Network Working Group                                        Jon Callas
Category: INTERNET-DRAFT                            Pretty Good Privacy
draft-ietf-openpgp-formats-00.txt                             Network Associates
draft-ietf-openpgp-formats-01.txt                      Lutz Donnerhacke
Expires May Aug 1998                     IN-Root-CA Individual Network e.V.
November
March 1997                                                   Hal Finney
                                                    Pretty Good Privacy
                                                     Network Associates
                                                          Rodney Thayer
                                                       Sable Technology

                  OP Formats - OpenPGP Message Format
                   draft-ietf-openpgp-formats-00.txt
                   draft-ietf-openpgp-formats-01.txt

Copyright 1997 1998 by The Internet Society.  All Rights Reserved.

Status of this Memo

This document is an Internet-Draft.  Internet-Drafts are working
documents of the Internet Engineering Task Force (IETF), its areas, and
its working groups.  Note that other groups may also distribute working
documents as Internet-Drafts.

Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time.  It is inappropriate to use Internet-Drafts Internet- Drafts as reference
material or to cite them other than as "work in progress."

To learn view the current status entire list of any Internet-Draft, current Internet-Drafts, please check the
"1id-abstracts.txt" listing contained in the Internet-Drafts Shadow
Directories on ftp.is.co.za (Africa), nic.nordu.net (Europe), ftp.nordu.net (Northern Europe),
ftp.nis.garr.it (Southern Europe), munnari.oz.au (Pacific Rim), ds.internic.net
ftp.ietf.org (US East Coast), or ftp.isi.edu (US West Coast).

Abstract

This document is maintained in order to publish all necessary
information needed to develop interoperable applications based on the
OP format.  It is not a step-by-step cookbook for writing an
application, it describes only the format and methods needed to read,
check, generate and write conforming packets crossing any network.  It
does not deal with storing and implementation questions albeit it is
necessary to avoid security flaws.

OP (Open-PGP)

Open-PGP software uses a combination of strong public-key and
conventional cryptography to provide security services for electronic
communications and data storage.  These services include
confidentiality, key management, authentication and digital signatures.
This document specifies the message formats used in OP.

Table of Contents

1.  Introduction
1.1 Terms
2.  General functions
2.1 Confidentiality via Encryption
2.2 Authentication via Digital signature
2.3 Compression
2.4 Conversion to Radix-64
2.4.1 Forming ASCII Armor
2.4.2 Encoding Binary in Radix-64
2.4.3 Decoding Radix-64
2.4.4 Examples of Radix-64
2.5 Example of an ASCII Armored Message
2.6 Cleartext signature framework
3.0
3.  Data Element Formats
3.1 Scalar numbers
3.2 Multi-Precision Integers
3.3 Counted Strings Key IDs
3.4 Text
3.5 Time fields
3.5
3.6 String-to-key (S2K) specifiers
3.5.1
3.6.1 String-to-key (S2k) specifier types
3.5.1.1
3.6.1.1 Simple S2K
3.5.1.2
3.6.1.2 Salted S2K
3.5.1.3
3.6.1.3 Iterated and Salted S2K
3.5.2
3.6.2 String-to-key usage
3.5.2.1
3.6.2.1 Secret key encryption
3.5.2.2
3.6.2.2 Conventional message encryption
3.5.3
3.6.3 String-to-key algorithms
3.5.3.1
3.6.3.1 Simple S2K algorithm
3.5.3.2
3.6.3.2 Salted S2K algorithm
3.5.3.3
3.6.3.3 Iterated-Salted S2K algorithm
4.0
4.  Packet Syntax
4.1 Overview
4.2 Packet Headers
4.3 Packet Tags
5.0
5.  Packet Types
5.1 Public-Key Encrypted Session Key Packets (Tag 1)
5.2 Signature Packet (Tag 2)
5.2.1 Version 3 Signature Packet Format
5.2.2 Version 4 Signature Packet Format
5.2.2.1 Signature Subpacket Specification
5.2.2.2 Signature Subpacket Types
5.2.3 Signature Types
5.2.4 Computing Signatures
5.3 Conventional Symmetric-Key Encrypted Session-Key Packets (Tag 3)
5.4 One-Pass Signature Packets (Tag 4)
5.5 Key Material Packet
5.5.1 Key Packet Variants
5.5.1.1 Public Key Packet (Tag 6)
5.5.1.2 Public Subkey Packet (Tag 14)
5.5.1.3 Secret Key Packet (Tag 5)
5.5.1.4 Secret Subkey Packet (Tag 7)
5.5.2 Public Key Packet Formats
5.5.3 Secret Key Packet Formats

5.6 Compressed Data Packet (Tag 8)
5.7 Symmetrically Encrypted Data Packet (Tag 9)
5.8 Marker Packet (Obsolete Literal Packet) (Tag 10)
5.9 Literal Data Packet (Tag 11)
5.10 Trust Packet (Tag 12)
5.11 User ID Packet (Tag 13)
5.12 Comment Packet (Tag 16)
6. Constants  Radix-64 Conversions
6.1 An Implementation of the CRC-24 in "C"
6.2 Forming ASCII Armor
6.3 Encoding Binary in Radix-64
6.4 Decoding Radix-64
6.5 Examples of Radix-64
6.6 Example of an ASCII Armored Message
7.  Cleartext signature framework
8.  Regular expressions
9.  Constants
9.1 Public Key Algorithms
6.2
9.2 Symmetric Key Algorithms
6.3
9.3 Compression Algorithms
6.4
9.4 Hash Algorithms
7.
10.  Packet Composition
7.1
10.1 Transferable Public Keys
7.2
10.2 OP Messages
8.
11.  Enhanced Key Formats
8.1
11.1 Key Structures
8.4
11.2 V4 Key IDs and Fingerprints
9.
12.  Security Considerations
10.
13.  Authors and Working Group Chair
11.
14.  References
12.
15.  Full Copyright Statement

1.  Introduction

This document provides information on the message-exchange packet
formats used by OP to provide encryption, decryption, signing, key
management and functions.  It builds on the foundation provided RFC
1991 "PGP Message Exchange Formats" [1]. Formats."

1.1 Terms

OP - OpenPGP.  This is a definition for security software that uses PGP
5.x as a basis.

PGP - Pretty Good Privacy.  PGP is a family of software systems
developed by Philip R.  Zimmermann from which OP is based.

PGP 2.6.x - This version of PGP has many variants, hence the term PGP
2.6.x.  It used only RSA and IDEA for its cryptography.

PGP 5.x - This version of PGP is formerly known as "PGP 3" in the
community and also in the predecessor of this document, RFC1991.  It
has new formats and corrects a number of problems in the PGP 2.6.x.  It
is referred to here as PGP 5.x because that software was the first
release of the "PGP 3" code base.

"PGP", "Pretty Good", and "Pretty Good Privacy" are trademarks of
Pretty Good Privacy,
Network Associates, Inc.

2.  General functions

OP provides data integrity services for messages and data files by
using these core technologies:

      -digital signature
      -encryption
      -compression
      -radix-64 conversion

In addition, OP provides key management and certificate services.

2.1 Confidentiality via Encryption

OP offers two encryption options to provide confidentiality:
conventional (symmetric-key) encryption and public key encryption.
With public-key encryption, the message is actually encrypted using a
conventional encryption algorithm.  In this mode, each conventional key
is used only once.  That is, a new key is generated as a random number
for each message.  Since it is used only once, the "session key" is
bound to the message and transmitted with it.  To protect the key, it
is encrypted with the receiver's public key.  The sequence is as
follows:

  1. The sender creates a message.
  2. The sending OP generates a random number to be used as a
        session key for this message only.
  3. The session key is encrypted using each recipient's public key.
        These "encrypted session keys" start the message.
  4. The sending OP encrypts the message using the session key, which
        forms the remainder of the message. Note that the message is
        also usually compressed.
  5. The receiving OP decrypts the session key using the recipient's
        private key.
  6. The receiving OP decrypts the message using the session key.
        If the message was compressed, it will be decompressed.

