keyword value is a MIME character set to use when searching strings. The FLAG keyword value can be a array reference or a string. Flags: "uid return a message uid's instead of sequence numbers; searchfree return the search program to free storage after finishing (internal use only); noprefetch" don't prefetch searched messages. The SEARCH keyword/value is not optional. All others keywords are optional.

sort(KEY => VALUE [, KEY1 => VALUE1 ...])

Returns an array reference of message numbers sorted by the given pairs of parameters (KEY => VALUE). The SORT keyword value is a array reference, and the argument crit_1, ..., crit_n, is a string and can be one of the following: date, arrival, from, subject, to, cc, size. The argument rev_1, ... rev_n is 0 or 1 if reverse sorting. The CHARSET keyword value is a MIME character set to use when sorting strings. The SEARCH keyword value is a string like, ALL, SEEN, UNSEEN, ANSWERED, UNANSWERED, FLAGGED, UNFLAGGED, SEARCHED or like SEARCH keyword in search method and return only messages that meet specified search criteria (See the searching criteria section for details). The FLAG keyword value can be a array reference or a string. Flags: "uid return uid's instead of sequence numbers; searchfree return the search program to free storage after finishing (internal use only); noprefetch don't prefetch searched messages; sortfree" return the sort program to free storage after finishing (internal use only). The SORT keyword/value is not optional. All others keywords are optional.

thread(KEY => VALUE [, KEY1 => VALUE1 ...])

This method returns a array reference of message sequence numbers and/or lists of lists of message numbers. The THREADING keyword value can take one of following strings values: "orderedsubject or references. The orderedsubject algorithm sorts by subject with a secondary sort of message date, and then for sets of messages with identical subjects. The references" algorithm threads the searched messages by grouping them together in parent/child relationships based on which messages are replies to others. The CHARSET keyword value is a MIME character set to use when searching strings. The SEARCH keyword value accepts a string ala search() method. The FLAG keyword value is a string. Flags: "uid return a message uid's instead of sequence numbers. All keywords are optional. If you don't use any keyword, in that case the default value is orderedsubject". Example: ARRAYREF = [2, [3, 6, [4, 23], [44, 7, 96]]

The following are miscellaneous methods.

ping

Checks where the mailstream is still alive: used as a keep-alive and to check for new mail.

check

Performs a (driver-dependent) checkpoint of the mailstream (not a check for new mail). Information about the checkpoint is passed to the log callback (q.v.).

expunge

Expunges all message marked as deleted in the mailbox. Calls the expunged callback (q.v.) on each such message and logging information is passed to the log callback. Decrementing message numbers happens after each and every message is expunged. As the example in the c-client documentation for mail_expunge says, if three consecutive messages starting at msgno 5 are expunged, the expunged callback will be called with a msgno of 5 three times.

! A message may be copied or moved into another mailbox with the methods CWcopy and CWmove. These methods only allow the destination mailbox to be of the same type as (and on the same host as) the mailstream object on which the methods are called.

copy(MSGNO, MAILBOX [, FLAGS])

This copies message MSGNO (a message uid if the "uid flag is included) to mailbox MAILBOX. If the move" flag is included then the message is actually moved instead (for compatibility with the CP_MOVE flag of the underlying c-client CWmail_copy_full function).

move(MSGNO, MAILBOX [, FLAGS])

This moved message MSGNO (a message uid if the "uid" flag is included) to mailbox MAILBOX.

append(MAILBOX, MESSAGE [, DATE [, FLAGS])

Append a raw message (MESSAGE is an ordinary string) to MAILBOX, giving it an optional date and FLAGS (again, simply strings).

gc( [FLAG, ...] )

Garbage collects the cache for the mailstream. The FLAG arguments, "elt, env, texts", determine what is garbage collected.

debug

Enables debugging for the mailstream, logged via the dlog callback (q.v.).

nodebug

Disables debugging for the mailstream.

set_sequence(SEQUENCE)

Sets the sequence bit for each message in SEQUENCE (and turns it off for all other messages). This has been renamed for Perl from the underlying c-client function CWmail_sequence to avoid clashing with the sequence field member of the mailstream object.

uid_set_sequence(SEQUENCE)

Sets the sequence bit for each message referenced by uid in SEQUENCE (and turns it off for all other messages). This has been renamed for Perl from the underlying c-client function CWmail_uid_sequence for consistency with CWset_sequence above.

search_msg(MSGNO, CRITERIA [, CHARSET])

It inspects the message MSGNO on that stream to see if it matches the CRITERIA or not. If it matches, 1 is returned, else 0.

msgno(UID)

Returns the msgno associated with message UID.

parameters(PARAM [, => NEWVALUE [, PARAM2 => NEWVALUE2 ...]])

