mailutils, Programming Manual

Short Contents


Node:Top, Next:, Previous:(dir), Up:(dir)


Node:Introduction, Next:, Previous:Top, Up:Top

Introduction

GNU Mailutils offers a general purpose library whose aim is to provide a rich set of functions for accessing different mailbox formats and mailers.

References

For more information on,


Node:libmailbox, Next:, Previous:Introduction, Up:Top

Framework

Wherever the mail is and whatever format it is stored in, it is operated upon using the same set of functions. To unified the C API, GNU Mailutils offers a heteroclite set of objects that work in aggregation to do operations on emails. Each object does a specific task and delegates non-related tasks to others. The object comes alive by specifying a URL parameter when created, it will indicate the storage format or protocol (POP3, IMAP4, MH, MAILDIR, etc ..).

                          folder_t                  url_t
  -/var/mail-    +- .. ->+-----------------+   +-->+------------+
 (  alain *-)-+  |       |   url_t       *-|---+   |   port     |
  ----------- |  |       |-----------------|       |   hostname |
 (  jakob *-)-+--+       |   auth_t      *-|---+   |   file     |
  ----------- |          |-----------------|   |   |   ...      |
 (  jeff  *-)-+          |   stream_t      |   |   +------------+
  ----------- |          |-----------------|   |
 (  shaleh*-)-+          |   .....         |   |    auth_t
  ----------             |-----------------|   +-->+------------+
                     +---|-* mailbox_t[]   |       |   ticket_t |
  mailbox_t          |   +-----------------+       +------------+
 +----------------+<-+
 |   locker_t  *--|-------------+
 |----------------|             |
 |   url_t        |             |          locker_t
 |----------------|             +-------->+---------+
 |   stream_t     |                       | lock    |
 |----------------|                       | unlock  |
 |  message_t[] *-|-------+               +---------+
 +----------------+       |                     envelope_t
                          |          +-------->+-----------+
  message_t               |          |         | date      |
 +----------------+<------+          |         | from      |
 |   envelope_t *-|------------------+         | to        |
 |----------------|              header_t      +-----------+
 |   header_t   *-|------------>+--------------+
 |----------------|             |   stream_t   |
 |   body_t     *-|----+        +--------------+
 +----------------+    |    body_t
                       +-->+--------------+
                           |   stream_t   |
                           +--------------+

As an example, here is a simplified version of from command. It lists the From and Subject headers of every mail in a mailbox.


/* sfrom, Simple From */
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <string.h>

#include <mailutils/mailutils.h>

int
main (int argc, const char **argv)
{
  char *from;
  char *subject;
  mailbox_t mbox;
  int status;
  size_t msgno, total = 0;

  /* Register the type of mailbox. IMAP4, POP3 and local format  */
  {
    list_t registrar;
    registrar_get_list (&registrar);
    list_append (registrar, imap_record);
    list_append (registrar, path_record);
    list_append (registrar, pop_record);
  }

  status = mailbox_create_default (&mbox, argv[1]);

  if (status != 0)
    {
      fprintf (stderr, "mailbox_create: %s\n", mu_errstring (status));
      exit (EXIT_FAILURE);
    }

  status = mailbox_open (mbox, MU_STREAM_READ);
  if (status != 0)
    {
      fprintf (stderr, "mailbox_open: %s\n", mu_errstring (status));
      exit (EXIT_FAILURE);
    }

  mailbox_messages_count (mbox, &total);

  for (msgno = 1; msgno <= total; msgno++)
    {
      message_t msg;
      header_t hdr;

      if ((status = mailbox_get_message (mbox, msgno, &msg)) != 0
          || (status = message_get_header (msg, &hdr)) != 0)
        {
          fprintf (stderr, "Error message: %s\n",
                   mu_errstring (status));
          exit (EXIT_FAILURE);
        }

      if (header_aget_value (hdr, MU_HEADER_FROM, &from))
        from = strdup ("(NO FROM)");

      if (header_aget_value (hdr, MU_HEADER_SUBJECT, &subject))
        subject = strdup("(NO SUBJECT)");

      printf ("%s\t%s\n", from, subject);
      free (from);
      free (subject);
    }

  mailbox_close (mbox);
  mailbox_destroy (&mbox);
  return 0;
}

Here is a sample output produced by this program:

% MAIL=pop://[email protected] ./sfrom
Passwd: xxxx
Jim Meyering <[email protected]>      fetish(shellutils) beta
François Pinard <[email protected]> recode new alpha
...


Node:Folder, Next:, Up:libmailbox

Folder


/* Prefix folder_ is reserve */
#include <mailutils/folder.h>


                             folder_t                  url_t
  -/var/mail-    +---//---->/-----------------\    +-->/-----------\
 (  alain *-)-+  |          |  url_t        *-|----+   |  port     |
  ----------- |  |          |-----------------+        |  hostname |
 (  jakob *-)-+--+          |  observer_t   *-|        |  file     |
  ----------- |             |-----------------+        |  ...      |
 (  jeff  *-)-+             |  stream_t       |        \-----------/
  ----------- |             |-----------------|
 (  sean  *-)-+             |  auth_t         |
  ----------                |-----------------|
                            |  mailbox_t(1)   |
                            |-----------------|
                            |  mailbox_t(2)   |
                            |  ......         |
                            |  mailbox_t(n)   |
                            \-----------------/

int folder_create (folder_t *, const char *url) Function

void folder_destroy (folder_t *) Function

int folder_open (folder_t, int flag) Function

int folder_close (folder_t) Function

int folder_delete (folder_t, const char *mailbox) Function

int folder_rename (folder_t, const char *, const char *mailbox) Function

int folder_subscribe (folder_t, const char *mailbox) Function

int folder_unsubscribe (folder_t, const char *mailbox) Function

int folder_list (folder_t, const char *ref, const char *wcard, iterator_t *) Function

int folder_lsub (folder_t, const char *ref, const char *wcar, iterator_t *) Function

int folder_get_stream (folder_t, stream_t *) Function

int folder_set_stream (folder_t, stream_t) Function

int folder_get_observable (folder_t, observable_t *) Function

int folder_get_debug (folder_t, debug_t *) Function

int folder_set_debug (folder_t, debug_t) Function

int folder_get_authority (folder_t, authority_t *) Function

int folder_set_authority (folder_t, authority_t) Function

int folder_get_url (folder_t, url_t *) Function

int folder_set_url (folder_t, url_t) Function


Node:Mailbox, Next:, Previous:Folder, Up:libmailbox

Mailbox


/* Prefix mailbox_ is reserved */
#include <mailutils/mailbox.h>

mailbox_t Data Type
The mailbox_t object is used to hold information and it is an opaque data structure to the user. Functions are provided to retrieve information from the data structure.

                             mailbox_t                  url_t
  -/var/mail-    +---//---->/-----------------\    +-->/-----------\
 (  alain   )    |          |  url_t        *-|----+   |  port     |
  -----------    |          |-----------------+        |  hostname |
 (  jakob *-)----+          |  observer_t   *-|        |  file     |
  -----------               |-----------------+        |  ...      |
 (  jeff    )               |  stream_t       |        \-----------/
  -----------               |-----------------|
 (  sean    )               |  locker_t       |
  ----------                |-----------------|
                            |  message_t(1)   |
                            |-----------------|
                            |  message_t(2)   |
                            |  ......         |
                            |  message_t(n)   |
                            \-----------------/

int mailbox_append_message (mailbox_t mbox, message_t message) Function
The message is appended to the mailbox mbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null or message is invalid.

int mailbox_close (mailbox_t mbox) Function
The stream attach to mbox is closed.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_create (mailbox_t *pmbox, const char *name) Function
The function mailbox_create allocates and initializes pmbox. The concrete mailbox type instantiate is based on the scheme of the url name.

The return value is 0 on success and a code number on error conditions:


MU_ERR_OUT_PTR_NULL
The pointer pmbox supplied is NULL.
MU_ERR_NO_HANDLER
The url name supplied is invalid or not supported.
EINVAL
ENOMEM
Not enough memory to allocate resources.

int mailbox_create_default (mailbox_t *pmbox, const char *name) Function
Create a mailbox with mailbox_create () based on the environment variable $MAIL or the string formed by _PATH_MAILDIR/user" or $LOGNAME if user is null,

void mailbox_destroy (mailbox_t *pmbox) Function
Destroys and releases resources held by pmbox.

int mailbox_expunge (mailbox_t mbox) Function
All messages marked for deletion are removed.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_get_folder (mailbox_t mbox, folder_t *folder) Function
Get the folder.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_get_debug (mailbox_t mbox, debug_t *debug) Function
Get a debug object. The return value is 0 on success and a code number on error conditions:
MU_ERROR_INVALID_PARAMETER
mbox is null.
ENOMEM

int mailbox_get_locker (mailbox_t mbox, locker_t *plocker) Function
Return the locker object.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_get_message (mailbox_t mbox, size_t msgno, message_t *pmessage) Function
Retrieve message number msgno, pmessage is allocated and initialized.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null or msgno is invalid.
ENOMEM
Not enough memory.

int mailbox_get_observable (mailbox_t mbox mbox, observable_t*observable) Function
Get the observable object.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.
ENOMEM
Not enough memory.

int mailbox_get_property (mailbox_t mbox, property_t *property) Function
Get the property object. The return value is 0 on success and a code number on error conditions:
MU_ERROR_INVALID_PARAMETER
mbox is null.
ENOMEM

int mailbox_get_size (mailbox_t mbox, off_t *psize) Function
Gives the mbox size.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_get_stream (mailbox_t mbox, stream_t *pstream) Function
The mailbox stream is put in pstream.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is invalid or pstream is NULL.

int mailbox_get_ticket (mailbox_t mbox, ticket_t ticket) Function
The return value is 0 on success and a code number on error conditions:
MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_get_url (mailbox_t mbox, url_t *purl) Function
Gives the constructed url.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_is_modified (mailbox_t mbox) Function
Check if the mailbox been modified by an external source.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_message_unseen (mailbox_t mbox, size_t *pnumber); Function
Give the number of first unseen message in mbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_messages_count (mailbox_t mbox, size_t *pnumber); Function
Give the number of messages in mbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_messages_recent (mailbox_t mbox, size_t *pnumber); Function
Give the number of recent messages in mbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_open (mailbox_t mbox, int flag) Function
A connection is open, if no stream was provided, a stream is created based on the mbox type. The flag can be OR'ed. See stream_create for flag's description.

The return value is 0 on success and a code number on error conditions:

EAGAIN
EINPROGRESS
Operation in progress.
EBUSY
Resource busy.
MU_ERROR_INVALID_PARAMETER
mbox is null or flag is invalid.
ENOMEM
Not enough memory.

int mailbox_scan (mailbox_t mbox, size_t msgno, size_t *pcount); Function
Scan the mailbox for new messages starting at message msgno.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.
ENOMEM
Not enough memory.

int mailbox_set_locker (mailbox_t mbox, locker_t locker) Function
Set the type of locking done by the mbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_set_stream (mailbox_t mbox, stream_t stream) Function
Set the stream connection to use for the mailbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox or stream is NULL.

int mailbox_set_ticket (mailbox_t mbox, ticket_t ticket) Function
The ticket will be set on the auth_t object on creation.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_uidnext (mailbox_t mbox, size_t *pnumber); Function
Give the next predicted uid for mbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.

int mailbox_uidvalidity (mailbox_t mbox, size_t *pnumber); Function
Give the uid validity of mbox.

The return value is 0 on success and a code number on error conditions:

MU_ERROR_INVALID_PARAMETER
mbox is null.


Node:Mailer, Next:, Previous:Mailbox, Up:libmailbox

Mailer


/* Prefix mailer_ is reserved */
#include <mailutils/mailer.h>

The API is still changing.

int mailer_create (mailer_t *, const char *) Function

void mailer_destroy (mailer_t *) Function

int mailer_open (mailer_t, int flags) Function

int mailer_close (mailer_t) Function

int mailer_send_message (mailer_t mailer, message_t msg, address_t from, address_t to); Function

If from is not NULL, it must contain a single fully qualified RFC2822 email address which will be used as the envelope from address. This is the address to which delivery status notifications are sent, so it never matters what it is set to until it REALLY matters. This is equivalent to sendmail's -f flag.

The default for from is provided by the specific mailer.

If to is not NULL, then the message will be sent to the list of addresses that it specifies.

The default for to is to use the contents of the standard "To:", "Cc:", and "Bcc:" fields, this is equivalent to sendmail's -t flag.

Note: the previous implementation of mailer_send_message() was equivalent to having both from and to be NULL.

int mailer_get_property (mailer_t, property_t *) Function

int mailer_get_stream (mailer_t, stream_t *) Function

int mailer_set_stream (mailer_t, stream_t) Function

int mailer_get_debug (mailer_t, debug_t *) Function

int mailer_set_debug (mailer_t, debug_t) Function

int mailer_get_observable (mailer_t, observable_t *) Function

int mailer_get_url (mailer_t, url_t *) Function

Some usage notes.

Some possible use cases the API must support are:

- original submission

1 - fill in header addresses

2 - mailer_send_message(mailer, msg, NULL, NULL);

- from will be filled in if missing,

- bcc's will be deleted before delivery to a non-bcc address,

- message-id and date will be added, if missing,

- a to: or apparently-to: will be added if non is present (for RFC compliance)

- MTA-style .forward ( and sieve-style redirect )

1 - get the envelope from of the message to be forwarded

2 - mailer_send_message(mailer, msg, from, to)

- MUA-style bounce

1 - add Resent-[to,from,....]

2 - mailer_send_message(mailer, msg, NULL, to)

- DSN "bounce"

1 - compose DSN

2 - mailer_deliver(mailer, msg, address_t( "<>" ), to)

Don't want mail loops, so the null but valid SMTP address of <> is the envelope from.

The sendmail mailer.

/sbin/sendmail isn't always sendmail... sometimes its a sendmail-compatible wrapper, so assume /sbin/sendmail understands only a recipient list, -f and -oi, these seem to be pretty basic. Cross fingers.

Pipe to "/sbin/sendmail -oi [-f from] [to...]", supplying -f if there was a from, and supplying the recipient list from the to (if there is no recipient list, assume it will read the message contents for the recipients).

Note: since the stdout and stderr of sendmail is closed, we have no way of ever giving feedback on failure. Also, what should the return code be from mailer_send_message() when sendmail returns 1? 1 maps to EPERM, which is less than descriptive!

The SMTP mailer.

This mailer does NOT canonicalize the message. This must be done before sending the message, or it may be assumed that the MTA will do so.

It does blind out the Bcc: header before sending, though.

Note: mutt always puts the recipient addresses on the command line, even bcc ones, do we strip the bcc before forwarding with SMTP?

Non-RFC822 addresses.

An address that has no domain is not and RFC822 email address. What do I do with them? Should the user of the API be responsible for determining what is mean by email to "john" means? Or should the be able to configure sendmail to decide globally what this means. If so, we can pass the address to sendmail, but we have to decide for SMTP! So, right now these addresses are rejected. This could be changed.


Node:Message, Next:, Previous:Mailer, Up:libmailbox

Message


#include <mailutils/message.h>
/* Prefix message_ is reserve */

The message_t object is a convenient way to manipulate messages. It encapsulates the envelope_t, the header_t and the body_t.

    mailbox_t
    __________                   message_t
   (message[1])        +------>+-----------------------+
    ----------         |       |  envelope_t           |
   (message[2])        |       |-----------------------|
    ----------         |       |  header_t             |
   (message[3])--------+       |-----------------------|
    ----------                 |  body_t               |
   (message[n])                |-----------------------|
    ----------                 |  attribute_t          |
                               |-----------------------|
                               |  stream_t             |
                               +-----------------------+

void message_create (message_t *msg, void *owner) Function

void message_destroy (message_t *msg, void *owner) Function
The resources allocate for msg are freed.

int message_get_header (message_t msg, header_t *pheader) Function
Retrieve msg header.

int message_set_header (message_t msg, header_t header, void *owner) Function

int message_get_body (message_t msg, body_t *pbody) Function

int message_set_body (message_t msg, body_t body, void *owner) Function

int message_is_multipart (message_t msg, int *pmulti) Function
Set *pmulti to non-zero value if msg is multi-part.

int message_get_num_parts (message_t msg, size_t *nparts) Function

int message_get_part (message_t msg, size_t part, message_t *msg) Function

int message_get_stream (message_t msg, stream_t *pstream) Function

int message_set_stream (message_t msg, stream_t stream,void *owner ) Function

int message_get_attribute (message_t msg, attribute_t *pattribute) Function

int message_set_attribute (message_t msg, attribute_t attribute, void *owner) Function

int message_get_envelope (message_t msg, envelope_t *penvelope) Function

int message_set_envelope (message_t msg, envelope_t envelope, void *owner) Function

int message_get_uid (message_t msg, size_t *uid) Function

int message_get_uidl (message_t msg, char *buffer, size_t buflen, size_t *pwriten) Function

int message_set_uidl (message_t msg, int (*_get_uidl)(message_t, char *, size_t, size_t *), void *owner) Function

int message_get_observable (message_t msg, observable_t *observable) Function

int message_create_attachment (const char *content_type, const char *encoding, const char *filename, message_t *newmsg) Function

int message_save_attachment (message_t msg, const char *filename, void **data) Function

int message_encapsulate (message_t msg, message_t *newmsg, void **data) Function

int message_unencapsulate (message_t msg, message_t *newmsg, void **data); Function


Node:Envelope, Next:, Previous:Message, Up:libmailbox

Envelope


/* Prefix envelope_ is reserved */
#include <mailutils/envelope.h>

int envelope_date (envelope_t, char *, size_t, size_t *); Function

Get the date that the message was delivered to the mailbox, in something close to ANSI ctime() format: Mon Jul 05 13:08:27 1999.

int envelope_sender (envelope_t, char *, size_t, size_t *); Function

Get the address that this message was reportedly received from. This would be the "mail from" argument if the message was delivered or received via SMTP, for example.

int envelope_get_message (envelope_t, message_t *); Function

int envelope_create (envelope_t *, void *); Function
Primarily for internal use.

void envelope_destroy (envelope_t *, void *); Function
Primarily for internal use.

int envelope_set_sender (envelope_t, int (*_sender) __P ((envelope_t, char *, size_t, size_t*)), void *); Function

Primarily for internal use. The implementation of envelope_t depends on the mailbox type, this allows the function which actually gets the sender to be set by the creator of an envelope_t.

int envelope_set_date (envelope_t, int (*_date) __P ((envelope_t, char *, size_t, size_t *)), void *); Function

Primarily for internal use. The implementation of envelope_t depends on the mailbox type, this allows the function which actually gets the date to be set by the creator of an envelope_t.


Node:Headers, Next:, Previous:Envelope, Up:libmailbox

Headers


/* Prefix header_ is reserved */
#include <mailutils/header.h>

So far we plan support for RFC822 and plan for RFC1522. with RFC1522 non ASCII characters will be encoded.

int header_create (header_t *hdr, const char *blurb, size_t len, void *owner) Function
Initialize a hdr to a supported type. If blurb is not NULL, it is parsed.

void header_destroy (header_t *hdr, void *owner) Function
The resources allocated for hdr are freed.

int header_set_value (header_t hdr, const char *fn, const char *fv, size_t n, int replace) Function
Set the field-name fn to field-value fv of size n in hdr. If replace is non-zero the initial value is replaced, if zero it is appended.

Some basic macros are already provided for rfc822.