Both digital signature and confidentiality services may be applied to
the same message.  First, a signature is generated for the message and
attached to the message.  Then, the message plus signature is encrypted
using a conventional session key.  Finally, the session key is
encrypted using public-key encryption and prepended to the encrypted
block.

2.2 Authentication via Digital signature

The digital signature uses a hash code or message digest algorithm, and
a public-key signature algorithm.  The sequence is as follows:

  1. The sender creates a message.
  2. The sending software generates a hash code of the message
  3. The sending software generates a signature from the hash code using
     the sender's private key.
  4. The binary signature is attached to the message.
  5. The receiving software keeps a copy of the message signature.
  6. The receiving software generates a new hash code for the received
     message and verifies it using the message's signature. If the
     verification is successful, the message is accepted as authentic.

2.3 Compression

OP implementations MAY compress the message after applying the
signature but before encryption.

2.4 Conversion to Radix-64

OP's underlying native representation for encrypted messages, signature
certificates, and keys is a stream of arbitrary octets.  Some systems
only permit the use of blocks consisting of seven-bit, printable text.
For transporting OP's native raw binary octets through email channels, channels that
are not safe to raw binary data, a printable encoding of these binary
octets is needed.  OP provides the service of converting the raw 8-bit
binary octet stream to a stream of printable ASCII characters, called
Radix-64 encoding or ASCII Armor.

In principle, any printable encoding scheme

Implementations SHOULD provide Radix-64 conversions.

Note that met the requirements
of the email channel would suffice, since it would not change the
underlying binary bit streams of many applications, particularly messaging applications, will
want more advanced features as described in the native OpenPGP-MIME document,
RFC2015.  An application that implements OP for messaging SHOULD also
implement OpenPGP-MIME.

3.  Data Element Formats

This section describes the data structures.  The OP
standard specifies one such printable encoding scheme elements used by OP.

3.1 Scalar numbers

Scalar numbers are unsigned, and are always stored in big-endian
format. Using n[k] to ensure
interoperability.

OP's Radix-64 encoding is composed refer to the kth octet being interpreted, the
value of two parts: a base64 encoding two-octet scalar is ((n[0] << 8) + n[1]).  The value of
the binary data, and a checksum.  The base64 encoding
four-octet scalar is identical to
the MIME base64 content-transfer-encoding [RFC 2045, Section 6.8].  An
OP implementation MAY use ASCII Armor ((n[0] << 24) + (n[1] << 16) + (n[2] << 8) +
n[3]).

3.2 Multi-Precision Integers

Multi-Precision Integers (also called MPIs) are unsigned integers used
to protect hold large integers such as the raw binary data.

The checksum is a 24-bit CRC converted to four characters ones used in cryptographic
calculations.

An MPI consists of radix-64
encoding by the same MIME base64 transformation, preceded by an equals
sign (=).  The CRC two pieces: a two-octet scalar that is computed by using the generator 0x864CFB and an
initialization length of 0xB704CE.  The accumulation is done on
the data
before it is converted to radix-64, rather than on the converted data.
(For more information on CRC functions, see chapter 19 MPI in bits followed by a string of [CAMPBELL].)

{{Editor's note:  This is old text, dating back to RFC 1991.  I have
never liked the glib way the CRC has been dismissed, but I also know octets that this is no place to start contain the actual
integer.

These octets form a discussion of CRC theory.  Should we
construct big-endian number; a sample implementation in C and put big-endian number can be made
into an MPI by prefixing it with the appropriate length.

Examples:

(all numbers are in an appendix? --
jdcc}} hexadecimal)

The checksum string of octets [00 01 01] forms an MPI with its leading equal sign MAY appear on the first line
after value 1.  The
string [00 09 01 FF] forms an MPI with the Base64 encoded data.

Rationale for CRC-24: value of 511.

Additional rules:

The size of 24 bits fits evenly into printable
base64. an MPI is ((MPI.length + 7) / 8) + 2.

The nonzero initialization can detect more errors than a zero
initialization.

2.4.1 Forming ASCII Armor

When OP encodes data into ASCII Armor, it puts specific headers around
the data, so OP can reconstruct the data later.  OP informs the user
what kind of data is encoded in the ASCII armor through the use length field of an MPI describes the

headers.

Concatenating the following data creates ASCII Armor:

- An Armor Header Line, appropriate for length starting from its most
significant non-zero bit.  Thus, the type of data - Armor
Headers - MPI [00 02 01] is not formed
correctly.  It should be [00 01 01].

3.3 Key IDs

A blank (zero-length, or containing only whitespace) line -
The ASCII-Armored data - An Armor Checksum - Key ID is an eight-octet number that identifies a key.
Implementations SHOULD NOT assume that Key IDs are unique.  The Armor Tail, which
depends on the Armor Header Line.

An Armor Header Line consists of the appropriate header line text
surrounded by five (5) dashes ('-', 0x2D) on either side of the header
line text.
section, "Enhanced Key Formats" below describes how Key IDs are formed.

3.4 Text

The header line default character set for text is chosen based upon the type UTF-8 [RFC2044] encoding of data
that is being encoded in Armor, and how it
Unicode [ISO10646].

3.5 Time fields

A time field is being encoded.  Header
line texts include an unsigned four-octet number containing the following strings:

BEGIN PGP MESSAGE number of
seconds elapsed since midnight, 1 January 1970 UTC.

3.6 String-to-key (S2K) specifiers

String-to-key (S2K) specifiers are used for signed, encrypted, or compressed files

BEGIN PGP PUBLIC KEY BLOCK to convert passphrase strings
into conventional encryption/decryption keys.  They are used for armoring public in two
places, currently: to encrypt the secret part of private keys

BEGIN PGP PRIVATE KEY BLOCK used for armoring in the
private keyring, and to convert passphrases to encryption keys

BEGIN PGP MESSAGE, PART X/Y used for multi-part messages, where
conventionally encrypted messages.

3.6.1 String-to-key (S2k) specifier types

There are three types of S2K specifiers currently supported, as
follows:

3.6.1.1 Simple S2K

This directly hashes the
armor is split amongst Y parts, and this is string to produce the Xth part out of Y.

BEGIN PGP MESSAGE, PART X used key data.  See below for multi-part messages, where
how this hashing is done.

    Octet 0:        0x00
    Octet 1:        hash algorithm

3.6.1.2 Salted S2K

This includes a "salt" value in the Xth part of an unspecified number of parts. Requires S2K specifier -- some arbitrary
data -- that gets hashed along with the MESSAGE-ID
Armor Header passphrase string, to be used.

BEGIN PGP SIGNATURE used for detached signatures, OP/MIME signatures, help
prevent dictionary attacks.

    Octet 0:        0x01
    Octet 1:        hash algorithm
    Octets 2-9:     8-octet salt value

3.6.1.3 Iterated and signatures following clearsigned messages Salted S2K

This includes both a salt and an octet count.  The Armor Headers are pairs of strings that can give salt is combined
with the user or passphrase and the
receiving OP message block some information about how to decode or use resulting value is hashed repeatedly.  This
further increases the message.  The Armor Headers are a part amount of the armor, not work an attacker must do to try
dictionary attacks.

    Octet 0:        0x04
    Octet 1:        hash algorithm
    Octets 2-9:     8-octet salt value
    Octets 10-13:   count, a part of four-octet, unsigned value

Note that the message, and hence are not protected by any signatures applied to
the message.

The format value 0x03 for octet 0 of an Armor Header a S2K specifier is that reserved; it
denotes an obsolete form of a key-value pair.  A colon
(':' 0x38) and a single space (0x20) separate the key Interated and value.  OP
should consider improperly formatted Armor Headers Salted S2K.