With a single argument, gets the current value of parameter PARAM. With one or more pairs of PARAM => VALUE arguments, sets those PARAM values to the given new values. PARAM can be one of the following strings: USERNAME, HOMEDIR, LOCALHOST, SYSINBOX, OPENTIMEOUT, READTIMEOUT, WRITETIMEOUT, CLOSETIMEOUT, RSHTIMEOUT, SSHTIMEOUT, SSLFAILURE, MAXLOGINTRIALS, LOOKAHEAD, IMAPPORT, PREFETCH, CLOSEONERROR, POP3PORT, UIDLOOKAHEAD, MBXPROTECTION, DIRPROTECTION, LOCKPROTECTION, FROMWIDGET, NEWSACTIVE, NEWSSPOOL, NEWSRC, DISABLEFCNTLLOCK, LOCKEACCESERROR, LISTMAXLEVEL, ANONYMOUSHOME.

new(KEY => VALUE [, KEY1 => VALUE1 ...])

Creates the SMTP object and opens an SMTP connection to a one of the hosts in the hostlist and if successful returns a stream suitable for use by the other SMTP methods. The hostlist keyword/value is not optional. All others keywords are optional. The following SMTP open options are available: * debug - Debug protocol negotiations. If debug is defined, protocol telemetry is logged via mm_dlog(). * dsn - DSN notification, none set mean NEVER * dsn_notify_failure * dsn_notify_delay * dsn_notify_success - DSN return full msg vs. header * dsn_return_full * 8bitmime - 8-bit MIME requested * secure - Don't do non-secure authentication * tryssl - Try SSL first * tryalt - Old name for tryssl

mail(OPTION => VALUE, ...)

The options which can be passed to the mail are as follows This option set the type of SMTP transaction (one of MAIL, SEND, SAML, or SOML) to deliver the specified message. Default host name if an address lacks an CW@host.

filehandle => HANDLE

Filehandle to write output message described with the given envelope and body.

envelope => { header => value, ... }

The following headers are available for envelope message: from, to, cc, bcc, sender, reply_to, return_path, in_reply_to, message_id, subject, remail, date, newsgroups, followup_to, references

body => { option => value, ... }

The following fields are available for body message:

content_type

The content_type is a string like type/subtype. Type is the MIME type of the message. For example, text, multipart, message, application, audio, image, video, model, ... Subtype is the MIME subtype of the message. For example, PLAIN, HTML, MIXED, RELATED, or ALTERNATIVE.

encoding

Encoding of a single part message or the body of a multipart message. The MIME encoding can be one of the following strings: 7bit, 8bit, binary, base64, quoted-printable.

part => [{part1}, {part2}, ...]

For MIME type multipart messages The content disposition of the body: a reference to a list consisting of the disposition type followed by a (possibly empty) list of parameter.

parameter => [{attribute => attr_val1 }, {value => val_val1}, ...]

List of hash references.

description

The MIME description of the body part. body language list reference (header Content-Language: ).

location

body content URI (header Content-Location: ).

id

The message ID (header Content-ID: )

md5

The MD5 checksum of the body (header Content-MD5: ).

data

Message data of part body.

path

Pathname of file to attach. For example /usr/local/images/image.jpg. If you use data in hash then don't use path.

close()

This method closes the SMTP stream.

debug

This method enables SMTP protocol telemetry logging for this stream. All SMTP protocol operations are passed to the application via the dlog() facility.

nodebug

This method disables SMTP protocol telemetry

The following are utility functions (not methods).

Mail::Cclient::rfc822_base64(SOURCE)

Returns the SOURCE text converted to base64 format.

Mail::Cclient::rfc822_binary(SOURCE)

Convert binary contents to base64

Mail::Cclient::rfc822_qprint(SOURCE)

Returns the SOURCE text converted to quoted printable format.

Mail::Cclient::rfc822_8bit(SOURCE)

Convert 8bit contents to quoted-printable text

Mail::Cclient::utf8_mime2text(SOURCE)

Returns the MIME-2 SOURCE text converted to UTF-8.

Mail::Cclient::rfc822_date()

Returns the current date in RFC822 format.

Mail::Cclient::rfc822_write_address(MAILBOX, HOST, PERSONAL)

This function return an RFC 822 format address string based on the information from MAILBOX, HOST and PERSONAL.

Mail::Cclient::rfc822_parse_adrlist(ADDRESSES, HOST)

This function parses the string of ADDRESSES into an address list array ref. Any addresses missing a host name are have the host name defaulted from the HOST argument. Any parsing errors are noted via the log() callback.