MU_HEADER_RETURN_PATH
Return-Path
MU_HEADER_RECEIVED
Received
MU_HEADER_DATE
Date
MU_HEADER_FROM
From
MU_HEADER_RESENT_FROM
Resent-From
MU_HEADER_SUBJECT
Subject
MU_HEADER_SENDER
Sender
MU_HEADER_RESENT_SENDER
Resent-SENDER
MU_HEADER_TO
To
MU_HEADER_RESENT_TO
Resent-To
MU_HEADER_CC
Cc
MU_HEADER_RESENT_CC
Resent-Cc
MU_HEADER_BCC
Bcc
MU_HEADER_RESENT_BCC
Resent-Bcc
MU_HEADER_REPLY_TO
Reply-To
MU_HEADER_RESENT_REPLY_TO
Resent-Reply-To
MU_HEADER_MESSAGE_ID
Message-ID
MU_HEADER_RESENT_MESSAGE_ID
Resent-Message-ID
MU_HEADER_IN_REPLY_TO
In-Reply-To
MU_HEADER_ENCRYPTED
Encrypted
MU_HEADER_PRECEDENCE
Precedence
MU_HEADER_STATUS
Status
MU_HEADER_CONTENT_LENGTH
Content-Length
MU_HEADER_CONTENT_TYPE
Content-Type
MU_HEADER_MIME_VERSION
MIME-Version

int header_get_value (header_t hdr, const char *fn, char *fv, size_t len, size_t *n) Function
Value of field-name fn is returned in buffer fv of size len. The number of bytes written is put in n.

int header_aget_value (header_t hdr, const char *fn, char **fv) Function
The value is allocated.

int header_get_stream (header_t hdr, stream_t *pstream) Function

int header_set_size (header_t hdr, size_t *size) Function

int header_set_lines (header_t hdr, size_t *lpines) Function


Node:Body, Next:, Previous:Headers, Up:libmailbox

Body


/* Prefix body_ is reserved */
#include <mailutils/body.h>

int body_create (body_t *body, void *owner) Function
Initialize an object bdy.

void body_destroy (body_t *pbody) Function
The resources allocate are release.

int body_get_stream (body_t body, stream_t *pstream) Function

int body_set_stream (body_t body, stream_t stream, void *owner) Function

int body_get_filename __P ((body_t body, char *buffer, size_tbuflen, size_t *pwriten) Function

int body_set_filename (body_t body, const char*buffer) Function

int body_size (body_t body, size_t*psize) Function

int body_lines (body_t body, size_t *plines) Function


Node:Attribute, Next:, Previous:Body, Up:libmailbox

Attribute


/* Prefix attribute_ is reserved */
#include <mailutils/attribute.h>

int attribute_create (attribute_t *pattribute) Function

void attribute_destroy (attribute_t *pattribute) Function

int attribute_is_seen (attribute_t attribute) Function

int attribute_is_answered (attribute_t attribute) Function

int attribute_is_flagged (attribute_t attribute) Function

int attribute_is_deleted (attribute_t attribute) Function

int attribute_is_draft (attribute_t attribute) Function

int attribute_is_recent (attribute_t attribute) Function

int attribute_is_read (attribute_t attribute) Function

int attribute_set_seen (attribute_t attribute) Function

int attribute_set_answered (attribute_t attribute) Function

int attribute_set_flagged (attribute_t attribute) Function

int attribute_set_deleted (attribute_t attribute) Function

int attribute_set_draft (attribute_t attribute) Function

int attribute_set_recent (attribute_t attribute) Function

int attribute_set_read (attribute_t attribute) Function

int attribute_unset_seen (attribute_t attribute) Function

int attribute_unset_answered (attribute_t attribute) Function

int attribute_unset_flagged (attribute_t attribute) Function

int attribute_unset_deleted (attribute_t attribute) Function

int attribute_unset_draft (attribute_t attribute) Function

int attribute_unset_recent (attribute_t attribute) Function

int attribute_unset_read (attribute_t attribute) Function

int attribute_is_equal (attribute_t att1, attribute_t att2) Function

int attribute_copy (attribute_t dst, attribute_t src) Function

int string_to_attribute (const char *buf, attribute_t *pattr) Function

int attribute_to_string (attribute_t attr, char *buf, size_t len, size_t *pwriten) Function


Node:Stream, Next:, Previous:Attribute, Up:libmailbox

Stream

#include <mailutils/stream.h>

These generic flags are interpreted as appropriate to the specific streams.

MU_STREAM_READ
The stream is open read only.
MU_STREAM_WRITE
The stream is open write only.
MU_STREAM_RDWR
The stream is open read and write.
MU_STREAM_APPEND
The stream is open in append mode for writing.
MU_STREAM_CREAT
The stream open will create the underlying resource (such as a file) if it doesn't exist already.
MU_STREAM_NONBLOCK
The stream is set non blocking.
MU_STREAM_NO_CHECK
Stream is destroyed without checking for the owner.
MU_STREAM_NO_CLOSE
Stream doesn't close it's underlying resource when it is closed or destroyed.

int file_stream_create (stream_t *pstream, const char *filename, int flags) Function

int tcp_stream_create (stream_t *pstream, const char *host, int port, int flags) Function

int mapfile_stream_create (stream_t *pstream, const char *filename, int flags) Function

int memory_stream_create (stream_t *pstream, const char *filename, int flags) Function

int encoder_stream_create (stream_t *pstream, stream_t iostream, const char *encoding) Function

int decoder_stream_create (stream_t *pstream, stream_t iostream, const char *encoding) Function

int stdio_stream_create (stream_t *pstream, FILE* stdio, int flags) Function
If MU_STREAM_NO_CLOSE is specified, fclose() will not be called on stdio when the stream is closed.

void stream_destroy (stream_t *pstream, void *owner) Function

int stream_open (stream_t stream) Function

int stream_close (stream_t stream) Function

int stream_is_seekable (stream_t stream) Function

int stream_get_fd (stream_t stream, int *pfd) Function

int stream_read (stream_t stream, char *buffer, size_t buflen, off_t offset, size_t *pwriten) Function

int stream_readline (stream_t stream, char *buffer, size_t buflen, off_t offset, size_t *pwriten) Function

int stream_size (stream_t stream, off_t *psize) Function

int stream_truncate (stream_t stream, off_t size) Function

int stream_write (stream_t stream, const char *buffer, size_t buflen, off_t offset, size_t *pwriten) Function

int stream_setbufsiz (stream_t stream, size_t size) Function

int stream_flush (stream_t stream) Function

These functions will typically only be useful to implementors of streams.

int stream_create (stream_t *pstream, int flags, void *owner) Function
Used to implement a new kind of stream.

int stream_get_flags (stream_t stream, int *pflags) Function

int stream_get_state (stream_t stream, int *pstate) Function
MU_STREAM_STATE_OPEN
Last action was stream_open.
MU_STREAM_STATE_READ
Last action was stream_read or stream_readline.
MU_STREAM_STATE_WRITE
Last action was stream_write.
MU_STREAM_STATE_CLOSE
Last action was stream_close.

An example using tcp_stream_create to make a simple web client:


/* This is an example program to illustrate the use of stream functions.
   It connects to a remote HTTP server and prints the contents of its
   index page */
#include <stdlib.h>
#include <stdio.h>
#include <errno.h>
#include <string.h>
#include <unistd.h>

#include <mailutils/mailutils.h>

const char *wbuf = "GET / HTTP/1.0\r\n\r\n";
char rbuf[1024];

int
main ()
{
  int ret, off = 0, fd;
  stream_t stream;
  size_t nb;
  fd_set fds;

  ret = tcp_stream_create (&stream, "www.gnu.org", 80, MU_STREAM_NONBLOCK);
  if (ret != 0)
    {
      mu_error ( "tcp_stream_create: %s\n", mu_errstring (ret));
      exit (EXIT_FAILURE);
    }

connect_again:
  ret = stream_open (stream);
  if (ret != 0)
    {
      if (ret == EAGAIN)
        {
          ret = stream_get_fd (stream, &fd);
          if (ret != 0)
            {
              mu_error ( "stream_get_fd: %s\n", mu_errstring (ret));
              exit (EXIT_FAILURE);
            }
          FD_ZERO (&fds);
          FD_SET (fd, &fds);
          select (fd + 1, NULL, &fds, NULL, NULL);
          goto connect_again;
        }
      mu_error ( "stream_open: %s\n", mu_errstring (ret));
      exit (EXIT_FAILURE);
    }

  ret = stream_get_fd (stream, &fd);
  if (ret != 0)
    {
      mu_error ( "stream_get_fd: %s\n", mu_errstring (ret));
      exit (EXIT_FAILURE);
    }

write_again:
  ret = stream_write (stream, wbuf + off, strlen (wbuf), 0, &nb);
  if (ret != 0)
    {
      if (ret == EAGAIN)
        {
          FD_ZERO (&fds);
          FD_SET (fd, &fds);
          select (fd + 1, NULL, &fds, NULL, NULL);
          off += nb;
          goto write_again;
        }
      mu_error ( "stream_write: %s\n", mu_errstring (ret));
      exit (EXIT_FAILURE);
    }

  if (nb != strlen (wbuf))
    {
      mu_error ( "stream_write: %s\n", "nb != wbuf length");
      exit (EXIT_FAILURE);
    }

  do
    {
      ret = stream_read (stream, rbuf, sizeof (rbuf), 0, &nb);
      if (ret != 0)
        {
          if (ret == EAGAIN)
            {
              FD_ZERO (&fds);
              FD_SET (fd, &fds);
              select (fd + 1, &fds, NULL, NULL, NULL);
            }
          else
            {
              mu_error ( "stream_read: %s\n", mu_errstring (ret));
              exit (EXIT_FAILURE);
            }
        }
      write (2, rbuf, nb);
    }
  while (nb || ret == EAGAIN);

  ret = stream_close (stream);
  if (ret != 0)
    {
      mu_error ( "stream_close: %s\n", mu_errstring (ret));
      exit (EXIT_FAILURE);
    }

  stream_destroy (&stream, NULL);
  exit (EXIT_SUCCESS);
}


Node:Iterator, Next:, Previous:Stream, Up:libmailbox

Iterator


/* Prefix iterator_ is reserved */
#include <mailutils/iterator.h>

int iterator_create (iterator_t *) Function

void iterator_destroy (iterator_t *) Function

int iterator_first (iterator_t) Function

int iterator_next (iterator_t) Function

int iterator_current (iterator_t, void **pitem) Function

int iterator_is_done (iterator_t) Function


Node:Authenticator, Next:, Previous:Iterator, Up:libmailbox

Authenticator


/* Prefix auth_ is reserved */
#include <mailutils/auth.h>

There are many ways to authenticate to a server. To be flexible the authentication process is provided by two objects auth_t and ticket_t. The auth_t can implement different protocol like APOP, MD5-AUTH, One Time Passwd etc .. By default if a mailbox does not understand or know how to authenticate it falls back to user/passwd authentication. The ticket_t is a way for Mailboxes and Mailers provide a way to authenticate when the URL does not contain enough information. The default action is to call the function auth_authenticate which will get the user and passwd if not set, this function can be overridden by a custom method.

int auth_create (auth_t *pauth, void *owner) Function

void auth_destroy (auth_t *pauth, void *owner) Function

int auth_prologue (auth_t auth) Function

int auth_authenticate (auth_t auth, char **user, char **passwd) Function

int auth_epilogue (auth_t auth) Function

A simple example of an authenticate function:

#include <mailutils/auth.h>
#include <stdio.h>
#include <string.h>

int
my_authenticate (auth_t auth, char **user, char **passwd)
{
  char u[128] = "";
  char p[128] = "";

  /* prompt the user name */
  printf ("User: ");
  fflush (stdout);
  fgets (u, sizeof (u), stdin);
  u[strlen (u) - 1] = '\0'; /* nuke the trailing NL */

  /* prompt the passwd */
  printf ("Passwd: "); fflush (stdout);
  echo_off ();
  fgets (p, sizeof(p), stdin);
  echo_on ();
  p[strlen (p) - 1] = '\0';

  /* duplicate */
  *user = strdup (u);
  *passwd = strdup (p);
  return 0;
}


Node:Address, Next:, Previous:Authenticator, Up:libmailbox

Address


/* Prefix address_ is reserved */
#include <mailutils/address.h>

The Internet address format is defined in RFC 822. RFC 822 has been updated, and is now superceeded by RFC 2822, which makes some corrections and clarifications. References to RFC 822 here apply equally to RFC 2822.

The RFC 822 format is more flexible than many people realize, here is a quick summary of the syntax this parser implements, see RFC 822 for the details. "[]" pairs mean "optional", "/" means "one or the other", and double-quoted characters are literals.

addr-spec    = local-part "¨domain
mailbox      = addr-spec ["(" display-name ")"] /
               [display-name] "<" [route] addr-spec ">"
mailbox-list = mailbox ["," mailbox-list]
group        = display-name ":" [mailbox-list] ";"
address      = mailbox / group / unix-mbox
address-list = address ["," address-list]

unix-mbox is a non-standard extension meant to deal with the common practice of using user names as addresses in mail utilities. It allows addresses such as "root" to be parsed correctly. These are NOT valid internet email addresses, they must be qualified before use.

Several address functions have a set of common arguments with consistent semantics, these are described here to avoid repetition.

Since an address-list may contain multiple addresses, they are accessed by a one-based index number, no. The index is one-based because pop, imap, and other message stores commonly use one-based counts to access messages and attributes of messages.

If len is greater than 0 it is the length of the buffer buf, and as much of the component as possible will be copied into the buffer. The buffer will be null terminated.

The size of a particular component may be queried by providing 0 for the len of the buffer, in which case the buffer is optional. In this case, if n is provided *n is assigned the length of the component string.

address_t Data Type
The address_t object is used to hold information about a parsed RFC822 address list, and is an opaque data structure to the user. Functions are provided to retrieve information about an address in the address list.

int address_create (address_t *addr, const char *string) Function
This function allocates and initializes addr by parsing the RFC822 address-list string.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOMEM
Not enough memory to allocate resources.
ENOENT
Invalid RFC822 syntax, parsing failed.

int address_createv (address_t *addr, const char *sv, size_t len) Function
This function allocates and initializes addr by parsing the array of pointers to RFC822 address-lists in sv. If len is -1, then sv must be null-terminated in the fashion of argv, otherwise len is the length of the array.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOMEM
Not enough memory to allocate resources.
ENOENT
Invalid RFC822 syntax, parsing failed.

void address_destroy (address_t *addr) Function
The addr is destroyed.

int address_get_email (address_t *addr, size_t no, char* buf, size_t len, size_t* n) Function

Accesses the noth email address component of the address list. This address is the plain email address, correctly quoted, suitable for using in an smtp dialog, for example, or as the address part of a contact book entry.

Note that the entry may be valid, but be a group name. In this case success is returned, but the length of the address is 0.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_aget_email (address_t *addr, size_t no, char** bufp) Function

As above, but mallocs the email address, if present,

and write a pointer to it into bufp. bufp will be NULL if there is no email address to return.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_get_personal (address_t *addr, size_t no, char* buf, size_t len, size_t* n) Function

Accesses the display-name describing the noth email address. This display-name is optional, so may not be present. If it is not present, but there is an RFC822 comment after the address, that comment will be returned as the personal phrase, as this is a common usage of the comment even though it is not defined in the internet mail standard.

A group is a kind of a special case. It has a display-name, followed by an optional mailbox-list. The display-name will be allocated an address all it's own, but all the other elements (local-part, domain, etc.) will be zero-length. So "a group: ;" is valid, will have a count of 1, but address_get_email(), and all the rest, will return zero-length output.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_get_comments (address_t *addr, size_t no, char* buf, size_t len, size_t* n) Function

Accesses the comments extracted while parsing the noth email address. These comments have no defined meaning, and are not currently collected.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_get_email (address_t *addr, size_t no, char* buf, size_t len, size_t* n) Function

Accesses the email addr-spec extracted while parsing the noth email address. This will be 0 length for a unix-mbox.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_get_local_part (address_t *addr, size_t no, char* buf, size_t len, size_t* n) Function

Accesses the local-part of an email addr-spec extracted while parsing the noth email address.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_get_domain (address_t *addr, size_t no, char* buf, size_t len, size_t* n) Function

Accesses the domain of an email addr-spec extracted while parsing the noth email address. This will be 0 length for a unix-mbox.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_get_route (address_t *addr, size_t no, char* buf, size_t len, size_t* n) Function

Accesses the route of an email addr-spec extracted while parsing the noth email address. This is a rarely used RFC822 address syntax, but is legal in SMTP as well. The entire route is returned as a string, those wishing to parse it should look at <mailutils/parse822.h>.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_is_group (address_t *addr, size_t no, size_t len, int* yes) Function

Sets *yes to 1 if this address is just the name of a group, 0 otherwise. This is faster than checking if the address has a non-zero length personal, and a zero-length local_part and domain.

yes can be null, though that doesn't serve much purpose other than determining that no refers to an address.

Currently, there is no way to determine the end of the group.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOENT
The index no is outside of the range of available addresses.

int address_to_string (address_t *addr, char* buf, size_t len, size_t* n) Function

Returns the entire address list as a single RFC822 formatted address list.

The return value is 0 on success and a code number on error conditions:

EINVAL
Invalid usage, usually a required argument was null.
ENOMEM
Not enough memory to allocate resources.

int address_get_count (address_t addr, size_t* count) Function

Returns a count of the addresses in the address list.

If addr is null, the count is 0. If count is not null, the count will be written to *count.

The return value is 0.

Example



#include <stdio.h>
#include <errno.h>
#include <string.h>

#include <mailutils/address.h>
#include <mailutils/errno.h>
#include <mailutils/mutil.h>

#define EPARSE ENOENT

static int
parse (const char *str)
{
  size_t no = 0;
  size_t pcount = 0;
  int status;
  char buf[BUFSIZ];
  address_t address = NULL;

  mu_set_user_email_domain ("localhost");

  status = address_create (&address, str);

  address_get_count (address, &pcount);

  if (status)
    {
      printf ("%s=> error %s\n\n", str, mu_errname (status));
      return 0;
    }
  else
    {
      printf ("%s=> pcount %lu\n", str, (unsigned long) pcount);
    }

  for (no = 1; no <= pcount; no++)
    {
      size_t got = 0;
      int isgroup;

      address_is_group (address, no, &isgroup);

      printf ("%lu ", (unsigned long) no);

      if (isgroup)
        {
          address_get_personal (address, no, buf, sizeof (buf), &got);

          printf ("group <%s>\n", buf);
        }
      else
        {
          address_get_email (address, no, buf, sizeof (buf), 0);

          printf ("email <%s>\n", buf);
        }

      address_get_personal (address, no, buf, sizeof (buf), &got);

      if (got && !isgroup)
        printf ("   personal <%s>\n", buf);

      address_get_comments (address, no, buf, sizeof (buf), &got);

      if (got)
        printf ("   comments <%s>\n", buf);

      address_get_local_part (address, no, buf, sizeof (buf), &got);

      if (got)
        {
          printf ("   local-part <%s>", buf);

          address_get_domain (address, no, buf, sizeof (buf), &got);

          if (got)
            printf (" domain <%s>", buf);

          printf ("\n");
        }

      address_get_route (address, no, buf, sizeof (buf), &got);

      if (got)
        printf ("   route <%s>\n", buf);
    }
  address_destroy (&address);

  printf ("\n");

  return 0;
}

static int
parseinput (void)
{
  char buf[BUFSIZ];

  while (fgets (buf, sizeof (buf), stdin) != 0)
    {
      buf[strlen (buf) - 1] = 0;
      parse (buf);
    }

  return 0;
}

int
main (int argc, const char *argv[])
{
  argc = 1;

  if (!argv[argc])
    {
      return parseinput ();
    }
  for (; argv[argc]; argc++)
    {
      if (strcmp (argv[argc], "-") == 0)
        {
          parseinput ();
        }
      else
        {
          parse (argv[argc]);
        }
    }

  return 0;
}


Node:Locker, Next:, Previous:Address, Up:libmailbox

Locker


/* Prefix locker_ is reserved */
#include <mailutils/locker.h>

int locker_create (locker_t * plocker, char *filename, size_t len, int flags) Function

void locker_destroy (locker_t * plocker) Function

int locker_lock (locker_t locker, int flag) Function
MU_LOCKER_RDLOCK
MU_LOCKER_WRLOCK
MU_LOCKER_PID
MU_LOCKER_FCNTL
MU_LOCKER_TIME

int locker_touchlock (locker_t locker) Function

int locker_unlock (locker_t locker) Function


Node:URL, Next:, Previous:Locker, Up:libmailbox

URL

A mailbox or a mailer can be described in a URL, the string will contain the necessary information to initialize mailbox_t, or mailer_t properly.

POP3

The POP URL scheme contains a POP server, optional port number and the authentication mechanism. The general form is

<pop://[<user>[;AUTH=<auth>]@]<host>[:<port>]>
 or
<pop://[<user>[:<passwd>]@]<host>[:<port>]>

If :port is omitted the default value is 110. Different forms of authentication can be specified with ;AUTH=type. The special string ;AUTH=* indicates that the client will use a default scheme base on the capability of the server.

<pop://[email protected]>
<pop://asterix;AUTH=*@france.com>
<pop://falbala;[email protected]>
<pop://obelix;[email protected]:2000>
<pop://obelix:[email protected]:2000>

For more complete information see rfc2368.

IMAP

The IMAP URL scheme contains an IMAP server, optional port number and the authentication mechanism. The general form is

<imap://[<user>[;AUTH=<type>]]@<host>[:port][/<mailbox>]>
 or
<imap://[<user>[:<passwd>]]@<host>[:port][/<mailbox>]>

If :port is omitted the default value is 220. Different forms of authentication can be specified with ;AUTH=type. The special string ;AUTH=* indicates that the client will use a default scheme base on the capability of the server.

<imap://[email protected]>
<imap://asterix;AUTH=*@imap.france.com>
<imap://asterix:[email protected]>

For more complete information see rfc2192.

File

Local folder should be handle by this URL. It is preferable to let the mailbox recognize the type of mailbox and take the appropriate action.

<file://path>
<file://var/mail/user>
<file://home/obelix/Mail>

For MMDF, MH local mailboxes URLs are provided, but it is preferable to use <file://path> and let the library figure out which one.