3.6.2 String-to-key usage

Implementations SHOULD use salted or iterated-and-salted S2K
specifiers, as simple S2K specifiers are more vulnerable to dictionary
attacks.

3.6.2.1 Secret key encryption

An S2K specifier can be corruption of stored in the ASCII Armor.  Unknown keys should be reported secret keyring to the user, but OP
should continue specify how to process the message.  Currently defined Armor Header
Keys include "Version" and "Comment", which define
convert the OP Version used passphrase to encode the message and a user-defined comment.

The "MessageID" Armor Header specifies a 32-character string of
printable characters.  The string must be key that unlocks the same for all parts secret data.  Older
versions of PGP just stored a
multi-part message that uses cipher algorithm octet preceding the "PART X" Armor Header.  MessageID
strings should be chosen with enough internal randomness
secret data or a zero to indicate that no two
messages would have the same MessageID string. secret data was unencrypted.
The MessageID should not appear unless it is in a multi-part message.
If it appears at all, it should be computed from the message in a

deterministic fashion, rather than contain a purely random value.  This
is to allow anyone MD5 hash function was always used to determine that convert the MessageID cannot serve as a
covert means of leaking cryptographic key information.

{{Editor's note:  This needs passphrase to be cleaned up, with a table of the
defined headers.  Also,
key for the MessageID description specified cipher algorithm.

For compatibility, when an S2K specifier is too vague about
how random used, the id has to be.}}

The Armor Tail Line special value 255
is composed stored in the same manner as position where the Armor Header
Line, except hash algorithm octet would have
been in the string "BEGIN" old data structure.  This is replaced then followed immediately by a
one-octet algorithm identifier, and then by the string "END."

2.4.2 Encoding Binary in Radix-64

The encoding process represents 24-bit groups of input bits S2K specifier as output
strings of 4
encoded characters.  Proceeding from left to right, a
24-bit input group above.

Therefore, preceding the secret data there will be one of these
possibilities:

    0           secret data is formed unencrypted (no pass phrase)
    255         followed by concatenating three 8-bit input groups. algorithm octet and S2K specifier
    Cipher alg  use Simple S2K algorithm using MD5 hash

This last possibility, the cipher algorithm number with an implicit use
of MD5 is provided for backward compatibility; it should be understood,
but not generated.

These 24 bits are then treated as four concatenated 6-bit groups, each followed by an 8-octet Initial Vector for the decryption of which is translated into a single digit in
the Radix-64 alphabet.
When encoding a bit stream with secret values, if they are encrypted, and then the Radix-64 encoding, secret key
values themselves.

3.6.2.2 Conventional message encryption

PGP 2.X always used IDEA with Simple string-to-key conversion when
conventionally encrypting a message.  PGP 5 can create a Conventional
Encrypted Session Key packet at the bit stream
must front of a message.  This can be presumed
used to allow S2K specifiers to be ordered used for the passphrase conversion,
to allow other ciphers than IDEA to be used, or to create messages with
a mix of conventional ESKs and public key ESKs.  This allows a message
to be decrypted either with a passphrase or a public key.

3.6.3 String-to-key algorithms

3.6.3.1 Simple S2K algorithm

Simple S2K hashes the most-significant-bit first.
That is, passphrase to produce the first bit session key.  The
manner in which this is done depends on the stream will be size of the high-order bit in session key
(which will depend on the
first 8-bit byte, cipher used) and the eighth bit will be size of the low-order bit in hash
algorithm's output. If the
first 8-bit byte, and so on.

	     +--first octet--+-second octet--+--third octet--+
	     |7 6 5 4 3 2 1 0|7 6 5 4 3 2 1 0|7 6 5 4 3 2 1 0|
	     +-----------+---+-------+-------+---+-----------+
	     |5 4 3 2 1 0|5 4 3 2 1 0|5 4 3 2 1 0|5 4 3 2 1 0|
	     +--1.index--+--2.index--+--3.index--+--4.index--+

Each 6-bit group hash size is greater than or equal to the
session key size, the leftmost octets of the hash are used as an index into an array the key.

If the hash size is less than the key size, multiple instances of 64 printable
characters from the table below.  The character referenced by
hash context are created -- enough to produce the index required key data.
These instances are preloaded with 0, 1, 2, ... octets of zeros (that
is placed in to say, the output string.

     Value Encoding  Value Encoding  Value Encoding  Value Encoding
         0 A            17 R            34 i            51 z
         1 B            18 S            35 j            52 0
         2 C            19 T            36 k            53 1
         3 D            20 U            37 l            54 2
         4 E            21 V            38 m            55 3
         5 F            22 W            39 n            56 4
         6 G            23 X            40 o            57 5
         7 H            24 Y            41 p            58 6
         8 I            25 Z            42 q            59 7
         9 J            26 a            43 r            60 8
        10 K            27 b            44 s            61 9
        11 L            28 c            45 t            62 +
        12 M            29 d            46 u            63 /
        13 N            30 e            47 v
        14 O            31 f            48 w         (pad) =
        15 P            32 g            49 x
        16 Q            33 h            50 y

The encoded output stream must be represented in lines of first instance has no more than
76 characters each.

Special processing is performed if fewer than 24 bits are available at preloading, the end second gets
preloaded with 1 octet of zero, the data being encoded.  There are three possibilities:

- The last data group has 24 bits (3 octets).  No special processing is
needed.

- The last data group has 16 bits (2 octets).  The first two 6-bit
groups are processed as above.  The third (incomplete) data group has is preloaded with two zero-value bits added to it, octets
of zeros, and so forth).

As the data is processed as above.  A pad
character (=) hashed, it is added given independently to each hash context.
Since the contexts have been initialized differently, they will each
produce different hash output.

- The last data group has 8 bits (1 octet).  The first 6-bit group  Once the passphrase is
processed as above.  The second (incomplete) hashed, the
output data group has four
zero-value bits added from the multiple hashes is concatenated, first hash

leftmost, to it, and produce the key data, with any excess octets on the right
discarded.

3.6.3.2 Salted S2K algorithm

Salted S2K is processed as above.  Two pad
characters (=) are added exactly like Simple S2K, except that the input to the output.

2.4.3 Decoding Radix-64

Any characters outside
hash function(s) consists of the base64 alphabet are ignored in Radix-64
data.  Decoding software must ignore all line breaks or other
characters not found in 8 octets of salt from the table above.

In Radix-64 data, characters other than those in S2K
specifier, followed by the table, line
breaks, passphrase.

3.6.3.3 Iterated-Salted S2K algorithm

Iterated-Salted S2K hashes the passphrase and other white space probably indicate a transmission error,
about which a warning message or even a message rejection might salt data multiple times.
The total number of octets to be
appropriate under some circumstances.

Because it hashed is used only for padding at specified in the end of four-octet
count in the data, S2K specifier.  Note that the
occurrence resulting count value is an
octet count of any "=" characters may how many octets will be taken hashed, not an iteration count.

Initially, one or more hash contexts are set up as evidence that with the end other S2K
algorithms, depending on how many octets of key data are needed.  Then
the salt, followed by the passphrase data has been reached (without truncation in transit).  No such
assurance is possible, however, when repeatedly hashed until
the number of octets transmitted
was a multiple of three and no "=" characters are present.