Mail::Cclient::rfc822_output(HASHREF)

This function writes the message described with the given envelope and body. The options which can be passed to the rfc822_output are as follows:

filehandle => HANDLE

envelope => { header => value, ... }

body => { option => value, ... }

Certain methods mentioned above use callbacks to pass or receive extra information. Each callback has a particular name (e.g. log, dlog, list, login) and can be associated with a particular piece of Perl code via the CWMail::Cclient::set_callback function (available for export by the CWMail::Cclient class). The CWset_callback function takes pairs of arguments NAME, CODE for setting callback NAME to be the given CODE, a subroutine reference. The only callback which is required to be set and the only callback whose return value matters is the login callback (only used when the new method constructs an IMAP mailstream). Apart from that case, callbacks which have not been set are ignored. A callback set to undef is also ignored.

searched(STREAM, MSGNO)

This callback is invoked for each message number satifying the CRITERIA of the search method, defined above.

exists(STREAM, MSGNO)

expunged(STREAM, MSGNO)

flags(STREAM, MSGNO)

notify(STREAM, STRING, ERRFLAG)

list(STREAM, DELIMITER, MAILBOX [, ATTR ...])

lsub(STREAM, DELIMITER, MAILBOX [, ATTR ...])

status(STREAM, MAILBOX, [, ATTR, VALUE] ...)

Attribute values passed can be messages, recent, unseen, uidvalidity, uidnext.

log(STRING, ERRFLAG)

dlog(STRING)

fatal(STRING)

login(NETMBXINFO, TRIAL)

The login callback is invoked when the c-client library is opening an IMAP mailstream and needs to find out the username and password required. This callback must return precisely two values in the form (USERNAME, PASSWORD). TRIAL is the number of the current login attempt (starting at 1). NETMBXINFO is a hash reference with the following keys:

host

The hostname of the IMAP server.

user

The username requested.

mailbox

The mailbox name requested.

service

port

anoflag

Set to 1 if anonymous access has been requested otherwise this key is not created at all.

dbgflag

Set to 1 if debugging access has been requested otherwise this key is not created at all.

secflag

Set to 1 if secure access has been requested otherwise this key is not created at all.

sslflag

trysslflag

novalidate

critical(STREAM)

nocritical(STREAM)

diskerror(STREAM, ERRCODE, SERIOUS)

The results of the CWfetch_structure and CWelt methods involve objects in the classes CWMail::Cclient::Envelope, CWMail::Cclient::Body, CWMail::Cclient::Address and CWMail::Cclient::Elt. These will be referred to as Envelope, Body, Address and Elt objects respectively. These objects are all read-only and only have methods for picking out particular fields.

An Address object represents a single email address and has the following fields, available as methods or, if Perl 5.005 or later is being used, as pseudo-hash keys.

personal

The personal phrase of the address (i.e. the part contained in parentheses or outside the angle brackets).

adl

The at-domain-list or source route (not usually used).

mailbox

The mailbox name (i.e. the part before the @ which is usually a username or suchlike).

host

The hostname (i.e. the part after the @).

error

Only set if the address has delivery errors when CWsmtp_mail is called. Since that function hasn't been implemented in the Perl module yet, this isn't any use.

An Envelope object represents a structured form of the header of a message. It has the following fields, available as methods or, if Perl 5.005 or later is being used, as pseudo-hash keys.

remail, date, subject, in_reply_to, message_id, newsgroups, followup_to, references,

These are all strings.

return_path, from, sender, reply_to, to, cc, bcc

These are all references to lists which contain one or more Address objects.

A Body object represents the structure of a message body (not its contents).It has the following fields, available as methods or, if Perl 5.005 or later is being used, as pseudo-hash keys.

type

The MIME type (as a string) of the message (currently in uppercase as returned from the c-client library). For example, TEXT or MULTIPART.

encoding

The MIME encoding (as a string) of the message.

subtype

The MIME subtype (as a string) of the message. For example, PLAIN, HTML or MIXED.

parameter

A reference to a list of MIME parameter key/value pairs.

id

The message ID.

description

The MIME description of the body part.

nested

If (and only if) the body is of MIME type multipart, then this field is a reference to a list of Body objects, each representing one of the sub parts of the message. If (and only if) the body is of MIME type message/rfc822, then this field is a reference to a list of the form (ENVELOPE, BODY) which are, respectively, the Body and Envelope objects referring to the encapsulated message. If the message is not of MIME type multipart or message/rfc822 then this field is undef.

lines

The size in lines of the body.

bytes

The size in bytes of the body.

md5

The MD5 checksum of the body.

disposition

The content disposition of the body: a reference to a list consisting of the disposition type followed by a (possibly empty) list of parameter key/value pairs.

These have fields containing flag information for a given message, along with internal date information and the RFC822 message size.

msgno

The message number.

date

This contains the internal date information (spread about a series of bitfields in the underlying c-client library C structure) in the form of a string:

    yyyy-mm-dd hh:mm:ss [+-]hhmm

flags

A reference to a list of flags associated with the message. The flags are in the forms of their RFC2060 names (e.g. \Deleted, \Seen) for official flags and the user-chosen name for user-defined flags.

rfc822_size

The RFC822 size of the message.

imapdate

This contains the internal date information (spread about a series of bitfields in the underlying c-client library C structure) in the form of a string:

    dd-mmm-yyyy hh:mm:ss [+-]hhmm

This form is needed by IMAP servers.