<mmdf://path>
<mh://path>

Mailto

After setting a mailer, <mailto:> is used to tell the mailer where and to whom the message is for.

<mailto://hostname>

Mailto can be used to generate short messages, for example to subscribe to mailing lists.

<mailto://[email protected]?body=subscribe>
<mailto://[email protected]?Subject=hello&body=subscribe>

For more complete information see rfc2368.

URL functions

Helper functions are provided to retrieve and set the URL fields.

int url_create (url_t *url, const char *name) Function
Create and the url data structure, but do not parse it.

void url_destroy (url_t *) Function
Destroy the url and free it's resources.

int url_parse (url_t url) Function
Parses the url, after calling this the get functions can be called.

The syntax, condensed from RFC 1738, and extended with the ;auth= of RFC 2384 (for POP) and RFC 2192 (for IMAP) is:

url =
    scheme ":" [ "//"

    [ user [ ( ":" password ) | ( ";auth=" auth ) ] "¨]

    host [ ":" port ]

    [ ( "/" urlpath ) | ( "?" query ) ] ]

This is a generalized URL syntax, and may not be exactly appropriate for any particular scheme.

const char* url_to_string (const url_t url) Function

int url_get_scheme (const url_t url, char *schem, size_t len, size_t *n) Function

int url_get_user (const url_t url, char *usr, size_t len, size_t *n) Function

int url_get_passwd (const url_t url, char *passwd, size_t len, size_t *n) Function

int url_get_host (const url_t url, char *host, size_t len, size_t *n) Function

int url_get_port (const url_t url, long *port) Function

int url_get_path (const url_t url, char *path, size_t len, size_t *n) Function

int url_get_query (const url_t url, char *query, size_tlen, size_t *n) Function

char* url_decode (const char* string) Function
Decodes an RFC 1738 % encoded string, returning the decoded string in allocated memory. If the string is not encoded, this degenerates to a strdup().

Example



#include <stdio.h>
#include <string.h>
#include <mailutils/errno.h>
#include <mailutils/url.h>

int
main ()
{
  char str[1024];
  char buffer[1024];
  long port = 0;
  int len = sizeof (buffer);
  url_t u = NULL;

  while (fgets (str, sizeof (str), stdin) != NULL)
    {
      int rc;

      str[strlen (str) - 1] = '\0';     /* chop newline */
      if (strspn (str, " \t") == strlen (str))
        continue;               /* skip empty lines */
      if ((rc = url_create (&u, str)) != 0)
        {
          fprintf (stderr, "url_create %s ERROR: [%d] %s",
                   str, rc, mu_errstring (rc));
          exit (1);
        }
      if ((rc = url_parse (u)) != 0)
        {
          printf ("%s => FAILED: [%d] %s\n",
                  str, rc, mu_errstring (rc));
          continue;
        }
      printf ("%s => SUCCESS\n", str);

      url_get_scheme (u, buffer, len, NULL);
      printf ("\tscheme <%s>\n", buffer);

      url_get_user (u, buffer, len, NULL);
      printf ("\tuser <%s>\n", buffer);

      url_get_passwd (u, buffer, len, NULL);
      printf ("\tpasswd <%s>\n", buffer);

      url_get_auth (u, buffer, len, NULL);
      printf ("\tauth <%s>\n", buffer);

      url_get_host (u, buffer, len, NULL);
      printf ("\thost <%s>\n", buffer);

      url_get_port (u, &port);
      printf ("\tport %ld\n", port);

      url_get_path (u, buffer, len, NULL);
      printf ("\tpath <%s>\n", buffer);

      url_get_query (u, buffer, len, NULL);
      printf ("\tquery <%s>\n", buffer);

      url_destroy (&u);

    }
  return 0;
}


Node:Parse822, Previous:URL, Up:libmailbox

Parse822 -- Functions for Parsing RFC 822 Headers


/* Prefix parse822_ is reserved */
#include <mailutils/parse822.h>

Internet Message Format, see Address node for the discussion.

int parse822_address_list (address_t* a, const char* s) Function

int parse822_mail_box (const char** p, const char* e, address_t* a) Function

int parse822_group (const char** p, const char* e, address_t* a) Function

int parse822_address (const char** p, const char* e, address_t* a) Function

int parse822_route_addr (const char** p, const char* e, address_t* a) Function

int parse822_route (const char** p, const char* e, char** route) Function

int parse822_addr_spec (const char** p, const char* e, address_t* a) Function

int parse822_unix_mbox (const char** p, const char* e, address_t* a) Function

int parse822_local_part (const char** p, const char* e, char** local_part) Function

int parse822_domain (const char** p, const char* e, char** domain) Function

int parse822_sub_domain (const char** p, const char* e, char** sub_domain) Function

int parse822_domain_ref (const char** p, const char* e, char** domain_ref) Function

int parse822_domain_literal (const char** p, const char* e, char** domain_literal) Function

int parse822_quote_string (char** quoted, const char* raw) Function

int parse822_quote_local_part (char** quoted, const char* raw) Function

int parse822_field_body (const char** p, const char *e, char** fieldbody) Function

int parse822_field_name (const char** p, const char *e, char** fieldname) Function


Node:libmuauth, Next:, Previous:libmailbox, Up:Top

Authentication Library

The functions from libmailbox library get user information from the system user database. The library libmuauth extends this functionality, allowing libmailbox functions to obtain information about a user from several places, like SQL database, etc. The method used is described in detail in authentication. This chapter contains a very succinct description of the underlying library mechanism.


Node:Data Types, Next:, Up:libmuauth

Data Types

mu_auth_fp Data Type

This is a pointer to authentication or authorization data. It is defined as follows:

typedef int (*mu_auth_fp) (void *return_data,
                           void *key,
                           void *func_data,
                           void *call_data);

Its arguments are:

return_data
1. Upon successful return authorization handler leaves in this memory location a pointer to the filled mu_auth_data structure with the user's information.

For authentication handlers this argument is always NULL and should be ignored.

key
The search key value. Its actual type depends upon type of the handler.

For authorization handlers it is const char* if the handler is called by mu_get_auth_by_name() and uid_t * if it is called by mu_get_auth_by_uid().

For authentication handlers it is always struct mu_auth_data* representing the user's data obtained by a previous call to a mu_get_auth_by_... function.

func_data
Any data associated with this handler.
call_data
Any call specific data. This argument is not used at the moment.

mu_auth_data Data Type

The mu_auth_data is used to return the information about the user. It is similar to system struct passwd, except that it is more mailutils-specific. Its definition is:

struct mu_auth_data {
  /* These are from struct passwd */
  char    *name;       /* user name */
  char    *passwd;     /* user password */
  uid_t   uid;         /* user id */
  gid_t   gid;         /* group id */
  char    *gecos;      /* real name */
  char    *dir;        /* home directory */
  char    *shell;      /* shell program */
  /* */
  char    *mailbox;    /* Path to the user's system mailbox */
  int     change_uid;  /* Should the uid be changed? */
};

mu_auth_module Data Type

The mu_auth_module structure contains full information about a libmuauth module. It is declared as follows:

struct mu_auth_module {
  char           *name;              /* Module name */
  struct argp    *argp;              /* Corresponding argp structure */
  mu_auth_fp     authenticate;       /* Authentication function ... */
  void           *authenticate_data; /* ... and its specific data */
  mu_auth_fp     auth_by_name;       /* Get user info by user name */
  void           *auth_by_name_data; /* ... and its specific data */
  mu_auth_fp     auth_by_uid;        /* Get user info by user id */
  void           *auth_by_uid_data;  /* ... and its specific data */
};


Node:Initializing the libmuauth, Next:, Previous:Data Types, Up:libmuauth

Initializing the libmuauth

void mu_auth_init (void) Function

This function registers the command line capability "auth". It must be called after registering libmuauth modules and before calling mu_agrp_parse(). If an error occurs, this function prints diagnostic message and aborts the program.

void MU_AUTH_REGISTER_ALL_MODULES (void) Function

This macro registers all default modules and calls mu_auth_init().


Node:Module Creation and Destruction, Next:, Previous:Initializing the libmuauth, Up:libmuauth

Module Creation and Destruction

int mu_auth_data_alloc (struct mu_auth_data **ptr, const char *name, const char *passwd, uid_t uid, gid_t gid, const char *gecos, const char *dir, const char *shell, const char *mailbox, int change_uid) Function

Create a mu_auth_data structure and initialize it with the given values. Returns 0 on success and 1 otherwise.

void mu_auth_data_free (struct mu_auth_data *ptr) Function

Free the mu_auth_data structure allocated by a call to mu_auth_data_alloc().

void mu_auth_register_module (struct mu_auth_module *mod) Function

Register the module defined by the mod argument.


Node:Obtaining Authorization Information, Next:, Previous:Module Creation and Destruction, Up:libmuauth

Obtaining Authorization Information

int mu_auth_runlist (list_t flist, void *return_data, void *key, void *call_data); Function

The list is expected to contain mu_auth_fp pointers. Each of them is dereferenced and executed until either the list is exhausted or any of the functions returns non-zero, whichever occurs first. The return_data and key arguments are passed as the first two parameters to the function (see the definition of mu_auth_fp, notice the footnote), the call_data is passed as its last parameter.

The function returns 0 if none of the functions from list succeeded, i.e. returned non-zero value. Otherwise it returns the return code from the succeeded function.

struct mu_auth_data * mu_get_auth_by_name (const char *username) Function

Search the information about given user by its username. Similar to system's getpwnam call).

struct mu_auth_data * mu_get_auth_by_uid (uid_t uid) Function
Search the information about given user by its uid. Similar to system's getpwuid call).

int mu_authenticate (struct mu_auth_data *auth_data, char *pass) Function

Authenticate the user whose data are in auth_data using password pass. Return 0 if the user is authenticated.


Node:Existing Modules, Next:, Previous:Obtaining Authorization Information, Up:libmuauth

Existing Modules

int mu_auth_nosupport (void *return_data, void *key, void *func_data, void *call_data); Function

The "not-supported" module. Always returns ENOSYS.

mu_auth_system_module Variable

This module is always registered even if libmuauth is not linked. It performs usual authentication using system user database (/etc/password et al.)

mu_auth_generic_module Variable

This module is always registered even if libmuauth is not linked. Both its authorization handlers are mu_auth_nosupport. Its authentication handler computes the MD5 or DES hash over the supplied password with the seed taken from passwd member of its key argument. Then it compares the obtained hash with the passwd member itself and returns 1 if both strings match.

mu_auth_pam_module Variable

Implements PAM authentication. Both authorization handlers are mu_auth_nosupport().

mu_auth_sql_module Variable

Implements authentication and authorization via MySQL database. The credentials for accessing the database are taken from global variables sql_host, sql_port, sql_user, sql_passwd and sql_db. The SQL queries for retrieving user information from global variables sql_getpwnam_query and sql_getpwuid_query. The variable sql_getpass_query keeps the query used for retrieving user's password. See auth, for information on command line options used to set these variables.

mu_auth_virtual_module Variable

Implements mu_get_auth_by_name method using virtual mail domains. Neither mu_get_auth_by_uid nor mu_authenticate is implemented. This module must be used together with generic module.


Node:Using Libmuauth in Your Programs, Previous:Existing Modules, Up:libmuauth

Using Libmuauth in Your Programs

To link your program against libmuauth, obtain loader arguments by running mailutils-config as follows:

mailutils-config --link auth

See mailutils-config, for more information about this utility.

Here is a sample Makefile fragment:

MU_LDFLAGS=`mailutils-config --link auth`
MU_INCLUDES=`mailutils-config --include`

myprog: myprog.c
        $(CC) -omyprog $(CFLAGS) $(MU_INCLUDES) myprog.c $(MU_LDFLAGS)

If your program will be using only default modules provided by the library, then it will suffice to call MU_AUTH_REGISTER_ALL_MODULES() somewhere near the start of your program. As an example, consider the following code fragment (it is taken from the imap4d daemon):

int
main (int argc, char **argv)
{
  struct group *gr;
  int status = EXIT_SUCCESS;

  state = STATE_NONAUTH; /* Starting state in non-auth.  */

  MU_AUTH_REGISTER_ALL_MODULES ();
  mu_argp_parse (&argp, &argc, &argv, 0, imap4d_capa,
                 NULL, &daemon_param);
  ...

Otherwise, if your program will use it's own modules, first register them with mu_auth_register_module and then call mu_auth_init(), e.g.:

struct mu_auth_module radius_module = {
  ...
};

struct mu_auth_module ext_module = {
  ...
};

int
main (int argc, char **argv)
{
  mu_auth_register_module (&radius_module);
  mu_auth_register_module (&ext_module);
  mu_auth_init ();

  ...

These two approaches may be combined, allowing you to use both your modules and the ones provided by Mailutils. Consider the example below:

int
main (int argc, char **argv)
{
  mu_auth_register_module (&radius_module);
  mu_auth_register_module (&ext_module);
  MU_AUTH_REGISTER_ALL_MODULES ();

  ...
}


Node:libmu_scm, Next:, Previous:libmuauth, Up:Top

Mailutils to Scheme interface

The library libmu_scm provides an interface between Mailutils and Guile, allowing to access the Mailutils functionality from a Scheme program. For more information about Guile, refer to Overview. For information about Scheme programming language, See Top.

Functions Provided by libmu_scm


Node:Address Functions, Next:, Up:libmu_scm

Address Functions

mu-address-get-personal ADDRESS NUM Scheme Function
Return personal part of an email address.

mu-address-get-comments ADDRESS NUM Scheme Function

mu-address-get-email ADDRESS NUM Scheme Function
Return email part of an email address.

mu-address-get-domain ADDRESS NUM Scheme Function
Return domain part of an email address

mu-address-get-local ADDRESS NUM Scheme Function
Return local part of an email address.

mu-address-get-count ADDRESS Scheme Function
Return number of parts in email address.


Node:Mailbox Functions, Next:, Previous:Address Functions, Up:libmu_scm

Mailbox Functions

mu-mailbox-open URL MODE Scheme Function
Opens a mailbox specified by URL.

mu-mailbox-close MBOX Scheme Function
Closes mailbox MBOX

mu-mailbox-get-url MBOX Scheme Function
Returns the URL of the mailbox.

mu-mailbox-get-port MBOX MODE Scheme Function
Returns a port associated with the contents of the MBOX. MODE is a string defining operation mode of the stream. It may contain any of the two characters: r for reading, w for writing.

mu-mailbox-get-message MBOX MSGNO Scheme Function
Retrieve from MBOX message # MSGNO.

mu-mailbox-messages-count MBOX Scheme Function
Returns number of messages in the mailbox.

mu-mailbox-expunge MBOX Scheme Function
Expunges deleted messages from the mailbox.

mu-mailbox-url MBOX Scheme Function
Returns the URL of the mailbox

mu-mailbox-append-message MBOX MESG Scheme Function
Appends the message to the mailbox


Node:Message Functions, Next:, Previous:Mailbox Functions, Up:libmu_scm

Message Functions

mu-message-copy MESG Scheme Function
Creates the copy of the given message.

mu-message-set-header MESG HEADER VALUE REPLACE Scheme Function
Sets new VALUE to the header HEADER of the message MESG. If the HEADER is already present in the message its value is replaced with the supplied one if the optional REPLACE is #t. Otherwise new header is created and appended.

mu-message-get-size MESG Scheme Function
Returns the size of the given message.

mu-message-get-lines MESG Scheme Function
Returns number of lines in the given message.

mu-message-get-sender MESG Scheme Function
Returns the sender email address for the message MESG.

mu-message-get-header MESG HEADER Scheme Function
Returns the header value of the HEADER in the MESG.

mu-message-get-header-fields MESG HEADERS Scheme Function
Returns the list of headers in the MESG. If optional HEADERS is specified it should be a list of header names to restrict return value to.

mu-message-set-header-fields MESG LIST REPLACE Scheme Function
Set the headers in the message MESG from LIST LIST is a list of (cons HEADER VALUE) Optional parameter REPLACE specifies whether the new header values should replace the headers already present in the message.

mu-message-delete MESG FLAG Scheme Function
Mark given message as deleted. Optional FLAG allows to toggle deleted mark The message is deleted if it is #t and undeleted if it is #f

mu-message-get-flag MESG FLAG Scheme Function
Return value of the attribute FLAG.

mu-message-set-flag MESG FLAG VALUE Scheme Function
Set the given attribute of the message. If optional VALUE is #f, the attribute is unset.

mu-message-get-user-flag MESG FLAG Scheme Function
Returns value of the user attribute FLAG.

mu-message-set-user-flag MESG FLAG VALUE Scheme Function
Set the given user attribute of the message. If optional VALUE is #f, the attribute is unset.

mu-message-get-port MESG MODE FULL Scheme Function
Returns a port associated with the given MESG. MODE is a string defining operation mode of the stream. It may contain any of the two characters: r for reading, w for writing. If optional FULL argument specified, it should be a boolean value. If it is #t then the returned port will allow access to any part of the message (including headers). If it is #f then the port accesses only the message body (the default).

mu-message-get-body MESG Scheme Function
Returns the message body for the message MESG.

mu-message-send MESG MAILER Scheme Function
Sends the message MESG. Optional MAILER overrides default mailer settings in mu-mailer.


Node:MIME Functions, Next:, Previous:Message Functions, Up:libmu_scm

MIME Functions

mu-mime-create FLAGS MESG Scheme Function
Creates a new MIME object.

mu-mime-multipart? MIME Scheme Function
Returns #t if MIME is a multipart object.

mu-mime-get-num-parts MIME Scheme Function
Returns number of parts in a MIME object.

mu-mime-get-part MIME PART Scheme Function
Returns part number PART from a MIME object.

mu-mime-add-part MIME MESG Scheme Function
Adds MESG to the MIME object.

mu-mime-get-message MIME Scheme Function
Converts MIME object to a message.


Node:Log Functions, Next:, Previous:MIME Functions, Up:libmu_scm

Log Functions

mu-openlog IDENT OPTION FACILITY Scheme Function
Opens a connection to the system logger for Guile program.

mu-logger PRIO TEXT Scheme Function
Generates a log message to be distributed via syslogd.

mu-closelog Scheme Function
Closes the channel to the system logger open by mu-openlog.


Node:Direct Linking, Next:, Previous:Log Functions, Up:libmu_scm

Direct Linking

If you plan to link your program directly to libguile, it will probably make sense to link libmu_scm directly as well. The arguments to the program loader may be obtained by running

mailutils-config --link guile

See mailutils-config, for more information about this utility.

Here is a sample Makefile fragment:

MU_LDFLAGS=`mailutils-config --link guile`
MU_INCLUDES=`mailutils-config --include`

myprog: myprog.c
        $(CC) -omyprog $(CFLAGS) $(MU_INCLUDES) myprog.c $(MU_LDFLAGS)


Node:Dynamic Linking, Previous:Direct Linking, Up:libmu_scm

Dynamic Linking

Dynamic linking is the preferred method of using libmu_scm. It uses Guile "use-modules" mechanism. An interface module mailutils.scm is provided in order to facilitate using this method. This module is installed in the package data directory (by default it is prefix/share/mailutils). A sample use of this module is:

(set! %load-path (list "/usr/local/share/mailutils"))
(use-modules (mailutils))

# Now you may use mailutils functions:

(let ((mb (mu-mailbox-open "/var/spool/mail/gray" "r")))
...

Note, that you should explicitly modify the %load-path before calling use-modules, otherwise Guile will not be able to find mailutils.scm.


Node:libsieve, Next:, Previous:libmu_scm, Up:Top

Sieve Library

This chapter describes GNU Sieve library.


Node:Library Description, Next:, Up:libsieve

Library Description

Libsieve is GNU implementation of the mail filtering language Sieve.

The library is built around a Sieve Machine -- an abstract computer constructed specially to handle mail filtering tasks. This computer has two registers: program counter and numeric accumulator; a runtime stack of unlimited depth and the code segment. A set of functions is provided for creating and destroying instances of Sieve Machine, manipulating its internal data, compiling and executing a sieve program.

The following is a typical scenario of using libsieve:

  1. Application program creates the instance of sieve machine.
  2. Then sieve_compile function is called to translate the Sieve source into an equivalent program executable by the Machine
  3. A mailbox is opened and associated with the Machine
  4. The Machine executes the program over the mailbox
  5. When the execution of the program is finished, all messages upon which an action was executed other than keep are marked with the delete flag. Thus, running mailbox_expunge upon the mailbox finishes the job, leaving in the mailbox only those messages that were preserved by the filter.
  6. Finally, the instance of Sieve Machine is destroyed and the resources allocated for it are reclaimed.

The following sections describe in detail the functions from the Sieve Library.


Node:Sieve Data Types, Next:, Up:Library Description

Sieve Data Types

sieve_machine_t Data Type
This is an opaque data type representing a pointer to an instance of sieve machine. The sieve_machine_t keeps all information necessary for compiling and executing the script.