2.4.4 Examples of Radix-64

Input data:  0x14fb9c03d97e
Hex:     1   4    f   b    9   c     | 0   3    d   9    7   e
8-bit:   00010100 11111011 10011100  | 00000011 11011001 11111110
6-bit:   000101 001111 101110 011100 | 000000 111101 100111 111110
Decimal: 5      15     46     28       0      61     37     63
Output:  F      P      u      c        A      9      l      /

Input data:  0x14fb9c03d9
Hex:     1   4    f   b    9   c     | 0   3    d   9
8-bit:   00010100 11111011 10011100  | 00000011 11011001
                                                pad with 00
6-bit:   000101 001111 101110 011100 | 000000 111101 100100
Decimal: 5      15     46     28       0      61     36
                                                   pad with =

Output:  F      P      u      c        A      9      k      =

Input data:  0x14fb9c03
Hex:     1   4    f   b    9   c     | 0   3
8-bit:   00010100 11111011 10011100  | 00000011
                                       pad with 0000
6-bit:   000101 001111 101110 011100 | 000000 110000
Decimal: 5      15     46     28       0      48
                                            pad with =      =
Output:  F      P      u      c        A      w      =      =

2.5 Example of an ASCII Armored Message

  -----BEGIN PGP MESSAGE-----
  Version: OP V0.0

  owFbx8DAYFTCWlySkpkHZDKEFCXmFedmFhdn5ucpZKdWFiv4hgaHKPj5hygUpSbn
  l6UWpabo8XIBAA==
  =3m1o
  -----END PGP MESSAGE-----

Note that this example is indented by two spaces.

2.6 Cleartext signature framework

Sometimes it is necessary to sign a textual octet stream without ASCII
armoring the stream itself, so the signed text is still readable
without special software.  In order to bind a signature to such a
cleartext, this framework is used. (Note that RFC 2015 defines another
way to clear sign messages for environments that support MIME.)

The cleartext signed message consists of:
  - The cleartext header '-----BEGIN PGP SIGNED MESSAGE-----' on a
    single line,
  - Zero or more "Hash" Armor Headers (3.1.2.4),
  - Exactly one empty line not included into the message digest,
  - The dash-escaped cleartext that is included into the message digest,
  - The ASCII armored signature(s) including the Armor Header and Armor
    Tail Lines.

If the "Hash" armor header is given, the specified message digest
algorithm is used for the signature.  If this header is missing, SHA-1
is assumed.  If more than one message digest is used in the signature,
the "Hash" armor header contains a comma-delimited list of used message
digests.  As an abbreviation, the "Hash" armor header may be placed on
the cleartext header line, inserting a comma after the word 'MESSAGE',
as follows:

'-----BEGIN PGP SIGNED MESSAGE, Hash:  MD5, SHA1'.

{{Editor's note:  Should the above armor header line stay or go?
There's no reason that the "Hash:" armor header can't have multiple

hashes in it.  I think anything that reduces parsing complexity is a
Good Thing. --jdcc}}

Current message digest names are:

	- "SHA1"
	- "MD5"
	- "RIPEMD160"

Dash escaped cleartext is the ordinary cleartext where every line
starting with a dash '-' (0x2D) is prepended by the sequence dash '-'
(0x2D) and space ' ' (0x20).  This prevents the parser from recognizing
armor headers of the cleartext itself.  The message digest is computed
using the cleartext itself, not the dash escaped form.

As with binary signatures on text documents (see below), the cleartext
signature is calculated on the text using canonical <CR><LF> line
endings.  The line ending (i.e. the <CR><LF>) before the '-----BEGIN
PGP SIGNATURE-----' line that terminates the signed text is not
considered part of the signed text.

Also, any trailing whitespace (spaces, and tabs, 0x09) at the end of
any line is ignored when the cleartext signature is calculated.

3.  Data Element Formats

This section describes the data elements used by OP.

3.1 Scalar numbers

Scalar numbers are unsigned, and are always stored in big-endian
format. Using n[k] to refer to the kth octet being interpreted, the
value of a two-octet scalar is ((n[0] << 8) + n[1]).  The value of a
four-octet scalar is ((n[0] << 24) + (n[1] << 16) + (n[2] << 8) +
n[3]).

3.2 Multi-Precision Integers

Multi-Precision Integers (also called MPIs) are unsigned integers used
to hold large integers such as the ones used in cryptographic
calculations.

An MPI consists of two pieces: a two-octet scalar that is the length of
the MPI in bits followed by a string of octets that contain the actual
integer.

These octets form a big-endian number; a big-endian number can be made
into an MPI by prefixing it with the appropriate length.

Examples:

(all numbers are in hexadecimal)

The string of octets [00 01 01] forms an MPI with the value 1.  The
string [00 09 01 FF] forms an MPI with the value of 511.

Additional rules:

The size of an MPI is ((MPI.length + 7) / 8) + 2.

The length field of an MPI describes the length starting from its most
significant non-zero bit.  Thus, the MPI [00 02 01] is not formed
correctly.  It should be [00 01 01].

3.3 Counted Strings

A counted string consists of a length and then N octets of string data.
Its default character set is UTF-8 [RFC2044] encoding of Unicode
[ISO10646].

3.4 Time fields

A time field is an unsigned four-octet number containing the number of
seconds elapsed since midnight, 1 January 1970 UTC.

3.5 String-to-key (S2K) specifiers

String-to-key (S2K) specifiers are used to convert passphrase strings
into conventional encryption/decryption keys.  They are used in two
places, currently: to encrypt the secret part of private keys in the
private keyring, and to convert passphrases to encryption keys for
conventionally encrypted messages.

3.5.1 String-to-key (S2k) specifier types

There are three types of S2K specifiers currently supported, as
follows:

3.5.1.1 Simple S2K

This directly hashes the string to produce the key data.  See below for
how this hashing is done.

	Octet 0:		0x00
	Octet 1:		hash algorithm

3.5.1.2 Salted S2K

This includes a "salt" value in the S2K specifier -- some arbitrary
data -- that gets hashed along with the passphrase string, to help
prevent dictionary attacks.

	Octet 0:		0x01
	Octet 1:		hash algorithm
	Octets 2-9:		8-octet salt value

3.5.1.3 Iterated and Salted S2K

This includes both a salt and an octet count.  The salt is combined
with the passphrase and the resulting value is hashed repeatedly.  This
further increases the amount of work an attacker must do to try
dictionary attacks.

	Octet 0:		0x03
	Octet 1:		hash algorithm
	Octets 2-9:		8-octet salt value
	Octet 10:		count, in special format (described below)

3.5.2 String-to-key usage

Implementations MUST implement simple S2K and salted S2K specifiers.
Implementations MAY implement iterated and salted S2K specifiers.
Implementations SHOULD use salted S2K specifiers, as simple S2K
specifiers are more vulnerable to dictionary attacks.

3.5.2.1 Secret key encryption

An S2K specifier can be stored in the secret keyring to specify how to
convert the passphrase to a key that unlocks the secret data.  Older
versions of PGP just stored a cipher algorithm octet preceding the
secret data or a zero to indicate that the secret data was unencrypted.
The MD5 hash function was always used to convert the passphrase to a
key for the specified cipher algorithm.

For compatibility, when an S2K specifier is used, the special value 255
is stored in the position where the hash algorithm octet would have
been in the old data structure.  This is then followed immediately by a
one-octet algorithm identifier, and then specified by the S2K specifier as
encoded above.

Therefore, preceding the secret data there will be one of these
possibilities:

	0		secret data is unencrypted (no pass phrase)
	255		followed by algorithm octet and S2K specifier
	Cipher alg	use Simple S2K algorithm using MD5 hash

This last possibility, the cipher algorithm number with an implicit use
of MD5 is provided for backward compatibility; it should be understood,
but not generated.

These are followed by an 8-octet Initial Vector for the decryption of
the secret values, if they are encrypted, and then the secret key
values themselves.

3.5.2.2 Conventional message encryption

PGP 2.X always used IDEA with Simple string-to-key conversion when

conventionally encrypting a message.  PGP 5 can create a Conventional
Encrypted Session Key packet at the front of a message.  This can be
used to allow S2K specifiers to be used for the passphrase conversion,
to allow other ciphers than IDEA to be used, or to create messages with
a mix of conventional ESKs and public key ESKs.  This allows a message
to be decrypted either with a passphrase or a public key.