Searching criteria consist of one or more search keys. When multiple keys are specified, the result is the intersection (AND function) of all the messages that match those keys. A search key can also be a parenthesized list of one or more search keys. In all search keys that use strings, a message matches the key if the string is a substring of the field. The matching is case-insensitive.

The defined search keys are as follows.

<message set>

Messages with message sequence numbers corresponding to the specified message sequence number set.

ALL

All messages in the mailbox; the default initial key for ANDing.

ANSWERED

Messages with the \Answered flag set.

BCC <string>

Messages that contain the specified string in the envelope structure's BCC field.

BEFORE <date>

Messages whose internal date is earlier than the specified date.

BODY <string>

Messages that contain the specified string in the body of the message.

CC <string>

Messages that contain the specified string in the envelope structure's CC field.

DELETED

Messages with the \Deleted flag set.

DRAFT

Messages with the \Draft flag set.

FLAGGED

Messages with the \Flagged flag set.

FROM <string>

Messages that contain the specified string in the envelope structure's FROM field.

HEADER <field-name> <string>

Messages that have a header with the specified field-name (as defined in [RFC-822]) and that contains the specified string in the [RFC-822] field-body.

KEYWORD <flag>

Messages with the specified keyword set.

LARGER <n>

Messages with an [RFC-822] size larger than the specified number of octets.

NEW

Messages that have the \Recent flag set but not the \Seen flag. This is functionally equivalent to (RECENT UNSEEN).

NOT <search-key>

Messages that do not match the specified search key.

OLD

Messages that do not have the \Recent flag set. This is functionally equivalent to NOT RECENT (as opposed to NOT NEW).

ON <date>

Messages whose internal date is within the specified date.

OR <search-key1> <search-key2>

Messages that match either search key.

RECENT

Messages that have the \Recent flag set.

SEEN

Messages that have the \Seen flag set.

SENTBEFORE <date>

Messages whose [RFC-822] Date: header is earlier than the specified date.

SENTON <date>

Messages whose [RFC-822] Date: header is within the specified date.

SENTSINCE <date>

Messages whose [RFC-822] Date: header is within or later than the specified date.

SINCE <date>

Messages whose internal date is within or later than the specified date.

SMALLER <n>

Messages with an [RFC-822] size smaller than the specified number of octets.

SUBJECT <string>

Messages that contain the specified string in the envelope structure's SUBJECT field.

TEXT <string>

Messages that contain the specified string in the header or body of the message.

TO <string>

Messages that contain the specified string in the envelope structure's TO field.

UID <message set>

Messages with unique identifiers corresponding to the specified unique identifier set.

UNANSWERED

Messages that do not have the \Answered flag set.

UNDELETED

Messages that do not have the \Deleted flag set.

UNDRAFT

Messages that do not have the \Draft flag set.

UNFLAGGED

Messages that do not have the \Flagged flag set.

UNKEYWORD <flag>

Messages that do not have the specified keyword set.

UNSEEN

Messages that do not have the \Seen flag set.

This CAVEATS section was contributed by Bruce Gingery <bgingery@gtcs.com>.

The Mail::Cclient::mailbox method returns the actual full path opened, which may not give an accurate string comparison with the mailbox that was requested to be opened. This is especially true with remote mailboxes.

The C-Client library is VERY intolerant of logic errors, and does not automatically garbage collect. Use the CWgc method as it makes sense for your application.

Some POP3 servers delete and expunge WITHOUT instruction to do so. This is not a malfunction in either the C-Client code nor the Mail::Cclient modules.

The CWopen method can be used to extend a CWhalfopen connection (e.g. use the same c-client instance to read a mailbox that was previously halfopened for a list of mailboxes. This may or may not be a good idea, depending upon your needs. It does, however, eliminate the problem of opening multiple connections, such as has been noted in Netscape 4.x mail handling, and which plagues some servers badly. It may be better, however, to CWclose the connection used for CWlist, and re-instantiate to process the mailbox.

C-Client may not support headers you need for send. Note that other modules can be used in place of sending with the c-client. These include Net::SMTP, local invocation of piped sendmail (or other E-mail insertion software), or sendto: URLs under libwww POST.

CWElt information for remote mailboxes is server dependent, as well. You may or may not get rfc822_size in elt returns, for example.

Multiple c-client instances open simultaneously may not work as expected.

Malcolm Beattie, mbeattie@sable.ox.ac.uk.

Maintained by Henrique Dias, hdias@aesbuc.pt.

version 1.12

perl(1) IMAP::Admin Mail::Send