It is created by sieve_machine_create() and destroyed by sieve_machine_destroy(). The functions for manipulating this data type are described in Manipulating the Sieve Machine.

sieve_data_type Data Type
This enumeration keeps the possible types of sieve data. These are:
SVT_VOID
No datatype.
SVT_NUMBER
Numeric type.
SVT_STRING
Character string.
SVT_STRING_LIST
A list_t. Each item in this list represents a character string.
SVT_TAG
A sieve tag. See sieve_runtime_tag_t below.
SVT_IDENT
A character string representing an identifier.
SVT_VALUE_LIST
A list_t. Each item in this list is of sieve_value_t.
SVT_POINTER
An opaque pointer.

sieve_value_t Data Type
The sieve_value_t keeps an instance of sieve data. It is defined as follows:
typedef struct {
  sieve_data_type type;        /* Type of the data */
  union {
    char *string;              /* String value or identifier */
    size_t number;               /* Numeric value */
    list_t list;               /* List value */
    sieve_runtime_tag_t *tag;  /* Tag value */
    void *ptr;                 /* Pointer value */
  } v;
} sieve_value_t;

Depending on the value of type member, following members of the union v keep the actual value:

SVT_VOID
Never appears.
SVT_NUMBER
The numeric value is kept in number member.
SVT_STRING
The string is kept in string member.
SVT_STRING_LIST
SVT_VALUE_LIST
The list itself is pointed to by list member
SVT_TAG
The tag value is pointed to by tag member.
SVT_IDENT
The string member points to the identifier name.
SVT_POINTER
The data are pointed to by ptr member.

sieve_tag_def_t Data Type
This structure represents a definition of a tagged (optional) argument to a sieve action or test. It is defined as follows:
typedef struct {
  char *name;              /* Tag name */
  sieve_data_type argtype; /* Type of tag argument. */
} sieve_tag_def_t;

The name member points to the tag's name without leading colon. The argtype is set to SVT_VOID if the tag does not take argument, or to the type of the argument otherwise.

sieve_runtime_tag_t Data Type
This structure represents the tagged (optional) argument at a runtime. It is defined as:
struct sieve_runtime_tag {
  char *tag;                /* Tag name */
  sieve_value_t *arg;       /* Tag argument (if any) */
};

The arg member is NULL if the tag does not take an argument.

sieve_handler_t Data Type

This is a pointer to function handler for a sieve action or test. It is defined as follows:

typedef int (*sieve_handler_t) (sieve_machine_t mach,
                                list_t args, list_t tags);

The arguments to the handler have the following meaning:

mach
Sieve machine being processed.
args
A list of required arguments to the handler
tags
A list of optional arguments (tags).

sieve_printf_t Data Type
A pointer to a diagnostic output function. It is defined as follows:
typedef int (*sieve_printf_t) (void *data, const char *fmt, va_list ap);

data
A pointer to application specific data. These data are passed as second argument to sieve_machine_init().
fmt
Printf-like format string.
ap
Other arguments.

sieve_parse_error_t Data Type
This data type is declared as follows:
typedef int (*sieve_parse_error_t) (void *data,
                                    const char *filename, int lineno,
                                    const char *fmt, va_list ap);

It is used to declare error handlers for parsing errors. The application-specific data are passed in the data argument. Arguments filename and line indicate the location of the error in the source text, while fmt and ap give verbose description of the error.

sieve_action_log_t Data Type
A pointer to the application-specific logging function:
typedef void (*sieve_action_log_t) (void *data,
                                    const char *script,
                                    size_t msgno, message_t msg,
                                    const char *action,
                                    const char *fmt, va_list ap);

data
Application-specific data.
script
Name of the sieve script being executed.
msgno
Ordinal number of the message in mailbox, if appropriate. When execution is started using sieve_message(), this argument is zero.
msg
The message this action is executed upon.
action
The name of the action.
fmt
var
These two arguments give the detailed description of the action.

sieve_comparator_t Data Type

typedef int (*sieve_comparator_t) (const char *, const char *);

A pointer to the comparator handler function. The function compares its two operands and returns 1 if they are equal, and 0 otherwise. Notice, that the sense of the return value is inverted in comparison with most standard libc functions like stcmp(), etc.

sieve_retrieve_t Data Type

typedef int (*sieve_retrieve_t) (void *item, void *data, int idx,
                                 char **pval);

A pointer to generic retriever function. See description of sieve_vlist_compare() for details of its usage.

sieve_destructor_t Data Type

typedef void (*sieve_destructor_t) (void *data);

A pointer to destructor function. The function frees any resources associated with data. See the description of sieve_machine_add_destructor() for more information.

sieve_tag_checker_t Data Type

typedef int (*sieve_tag_checker_t) (const char *name,
                                    list_t tags,
                                    list_t args)

A pointer to tag checker function. The purpose of the function is to perform compilation-time consistency test on tags. Its arguments are:

name
Name of the test or action whose tags are being checked.
tags
A list of sieve_runtime_tag_t representing tags.
args
A list of sieve_value_t representing required arguments to name.

The function is allowed to make any changes in tags and args. It should return 0 if the syntax is correct and non-zero otherwise. It is responsible for issuing the diagnostics in the latter case. [FIXME: describe how to do that]


Node:Manipulating the Sieve Machine, Next:, Previous:Sieve Data Types, Up:Library Description

Manipulating the Sieve Machine

This subsection describes functions used to create an instance of the sieve machine, read or alter its internal fields and destroy it.

int sieve_machine_init (sieve_machine_t *mach, void *data) Function

The sieve_machine_init() function creates an instance of a sieve machine. A pointer to the instance itself is returned in the argument mach. The user-specific data to be associated with the new machine are passed in data argument. The function returns 0 on success, non-zero error code otherwise,

void sieve_machine_destroy (sieve_machine_t *pmach) Function

This function destroys the instance of sieve machine pointed to by mach parameter. After execution of sieve_machine_destroy() pmach contains NULL. The destructors registered with sieve_machine_add_destructor() are executed in LIFO order.

int sieve_machine_add_destructor (sieve_machine_t mach, sieve_destructor_t destr, void *ptr); Function

This function registers a destructor function dest. The purpose of the destructor is to free any resources associated with the item ptr. The destructor function takes a single argument -- a pointer to the data being destroyed. All registered destructors are called in reverse order upon execution of sieve_machine_destroy(). Here's a short example of the use of this function:

static void
free_regex (void *data)
{
  regfree ((regex_t*)data);
}

int
match_part_checker (const char *name, list_t tags, list_t args)
{
  regex_t *regex;

  /* Initialise the regex: */
  regex = sieve_malloc (mach, sizeof (*regex));
  /* Make sure it will be freed when necessary */
  sieve_machine_add_destructor (sieve_machine, free_regex, regex);
  .
  .
  .
}

void *sieve_get_data (sieve_machine_t mach) Function
This function returns the application-specific data associated with the instance of sieve machine. See sieve_machine_init().

message_t sieve_get_message (sieve_machine_t mach) Function
This function returns the current message.

size_t sieve_get_message_num (sieve_machine_t mach); Function
This function returns the current message number in the mailbox. If there are no mailbox, i.e. the execution of the sieve code is started with sieve_message, this function returns 1.

int sieve_get_debug_level (sieve_machine_t mach) Function
Returns the debug level set for this instance of sieve machine.

ticket_t sieve_get_ticket (sieve_machine_t mach) Function
Returns the authentication ticket for this machine.

mailer_t sieve_get_mailer (sieve_machine_t mach) Function
Returns the mailer.

char * sieve_get_daemon_email __P((sieve_machine_t mach) Function
This function returns the daemon email associated with this instance of sieve machine. The daemon email is an email address used in envelope from addresses of automatic reply messages. By default its local part is <MAILER-DAEMON> and the domain part is the machine name.

void sieve_set_error (sieve_machine_t mach, sieve_printf_t error_printer) Function
This function sets the error printer function for the machine. If it is not set, the default error printer will be used. It is defined as follows:
int
_sieve_default_error_printer (void *unused, const char *fmt,
                              va_list ap)
{
  return mu_verror (fmt, ap);
}

void sieve_set_parse_error (sieve_machine_t mach, sieve_parse_error_t p) Function
This function sets the parse error printer function for the machine. If it is not set, the default parse error printer will be used. It is defined as follows:
int
_sieve_default_parse_error (void *unused,
                            const char *filename, int lineno,
			    const char *fmt, va_list ap)
{
  if (filename)
    fprintf (stderr, "%s:%d: ", filename, lineno);
  vfprintf (stderr, fmt, ap);
  fprintf (stderr, "\n");
  return 0;
}

void sieve_set_debug (sieve_machine_t mach, sieve_printf_t debug)); Function
This function sets the debug printer function for the machine. If it is not set, the default debug printer is NULL which means no debugging information will be displayed.

void sieve_set_debug_level (sieve_machine_t mach, mu_debug_t dbg, int level) Function
This function sets the debug level for the given instance of sieve machine. The dbg argument is the mu_debug_t object to be used with mailutils library, the level argument specifies the debugging level for the sieve library itself. It is a bitwise or of the following values:
MU_SIEVE_DEBUG_TRACE
Trace the execution of the sieve script.
MU_SIEVE_DEBUG_INSTR
Print the sieve machine instructions as they are executed.
MU_SIEVE_DEBUG_DISAS
Dump the disassembled code of the sieve machine. Do not run it.
MU_SIEVE_DRY_RUN
Do not executed the actions, only show what would have been done.

void sieve_set_logger (sieve_machine_t mach, sieve_action_log_t logger) Function
This function sets the logger function. By default the logger function is NULL, which means that the executed actions are not logged.

void sieve_set_ticket (sieve_machine_t mach, ticket_t ticket) Function
This function sets the authentication ticket to be used with this machine.

void sieve_set_mailer (sieve_machine_t mach, mailer_t mailer) Function
This function sets the mailer. The default mailer is "sendmail:".

void sieve_set_daemon_email (sieve_machine_t mach, const char *email) Function
This functions sets the daemon email for reject and redirect actions.

int sieve_is_dry_run (sieve_machine_t mach) Function
The sieve_is_dry_run() returns 1 if the machine is in dry run state, i.e. it will only log the actions that would have been executed without actually executing them. The dry run state is set by calling sieve_set_debug_level() if its last argument has the MU_SIEVE_DRY_RUN bit set.

const char * sieve_type_str (sieve_data_type type) Function
Returns the string representation for the given sieve data type. The return value is a pointer to a static constant string.


Node:Logging and Diagnostic Functions, Next:, Previous:Manipulating the Sieve Machine, Up:Library Description

Logging and Diagnostic Functions

void sieve_error (sieve_machine_t mach, const char *fmt, ...) Function
Format and output an error message using error printer of the machine mach.

void sieve_debug (sieve_machine_t mach, const char *fmt, ...) Function
Format and output a debug message using debug printer of the machine mach.

void sieve_log_action (sieve_machine_t mach, const char *action, const char *fmt, ...) Function
Log a sieve action using logger function associated with the machine mach.

void sieve_abort (sieve_machine_t mach) Function
Immediately abort the execution of the script.


Node:Symbol Space Functions, Next:, Previous:Logging and Diagnostic Functions, Up:Library Description

Symbol Space Functions

sieve_register_t * sieve_test_lookup (sieve_machine_t mach, const char *name) Function
Find a register object describing the test name. Returns NULL if no such test exists.

sieve_register_t * sieve_action_lookup (sieve_machine_t mach, const char *name) Function
Find a register object describing the action name. Returns NULL if no such action exists.

int sieve_register_test (sieve_machine_t mach, const char *name, sieve_handler_t handler, sieve_data_type *arg_types, sieve_tag_group_t *tags, int required) Function

int sieve_register_action (sieve_machine_t mach, const char *name, sieve_handler_t handler, sieve_data_type *arg_types, sieve_tag_group_t *tags, int required) Function

int sieve_register_comparator (sieve_machine_t mach, const char *name, int required, sieve_comparator_t is, sieve_comparator_t contains, sieve_comparator_t matches, sieve_comparator_t regex) Function

int sieve_tag_lookup (list_t taglist, char *name, sieve_value_t **arg) Function

int sieve_load_ext (sieve_machine_t mach, const char *name) Function


Node:Memory Allocation, Next:, Previous:Symbol Space Functions, Up:Library Description

Memory Allocation

The following functions act as their libc counterparts. The allocated memory is associated with the mach argument and is automatically freed upon the call to sieve_machine_destroy (mach).

void * sieve_malloc (sieve_machine_t mach, size_t size) Function
Allocates size bytes and returns a pointer to the allocated memory.

char * sieve_mstrdup (sieve_machine_t mach, const char *str) Function
This function returns a pointer to a new string which is a duplicate of the string str.

void * sieve_mrealloc (sieve_machine_t mach, void *ptr, size_t size) Function
Changes the size of the memory block pointed to by ptr to size bytes. The contents will be unchanged to the minimum of the old and new sizes; newly allocated memory will be uninitialized. If ptr is NULL, the call is equivalent to sieve_malloc(mach, size); if size is equal to zero, the call is equivalent to sieve_mfree(ptr). Unless ptr is NULL, it must have been returned by an earlier call to sieve_malloc() or sieve_mrealloc().

void sieve_mfree (sieve_machine_t mach, void *ptr) Function
sieve_mfree() frees the memory space pointed to by ptr and detaches it from the destructor list of mach. The ptr must have been returned by a previous call to sieve_malloc() or sieve_mrealloc(). Otherwise, or if sieve_mfree(ptr) has already been called before, undefined behaviour occurs.

If ptr is NULL, no operation is performed.


Node:Compiling and Executing the Script, Previous:Memory Allocation, Up:Library Description

Compiling and Executing the Script

int sieve_compile (sieve_machine_t mach, const char *name) Function
Compile the sieve script from the file name.

int sieve_mailbox (sieve_machine_t mach, mailbox_t mbox) Function
Execute the code from the given instance of sieve machine mach over each message in the mailbox mbox.

int sieve_message (sieve_machine_t mach, message_t message) Function
Execute the code from the given instance of sieve machine mach over the message.

int sieve_disass (sieve_machine_t mach) Function
Dump the disassembled code of the sieve machine mach.


Node:Input Language, Next:, Previous:Library Description, Up:libsieve

Input Language

The input language understood by the GNU Sieve Library is a superset of the Sieve language as described in RFC 3028.


Node:Lexical Structure, Next:, Up:Input Language

Lexical Structure

Whitespace and Comments

Comments are semantically equivalent to whitespace and can be used anyplace that whitespace is (with one exception in multi-line strings, as described below).

There are two kinds of comments: hash comments, that begin with a # character that is not contained within a string and continue until the next newline, and C-style or bracketed comments, that are delimited by /* and */ tokens. The bracketed comments may span multiple lines. E.g.:

if size :over 100K
  { # this is a comment
    discard;
  }

if size :over 100K
  { /* this is a comment
       this is still a comment */ discard /* this is a comment again
     */ ;
  }

Like in C, bracketed comments do not nest.

Lexical Tokens

The basic lexical entities are identifiers and literals.

An identifier is a sequence of letters, digits and underscores, started with a letter or underscore. For example, header and check_822_again are valid identifiers, whereas 1st is not. A special form of identifier is tag: it is an identifier prefixed with a colon (:), e.g.: :comparator.

A literal is a data that is not executed, merely evaluated "as is", to be used as arguments to commands. There are four kinds of literals:


Node:Syntax, Next:, Previous:Lexical Structure, Up:Input Language

Syntax

Being designed for the sole purpose of filtering mail, Sieve has a very simple syntax.


Node:Commands, Next:, Up:Syntax
Commands

The basic syntax element is a command. It is defined as follows:

command-name [tags] args

where command-name is an identifier representing the name of the command, tags is an optional list of optional or tagged arguments and args is a list of required or positional arguments.

Positional arguments are literals delimited with whitespace. They provide the command with the information necessary to its proper functioning. Each command has a fixed number of positional arguments. It is an error to supply more arguments to the command or to give it fewer arguments than it accepts.

Optional arguments allow to modify the behaviour of the command, like command line options in UNIX do. They are a list of tags (see Lexical Structure) separated by whitespace. An optional argument may have at most one parameter.

Each command understands a set of optional arguments. Supplying it tags that it does not understand results in an error.

For example, consider the following command

header :mime :comparator "i;octet" ["to", "from"] "[email protected]"

Here, given that header takes two positional arguments: header is command name, the list ["to", "from"] is first positional argument and the string "[email protected]" is second positional argument. There are two optional arguments: :mime and :comparator. The latter has a string "i;octet" as its parameter.


Node:Actions Described, Next:, Previous:Commands, Up:Syntax
Actions Described

An action is a Sieve command that performs some operation over the message. Actions do the main job in any Sieve program. Syntactically, an action is a command terminated with semicolon, e.g.:

keep;

fileinto "mbox";

GNU Sieve provides the full set of actions described in RFC 3028. It also allows to extend this set using loadable actions. See Actions, for detailed discussion of actions.


Node:Control Flow, Next:, Previous:Actions Described, Up:Syntax
Control Flow

The only control flow statement Sieve has is "if" statement. In its simplest form it is:

if condition { ... }

The effect of this statement is that the sequence of actions between the curly braces is executed only if the condition evaluates to true.

A more elaborate form of this statement allows to execute two different sets of actions depending on whether the condition is true or not:

if condition { ... } else { ... }

The most advanced form of the "if" statement allows to select an action depending on what condition from the set of conditions is met.

if cond1 { ... } elsif cond2 { ... } else { ... }

There may be any number of "elsif" branches in an "if" statement. However it may have at most one "else" branch. Notes for C programmers:

  1. The braces surrounding each branch of an "if" statement are required.
  2. The "else if" construct is disallowed. Use "elsif" keyword instead.

Here's an example of "if" statement:

if header :contains "from" "coyote"
  {
    discard;
  }
elsif header :contains ["subject"] ["$$$"]
  {
    discard;
  }
else
  {
    fileinto "INBOX";
  }

The following section describes in detail conditions used in "if" statements.


Node:Tests and Conditions, Previous:Control Flow, Up:Syntax
Tests and Conditions

Tests are Sieve commands that return boolean value. E.g. the test

header :contains "from" "coyote"

returns true only if the header "From" of the current message contains substring "coyote".

The tests shipped with the GNU Sieve are described in Tests.

Condition is a Sieve expression that evaluates to true or false. In its simplest form, condition is just a Sieve test.

To reverse the sense of a condition use keyword not, e.g.:

not header :contains "from" "coyote"

The results of several conditions may be joined together by logical and and or operations. The special form allof takes several tests as its arguments and computes the logical and of their results. Similarly, the form anyof performs logical or over the results of its arguments. E.g.:

if anyof (not exists ["From", "Date"],
          header :contains "from" "[email protected]")
  {
    discard;
  }


Node:Preprocessor, Next:, Previous:Syntax, Up:Input Language

Preprocessor

The preprocessor statements are a GNU extension to the Sieve language. The syntax for a preprocessor statement is similar to that used in C programming language, i.e.: a pound character (#) followed by a preprocessor directive and its arguments. Any amount of whitespace can be inserted between the # and the directive. Currently implemented directives are include and searchpath.


Node:#include, Next:, Up:Preprocessor
Sieve #include directive

The #include directive reads in the contents of the given file. The contents is "inserted" into the text being parsed starting at the line where the directive appears. The directive takes two forms:

#include "filename"
The filename is taken relative to the current directory.
#include <filename>"
The filename is searched in the list of include directories as specified by the -I command line options.

If filename starts with a directory separator character (/) both forms have the same effect.


Node:#searchpath, Previous:#include, Up:Preprocessor
Sieve #searchpath directive

The #searchpath directive adds its argument to the list of directories searched for loadable modules. It has the same effect as -L command line switch used by GNU sieve utility (see sieve group).


Node:Require Statement, Next:, Previous:Preprocessor, Up:Input Language

Require Statement


Syntax:   require string;
          require string-list;

The require statement informs the parser that a script makes use of a certain extension. Multiple capabilities can be declared using the second form of the statement. The actual handling of a capability name depends on its suffix.

If the name starts with comparator-, it is understood as a request to use the specified comparator. The comparator name consists of the characters following the suffix.

If the name starts with test-, it means a request to use the given test. The test name consists of the characters following the suffix.

Otherwise, the capability is understood as a name of an action to be used.

The require statement, if present, must be used before any other statement that is using the required capability. As an extension, the GNU sieve allows the require and any other statements to be interspersed.

By default the following actions and comparators are always required:

Example:

require ["fileinto", "reject"];

require "fileinto";

require "comparator-i;ascii-numeric";

When processing arguments for require statement, GNU libsieve uses the following algorithm:

  1. Look up the name in a symbol table. If the name begins with comparator- it is looked up in the comparator table. If it begins with test-, the test table is used instead. Otherwise the name is looked up in the action table.
  2. If the name is found, the search is terminated.
  3. Otherwise, transform the name. First, any comparator- or test- prefix is stripped. Then, any character other than alphanumeric characters, . and , is replaced with dash (-). The name thus obtained is used as a file name of an external loadable module.
  4. Try to load the module. The module is searched in the following search paths (in the order given):
    1. Mailutils module directory. By default it is $prefix/lib/mailutils.
    2. Sieve library path as given with the -L options in the command line
    3. Additional search directories specified with the #searchpath directive.
    4. The value of the environment variable LTDL_LIBRARY_PATH.
    5. System library search path: The system dependent library search path (e.g. on Linux it is set by the contents of the file /etc/ld.so.conf and the value of the environment variable LD_LIBRARY_PATH).