3.5.3 String-to-key algorithms

3.5.3.1 Simple S2K algorithm

Simple S2K hashes the passphrase to produce the session key.  The
manner in which this is done depends on the size of the session key
(which will depend on the cipher used) and the size of the hash
algorithm's output. If the hash size is greater than or equal to the
session key size, the leftmost octets of the hash are used as the key.

If count has been hashed.  The
one exception is that if the hash size octet count is less than the key size, multiple instances size of the
hash context are created -- enough to produce
salt plus passphrase, the required key data.
These instances are preloaded with 0, 1, 2, ... octets of zeros (that full salt plus passphrase will be hashed even
though that is to say, the first instance has no preloading, greater than the second gets
preloaded with 1 octet of zero, count.  After the third hashing is preloaded with two octets
of zeros, and so forth).

As done
the data is hashed, it is given independently to each hash context.
Since unloaded from the contexts have been initialized differently, they will each
produce different hash output.  Once context(s) as with the passphrase is hashed, other S2K
algorithms.

4.  Packet Syntax

This section describes the
output data packets used by OP.

4.1 Overview

An OP message is constructed from a number of records that are
traditionally called packets.  A packet is a chunk of data that has a
tag specifying its meaning.  An OP message, keyring, certificate, and
so forth consists of a number of packets.  Some of those packets may
contain other OP packets (for example, a compressed data packet, when
uncompressed, contains OP packets).

Each packet consists of a packet header, followed by the multiple hashes packet body.
The packet header is concatenated, of variable length.

4.2 Packet Headers

The first hash
leftmost, to produce the key data, with any excess octets on octet of the right
discarded.

3.5.3.2 Salted S2K algorithm

Salted S2K packet header is exactly like Simple S2K, except that called the "Packet Tag." It
determines the format of the input to header and denotes the
hash function(s) consists packet contents.
The remainder of the 8 octets packet header is the length of salt from the S2K
specifier, followed by packet.

Note that the passphrase.

3.5.3.3 Iterated-Salted S2K algorithm

{{Editor's note:  This most significant bit is very complex, the left-most bit, called bit 7.
A mask for this bit is 0x80 in hexadecimal.

       +---------------+
  PTag |7 6 5 4 3 2 1 0|
       +---------------+
  Bit 7 -- Always one
  Bit 6 -- New packet format if set

PGP 2.6.X only uses old format packets.  Thus, software that
interoperates with bizarre things like those versions of PGP must only use old format
packets.  If interoperability is not an
8-bit floating point format.  Should we just drop it? --jdcc}}

Iterated-Salted S2K hashes the passphrase issue, either format may be
used.  Note that old format packets have four bits of content tags, and salt data multiple times.
new format packets have six; some features cannot be used and still be
backwards-compatible.

Old format packets contain:
  Bits 5-2 -- content tag
  Bits 1-0 - length-type

New format packets contain:
  Bits 5-0 -- content tag

The total number meaning of the length-type in old-format packets is:

0 - The packet has a one-octet length.  The header is 2 octets to be hashed long.

1 - The packet has a two-octet length.  The header is encoded in 3 octets long.

2 - The packet has a four-octet length.  The header is 5 octets long.

3 - The packet is of indeterminate length.  The header is 1 byte long,
and the count octet
that follows application must determine how long the salt in packet is.  If the S2K specifier.  The count value
packet is stored
as in a normalized floating-point value with 4 bits of exponent and 4 bits
of mantissa.  The formula to convert from file, this means that the count octet to a count of packet extends until the number end
of octets to be hashed is as follows, letting the high 4
bits file.  In general, an application should not use indeterminate
length packets except where the end of the count octet data will be CEXP and clear from the low four bits be CMANT:

	count
context.

New format packets have three possible ways of octets to be hashed = (16 + CMANT) << (CEXP + 6)

This allows encoding hash counts as low as 16 << 6 or 1024 (using an

octet value length.  A
one-octet Body Length header encodes packet lengths of 0), up to 191
octets, and as high as 31 << 21 a two-octet Body Length header encodes packet lengths of
192 to 8383 octets.  For cases where longer packet body lengths are
needed, or 65011712 (using an octet
value where the length of 0xff).  Note that the resulting count value packet body is an octet count
of how many octets will be hashed, not an iteration count.

Initially, one or more hash contexts known in advance
by the issuer, Partial Body Length headers can be used.  These are set up as with
one-octet length headers that encode the other S2K
algorithms, depending on how many octets length of only part of key data are needed.  Then the salt,
data packet.

Each Partial Body Length header is followed by a portion of the passphrase data is repeatedly hashed until packet
body data.  The Partial Body Length header specifies this portion's
length.  Another length header (of one of the three types) follows that
portion.  The last length header in the number packet must always be a regular
Body Length header.  Partial Body Length headers may only be used for
the non-final parts of octets specified by the octet count has been hashed.  The
one exception packet.

A one-octet Body Length header encodes a length of from 0 to 191
octets. This type of length header is that if recognized because the one octet count
value is less than the size 192.  The body length is equal to:

bodyLen = length_octet;

A two-octet Body Length header encodes a length of the
salt plus passphrase, the full salt plus passphrase will be hashed even
though that from 192 to 8383
octets.  It is greater than the recognized because its first octet count.  After the hashing is done
the data is unloaded from the hash context(s) as with the other S2K
algorithms.

4.  Packet Syntax

This section describes in the packets used by OP.

4.1 Overview

An OP message range 192
to 223.  The body length is constructed from a number of records that are
traditionally called packets. equal to:

bodyLen = (1st_octet - 192) * 256 + (2nd_octet) + 192

A packet Partial Body Length header is a chunk of data that has a
tag specifying its meaning.  An OP message, keyring, certificate, one octet long and
so forth consists of encodes a number of packets.  Some of those packets may
contain other OP packets (for example, length
which is a compressed data packet, when
uncompressed, contains OP packets).

Each packet consists power of a 2, from 1 to 2147483648 (2 to the 31st power).  It
is recognized because its one octet value is greater than or equal to
224.  The partial body length is equal to:

partialBodyLen = 1 << (length_octet & 0x1f);

Examples:

A packet header, with length 100 may have its length encoded in one octet:
0x64. This is followed by the packet body.
The 100 octets of data.

A packet with length 1723 may have its length coded in two octets:
0xC5, 0xFB.  This header is followed by the 1723 octets of variable length.

4.2 Packet Headers

The data.

A packet with length 100000 might be encoded in the following octet
stream: 0xE1, first two octets of data, 0xE0, next one octet of the packet header data,
0xEF, next 32768 octets of data, 0xF0, next 65536 octets of data, 0xC5,
0xDD, last 1693 octets of data.  This is called just one possible encoding,
and many variations are possible on the "Packet Tag." It
determines size of the format Partial Body Length
headers, as long as a regular Body Length header encodes the last
portion of the data.  Note also that the last Body Length header and denotes can be
a zero-length header.

Please note that in all of these explanations, the packet contents.
The remainder total length of the
packet header is the length of the packet.

Note that header(s) plus the most significant bit is length of the left-most bit, called bit 7.
A mask for this bit is 0x80 in hexadecimal.

       +---------------+
  PTag |7 6 5 4 3 2 1 0|
       +---------------+
  Bit 7 -- Always one
  Bit 6 -- New body.

4.3 Packet Tags

The packet format if set

PGP 2.6.X only uses old format packets.  Thus, software that
interoperates with those versions tag denotes what type of PGP must only use packet the body holds.  Note that
old format
packets.  If interoperability is not an issue, either format may be
used.