    The value of LTDL_LIBRARY_PATH and LD_LIBRARY_PATH must be a colon-separated list of absolute directories, for example, "/usr/lib/mypkg:/lib/foo".

    In any of these directories, libsieve first attempts to find and load the given filename. If this fails, it tries to append the following suffixes to the file name:

    1. the libtool archive extension .la
    2. the extension used for native dynamic libraries on the host platform, e.g., .so, .sl, etc.
  5. If the module is found, libsieve executes its initialization function (see below) and again looks up the name in the symbol table. If found, search terminates successfully.
  6. If either the module is not found, or the symbol wasn't found after execution of the module initialization function, search is terminated with an error status. libsieve then issues the following diagnostic message:
    source for the required action NAME is not available
    


Node:Comparators, Next:, Previous:Require Statement, Up:Input Language

Comparators

GNU libsieve supports the following built-in comparators:

i;octet
This comparator simply compares the two arguments octet by octet
i;ascii-casemap
It treats uppercase and lowercase characters in the ASCII subset of UTF-8 as the same. This is the default comparator.
i;ascii-numeric
Treats the two arguments as ASCII representation of decimal numbers and compares their numeric values. This comparator must be explicitly required prior to use.


Node:Tests, Next:, Previous:Comparators, Up:Input Language

Tests

This section describes the built-in tests supported by GNU libsieve. In the discussion below the following macro-notations are used:

match-type
This tag specifies the matching type to be used with the test. It can be one of the following:
:is
The :is match type describes an absolute match; if the contents of the first string are absolutely the same as the contents of the second string, they match. Only the string "frobnitzm" is the string "frobnitzm". The null key ":is" and only ":is" the null value. This is the default match-type.
:contains
The :contains match type describes a substring match. If the value argument contains the key argument as a substring, the match is true. For instance, the string "frobnitzm" contains "frob" and "nit", but not "fbm". The null key "" is contained in all values.
:matches
The :matches version specifies a wildcard match using the characters * and ?. * matches zero or more characters, and ? matches a single character. ? and * may be escaped as \\? and \\* in strings to match against themselves. The first backslash escapes the second backslash; together, they escape the *.
:regex
The :regex version specifies a match using POSIX Extended Regular Expressions.

comparator
A comparator syntax item is defined as follows:
:comparator "comparator-name"

It instructs sieve to use the given comparator with the test. If comparator-name is not one of i;octet, i;ascii-casemap it must be required prior to using it. For example:

require "comparator-i;ascii-numeric";

if header :comparator "i;ascii-numeric" :is "X-Num" "10"
  {
    ...

address-part
This syntax item is used when testing structured Internet addresses. It specifies which part of an address must be used in comparisons. Exactly one of the following tags may be used:
:all
Use the whole address. This is the default.
:localpart
Use local part of the address.
:domain
Use domain part of the address.

Notice, that match-type modifiers interact with comparators. Some comparators are not suitable for matching with :contains or :matches. If this occurs, sieve issues an appropriate error message. For example, the statement:

if header :matches :comparator "i;ascii-numeric"

would result in the following error message:

comparator `i;ascii-numeric' is incompatible with match type `:matches'
in call to `header'

false Test

This test always evaluates to "false".

true Test

This test always evaluates to "true".

address [address-part][comparator][match-type] header-names key-list Test

Tagged arguments:

address-part
Selects the address part to compare. Default is the whole email address (:all).
comparator
Specifies the comparator to be used instead of the default i;ascii-casemap.
match-type
Specifies the match type to be used instead of the default :is.

Required arguments:

header-names
A list of header names.
key-list
A list of address values.

The address test matches Internet addresses in structured headers that contain addresses. It returns true if any header contains any key in the specified part of the address, as modified by comparator and match-type optional arguments.

This test returns true if any combination of the header-names and key-list arguments match.

The address primitive never acts on the phrase part of an email address, nor on comments within that address. Use the header test instead. It also never acts on group names, although it does act on the addresses within the group construct.

Example:

if address :is :all "from" "[email protected]"
  {
     discard;
  }

size [:over|:under] number Test

The size test deals with the size of a message. The required argument number represents the size of the message in bytes. It may be suffixed with the following quantifiers:

k
K
The number is expressed in kilobytes.
m
M
The number is expressed in megabytes.
g
G
The number is expressed in gigabytes.

If the tagged argument is :over, and the size of the message is greater than number, the test is true; otherwise, it is false.

If the argument is :under, and the size of the message is less than the number, the test is true; otherwise, it is false.

Otherwise, the test is true only if the size of the message equals exactly number. This is a GNU extension.

The size of a message is defined to be the number of octets from the initial header until the last character in the message body.

envelope [address-part][comparator][match-type] envelope-part key-list Test

Tagged arguments:

address-part
Selects the address part to compare. Default is the whole email address (:all).
comparator
Specifies the comparator to be used instead of the default i;ascii-casemap.
match-type
Specifies the match type to be used instead of the default :is.

Required arguments:

envelope-parts
A list of envelope parts to operate upon.
key-list
A list of address values.

The envelope test is true if the specified part of the SMTP envelope matches the specified key.

If the envelope-part strings is (case insensitive) from, then matching occurs against the FROM address used in the SMTP MAIL command.

Notice, that due to the limitations imposed by SMTP envelope structure the use of any other values in envelope-parts header is meaningless.

exists header-names Test

Required arguments:

header-names
List of message header names.

The exists test is true if the headers listed in header-names argument exist within the message. All of the headers must exist or the test is false.

The following example throws out mail that doesn't have a From header and a Date header:

if not exists ["From","Date"]
  {
     discard;
  }

header [comparator] [match-type] [:mime] header-names key-list Test

Tagged arguments:

comparator
Specifies the comparator to be used instead of the default i;ascii-casemap.
match-type
Specifies the match type to be used instead of the default :is.
:mime
This tag instructs header to search through the mime headers in multipart messages as well.

Required arguments:

header-names
A list of header names.
key-list
A list of header values.

The header test evaluates to true if any header name matches any key. The type of match is specified by the optional match argument, which defaults to ":is" if not explicitly given.

The test returns true if any combination of the header-names and key-list arguments match.

If a header listed in header-names exists, it contains the null key (""). However, if the named header is not present, it does not contain the null key. So if a message contained the header

X-Caffeine: C8H10N4O2

these tests on that header evaluate as follows:

header :is ["X-Caffeine"] [""] => false
header :contains ["X-Caffeine"] [""] => true

numaddr [:over|:under] header-names number Test

This test is provided as an example of loadable extension tests. You must use require "test-numaddr" statement before actually using it.

The numaddr test counts Internet addresses in structured headers that contain addresses. It returns true if the total number of addresses satisfies the requested relation.

If the tagged argument is :over and the number of addresses is greater than number, the test is true; otherwise, it is false.

If the tagged argument is :under and the number of addresses is less than number, the test is true; otherwise, it is false.

If the tagged argument is not given, :over is assumed.


Node:Actions, Previous:Tests, Up:Input Language

Actions

The GNU libsieve supports the following default actions:

Among them the first three actions do not need to be explicitly required by a require statement, while the others do.

These actions are described in detail below.

stop Action

The stop action ends all processing. If no actions have been executed, then the keep action is taken.

keep Action

The effect of this action is to preserve the current message in the mailbox. This action is executed if no other action has been executed.

discard Action

Discard silently throws away the current message. No notification is returned to the sender, the message is deleted from the mailbox.

Example:

if header :contains ["from"] ["[email protected]"]
  {
    discard;
  }

fileinto folder Action

Required arguments:

folder
A string representing the folder name

The fileinto action delivers the message into the specified folder.

reject reason Action

The optional reject action refuses delivery of a message by sending back a message delivery notification to the sender. It resends the message to the sender, wrapping it in a "reject" form, noting that it was rejected by the recipient. The required argument reason is a string specifying the reason for rejecting the message.

Example:

If the message contained

Date: Tue, 1 Apr 1997 09:06:31 -0800 (PST)
From: [email protected]
To: [email protected]
Subject: I have a present for you

I've got some great birdseed over here at my place.
Want to buy it?

and the user's script contained:

if header :contains "from" "[email protected]"
  {
    reject "I am not taking mail from you, and I don't want
            your birdseed, either!";
  }

then the original sender <[email protected]> would receive the following notification:

To: <[email protected]>
X-Authentication-Warning: roadrunner set sender using -f flag
Content-Type: multipart/mixed; boundary=----- =_aaaaaaaaaa0
MIME-Version: 1.0
----- =_aaaaaaaaaa0
The original message was received at
Tue, 1 Apr 1997 09:07:15 -0800 from
[email protected]
Message was refused by recipient's mail filtering program.
Reason given was as follows:

I am not taking mail from you, and I don't want your birdseed, either!

----- =_aaaaaaaaaa0
Content-Type: message/delivery-status

Reporting-UA: sieve; GNU Mailutils 0.1.3
Arrival-Date: Tue, 1 Apr 1997 09:07:15 -0800
Final-Recipient: RFC822; [email protected]
Action: deleted
Disposition: automatic-action/MDN-sent-automatically;deleted
Last-Attempt-Date: Tue, 1 Apr 1997 09:07:15 -0800

----- =_aaaaaaaaaa0
Content-Type: message/rfc822

From: [email protected]
To: [email protected]
Subject: I have a present for you

I've got some great birdseed over here at my place.
Want to buy it?
----- =_aaaaaaaaaa0

If the reason argument is rather long, the common approach is to use the combination of the text: and #include keywords, e.g.:

if header :mime :matches "Content-Type"
          [ "*application/msword;*", "*audio/x-midi*" ]
  {
    reject text:
#include "nomsword.txt"
    .
    ;
  }

redirect address Action

The redirect action is used to send the message to another user at a supplied address, as a mail forwarding feature does. This action makes no changes to the message body or existing headers, but it may add new headers. It also modifies the envelope recipient.

The redirect command performs an MTA-style "forward" -- that is, what you get from a .forward file using sendmail under UNIX. The address on the SMTP envelope is replaced with the one on the redirect command and the message is sent back out. Notice, that it differs from the MUA-style forward, which creates a new message with a different sender and message ID, wrapping the old message in a new one.


Node:Extensions, Next:, Previous:Input Language, Up:libsieve

Extensions

This section summarizes the GNU extensions to the sieve language

  1. Handling of the require statement.
  2. header test

    The header takes an optional argument :mime, meaning to scan the headers from each part of a multipart message.

  3. size test

    The size test allows to omit the optional argument (:over|:under). In this case exact equality is assumed.

  4. envelope test

    The only value that can be meaningfully used as the first required argument of an envelope test is from. This limitation may disappear from the subsequent releases.

  5. Match type optional argument.

    Along with the usual :is, :matches and contains matching type, GNU sieve library understands :regex type. This matching type toggles POSIX Extended Regular Expression matching.


Node:Writing Loadable Commands, Previous:Extensions, Up:libsieve

Writing Loadable Commands

This section contains an example of how to write external loadable commands for GNU libsieve.


/*  This is an example on how to write extension tests for GNU sieve.
    It provides test "numaddr".

    Syntax:   numaddr [":over" / ":under"] <header-names: string-list>
              <limit: number>

    The "numaddr" test counts Internet addresses in structured headers
    that contain addresses.  It returns true if the total number of
    addresses satisfies the requested relation:

    If the argument is ":over" and the number of addresses is greater than
    the number provided, the test is true; otherwise, it is false.

    If the argument is ":under" and the number of addresses is less than
    the number provided, the test is true; otherwise, it is false.

    If the argument is empty, ":over" is assumed. */

#ifdef HAVE_CONFIG_H
# include <config.h>
#endif

#include <stdlib.h>
#include <mailutils/libsieve.h>

struct val_ctr {  /* Data passed to the counter function */
  header_t hdr;   /* Headers of the current message */
  size_t limit;   /* Limit for the number of addresses */
  size_t count;   /* Number of addresses counted so far */
};

/* Count addresses in a single header value.

   Input:
     ITEM is the name of the header to scan.
     DATA is a pointer to the val_ctr structure
   Return value:
     non-zero if the limit on the number of addresses has been reached. */

static int
_count_items (void *item, void *data)
{
  char *name = item;
  struct val_ctr *vp = data;
  char *val;
  address_t addr;
  size_t count = 0;

  if (header_aget_value (vp->hdr, name, &val))
    return 0;

  if (address_create (&addr, val) == 0)
    {
      address_get_count (addr, &count);
      address_destroy (&addr);
      vp->count += count;
    }
  free (val);
  return vp->count >= vp->limit;
}

/* Handler for the numaddr test */
static int
numaddr_test (sieve_machine_t mach, list_t args, list_t tags)
{
  sieve_value_t *h, *v;
  struct val_ctr vc;
  int rc;

  if (sieve_get_debug_level (mach) & MU_SIEVE_DEBUG_TRACE)
    sieve_debug (mach, "NUMADDR\n");

  /* Retrieve required arguments: */
  /* First argument: list of header names */
  h = sieve_value_get (args, 0);
  if (!h)
    {
      sieve_error (mach, "numaddr: can't get argument 1");
      sieve_abort (mach);
    }
  /* Second argument: Limit on the number of addresses */
  v = sieve_value_get (args, 1);
  if (!v)
    {
      sieve_error (mach, "numaddr: can't get argument 2");
      sieve_abort (mach);
    }

  /* Fill in the val_ctr structure */
  message_get_header (sieve_get_message (mach), &vc.hdr);
  vc.count = 0;
  vc.limit = v->v.number;

  /* Count the addresses */
  rc = sieve_vlist_do (h, _count_items, &vc);

  /* Here rc >= 1 iff the counted number of addresses is greater or equal
     to vc.limit. If `:under' tag was given we reverse the return value */
  if (sieve_tag_lookup (tags, "under", NULL))
    rc = !rc;
  return rc;
}

/* Syntactic definitions for the numaddr test */

/* Required arguments: */
static sieve_data_type numaddr_req_args[] = {
  SVT_STRING_LIST,
  SVT_NUMBER,
  SVT_VOID
};

/* Tagged arguments: */
static sieve_tag_def_t numaddr_tags[] = {
  { "over", SVT_VOID },
  { "under", SVT_VOID },
  { NULL }
};

static sieve_tag_group_t numaddr_tag_groups[] = {
  { numaddr_tags, NULL },
  { NULL }
};

/* Initialization function. It is the only function exported from this
   module. */
int
SIEVE_EXPORT(numaddr,init) (sieve_machine_t mach)
{
  return sieve_register_test (mach, "numaddr", numaddr_test,
                              numaddr_req_args, numaddr_tag_groups, 1);
}


Node:Programs, Next:, Previous:libsieve, Up:Top

Programs

GNU Mailutils provides a set of programs for handling the email.


Node:authentication, Next:, Up:Programs

Authorization and authentication principles.

Some mail utilities provide access to their services only after verifying that the user is actually the person he is claiming to be. Such programs are, for example, pop3d and imap4d. The process of the verification is broken down into two stages: authorization and authentication. In authorization stage the program retrieves the information about a particular user. In authentication stage, this information is compared against the user-supplied credentials. Only if both stages succeed is the user allowed to use the service.

A set of modules is involved in performing each stage. For example, the authorization stage can retrieve the user description from various sources: system database, sql database, virtual domain table, etc. Each module is responsible for retrieving the description from a particular source of information. The modules are arranged in a module list. The modules from the list are invoked in turn, until either a one of them succeeds or the list is exhausted. In latter case the authorization fails. Otherwise the data returned by the succeeded module are used in authentication.

Similarly, authentication may be performed in several ways. The authentication modules are also grouped in a list. Each module is tried in turn until either a module succeeds, in which case the authentication succeeds, or the end of the list is reached.

We represent the module lists as column-separated lists of module names. For example, the authorization list

        system:sql:virtdomains

means that first the system user database (/etc/password) is searched for a description of a user in question. If the search fails, the sql database is searched. Finally, if it also fails, the search is performed in the virtual domain database.

Note, that some authentication and/or authorization modules may be disabled when configuring the package before compilation. The names of the disabled modules are nevertheless available for use in runtime configuration options, but they represent a "fail-only" functionality, e.g. if the package was compiled without sql support then the module sql in the above example will always fail, thus passing the execution on to the next module.

The modules available for use in authorization list are:

system
User credentials are retrieved from the system user database (/etc/password).
sql
User credentials are retrieved from the sql database. The set of --sql- options (see auth) is used to configure access to the database.
virtdomain
User credentials are retrieved from a "virtual domain" user database.

The modules available for use in authentication list are:

generic
The generic authentication type. User password is hashed and compared against the hash value returned in authorization stage.
system
The hashed value of the user password is retrieved from /etc/shadow file on systems that support it.
sql
The hashed value of the user password is retrieved from the sql database using query supplied by --sql-getpass option (see auth).
pam
The user is authenticated via pluggable authentication module (pam). The pam service name to be used is configured via --pam-service option (see auth)

Unless overridden by --authentication command line option, the list of authentication modules is:

        generic:system:pam:sql

Unless overridden by --authorization command line option, the list of authorization modules is:

        system:sql:virtdomains


Node:configuration, Next:, Previous:authentication, Up:Programs

Mailutils configuration file

There are some command line options that are used so often that it is inconvenient to specify them in the command line each time you run a Mailutils utility. The configuration files provide a way to add default command line arguments without having to type them in the command line. Upon startup, each Mailutils utility scans and processes the contents of the three startup files, none of which are required to exist:

  1. the site-wide configuration file

    mailutils.rc, found in your your system configuration directory (usually /etc or /usr/local/etc).

  2. the user-specific configuration file

    Usually ~/.mailutils, unless ~/.mailutils is a directory, in which case ~/.mailutils/mailutils is used.

  3. the programs-specific configuration file

    Usually ~/.mu.programrc, unless ~/.mailutils is a directory, in which case ~/.mailutils/programrc is used (where program means the program name).

These files have simple line-oriented syntax. Comments begin with the pound sign (#) and extend through the end of the line 2. Very long lines may be split across several lines by escaping final newline with a backslash (\) character.

In the non-program-specific configuration files, any configuration line must start with a tag. In the program-specific configuration file the tag must not be present, all options are for that specific program.

A tag is either a name of a particular Mailutils utility or option group, prefixed with colon (:). The command line options common for several Mailutils programs are divided into option groups or capabilities, e.g. the options --mail-spool and --lock-flags form group mailbox. These groups are discussed in detail below.

When processing the non-program-specific configuration files, a Mailutils utility selects those lines whose tag is either the name of that utility or the name of the option group supported by it. In the program-specific configuration file, all lines are selected. For each line found, its tag (if present) is stripped away, and the rest of the line is split up into words. These words are regarded as command line options and are inserted to the program arguments before any options from the command line. Thus the options from .mailutils take precedence over those from mailutils.rc, and the options from the command line take precedence over those from all three configuration files.

The word splitting occurs at whitespace characters and is similar to that performed by the shell. If an option must contain embedded whitespace, it should be enclosed in a pair of quotes (either double or single quotes).


Node:default, Next:, Up:configuration

default -- Options understood by most GNU utilities.

Each program also understands the following informational options:

-u
--usage
Display a short usage message and exit.
-h
--help
Display help message and exit.
-L
--license
Display GNU General Public License and exit.
-v
--version
Display program version and exit.


Node:mailbox, Next:, Previous:default, Up:configuration

mailbox -- Specifies the mail spool location, and locking strategy.

Option group mailbox consists of options used to specify the location of the mail spool, and the locking strategy.

-m path
--mail-spool=path
Set path to the mailspool directory
--lock-flags=flags
Set the default mailbox lock flags (E=external, R=retry, T=time, P=pid).


Node:mailer, Next:, Previous:mailbox, Up:configuration

mailer -- Sets the default mailer URL.

This option group overrides the default mailer URL (<sendmail:>).

-m url
--mailer url


Node:address, Next:, Previous:mailer, Up:configuration

address -- Specifies the default email address and domain.

Option group address consists of options used to specify how to qualify email addresses.

An unqualified address (one without an @) is qualified by appending @defaultdomain. defaultdomain is the return of gethostname(), or the result of gethostbyname() on the return of gethostname() (if the DNS lookup is successful).

If the email address of the current user is needed, either the address set by -email-addr is returned, or the current uid is looked up in the user database, and qualified with the defaultdomain.

-E email
--email-addr=email
Set the current user's email address, this it makes more sense to use this in one of the per-user configuration files.
-D domain
--email-domain=domain
Set the default email domain, this might make sense to use in either the global or one of the per-user configuration files.


Node:daemon, Next:, Previous:address, Up:configuration

daemon -- Options common for daemon programs.

-d[number]
--daemon[=number]
Run in standalone mode. An optional number specifies the maximum number of child processes the daemon is allowed to fork. When it is omitted, it defaults to 20 processes. Please note, that there should be no whitespace between the -d and its parameter.
-i
--inetd
Run in inetd mode.
-p number
--port number
Listen on given port number. This option is meaningful only in standalone mode. It defaults to port 143.
-t number
--timeout number
Set idle timeout to given number of seconds. The daemon breaks the connection if it receives no commands from the client within that number of seconds.


Node:auth, Next:, Previous:daemon, Up:configuration

auth -- Authentication-specific options.

These options control the authorization and authentication module lists. For a description of authentication concepts, refer to See authentication.

--authorization modlist
This option allows to set up a list of modules to be used for authorization. modlist is a colon-separated list of modules. Valid modules are:
system
User credentials are retrieved from the system user database (/etc/password).
sql
User credentials are retrieved from the sql database. The set of --sql- options (see below) is used to configure access to the database.
virtdomain
User credentials are retrieved from a "virtual domain" user database.

--authentication modlist
This option allows to set up a list of modules to be used for authentication. modlist is a colon-separated list of modules. Valid modules are:
generic
The generic authentication type. User password is hashed and compared against the hash value returned in authorization stage.
system
The hashed value of the user password is retrieved from /etc/shadow file on systems that support it.
sql
The hashed value of the user password is retrieved from the sql database using query supplied by --sql-getpass option (see below).
pam
The user is authenticated via pluggable authentication module (pam). The pam service name to be used is configured via --pam-service option (see below)

--pam-service name
When compiled with pam support, this option specifies the name of pam service to be used when authenticating.

The following options exist in this group if the package was configured with --enable-sql option. They take effect only if the sql module is used in authentication and/or authorization. Currently only MySQL is supported.

--sql-getpwnam query
sql query to retrieve a passwd entry based on username
--sql-getpwuid query
--sql-getpass query
sql query to retrieve a password from the database
--sql-host name
Name or IP of MySQL server to connect to.
--sql-user name
sql user name
--sql-passwd string
sql connection password
--sql-db string
Name of the database to connect to.
--sql-port number
Port to use


Node:logging, Next:, Previous:auth, Up:configuration

logging -- Logging control options.

--log-facility facility
Output logs to the specified syslog facility. The following facility names are recognized: user, daemon, mail, auth and local0 through local7. These names are case-insensitive.


Node:sieve group, Next:, Previous:logging, Up:configuration

Sieve specific options

The following options comprise this group:

-I dir
--includedir=dir
Append directory dir to the list of directories searched for include files.
-L dir
--libdir=dir
Append directory dir to the list of directories searched for library files.


Node:config sample, Previous:sieve group, Up:configuration

A sample configuration file.

The following configuration file specifies that all Mailutils programs should use /var/spool/mail as a local mailspool directory. Programs performing authentication will use pam service mailutils. All programs, except imap4d will issue log messages via mail facility, imap4d will use facility local1.

:mailbox --mail-spool /var/spool/mail
:auth --authentication pam --pam-service mailutils
:logging --log-facility mail
imap4d --daemon=20 --timeout=1800 --log-facility local1


Node:imap4d, Next:, Previous:configuration, Up:Programs

IMAP4 daemon

GNU imap4d is a daemon implementing IMAP4 rev1 protocol for accessing and handling electronic mail messages on a server. It can be run either as a standalone program or from inetd.conf file.


Node:Namespace, Next:, Up:imap4d

Imap4d Namespace

GNU imap4d supports a notion of namespaces defined in RFC 2342. A namespace is a set of directories upon which the user has certain permissions. It should be understood that these permissions apply only if the underlying filesystem allows them.

The three namespaces supported by imap4d are:

Personal Namespace
A namespace that is within the personal scope of the authenticated user on a particular connection. The user has all permissions on this namespace.
Other Users' Namespace
A namespace that consists of mailboxes from the "Personal Namespaces" of other users. The user can read and list mailboxes from this namespace. However, he is not allowed to use % and * wildcards with LIST command, that is he can access a mailbox only if he knows exactly its location.
Shared Namespace
A namespace that consists of mailboxes that are intended to be shared amongst users and do not exist within a user's Personal Namespace. The user has all permissions on this namespace.

By default, imap4d starts with the following namespaces:

Personal Namespace
The home directory of the user, if exists.
Other Users' Namespace
Empty
Shared Namespace
Empty

Note, that this means that by default, a user won't be able to see or otherwise access mailboxes residing in the directories other than his own home.

To change these defaults, use --shared-namespace and --other-namespace options.


Node:Starting imap4d, Previous:Namespace, Up:imap4d

Starting imap4d

imap4d may run either in standalone or in inetd operation modes. When run in "standalone" mode, the server disconnects from the terminal and runs as a daemon, forking a child for each new connection.

The "inetd" mode allows to start the server from /etc/inetd.conf file. This is the default operation mode.

imap4  stream tcp nowait  root  /usr/local/sbin/imap4d imap4d

The program uses following option groups: See mailbox, See daemon, See logging, See auth.

Command line options

-d[number]
--daemon[=number]
Run in standalone mode. An optional number specifies the maximum number of child processes the daemon is allowed to fork. When it is omitted, it defaults to 20 processes. Please note, that there should be no whitespace between the -d and its parameter.
-h
--help
Display short help message and exit.
-i
--inetd
Run in inetd mode.
-m path
--mail-spool=path
Set path to the mailspool directory
-O pathlist
--other-namespace=pathlist
Set the list of directories forming the "Other User's" namespace. pathlist is a list of directory names separated by colons.
-p number
--port number
Listen on given port number. This option is meaningful only in standalone mode. It defaults to port 143.
-S pathlist
--shared-namespace=pathlist
Set the list of directories, forming the "Shared" namespace. pathlist is a list of directory names separated by colons.
-t number
--timeout number
Set idle timeout to given number of seconds. Default is 1800 seconds (30 minutes). The daemon breaks the connection if it receives no commands from the client within that number of seconds.
-v
--version
Display program version and exit.


Node:pop3d, Next:, Previous:imap4d, Up:Programs

POP3 daemon

The pop3d daemon implements the Post Office Protocol server.

pop3d has two operation modes:

Inetd
The server is started from /etc/inetd.conf file:
pop3  stream tcp nowait  root  /usr/local/sbin/pop3d pop3d

This is the default operation mode.

Standalone
The server runs as daemon, forking a child for each new connection. This mode is triggered by -d command line switch.

The program uses following option groups: See mailbox, See daemon, See logging, See auth.

Command line options

-d[number]
--daemon[=number]
Run in standalone mode. An optional number specifies the maximum number of child processes the daemon is allowed to fork. When it is omitted, it defaults to 10 processes. Please note, that there should be no whitespace between the -d and its parameter.
-h
--help
Display short help message and exit.
-i
--inetd
Run in inetd mode.
-m path
--mail-spool=path
Set path to the mailspool directory
-p number
--port number
Listen on given port number. This option is meaningful only in standalone mode. It defaults to port 110.
-t number
--timeout number
Set idle timeout to given number of seconds. Default is 600 seconds (10 minutes). The daemon breaks the connection if it receives no commands from the client within that number of seconds.
-v
--version
Display program version and exit.


Node:frm, Next:, Previous:pop3d, Up:Programs

frm -- List headers from a mailbox.

The frm command outputs a header information of the selected messages in a mailbox. By default, frm reads the user's system mailbox and outputs the contents of From and Subject headers for each message. If a folder is specified in the command line, the program reads that folder rather than the default mailbox.

The program uses following option groups: See mailbox.

The following command line options alter the behavior of the program:

-f STRING
--field STRING
Display the header named by STRING instead of From Subject pair.
-l
--to
Include the contents of To header to the output. The output field order is then: To From Subject.
-n
--number
Prefix each line with corresponding message number.
-Q
--Quiet
Be very quiet. Nothing is output except error messages. This is useful in shell scripts where only the return status of the program is important.
-q
--query
Print a message only if there are unread messages in the mailbox.
-S
--summary
Print a summary line.
-s ATTR
--status ATTR
Only display headers from messages with the given status. ATTR may be one of the following: new, read, unread. It is sufficient to specify only first letter of an ATTR. Multiple -s options are allowed.
-t
--align
Tidy mode. Currently is not implemented. Included for compatibility with frm program from Elm package.


Node:mail, Next:, Previous:frm, Up:Programs

mail -- send and receive mail.

Mail is an enhanced version of standard /bin/mail program. As well as its predecessor, it can be used either in sending mode or in reading mode. Mail enters sending mode when one or more email addresses were specified in this command line. In this mode the program waits until user finishes composing the message, then attempts to send it to the specified addresses and exits. See Composing Mail, for a detailed description of this behavior.

If the command line contained no email addresses, mail switches to reading mode. In this mode it allows to read and manipulate the contents of a mailbox. The URL of the mailbox to operate upon is taken from the argument of --file command line option. If it is not specified, the user's system mailbox is assumed. For more detail, see Reading Mail.


Node:Invoking Mail, Next:, Up:mail

Command line options

General usage of mail program is:

      mail [OPTION...] [address...]

If [address...] part is present, mail switches to mail sending mode, otherwise it operates in mail reading mode.

The program uses following option groups: See mailbox.

Mail understands following command line options:

-e
--exist
Return true if the mailbox contains some messages. Return false otherwise. This is useful for writing shell scripts.
-f[FILE]
--file[=FILE]
Operate on mailbox FILE. If this option is not specified, the default is user's system mailbox. If it is specified without argument, the default is $HOME/mbox. Please note, that there should be no whitespace between the short variant of the option (-f), and its parameter. Similarly, when using long option (--file), its argument must be preceded by equal sign.
-F
--byname
Save messages according to sender. Currently this option is not implemented.
-H
--headers
Print header summary to stdout and exit.
-i
--ignore
Ignore interrupts.
-m path
--mail-spool=path
Set path to the mailspool directory
-n
--norc
Do not read the system-wide mailrc file. See Mail Configuration Files.
-N
--nosum
Do not display initial header summary.
-p
--print
-r
--read
Print all mail to standard output. It is equivalent to issuing following commands after starting mail -N:
 print *
 quit

-q
--quit
Cause interrupts to terminate program.
-s SUBJ
--subject=SUBJ
Send a message with a Subject of SUBJ. Valid only in sending mode.
-t
--to
Switch to sending mode.
-u USER
--user=USER
Operate on USER's mailbox. This is equivalent to:
        mail -f/spool_path/USER

with spool_path being the full path to your mailspool directory
(/var/spool/mail or /var/mail on most systems).

-?
--help
Display a help message.
--usage
Display a short usage summary.
-V
--version
Print program version and exit.


Node:Specifying Messages, Next:, Previous:Invoking Mail, Up:mail

How to specify message sets

Many mail commands such as print and delete can be given a message list to operate upon. Wherever the message list is omitted, the command operates on the current message.

The message list in its simplest form is one of:

.
Selects current message. It is equivalent to empty message list.
*
Selects all messages in the mailbox.
^
Selects first non-deleted message.
$
Selects last non-deleted message.

In its complex form, the message list is a comma or whitespace-separated list of message specifiers. A message specifier is one of

Message Number
This specifier addresses the message with the given ordinal number in the mailbox.
Message range
Message range is specified as two message numbers separated by a dash. It selects all messages with the number lying within that range.
Attribute specifier
An Attribute specifier is a colon followed by a single letter. The Attribute specifier addresses all messages in the mailbox that have the given attribute. These are the valid attribute specifiers:
:d
Selects all deleted messages.
:n
Selects all recent messages, i.e. the messages that have not been neither read not seen so far.
:o
Selects all messages that have been seen.
:r
Selects all messages that have been read.
:u
Selects all messages that have not been read.
:t
Selects all tagged messages (see Marking Messages).
:T
Selects all untagged messages.

Header match
The header match is a string in the form:
[header:]/string/

It selects all messages that contain header field header matching given regexp. If the variable regexp is set, the string is assumed to be a POSIX regexp. Otherwise, a header is considered to match string if the latter constitutes a substring of the former (comparison is case-insensitive).

If header: part is omitted, it is assumed to be Subject:.

Message body match
The message body match is a string in the form:
:/string/

It selects all messages whose body matches the string. The matching rules are the same as described under "Header match".

A message specifier can be followed by message part specifier, enclosed in a pair of brackets. A message part specifier controls which part of a message should be operated upon. It is meaningful only for multipart messages. A message part specifier is a comma or whitespace - separated list of part numbers or ranges. Each part number can in turn be message part specifier, thus allowing for operating upon multiply-encoded messages.

The following are the examples of valid message lists:


Node:Composing Mail, Next:, Previous:Specifying Messages, Up:mail

Composing mail

You can compose the message by simply typing the contents of it, line by line. But usually this is not enough, you would need to edit your text, to quote some messages, etc. Mail provides these capabilities through compose escapes. The compose escapes are single-character commands, preceded by special escape character, which defaults to ~. The combination escape character + command is recognized as a compose escape only if it occurs at the beginning of a line. If the escape character must appear at the beginning of a line, enter it twice. The actual escape character may be changed by setting the value of escape mail variable (see Mail Variables).


Node:Quitting Compose Mode, Next:, Up:Composing Mail
Quitting Compose Mode

There are several commands allowing you to quit the compose mode.

Typing the end-of-file character (C-D) on a line alone finishes compose mode and sends the message to its destination. The C-D character looses its special meaning if ignoreeof mail variable is set.

If mail variable dot is set, typing dot (.) on a line alone achieves the same effect as C-D above.

Finally, using ~. escape always quits compose mode and sends out the composed message.

To abort composing of a message without sending it, type interrupt character (by default, C-C) twice. This behavior is disabled when mail variable ignore is set. In this case, you can use ~x escape to achieve the same effect.


Node:Getting Help on Compose Escapes, Next:, Previous:Quitting Compose Mode, Up:Composing Mail
Getting Help on Compose Escapes: ~?

The ~? escape prints on screen a brief summary of the available compose escapes. Please note, that ~h escape prompts for changing the header values, and does not give help.


Node:Editing the Message, Next:, Previous:Getting Help on Compose Escapes, Up:Composing Mail
Editing the Message: ~e and ~v.

If you are not satisfied with the message as it is, you can edit it using a text editor specified either by EDITOR or by VISUAL environment variables. The ~e uses the former, and ~v uses the latter.

By default both escapes allow you to edit only the body of the message. However, if the editheaders variable is set, mail will load into the editor the complete text of the message with headers included, thus allowing you to change the headers as well.


Node:Modifying the Headers, Next:, Previous:Editing the Message, Up:Composing Mail
Modifying the Headers: ~h, ~t, ~c, ~b, ~s

To add new addresses to the list of message recipients, use ~t command, e.g.:

~t [email protected] name2

To add addresses to Cc or Bcc, use ~c or ~b escapes respectively.

To change the Subject header, use ~s escape, e.g.:

~s "Re: your message"

Finally, to edit all headers, type ~h escape. This will present you with the values of To, Cc, Bcc, and Subject headers allowing to edit them with normal text editing commands.


Node:Enclosing Another Message, Next:, Previous:Modifying the Headers, Up:Composing Mail
Enclosing Another Message: ~m and ~M

If you are sending mail from within mail command mode, you can enclose the contents of any message sent to you by using ~m or ~M commands. Typing ~m alone will enclose the contents of the current message, typing ~m 12 will enclose the contents of message #12 and so on.

The ~m uses retained and ignored lists when enclosing headers, the ~M encloses all header fields (see Controlling Header Display).

In both cases, the contents of indentprefix mail variable is prepended to each line enclosed.


Node:Adding a File to the Message, Next:, Previous:Enclosing Another Message, Up:Composing Mail
Adding a File to the Message: ~r and ~d

To append the contents of file filename to the message, type

~r filename

or

~< filename

The ~d escape is a shorthand for

~r dead.letter


Node:Printing And Saving the Message, Next:, Previous:Adding a File to the Message, Up:Composing Mail
Printing And Saving the Message

The ~p escape types the contents of the message entered so far, including headers, on your terminal. You can save the message to an arbitrary file using ~w escape. It takes the filename as its argument.


Node:Signing the Message, Next:, Previous:Printing And Saving the Message, Up:Composing Mail
Signing the Message: ~a and ~A

To save you the effort of typing your signature at the end of each message, you can use ~a or ~A escapes. If your signature occupies one line only, save it to the variable sign and use ~a escape to insert it. Otherwise, if it is longer than one line, save it to a file, store the name of this file in the variable Sign, and use ~A escape to insert it into the message.


Node:Printing Another Message, Next:, Previous:Signing the Message, Up:Composing Mail
Printing Another Message: ~f and ~F

Sometimes it is necessary to view the contents of another message, while composing. These two escapes allow it. Both take the message list as their argument. If they are used without argument, the contents of the current message is printed. The difference between ~f and ~F is that the former uses ignored and retained lists to select headers to be displayed, whereas the latter prints all headers (see Controlling Header Display).


Node:Inserting Value of a Mail Variable, Next:, Previous:Printing Another Message, Up:Composing Mail
Inserting Value of a Mail Variable: ~i

The ~i escape enters the value of the named mail variable into the body of the message being composed.


Node:Executing Other Mail Commands, Next:, Previous:Inserting Value of a Mail Variable, Up:Composing Mail
Executing Other Mail Commands: ~: and ~-

You can execute a mail command from within compose mode using ~: or ~- escapes. For example, typing

~: from :t

will display the from lines of all tagged messages. Note, that executing mail-sending commands (see Replying) from within the compose mode is not allowed. An attempt to execute such a command will result in diagnostic message "Command not allowed in an escape sequence" being displayed. Also, when starting compose mode immediately from the shell (e.g. running mail [email protected]), most mail commands are meaningless, since there is no mailbox to operate upon. In this case, the only commands that can reasonably be used are: alias, unalias, alternate, set, and unset.


Node:Executing Shell Commands, Previous:Executing Other Mail Commands, Up:Composing Mail
Executing Shell Commands: ~! and ~|

The ~! escape executes specified command and returns you to mail compose mode without altering your message. When used without arguments, it starts your login shell. The ~| escape pipes the message composed so far through the given shell command and replaces the message with the output the command produced. If the command produced no output, mail assumes that something went wrong and retains the old contents of your message.


Node:Reading Mail, Next:, Previous:Composing Mail, Up:mail

Reading mail

To read messages from a given mailbox, use one of the following ways of invoking mail:

mail
To read messages from your system mailbox.
mail --file
To read messages from your mailbox ($HOME/mbox).
mail --file=path_to_mailbox
To read messages from the specified mailbox.
mail --user=user
To read messages from the system mailbox belonging to user.

Please note, that usual mailbox permissions won't allow you to use the last variant of invocation, unless you are a super-user. Similarly, the last but one variant is also greatly affected by the permissions the target mailbox has.

Unless you have started mail with --norc command line option, it will read the contents of the system-wide configuration file. Then it reads the contents of user configuration file, if any. For detailed description of these files, see Mail Configuration Files. After this initial setup, mail displays the first page of header lines and enters interactive mode. In interactive mode, mail displays its prompt (?, if not set otherwise) and executes the commands the user enters.


Node:Quitting the Program, Next:, Up:Reading Mail
Quitting the program

Following commands quit the program:

quit
Terminates the session. If mail was operating upon user's system mailbox, then all undeleted and unsaved messages that have been read and are not marked with hold flag are saved to the user's mbox file ($HOME/mbox). The messages, marked with delete are removed. The program exits to the Shell, unless saving the mailbox fails, in which case user can escape with the exit command.
exit
ex
xit
Program exits to the Shell without modifying the mailbox it operates upon.

Typing EOF (C-D) alone is equivalent to quit.


Node:Obtaining Online Help, Next:, Previous:Quitting the Program, Up:Reading Mail
Obtaining online help

Following commands can be used during the session to request online help:

help [command]
hel [command]
? [command]
Display detailed command synopsis. If no command is given, help for all available commands is displayed.
list
*
Print a list of available commands.
version
ve
Display program version.
warranty
wa
Display program warranty statement.


Node:Moving Within a Mailbox, Next:, Previous:Obtaining Online Help, Up:Reading Mail
Moving within a mailbox
next
n
Move to the next message.
previous
prev
Move to the previous message.