Old format packets contain:
  Bits 5-2 -- content tag
  Bits 1-0 - length-type

New can only have tags less than 16, whereas new format
packets contain:
  Bits 5-0 -- content tag can have tags as great as 63.  The meaning of the length-type in old-format packets is: defined tags (in decimal)
are:

0 - The        -- Reserved. A packet has must not have a one-octet length.  The header is 2 octets long. tag with this value.
1 - The packet has a two-octet length.  The header is 3 octets long.        -- Public-Key Encrypted Session Key Packet
2 - The packet has a four-octet length.  The header is 5 octets long.        -- Signature Packet

3 - The        -- Symmetric-Key Encrypted Session Key Packet
4        -- One-Pass Signature Packet
5        -- Secret Key Packet
6        -- Public Key Packet
7        -- Secret Subkey Packet
8        -- Compressed Data Packet
9        -- Symmetrically Encrypted Data Packet
10       -- Marker Packet
11       -- Literal Data Packet
12       -- Trust Packet
13       -- Name Packet
14       -- Subkey Packet
15       -- Reserved
60 to 63 -- Private or Experimental Values

5.  Packet Types

5.1 Public-Key Encrypted Session Key Packets (Tag 1)

A Public-Key Encrypted Session Key packet holds the key used to encrypt
a message that is of indeterminate length. itself encrypted with a public key.  Zero or more
Encrypted Session Key packets and/or Conventional Encrypted Session Key
packets may precede a Symmetrically Encrypted Data Packet, which holds
an encrypted message.  The header message is 1 byte long, encrypted with a session key, and
the application must determine how long session key is itself encrypted and stored in the Encrypted Session
Key packet(s).  The Symmetrically Encrypted Data Packet is preceded by
one Public-Key Encrypted Session Key packet is.  If for each OP key to which
the
packet message is in encrypted.  The recipient of the message finds a file, this means session
key that is encrypted to their public key, decrypts the session key,
and then uses the session key to decrypt the message.

The body of this packet extends until consists of:

    - A one-octet number giving the end version number of the file.  In general, an application packet type.
      The currently defined value for packet version is 3. An
        implementation should accept, but not use indeterminate
length packets except where generate a version of 2,
        which is equivalent to V3 in all other respects.
    - An eight-octet number that gives the end key ID of the data will be clear from public key that
      the
context.

New format packets have three possible ways of encoding length. session key is encrypted to.
    - A one-octet Body Length header encodes packet lengths number giving the public key algorithm used.
    - A string of octets that is the encrypted session key. This
      string takes up to 191
octets, and a two-octet Body Length header encodes packet lengths the remainder of
192 to 8383 octets.  For cases where longer packet body lengths the packet, and its contents are
needed, or where
      dependent on the length public key algorithm used.

    Algorithm Specific Fields for RSA encryption
    - multiprecision integer (MPI) of RSA encrypted value m**e mod n.

    Algorithm Specific Fields for Elgamal encryption:
    - MPI of DSA value g**k mod p.
    - MPI of DSA value m * y**k mod p.

The encrypted value "m" in the packet body above formulas is not known in advance
by derived from the issuer, Partial Body Length headers can be used.  These are
session key as follows.  First the session key is prepended with a
one-octet length headers algorithm identifier that encode specifies the length of only part of conventional
encryption algorithm used to encrypt the
data packet.

Each Partial Body Length header is followed by following Symmetrically
Encrypted Data Packet.  Then a portion of two-octet checksum is appended which is
equal to the packet
body data.  The Partial Body Length header specifies this portion's
length.  Another length header (of one sum of the three types) follows that
portion.  The last length header preceding octets, including the algorithm
identifier and session key, modulo 65536.  This value is then padded as
described in PKCS-1 block type 02 [PKCS1] to form the packet must always be a regular
Body Length header.  Partial Body Length headers may only be "m" value used for in
the non-final parts formulas above.

An implementation MAY use a Key ID of zero as a "wild card" or
"speculative" Key ID.  In this case, the packet.

A one-octet Body Length header encodes implementation would try all
available private keys, checking for a length of from 0 to 191
octets. valid decrypted session key.
This type format helps reduce traffic analysis of length header is recognized because the one octet
value is less than 192.  The body length is equal to:

bodyLen = length_octet; messages.

5.2 Signature Packet (Tag 2)

A two-octet Body Length header encodes signature packet describes a binding between some public key and some
data.  The most common signatures are a length signature of from 192 to 8383
octets.  It is recognized because its first octet is in the range 192
to 223.  The body length is equal to:

bodyLen = (1st_octet - 192) * 256 + (2nd_octet) + 192

A Partial Body Length header is one octet long a file or a block
of text, and encodes a length
which signature that is a power certification of 2, from 1 to 2147483648 (2 to a user ID.

Two versions of signature packets are defined.  Version 3 provides
basic signature information, while version 4 provides an expandable
format with subpackets that can specify more information about the 31st power).  It
signature. PGP 2.6.X only accepts version 3 signatures.

Implementations MUST accept V3 signatures.  Implementations SHOULD
generate V4 signatures, unless there is recognized because its one octet value a need to generate a signature
that can be verified by old implementations.

Note that if an implementation is greater than or equal creating an encrypted and signed
message that is encrypted to
224.  The partial body length a V3 key, it is equal to:

partialBodyLen = 1 << (length_octet & 0x1f);

Examples: reasonable to create a V3
signature.

5.2.1 Version 3 Signature Packet Format

A version 3 Signature packet with length 100 may have its contains:
    - One-octet version number (3).
    - One-octet length encoded in one octet:
0x64. of following hashed material.  MUST be 5.
    - One-octet signature type.
    - Four-octet creation time.
    - Eight-octet key ID of signer.
    - One-octet public key algorithm.
    - One-octet hash algorithm.
    - Two-octet field holding left 16 bits of signed hash value.
    - One or more multi-precision integers comprising the signature.
      This portion is followed by 100 octets of data.

A packet with length 1723 may have its length coded in two octets:
0xC5, 0xFB.  This header algorithm specific, as described below.

The data being signed is followed by hashed, and then the 1723 octets of data.

A signature type and
creation time from the signature packet with length 100000 might be encoded are hashed (5 additional
octets).  The resulting hash value is used in the following octet
stream: 0xE1, first signature algorithm.
The high 16 bits (first two octets of data, 0xE0, next one octet octets) of data,
0xEF, next 32768 octets the hash are included in the

signature packet to provide a quick test to reject some invalid
signatures.

    Algorithm Specific Fields for RSA signatures:
    - multiprecision integer (MPI) of data, 0xF0, next 65536 octets RSA signature value m**d.

    Algorithm Specific Fields for DSA signatures:
    - MPI of data, 0xC5,
0xDD, last 1693 octets DSA value r.
    - MPI of data.  This DSA value s.

The signature calculation is just one possible encoding,
and many variations are possible based on the size of the Partial Body Length
headers, as long as a regular Body Length header encodes the last
portion hash of the data.  Note also that the last Body Length header can be
a zero-length header.

Please note that in all signed data, as
described above.  The details of these explanations, the total length of calculation are different for DSA
signature than for RSA signatures.

With RSA signatures, the
packet hash value is the length of the header(s) plus the length encoded as described in PKCS-1
section 10.1.2, "Data encoding", producing an ASN.1 value of type
DigestInfo, and then padded using PKCS-1 block type 01 [PKCS1].  This
requires inserting the body.