Node:Changing mailbox/directory, Next:, Previous:Moving Within a Mailbox, Up:Reading Mail
Changing mailbox/directory
cd [dir]
chdir [dir]
ch [dir]
Change to the specified directory. If dir is omitted, $HOME is assumed.
file [mailbox]
fi [mailbox]
folder [mailbox]
fold [mailbox]
Read in the contents of the specified mailbox. The current mailbox is updated as if quit command has been issued. If mailbox is omitted, the command prints the current mailbox name followed by the summary information regarding it, e.g.:
& fold
"/var/spool/mail/gray": 23 messages 22 unread


Node:Controlling Header Display, Next:, Previous:Changing mailbox/directory, Up:Reading Mail
Controlling header display

To control which headers in the message should be displayed, mail keeps two lists: a retained header list and an ignored header list. If retained header list is not empty, only the header fields listed in it are displayed when printing the message. Otherwise, if ignored header list is not empty, only the headers not listed in this list are displayed. The uppercase variants of message-displaying commands can be used to print all the headers.

The following commands modify and display the contents of both lists.

discard [header-field-list]
di [header-field-list]
ignore [header-field-list]
ig [header-field-list]
Add header-field-list to the ignored list. When used without arguments, this command prints the contents of ignored list.
retain [header-field-list]
ret [header-field-list]
Add header-field-list to the retained list. When used without arguments, this command prints the contents of retained list.


Node:Displaying Information, Next:, Previous:Controlling Header Display, Up:Reading Mail
Displaying information
=
Displays the current message number.
headers [msglist]
h [msglist]
Lists the current pageful of headers.
from [msglist]
f [msglist]
Lists the contents of From headers for a given set of messages.
z [arg]
Presents message headers in pagefuls as described for headers command. When arg is ., it is generally equivalent to headers. When arg is omitted or is +, the next pageful of headers is displayed. If arg is -, the previous pageful of headers is displayed. The latter two forms of z command may also take a numerical argument meaning the number of pages to skip before displaying the headers. For example:
& z +2

will skip two pages of messages before displaying the header summary.

size [msglist]
si [msglist]
Lists the message number and message size in bytes for each message in msglist.
folders
Displays the value of folder variable.
summary
su
Displays current mailbox summary. E.g.:
& summary
"/var/spool/mail/gray": 23 messages 22 unread


Node:Displaying Messages, Next:, Previous:Displaying Information, Up:Reading Mail
Displaying messages
print [msglist]
p [msglist]
type [msglist]
t [msglist]
Prints out the messages from msglist. The variable crt determines the minimum number of lines the body of the message must contain in order to be piped through pager command specified by environment variable PAGER. If crt is set to a numeric value, this value is taken as the minimum number of lines. Otherwise, if crt is set without a value then the height of the terminal screen is used to compute the threshold. The number of lines on screen is controlled by screen variable.
Print [msglist]
P [msglist]
Type [msglist]
T [msglist]
Like print but also prints out ignored header fields. For detailed description of ignored header lists, see Controlling Header Display.
decode [msglist]
dec [msglist]
Print a multipart message. The decode command decodes and prints out specified message parts. E.g.
& decode 15[2]
+---------------------------------------
| Message=15[2]
| Type=message/delivery-status
| encoding=7bit
+---------------------------------------
Content-Type: message/delivery-status
...

top [msglist]
to [msglist]
Prints the top few lines of each message in msglist. The number of lines printed is controlled by the variable toplines and defaults to five.
pipe [msglist] [shell-command]
| [msglist] [shell-command]
Pipe the contents of specified messages through shell-command. If shell-command is empty but the string variable cmd is set, the value of this variable is used as a command name.


Node:Marking Messages, Next:, Previous:Displaying Messages, Up:Reading Mail
Marking messages
tag [msglist]
ta [msglist]
Tag messages. The tagged messages can be referred to in message list using :t notation.
untag [msglist]
unt [msglist]
Clear tags from specified messages. To untag all messages tagged so far type
& untag :t

hold [msglist]
ho [msglist]
preserve [msglist]
pre [msglist]
Marks each message to be held in user's system mailbox. This command does not override the effect of delete command.


Node:Disposing of Messages, Next:, Previous:Marking Messages, Up:Reading Mail
Disposing of messages
delete [msglist]
d [msglist]
Mark messages as deleted. Upon exiting with quit command these messages will be deleted from the mailbox. Until the end of current session the deleted messages can be referred to in message lists using :d notation.
undelete [msglist]
u [msglist]
Clear delete mark from the specified messages.
dp [msglist]
dt [msglist]
Deletes the current message and prints the next message. If msglist is specified, deletes all messages from the list and prints the message, immediately following last deleted one.


Node:Saving Messages, Next:, Previous:Disposing of Messages, Up:Reading Mail
Saving messages
save [[msglist] file]
s [[msglist] file]
Takes a message list and a file name and appends each message in turn to the end of the file. The name of file and number of characters appended to it is echoed on the terminal. Each saved message is marked for deletion as if with delete command, unless the variable keepsave is set.
Save [msglist]
S [msglist]
Like save, but the file to append messages to is named after the sender of the first message in msglist. For example:
& from 14 15
 U  14 [email protected] Fri Jun 30 18:11  14/358   The Save c
 U  15 [email protected]  Fri Jun 30 18:30  8/245    Re: The Sa
& Save 14 15
"smith" 22/603

i.e., 22 lines (603 characters) have been appended to the file "smith". If the file does not exist, it is created.

write [[msglist] file]
w [[msglist] file]
Similar to save, except that only message body (without the header) is saved.
Write [msglist]
W [msglist]
Similar to Save, except that only message body (without the header) is saved.
mbox [msglist]
mb [msglist]
touch [msglist]
tou [msglist]
Mark list of messages to be saved in the user's mailbox ($HOME/mbox) upon exiting via quit command. This is the default action for all read messages, unless you have variable hold set.
copy [[msglist] file]
c [[msglist] file]
Similar to save, except that saved messages are not marked for deletion.
Copy [msglist]
C [msglist]
Similar to Save, except that saved messages are not marked for deletion.


Node:Editing Messages, Next:, Previous:Saving Messages, Up:Reading Mail
Editing messages

These command allow to edit messages in a mailbox. Please note, that modified messages currently do not replace original ones. i.e. you have to save them explicitly using your editor's save command if you do not want the effects of your editing to be lost.

edit [msglist]
e [msglist]
Edits each message in msglist with the editor, specified in EDITOR environment variable.
visual [msglist]
v [msglist]
Edits each message in msglist with the editor, specified in VISUAL environment variable.


Node:Scripting, Next:, Previous:Editing Messages, Up:Reading Mail
Scripting

Comments

The # character introduces an end-of-line comment. All characters until and including the end of line are ignored.

Displaying arbitrary text

The echo (ec) command prints its arguments to stdout.

Sourcing external command files

The command source filename reads commands from the named file. Its minimal abbreviation is so.

Setting and unsetting the variables.

The mail variables may be set using set (se) command. The command takes a list of assignments. The syntax of an assignment is

name=string
Assign a string value to the variable. If string contains whitespace characters it must be enclosed in a pair of double-quote characters (")
name=number
Assign a numeric value to the variable.
name
Assign boolean True value.
noname
Assign boolean False value.

Example:

& set askcc nocrt indentprefix="> "

This statement sets askcc to True, crt to False, and indentprefix to "> ".

To unset mail variables use unset(uns) command. The command takes a list of variable names to unset.

Example: To undo the effect of the previous example, do:

& unset askcc crt indentprefix

Setting and unsetting shell environment variables.

Shell environment may be modified using setenv (sete) command. The command takes a list of assignments. The syntax of an assignment is:

name=value
If variable name does not already exist in the environment, then it is added to the environment with the value value. If name does exist, then its value in the environment is changed to value.
name
Delete the variable name from the environment ("unset" it).

Conditional statements

The conditional statement allows to execute a set of mail commands depending on the mode the mail program is in. The conditional statement is:

if cond
...
else
...
endif

where ... represents the set of commands to be executed in each branch of the statement. cond can be one of the following:

s
True if mail is operating in mail sending mode.
r
True if mail is operating in mail reading mode.
t
True if stdout is a terminal device (as opposed to a regular file).

The conditional statements can be nested to arbitrary depth. The minimal abbreviations for if, else and endif commands are i, el and en.

Example:

if t
set crt prompt="& "
else
unset prompt
endif
if s
alt [email protected] [email protected]
set


Node:Aliasing, Next:, Previous:Scripting, Up:Reading Mail
Aliasing
alias [alias [address...]]
a [alias [address...]]
group [alias [address...]]
g [alias [address...]]
With no arguments, prints out all currently-defined aliases. With one argument, prints out that alias. With more than one argument, creates a new alias or changes an old one.
unalias [alias...]
una [alias...]
Takes a list of names defined by alias commands and discards the remembered groups of users. The alias names no longer have any significance.
alternates name...
alt name...
The alternates command is useful if you have accounts on several machines. It can be used to inform mail that the listed addresses are really you. When you reply to messages, mail will not send a copy of the message to any of the addresses listed on the alternates list. If the alternates command is given with no argument, the current set of alternate names is displayed.


Node:Replying, Next:, Previous:Aliasing, Up:Reading Mail
Replying
mail [address...]
m [address...]
Switches to compose mode. After composing the message, sends messages to the specified addresses.
reply [msglist]
respond [msglist]
r [msglist]
For each message in msglist, switches to compose mode and sends the composed message to the sender and all recipients of the message.
Reply [msglist]
Respond [msglist]
R [msglist]
Like reply, except that the composed message is sent only to originators of the specified messages.
followup [msglist]
fo [msglist]
Switches to compose mode. After composing, sends the message to the originators and recipients of all messages in msglist.
Followup [msglist]
F [msglist]
Similar to followup, but reply message is sent only to originators of messages in msglist.


Node:Incorporating New Mail, Next:, Previous:Replying, Up:Reading Mail
Incorporating new mail

The incorporate (inc) command incorporates newly arrived messages to the displayed list of messages. This is done automatically before returning to mail command prompt if the variable autoinc is set.


Node:Shell Escapes, Previous:Incorporating New Mail, Up:Reading Mail
Shell escapes

To run arbitrary shell command from mail command prompt, use shell (sh) command. If no arguments are specified, the command starts the user login shell. Otherwise, it uses its first argument as a file name to execute and all subsequent arguments are passed as positional parameters to this command. The shell command can also be spelled as !.


Node:Mail Variables, Next:, Previous:Reading Mail, Up:mail

How to alter the behavior of mail

Following variables control the behavior of GNU mail:

Sign

Type: String.
Default: Unset.

Contains the filename holding users signature. The contents of this file is appended to the end of a message being composed by ~A escape.

appenddeadletter

Type: Boolean.
Default: False.

If this variable is True, the contents of canceled letter is appended to the user's dead.letter file. Otherwise it overwrites its contents.

askbcc

Type: Boolean.
Default: False.

When set to True the user will be prompted to enter Bcc field before composing the message.

askcc

Type: Boolean.
Default: True.

When set to True the user will be prompted to enter Cc field before composing the message.

asksub

Type: Boolean.
Default: True in interactive mode, False otherwise.

When set to True the user will be prompted to enter Subject field before composing the message.

autoinc

Type: Boolean.
Default: True.

Automatically incorporate newly arrived messages.

autoprint

Type: Boolean.
Default: False.

Causes the delete command to behave like dp - thus, after deleting a message, the next one will be typed automatically.

cmd

Type: String.
Default: Unset.

Contains default shell command for pipe.

columns

Type: Numeric.
Default: Detected at startup by querying the terminal device. If this fails, the value of environment variable COLUMNS is used.

This variable contains the number of columns on terminal screen.

crt

Type: Boolean or Numeric
Default: True in interactive mode, False otherwise.

The variable crt determines the minimum number of lines the body of the message must contain in order to be piped through pager command specified by environment variable PAGER. If crt is set to a numeric value, this value is taken as the threshold. Otherwise, if crt is set without a value, then the height of the terminal screen is used to compute the threshold. The number of lines on screen is controlled by screen variable.

dot

Type: Boolean.
Default: False.

If True, causes mail to interpret a period alone on a line as the terminator of a message you are sending.

editheaders

Type: Boolean.
Default: False.

When set, mail will include message headers in the text to be the ~e and ~v escapes, thus allowing you to customize the headers.

escape

Type: String.
Default: ~

If defined, the first character of this option gives the character to denoting escapes.

folder

Type: String.
Default: Unset.

The name of the directory to use for storing folders of messages. If unset, $HOME is assumed.

header

Type: Boolean.
Default: True, unless started with --nosum (-N) option.

Whether to run headers command automatically after entering interactive mode.

hold

Type: Boolean.
Default: False.

When set to True, the read or saved messages will be stored in user's mailbox ($HOME/mbox). Otherwise, they will be held in system mailbox also. This option is in effect only when operating upon user's system mailbox.

ignore

Type: Boolean.
Default: False.

When set to True, mail will ignore keyboard interrupts when composing messages. Otherwise an interrupt will be taken as a signal to abort composing.

ignoreeof

Type: Boolean.
Default: False.

Controls whether typing EOF character terminates the letter being composed.

indentprefix

Type: String.
Default: "\t" (a tab character).

String used by the ~m tilde escape for indenting quoted messages.

keepsave

Type: Boolean.
Default: False.

Controls whether saved messages should be kept in system mailbox too. This variable is in effect only when operating upon a user's system mailbox.

metoo

Type: Boolean.
Default: False.

Usually, when an alias is expanded that contains the sender, the sender is removed from the expansion. Setting this option causes the sender to be included in the group.

mode

Type: String.
Default: The name of current operation mode.

Setting this variable does not affect the operation mode of the program.

noregex

Type: Boolean.
Default: False.

Setting this to True enables use of regular expressions in /.../ message specifications.

outfolder

Type: String.
Default: Unset.

Contains the directory in which files created by save, write, etc. commands will be stored. When unset, current directory is assumed.

page

Type: Boolean.
Default: False.

If set to True, the pipe command will emit a linefeed character after printing each message.

prompt

Type: String.
Default: "? "

Contains the command prompt sequence.

quiet

Type: Boolean.
Default: False.

When set, suppresses the output of the startup banner.

quit

Type: Boolean.
Default: False, unless started with --quit (-q) option.

When set, causes keyboard interrupts to terminate the program.

rc

Type: Boolean.
Default: True, unless started with --norc (-N) option.

When this variable is set, mail will read the system-wide configuration file upon startup. See Mail Configuration Files.

record

Type: String.
Default: Unset.

When set, any outgoing message will be saved to the named file.

save

Type: Boolean.
Default: True.

When set, the aborted messages will be stored in the user's dead.file. See also appenddeadletter.

screen

Type: Numeric.
Default: Detected at startup by querying the terminal device. If this fails, the value of environment variable LINES is used.

This variable contains the number of lines on terminal screen.

sendmail

Type: String.
Default: sendmail:/usr/lib/sendmail

Contains the URL of mail transport agent.

sign

Type: String.
Default: Unset.

Contains the user's signature. The contents of this variable is appended to the end of a message being composed by ~a escape. Use Sign variable, if your signature occupies more than one line.

subject

Type: String.
Default: Unset.

Contains default subject line. This will be used when asksub is off.

toplines

Type: Numeric.
Default: 5

Number of lines to be displayed by top and Top commands.

verbose

Type: Boolean.
Default: False.

When set, the actual delivery of messages is displayed on the user's terminal.


Node:Mail Configuration Files, Previous:Mail Variables, Up:mail

Personal and system-wide configuration files

Upon startup, mail reads the contents of the two command files: the system-wide configuration file, and the user's configuration file. Each line read from these files is processed like a usual mail command.

When run with --norc (-N) option, mail does not read the contents of system-wide configuration file. The user's file, if it exists, is always processed.

The user's configuration file is located in the user's home directory and is named .mailrc. The location and name of the system-wide configuration file is determined when configuring the package via --with-mail-rc option. It defaults to sysconfdir/mail.rc.


Node:mail.local, Next:, Previous:mail, Up:Programs

mail.local -- Deliver mail to the local mailbox.

mail.local reads the standard input up to an end-of-file and appends the received data to the local mailboxes.


Node:Invocation, Next:, Up:mail.local

Invoking mail.local

The program uses following option groups: See mailbox, See auth, See logging, See sieve.

-f addr
--from addr
Specify the sender's name. This option forces mail.local to add From envelope to the beginning of the message. If it is not specified, mail.local first looks into the first line from the standard input. If it starts with From , it is assumed to contain a valid envelope. If it does not, mail.local creates the envelope by using current user name and date.
-h
--help
Display this help and exit.
-L
--license
Display GNU General Public License and exit.
-m path
--mail-spool path
Specify path to mailspool directory.
-q
--quota-db file
Specify path to mailbox quota database (see Mailbox Quotas).
-s pattern
--source pattern
Set name pattern for user-defined mail filters written in Scheme (see Scheme Filters). The metacharacters %u and %h in the pattern are expanded to the current recipient user name and home directory correspondingly.

This option is available only if the package has been configured to use Guile extension language.

-S pattern
--sieve pattern
Set name pattern for user-defined mail filters written is Sieve (see Sieve Filters). The metacharacters %u and %h in the pattern are expanded to the current recipient user name and home directory correspondingly.
-t number
--timeout number
Wait number seconds for acquiring the lockfile. If it doesn't become available after that amount of time, return failure. The timeout defaults to 5 minutes.
-x flags
--debug flags
Enable debugging. The debugging information will be output using syslog. The flags is a string consisting of the following flags: Debug flags are:
g
Start with guile debugging evaluator and backtraces. This is convenient for debugging user-defined filters (see Scheme Filters).
T
Enable libmailutil traces (MU_DEBUG_TRACE).
P
Enable network protocol traces (MU_DEBUG_PROT)
t
Enable sieve trace (MU_SIEVE_DEBUG_TRACE)
l
Enable sieve action logs

The digits in the range 0 - 9 used in flags set mail.local debugging level.

-v
--version
Display program version and exit.
--ex-multiple-delivery-success
Don't return errors when delivering to multiple recipients.
--ex-quota-tempfail
Return temporary failure if disk or mailbox quota is exceeded. By default, 'service unavailable' is returned if the message exceeds the mailbox quota.


Node:MTA, Next:, Previous:Invocation, Up:mail.local

Using mail.local with various MTAs

This section explains how to invoke mail.local from configuration files of various Mail Transport Agents.

All examples in this section suppose that mail.local must receive following command line switches:

  -s %h/.filter.scm -q /etc/mail/userquota


Node:sendmail, Next:, Up:MTA
Using mail.local with sendmail.

The mail.local must be invoked from the local mailer definition in the sendmail.cf file. It must have the following flags set lswS, meaning the mailer is local, the quote characters should be stripped off the address before invoking the mailer, the user must have a valid account on this machine and the userid should not be reset before calling the mailer. Additionally, fn flags may be specified to allow mail.local to generate usual From envelope instead of the one supplied by sendmail.

Here is an example of mailer definition in sendmail.cf

Mlocal, P=/usr/local/libexec/mail.local,
        F=lsDFMAw5:/|@qSPfhn9,
        S=EnvFromL/HdrFromL, R=EnvToL/HdrToL,
        T=DNS/RFC822/X-Unix,
        A=mail -s %h/.filter.scm -q /etc/mail/userquota $u

To define local mailer in mc source file, it will suffice to set:

define(`LOCAL_MAILER_PATH', `/usr/local/libexec/mail.local')
define(`LOCAL_MAILER_ARGS',
       `mail -s %h/.filter.scm -q /etc/mail/userquota $u')


Node:exim, Previous:sendmail, Up:MTA
Using mail.local with exim.

Using mail.local with exim is quite straightforward. The following example illustrates the definition of appropriate transport and director in exim.conf:

# transport
mail_local_pipe:
  driver = pipe
  command = /usr/local/libexec/mail.local -s %h/.filter.scm \
              -q /etc/mail/userquota $local_part
  return_path_add
  delivery_date_add
  envelope_to_add

# director
mail_local:
  driver = localuser
  transport = mail_local_pipe


Node:Mailbox Quotas, Next:, Previous:MTA, Up:mail.local

Setting up mailbox quotas


Node:Sieve Filters, Next:, Previous:Mailbox Quotas, Up:mail.local

Implementing user-defined Sieve mail filters.


Node:Scheme Filters, Previous:Sieve Filters, Up:mail.local

Implementing user-defined Scheme mail filters.


Node:messages, Next:, Previous:mail.local, Up:Programs

messages -- Count the number of messages in a mailbox.

Messages prints on standard output the number of messages contained in each folder specified in command line. If no folders are specified, it operates upon user's system mailbox. For each folder, the following output line is produced:

     Number of messages in folder: number

where folder represents the folder name, number represents the number of messages.

The program uses following option groups: See mailbox.

The program accepts following command line options:

-q
--quite
-s
--silent
Be quiet. Display only number of messages per mailbox, without leading text.
-?
--help
Output help message and exit.
--usage
Output short usage summary and exit.
-V
--version
Output program version and exit.


Node:readmsg, Next:, Previous:messages, Up:Programs

readmsg -- Extract messages from a folder.

The program, readmsg, extracts with the selection argument messages from a mailbox. Selection can be specify by:

  1. A lone "*" means select all messages in the mailbox.
  2. A list of message numbers may be specified. Values of "0" and "$" in the list both mean the last message in the mailbox. For example:
         readmsg 1 3 0
    
    extracts three messages from the folder: the first, the third, and the last.
  3. Finally, the selection may be some text to match. This will select a mail message which exactly matches the specified text. For example,
         readmsg staff meeting
    
    extracts the message which contains the words "staff meeting." Note that it will not match a message containing "Staff Meeting" - the matching is case sensitive. Normally only the first message which matches the pattern will be printed.

Command line options


-a
--show-all
If a pattern is use for selection show all messages that match pattern by default only the first one is presented.
-d
--debug
Display mailbox debugging information.
-f MAILBOX
--folder=MAILBOX
Specified the default mailbox.
-h
--header
Show the entire header and ignore the weedlist.
-n
--no-header
Do not print the message header.
-p
--form-feed
Put form-feed (Control-L) between messages instead of newline.
-w weedlist
--weedlist=weedlist
A whitespace or coma separated list of header names to show per message. Default is -weedlist="From Subject Date To CC Apparently-"


Node:sieve, Next:, Previous:readmsg, Up:Programs

sieve

Sieve is a language for filtering e-mail messages at time of final delivery, described in RFC 3028. GNU Mailutils provides two implementations of this language: a stand-alone sieve interpreter and a sieve translator and filter. The following sections describe these utilities in detail.


Node:sieve interpreter, Next:, Up:sieve

A Sieve Interpreter

Sieve interpreter sieve allows to apply Sieve scripts to an arbitrary number of mailboxes. GNU sieve implements a superset of the Sieve language as described in RFC 3028. See Input Language, for a description of the Sieve language. See Extensions, for a discussion of differences between the GNU implementation of Sieve and its standard.


Node:Invoking Sieve, Next:, Up:sieve interpreter
Invocation

The sieve invocation syntax is:

        sieve [options] script

where script denotes the filename of the sieve program to parse, and options is one or more of the following:

-c
--compile-only
Compile script and exit.
-d[flags]
--debug[=flags]
Specify debug flags. The flags argument is a sequence of one or more of the following letters:

g Enable main parser traces
T Enable mailutil traces
P Trace network protocols
t Enable sieve trace
i Trace the program instructions

See Logging and Debugging, for detailed discussion of these.

-D
--dump
Compile the script, dump disassembled code on standard output and exit.
-e address
--email address
Override the user email address. This is useful for reject and redirect actions. By default, the user email address is deduced from the user name and the full name of the machine where sieve is executed.
-f
--mbox-url=mbox
Mailbox to sieve (defaults to user's system mailbox)
-k
--keep-going
Keep on going if execution fails on a message
-n
--no-actions
Dry run: do not execute any actions, just print what would be done.
-t ticket
--ticket=ticket
Ticket file for mailbox authentication
-v
--verbose
Log all actions executed.

Apart from these, sieve understands the options from the following groups: sieve, mailbox, mailer, logging.


Node:Logging and Debugging, Next:, Previous:Invoking Sieve, Up:sieve interpreter
Logging and debugging

The default behavior of sieve is to remain silent about anything except errors. However, it is sometimes necessary to see which actions are executed and on which messages. This is particularly useful when debugging the sieve scripts. The --verbose (-v) option outputs log of every action executed.

Option --debug allows to produce even more detailed debugging information. This option takes an argument specifying the debugging level to be enabled. The argument can consist of the following letters:

t
This flag enables sieve tracing. It means that every test will be logged when executed.
T
This flag enables debugging of underlying mailutils library.
P
Trace network protocols: produces log of network transactions executed while running the script.
g
Enable main parser traces. This is useful for debugging the sieve grammar.
i
Trace the program instructions. It is the most extensive debugging level. It produces the full execution log of a sieve program, showing each instruction and states of the sieve machine. It is only useful for debugging the code generator.

Note, that there should be no whitespace between the short variant of the option (-d), and its argument. Similarly, when using long option (--debug), its argument must be preceded by equal sign.

If the argument to --debug is omitted, it defaults to TPt.

Option --dump produces the disassembled dump of the compiled sieve program.

By default sieve output all diagnostics on standard error and verbose logs on standard output. This behaviour is changed when --log-facility is given in the command line (see logging). This option causes sieve to output its diagnostics to the given syslog facility.


Node:Extending Sieve, Previous:Logging and Debugging, Up:sieve interpreter
Extending Sieve

The basic set of sieve actions, tests and comparators may be extended using loadable extensions. Usual require mechanism is used for that.

When processing arguments for require statement, sieve uses the following algorithm:

  1. Look up the name in a symbol table. If the name begins with comparator- it is looked up in the comparator table. If it begins with test-, the test table is used instead. Otherwise the name is looked up in the action table.
  2. If the name is found, the search is terminated.
  3. Otherwise, transform the name. First, any comparator- or test- prefix is stripped. Then, any character other than alphanumeric characters, . and , is replaced with dash (-). The name thus obtained is used as a file name of an external loadable module.
  4. Try to load the module. The module is searched in the following search paths (in the order given):
    1. Mailutils module directory. By default it is $prefix/lib/mailutils.
    2. The value of the environment variable LTDL_LIBRARY_PATH.
    3. Additional search directories specified with the #searchpath directive.
    4. System library search path: The system dependent library search path (e.g. on Linux it is set by the contents of the file /etc/ld.so.conf and the value of the environment variable LD_LIBRARY_PATH).

    The value of LTDL_LIBRARY_PATH and LD_LIBRARY_PATH must be a colon-separated list of absolute directories, for example, "/usr/lib/mypkg:/lib/foo".

    In any of these directories, sieve first attempts to find and load the given filename. If this fails, it tries to append the following suffixes to the file name:

    1. the libtool archive extension .la
    2. the extension used for native dynamic libraries on the host platform, e.g., .so, .sl, etc.
  5. If the module is found, sieve executes its initialization function (see below) and again looks up the name in the symbol table. If found, search terminates successfully.
  6. If either the module is not found, or the symbol wasn't found after execution of the module initialization function, search is terminated with an error status. sieve then displays the following diagnostic message:
    source for the required action NAME is not available
    


Node:sieve.scm, Previous:sieve interpreter, Up:sieve

A Sieve to Scheme Translator and Filter

A Sieve to Scheme Translator sieve.scm translates a given Sieve script into an equivalent Scheme program and optionally executes it. The program itself is written in Scheme and requires presence of Guile 1.4 on the system. For more information on Guile refer to Overview.

-f filename
--file filename
Set input file name.
-o filename
--output filename
Set output file name
-L dirname
--lib-dir dirname
Set sieve library directory name
-d level
--debug level
Set debugging level

The Scheme programs produced by sieve.scm can be used with guimb or mail.local.


Node:guimb, Next:, Previous:sieve, Up:Programs

guimb -- A mailbox scanning and processing language.

Guimb is for mailboxes what awk is for text files. It processes mailboxes, applying the user-supplied scheme procedures to each of them in turn and saves the resulting output in mailbox format.

The program uses following option groups: See mailbox.


Node:Specifying Scheme Program to Execute, Next:, Up:guimb

Specifying Scheme Program to Execute

The Scheme program or expression to be executed is passed to guimb via the following options:

-s file
--source file
Load Scheme source code from file.
-c expr
--code expr
Execute given scheme expression.

The above switches stop further argument processing, and pass all remaining arguments as the value of (command-line).

If the remaining arguments must be processed by guimb itself, use following options:

-e expr
--expression expr
Execute scheme expression.
-f file
--file file
Load Scheme source code from file.

You can specify both of them. In this case, the file is read first, then expr is executed. You may still pass any additional arguments to the script using --guile-arg option (see Passing Options to Scheme).


Node:Specifying Mailboxes to Operate Upon, Next:, Previous:Specifying Scheme Program to Execute, Up:guimb

Specifying Mailboxes to Operate Upon

There are four basic ways of passing mailboxes to guimb.

guimb [options] [mailbox...]
The resulting mailbox is not saved, unless the user-supplied scheme program saves it.
guimb [options] --mailbox defmbox
The contents of defmbox is processed and is replaced with the resulting mailbox contents. Useful for applying filters to user's mailbox.
guimb [options] --mailbox defmbox mailbox [mailbox...]
The contents of specified mailboxes is processed, and the resulting mailbox contents is appended to defmbox.
guimb [options] --user username [mailbox...]
The contents of specified mailboxes is processed, and the resulting mailbox contents is appended to the user's system mailbox. This allows to use guimb as a mail delivery agent.

If no mailboxes are specified in the command line, guimb reads and processes the system mailbox of the current user.


Node:Passing Options to Scheme, Next:, Previous:Specifying Mailboxes to Operate Upon, Up:guimb

Passing Options to Scheme

Sometimes it is necessary to pass some command line options to the scheme procedure. There are three ways of doing so.

When using --source (-s) or --code (-c) options, all the rest of the command line following the option's argument is passed to Scheme program verbatim. This allows for making guimb scripts executable by the shell. If your system supports #! magic at the start of scripts, add the following two lines to the beginning of your script to allow for its immediate execution:

#! /usr/local/bin/guimb -s
!#

(replace /usr/local/bin/ with the actual path to the guimb).

Otherwise, if you use --file or --expression options, the additional arguments may be passed to the Scheme program -g (--guile-arg) command line option. For example:

guimb --guile-arg -opt --guile-arg 24 --file progfile

In this example, the scheme procedure will see the following command line:

progfile -opt 24

Finally, if there are many arguments to be passed to Scheme, it is more convenient to enclose them in -{ and -} escapes:

guimb -{ -opt 24 -} --file progfile


Node:Guimb Invocation Summary, Previous:Passing Options to Scheme, Up:guimb

Guimb Invocation Summary

This is a short summary of the command line options available to guimb.

-d
--debug
Start with debugging evaluator and backtraces.
-e EXPR
--expression EXPR
Execute given Scheme expression.
-m path
--mail-spool=path
Set path to the mailspool directory
-f PROGFILE
--file PROGFILE
Read Scheme program from PROGFILE.
-g ARG
--guile-command ARG
Append ARG to the command line passed to Scheme program.
-{ ... -}
Pass all command line options enclosed between -{ and -} to Scheme program.
-m
--mailbox MBOX
Set default mailbox name.
-u
--user NAME
Act as local MDA for user NAME.
-h
--help
Display help message.
-v
--version
Display program version.


Node:comsatd, Next:, Previous:guimb, Up:Programs

Comsat daemon

Comsatd is the server which receives reports of incoming mail and notifies users, wishing to get this service. It can be started either from inetd.conf or as a standalone daemon.


Node:Starting comsatd, Next:, Up:comsatd

Starting comsatd

Comsatd uses following option groups: See mailbox, See daemon, See logging.

-c file
--config file
Read configuration from given file. For more information about comsatd configuration files, see Configuring comsatd.
-d
--daemon
Run as a standalone daemon.
-i
--inetd
The server is started from /etc/inetd.conf file:
comsat dgram  udp wait  root  /usr/sbin/comsatd \
comsatd -c /etc/comsat.conf

This is the default operation mode.

-m path
--mail-spool=path
Set path to the mailspool directory
-p number
--port number
Specify the port number to listen on. Default is 512.
-v
--version
Output version and exit successfully.
-h
--help
Display short help message and exit.


Node:Configuring comsatd, Next:, Previous:Starting comsatd, Up:comsatd

Configuring comsatd

The configuration parameters for comsatd are kept in a single configuration file. The file uses line-oriented format: each line contains a single statement. Comments are introduced with the # sign and empty lines are ignored. You can specify the configuration file to use by using -c or --config command line switch.

The configuration file statements can logically be subdivided into General Settings, Security Settings and Access Control Lists. The following sections address each of these statement group in detail.


Node:General Settings, Next:, Up:Configuring comsatd
General Settings

These statements control the general behavior of the comsat daemon:

max-lines number
Set maximum number of message body lines to be output.
allow-biffrc ( yes | no )
Enable or disable processing of user's .biffrc file. By default, it is enabled.


Node:Security Settings, Next:, Previous:General Settings, Up:Configuring comsatd
Security Settings

These statements control the way comsatd fights possible flooding attacks.

max-requests number
Set maximum number of incoming requests per request-control-interval.
request-control-interval number
Set the request control interval (seconds).
overflow-delay-time number
Set the initial amount of time to sleep, after the first overflow occurs.
overflow-control-interval number
Set the overflow control interval. If two consecutive overflows happen within number seconds, the overflow-delay-time is doubled.


Node:Access Control Lists, Previous:Security Settings, Up:Configuring comsatd
Access Control Lists

Access control lists determine from which addresses comsatd will receive mail notification messages.

The access control lists are introduced in configuration file using keyword acl. General format for an ACL rule is

acl action netlist

Here, action specifies the action to be taken when a request arrives from one of the networks, listed in netlist. There are two possible actions: allow and deny.

The netlist is a whitespace-separated list of network numbers. Each network number may be specified in one of the following forms:

netnum
Means a single host with IP address netnum.
netnum/netmask
netnum/masklen
any
Denotes any IP address. It is equivalent to 0.0.0.0/0.

Upon receiving a notification message, comsatd compares its source address against each ACL rule in the order of their appearance in the configuration file. The first rule that matches the packet determines whether the message will be processed or rejected. If no matching rule was found, the default rule applies. Currently, default rule is

acl allow any

If you don't need such behavior, specify the default rule explicitly. For example, the common use would be:

acl allow 127.0.0.1
acl deny any

which makes comsatd receive the notification messages from localhost only.


Node:dot.biffrc, Previous:Configuring comsatd, Up:comsatd

A per-user configuration file.

By default, when a notification arrives, comsatd prints subject, from headers and the first five lines from the new message to the user's tty. The user is allowed to change this behavior by using his own configuration file. This file should be located in the user's home directory and should be named .biffrc. It must be owned by the user and have its permissions bits set to 0600. (Please note, that the use of per-user configuration files may be disabled, by specifying allow-biffrc no in the main configuration file, see see Configuring comsatd).

The .biffrc file consists of a series of statements. Each statement occupies one line and defines an action to be taken upon arrival of a new mail. Very long lines may be split using \ as the last character on the line. As usual, comments may be introduced with # character.

The actions specified in .biffrc file are executed in turn. The following actions are defined:

beep
Produce an audible signal.
echo string
Output string to user's terminal device.
exec prog arglist
Execute program prog with arguments from arglist. prog must be specified with absolute pathname. It may not be a setuid or setgid program.

In the description above, string denotes any sequence of characters. This sequence must be enclosed in a pair of double-quotes, if it contains whitespace characters. The \ character inside a string starts a C escape sequence. Following meta-characters may be used in strings:

$u
Expands to username
$h
Expands to hostname
$H{name}
Expands to value of message header name.
$B(c,l)
Expands to message body. c and l give maximum number of characters and lines in the expansion. When omitted, they default to 400, 5.
Example I.

Dump to the user's terminal the contents of From and Subject headers followed by at most 5 lines of message body.

echo "Mail to \[email protected]$h\a\n---\n\
From: $H{from}\n\
Subject: $H{Subject}\n\
---\n\
$B(,5)\
---\n"
Example II.

Produce a bell, then pop up the xmessage window on display :0.0 with the text formatted in the same manner as in the previous example.

beep
exec /usr/X11R6/bin/xmessage \
-display :0.0 -timeout 10 "Mail to [email protected]$h \n---\n\
From: $H{from}\n\
Subject: $H{Subject}\n\
---\n\
$B(,5)\
---\n"


Node:mailutils-config, Previous:comsatd, Up:Programs

mailutils-config -- Get the information about the Mailutils build.

This program is designed for developers wishing to link their programs against libmailbox. It allows to examine the particulars of the current build of Mailutils and to get the command line parameters necessary for compiling and linking an application with Mailutils libraries.


Node:Compiler Flags, Next:, Up:mailutils-config

Getting Compiler Flags.

When invoked with the option --compile, or its short form -c, mailutils-config prints the flags that must be given to the compiler for compiling the program using Mailutils functions. An example usage:

cc -omyprog.o `mailutils-config --compile` myprog.c


Node:Loader Flags, Next:, Previous:Compiler Flags, Up:mailutils-config

Getting Loader Flags

The --link, or its short form -l prints to the standard output the loader flags necessary to link a program against Mailutils libraries.

When invoked without arguments, it produces the flags necessary to link against the basic library of Mailutils: libmailbox. Up to two arguments may be given that alter this behavior. These are:

auth
Print flags to link against libmuauth, the library adding new authentication methods to libmailbox.
guile
Print flags to link against libmu_scm, the Guile interface library.

Both arguments may be given simultaneously, e.g.:

cc -omyprog myprog.o `mailutils-config --link auth guile`


Node:General Information, Previous:Loader Flags, Up:mailutils-config

Obtaining General Build Information

The --info, or -i retrieves the options (flags) used when building Mailutils. It may be used with or without arguments.

When used without arguments, it prints the list of all build flags, e.g.:

$ mailutils-config --info
VERSION=0.1.3
USE_LIBPAM
HAVE_LIBLTDL
WITH_GUILE
WITH_READLINE
HAVE_MYSQL
ENABLE_VIRTUAL_DOMAINS
ENABLE_IMAP
ENABLE_POP
ENABLE_MH
ENABLE_SMTP
ENABLE_SENDMAIL

This option also accepts any number of arguments. When these are given, each argument is treated as a name of a build flag. Mailutils-config checks if such a flag was defined and prints its full name if so. It exits with zero code if all the flags given on the command line are defined. Otherwise, it exits with code of 1.

The comparison of the flag names is case-insensitive. The arguments given need not include the leading prefix (i.e. the characters up to and including the first underscore character).

Given the previous example, the invocation

$ mailutils --info readline use_libpam pop

will produce the following output:

WITH_READLINE
USE_LIBPAM
ENABLE_POP

and will exit with a zero status.

The following command:

$ mailutils --info readline gssapi pop

will exit with status 1, and will print:

WITH_READLINE
ENABLE_POP

since WITH_GSSAPI flag is not defined.

The flags and their meanings are:

USE_LIBPAM
The Mailutils uses PAM libraries.
HAVE_LIBLTDL
The GNU wrapper library libltdl is present and is used by Mailutils. See Using libltdl, for more information on libltdl library.
WITH_BDB2
Support for Berkeley DB is compiled in (the package was configured with --with-db2 option).
WITH_NDBM
Support for NDBM is compiled in (the package was configured with --with-ndbm option).
WITH_OLD_DBM
Support for old style DBM is compiled in (the package was configured with --with-dbm option).
WITH_GDBM
Support for GDBM is compiled in (the package was configured with --with-gdbm option). See Introduction, for information about this library.
WITH_GSSAPI
Support for GSSAPI is compiled in (the package was configured with --with-gssapi option).
WITH_GUILE
Support for Guile extension language is built (the package was configured with --with-guile option). See Overview, for information about Guile.
WITH_PTHREAD
The POSIX thread support is compiled in.
WITH_READLINE
The readline support is enabled (the package was configured with --with-readline option). See Top, for more information.
HAVE_MYSQL
Authentication via MySQL is supported (the package was configured with --enable-mysql option).
ENABLE_VIRTUAL_DOMAINS
Support for mail virtual domains is enabled (the package was configured with --enable-virtual-domains option).
ENABLE_IMAP
Support for IMAP4 protocol is enabled.
ENABLE_POP
Support for POP3 protocol is enabled.
ENABLE_MH
Support for mailboxes in MH format is enabled.
ENABLE_SMTP
Support for SMTP mailer is enabled.
ENABLE_SENDMAIL
Support for Sendmail mailer is enabled.


Node:Reporting Bugs, Next:, Previous:Programs, Up:Top

Reporting Bugs

Email bug reports to [email protected]. Be sure to include the word "mailutils" somewhere in the "Subject:" field.

As the purpose of bug reporting is to improve software, please be sure to include maximum information when reporting a bug. The information needed is:

The archives of bug-mailutils mailing list are available from <http://mail.gnu.org/mailman/listinfo/bug-mailutils>.


Node:News, Next:, Previous:Reporting Bugs, Up:Top

Getting news about GNU mailutils

The two places to look for any news regarding GNU Mailutils are the Mailutils homepage at <http://www.gnu.org/software/mailutils>, and the project page at <http://savannah.gnu.org/projects/mailutils>.

The updated versions of this manual are available online from <http://www.gnu.org/software/mailutils/manual>.


Node:Acknowledgement, Next:, Previous:News, Up:Top

Acknowledgement

In no particular order,


Node:Function Index, Next:, Previous:Acknowledgement, Up:Top

Function Index

This is an alphabetical list of all Mailutils functions.


Node:Program Index, Next:, Previous:Function Index, Up:Top

Program Index


Node:Concept Index, Previous:Program Index, Up:Top

Concept Index

This is a general index of all issues discussed in this manual

Table of Contents


Footnotes

  1. Actually it shoud have been struct mu_auth_data** return_data. This will be fixed in the next release

  2. If # is not the first character on the line, it should be separated from the previous word by any amount of whitespace.