4.3 Packet Tags hash value as an octet string into an ASN.1
structure.  The packet tag denotes what object identifier for the type of packet hash being used is
included in the body holds.  Note that
old format packets can only have tags less than 16, whereas new format
packets can have tags as great as 63. structure.  The hexadecimal representations for the
currently defined tags (in decimal) hash algorithms are:

0        -- Reserved. A packet must not have a tag

    - MD5:          0x2a, 0x86, 0x48, 0x86, 0xf7, 0x0d, 0x02, 0x05
    - SHA-1:        0x2b, 0x0e, 0x03, 0x02, 0x1a
    - RIPEMD-160:   0x2b, 0x24, 0x03, 0x02, 0x01

The ASN.1 OIDs are:
    - MD5:       1.2.840.113549.2.5
    - SHA-1:     1.3.14.3.2.26
    - RIPEMD160: 1.3.36.3.2.1

DSA signatures SHOULD use hashes with this value.
1        -- Encrypted Session Key Packet
2        -- Signature Packet
3        -- Conventionally Encrypted Session Key Packet
4        -- One-Pass Signature Packet
5        -- Secret Key Packet
6        -- Public Key Packet
7        -- Secret Subkey Packet
8        -- Compressed Data Packet
9        -- Symmetrically Encrypted Data Packet
10       -- Marker Packet
11       -- Literal Data Packet
12       -- Trust Packet
13       -- Name Packet
14       -- Subkey Packet
15       -- Reserved
16       -- Comment Packet
60 a size of 160 bits, to 63 -- Private or Experimental Values

5.  Packet Types

5.1 Encrypted Session Key Packets (Tag 1)

An Encrypted Session Key packet holds match q,
the key used to encrypt a message
that size of the group generated by the DSA key's generator value.  The
hash function result is itself encrypted with treated as a 160 bit number and used directly
in the DSA signature algorithm.

5.2.2 Version 4 Signature Packet Format

A version 4 Signature packet contains:
    - One-octet version number (4).
    - One-octet signature type.
    - One-octet public key.  Zero key algorithm.
    - One-octet hash algorithm.
    - Two-octet octet count for following hashed subpacket data.
    - Hashed subpacket data. (zero or more Encrypted
Session Key packets and/or Conventional Encrypted Session Key packets
may precede a Symmetrically Encrypted Data Packet, which holds an
encrypted message. subpackets)
    - Two-octet octet count for following unhashed subpacket data.
    - Unhashed subpacket data. (zero or more subpackets)
    - Two-octet field holding left 16 bits of signed hash value.
    - One or more multi-precision integers comprising the signature.
      This portion is algorithm specific, as described above.

The message data being signed is encrypted with a session key, hashed, and then the session key signature data from the
version number through the hashed subpacket data is itself encrypted and stored hashed.  The
resulting hash value is what is signed.  The left 16 bits of the hash
are included in the Encrypted Session
Key signature packet or to provide a quick test to reject
some invalid signatures.

There are two fields consisting of signature subpackets.  The first
field is hashed with the Conventional Encrypted Session Key packet. rest of the signature data, while the second
is unhashed.  The second set of subpackets is not cryptographically
protected by the signature and should include only advisory
information.

The algorithms for converting the hash function result to a signature
are described above.

5.2.2.1 Signature Subpacket Specification

The
Symmetrically Encrypted Data Packet subpacket fields consist of zero or more signature subpackets.
Each set of subpackets is preceded by one Encrypted
Session Key packet for each OP key to which a two-octet count of the message is encrypted.
The recipient length
of the message finds set of subpackets.

Each subpacket consists of a session key that is encrypted to
their public key, decrypts the session key, subpacket header and then uses the session
key to decrypt the message. a body.  The body of this packet header
consists of:

    - A one-octet number giving subpacket length (1 or 2 octets):
      Length includes the type octet but not this length,
      1st octet <  192, then length is octet value
      1st octet >= 192, then length is 2 octets and equal to
        (1st octet - 192) * 256 + (2nd octet) + 192

    - subpacket type (1 octet):
      If bit 7 is set, subpacket understanding is critical,
       2 = signature creation time,
       3 = signature expiration time,
       4 = exportable,
       5 = trust signature,
       6 = regular expression,
       7 = revocable,
       9 = key expiration time,
      10 = placeholder for backwards compatibility
      11 = preferred symmetric algorithms,
      12 = revocation key,
      16 = issuer key ID,
      20 = notation data,
      21 = preferred hash algorithms,
      22 = preferred compression algorithms,
      23 = key server preferences,
      24 = preferred key server,
      25 = primary user id,
      26 = policy URL,

27 = key flags, 28 = Signer's user id

    - subpacket specific data:

An implementation SHOULD ignore any subpacket that it does not
recognize.

Bit 7 of the subpacket type is the "critical" bit.  If set, it denotes
that the version number subpacket is one which is critical that the evaluator of the packet type.
	  The currently defined value for packet version is 3. An
        implementation should accept, but not generate
signature recognize.  If a version of 2, subpacket is encountered which is equivalent marked
critical but is unknown to V3 the evaluating software, the evaluator
SHOULD consider the signature to be in all other respects.
	- error.

An eight-octet number that gives the key ID evaluator may "recognize" a subpacket, but not implement it.  The
purpose of the public key that
	  the session key critical bit is encrypted to.
	- A one-octet number giving to allow the public key algorithm used.
	- A string of octets signer to tell an evaluator
that is the encrypted session key. This
	  string takes up the remainder it would prefer a new, unknown feature to generate an error than
be ignored.

5.2.2.2 Signature Subpacket Types

Several types of subpackets are currently defined.  Some subpackets
apply to the packet, signature itself and its contents some are
	  dependent on the public key algorithm used.

    Algorithm Specific Fields for RSA encryption
	- multiprecision integer (MPI) of RSA encrypted value m**e.

    Algorithm Specific Fields for Elgamal encryption:
	- MPI of DSA value g**k.
	- MPI attributes of DSA value m * y**k.

The encrypted value "m" in the above formulas is derived from the
session key as follows.  First key.
Subpackets that are found on a self-signature are placed on a user name
certification made by the session key is prepended with a
one-octet algorithm identifier itself.  Note that specifies the conventional
encryption algorithm used to encrypt the following Symmetrically
Encrypted Data Packet.  Then a two-octet checksum is appended which key may have more
than one user name, and thus may have more than one self-signature, and
differing subpackets.

A self-signature is
equal to a binding signature made by the sum key the signature
refers to.  There are three types of self-signatures, the preceding octets, including certification
signatures (types 0x10-0x13), the algorithm
identifier direct-key signature (type 0x1f), and session key, modulo 65536.  This value is then padded as
described in PKCS-1 block type 02 [PKCS1] to form the "m" value used in
the formulas above.

5.2 Signature Packet (Tag 2)

A signature packet describes a subkey binding between some public key and some
data.  The most common signatures are a signature of a file or (type 0x18).  For certification
self-signatures, username may have a block
of text, self-signature, and thus different
subpackets in those self-signatures.  For subkey binding signatures,
each subkey in fact has a signature self-signature.  Subpackets that is appear in a
certification of a user ID.

Two versions of signature packets are defined.  Version 3 provides
basic self-signature apply to the username, and subpackets that
appear in the subkey self-signature apply to the subkey.  Lastly,
subpackets on the direct key signature information, while version 4 provides an expandable
format with apply to the entire key.

Implementing software should interpret a self-signature's preference
subpackets as narrowly as possible.  For example, suppose a key has two
usernames, Alice and Bob.  Suppose that can specify more information about Alice prefers the
signature. PGP 2.6.X only accepts version 3 signatures.

Implementations MUST accept V3 signatures.  Implementations SHOULD
generate V4 signatures, unless there symmetric
algorithm CAST5, and Bob prefers IDEA or Triple-DES.  If the software
locates this key via Alice's name, then the preferred algorithm is
CAST5, if software locates the key via Bob's name, then the preferred
algorithm is IDEA.  If the key is a need to generate a signature
that can be verified located by PGP 2.6.x.

5.2.1 Version 3 Signature Packet Format

A version 3 Signature packet contains:
	- One-octet version number (3).
	- One-octet length of following hashed material.  MUST be 5.
	- One-octet signature type.
	- Four-octet creation time.
	- Eight-octet key ID id, then algorithm of signer.
	- One-octet public
the default user name of the key provides the default symmetric
algorithm.
	- One-octet hash algorithm.
	- Two-octet field holding left 16 bits of signed hash value.
	- One or more multi-precision integers comprising

A subpacket may be found either in the hashed or unhashed subpacket
sections of a signature.
	  This portion is algorithm specific, as described below.

The data being signed  If a subpacket is not hashed, and then the
information in it cannot be considered definitive because it is not
part of the signature type and proper.

Subpacket types:

    Signature creation time from the signature packet are hashed (5 additional
octets). (4 octet time field)

        The resulting hash value is used in time the signature algorithm. was made.  Always included with new
        signatures.

    Issuer (8 octet key ID)

        The high 16 bits (first two octets) OP key ID of the hash are included in key issuing the
signature packet to provide a quick test to reject some invalid
signatures.

    Algorithm Specific Fields for RSA signatures:
	- multiprecision integer (MPI) signature.

    Key expiration time (4 octet time field)

        The validity period of RSA signature value m**d.

    Algorithm Specific Fields for DSA signatures:
	- MPI the key.  This is the number of DSA seconds
        after the key creation time that the key expires.  If this is
        not present or has a value r.
	- MPI of DSA value s.

The signature calculation zero, the key never expires. This
        is based found only on a hash self-signature.

    Preferred symmetric algorithms (array of one-octet values)

        Symmetric algorithm numbers that indicate which algorithms the signed data, as
described above.  The details
        key holder prefers to use.  This is an ordered list of octets
        with the calculation most preferred listed first.  It should be assumed
        that only algorithms listed are different for DSA
signature than for RSA signatures.

With RSA signatures, supported by the hash value is encoded as described recipient's
        software.  Algorithm numbers in PKCS-1 section 10.1.2, "Data encoding", producing an ASN.1 value of type
DigestInfo, and then padded using PKCS-1 block type 01 [PKCS1]. 6. This
requires inserting the is only found
        on a self-signature.

    Preferred hash value as an octet string into an ASN.1
structure.  The object identifier for the type algorithms (array of hash being used is
included in one-octet values)

        Message digest algorithm numbers that indicate which algorithms
        the structure.  The hexadecimal representations for key holder prefers to receive.  Like the
currently defined hash algorithms are:

	- SHA-1:	    0x2b, 0x0e, 0x03, 0x02, 0x1a
	- MD5:		    0x2a, 0x86, 0x48, 0x86, 0xf7, 0x0d, 0x02, 0x05
	- RIPEMD-160:	0x2b, 0x24, 0x03, 0x02, 0x01

The ASN.1 OIDs are:
	- MD5:       1.2.840.113549.2.5
	- SHA-1:     1.3.14.3.2.26
	- RIPEMD160: 1.3.36.3.2.1

DSA signatures SHOULD use hashes with preferred
        symmetric algorithms, the list is ordered. Algorithm numbers
        are in section 6. This is only found on a size self-signature.

    Preferred compression algorithms (array of 160 bits, one-octet values)

        Compression algorithm numbers that indicate which algorithms
        the key holder prefers to match q, use.  Like the size of preferred symmetric
        algorithms, the group generated by list is ordered.  Algorithm numbers are in
        section 6.  If this subpacket is not included, ZIP is
        preferred. A zero denotes that uncompressed data is preferred;
        the DSA key's generator value.  The
hash function result key holder's software may not have compression software.
        This is treated as only found on a 160 bit self-signature.

    Signature expiration time (4 octet time field)
        The validity period of the signature.  This is the number and used directly
in of
        seconds after the DSA signature algorithm.

5.2.2 Version 4 Signature Packet Format

A version 4 Signature packet contains:
	- One-octet version number (4).
	- One-octet creation time that the signature type.
	- One-octet public key algorithm.
	- One-octet hash algorithm.
	- Two-octet
        expires.  If this is not present or has a value of zero, it
        never expires.

    Exportable (1 octet count of exportability, 0 for following hashed subpacket data.
	- Hashed subpacket data.
	- Two-octet octet count not, 1 for following unhashed subpacket data.
	- Unhashed subpacket data.
	- Two-octet field holding left 16 bits of signed hash value.
	- One or more multi-precision integers comprising exportable)

        Signature's exportability status.  Packet body contains a
        boolean flag indicating whether the signature.
	  This portion is algorithm specific, as described above.

The data being signed signature is hashed, exportable.
        Signatures which are not exportable are ignored during export
        and then import operations.  If this packet is not present the
        signature is assumed to be exportable.

    Revocable (1 octet of revocability, 0 for not, 1 for revocable)

        Signature's revocability status.  Packet body contains a
        boolean flag indicating whether the signature data from is revocable.
        Signatures which are not revocable have any later revocation
        signatures ignored.  They represent a commitment by the
version number through signer
        that he cannot revoke his signature for the hashed subpacket data is hashed.  The
resulting hash value life of his key.
        If this packet is what not present, the signature is signed.  The left 16 bits revocable.

    Trust signature (1 octet "level" (depth), 1 octet of trust amount)

        Signer asserts that the hash
are included in key is not only valid, but also
        trustworthy, at the signature packet specified level.  Level 0 has the same
        meaning as an ordinary validity signature.  Level 1 means that
        the signed key is asserted to provide be a quick test to reject
some invalid signatures.

There are two fields consisting of signature subpackets.  The first
field is hashed valid trusted introducer,
        with the rest 2nd octet of the signature data, while body specifying the second
is unhashed.  The second set degree of subpackets is not cryptographically
protected by the signature and should include only advisory
information.

The algorithms for converting trust.
        Level 2 means that the hash function result signed key is asserted to be trusted to
        issue level 1 trust signatures, i.e. that it is a "meta
        introducer".  Generally, a level n trust signature
are described above.

5.2.2.1 Signature Subpacket Specification asserts that
        a key is trusted to issue level n-1 trust signatures.  The subpacket fields consist
        trust amount is in a range from 0-255, interpreted such that
        values less than 120 indicate partial trust and values of zero 120
        or more greater indicate complete trust.  Implementations SHOULD
        emit values of 60 for partial trust and 120 for complete trust.

    Regular expression (null-terminated regular expression)

        Used in conjunction with trust signature subpackets.
Each set packets (of level > 0)
        to limit the scope of subpackets trust which is preceded extended.  Only signatures
        by a two-octet count of the length
of target key on user IDs which match the set regular
        expression in the body of subpackets.

Each subpacket consists this packet have trust extended by
        the trust packet.  The regular expression uses the same syntax
        as the Henry Spencer's "almost public domain" regular
        expression package.  A description of the syntax in in a subpacket header and a body.  The header
consists of:

	- subpacket length
        section below.

    Revocation key (1 or 2 octets):
	  Length includes the type octet but not this length,
	  1st octet <  192, then length is octet value
	  1st of class, 1 octet >= 192, then length is 2 of algid, 20 octets and equal of
    fingerprint)

        Authorizes the specified key to
	    (1st issue revocation
        self-signatures for this key.  Class octet - 192) * 256 + (2nd octet) + 192
	- subpacket type (1 octet):
	  If must have bit 7 is 0x80
        set, subpacket understanding is critical,
	   2 = signature creation time,
	   3 = other bits are for future expansion to other kinds of
        signature expiration time,
	   4 = exportable,
	   5 = trust signature,
	   6 = regular expression,
	   7 = revocable,
	   9 = key expiration time,
	  10 = additional recipient request,
	  11 = preferred symmetric algorithms,
	  12 = revocation key,
	  16 = issuer key ID,
      20 = notation data,
      21 = preferred hash algorithms,
      22 = preferred compression algorithms,
      23 = key server preferences,
      24 = preferred authorizations. This is found on a self-signature.

        Authorizes the specified key server

	- subpacket specific data:

Bit 7 of to issue revocation signatures for
        this key.  Class octet must have bit 0x80 set.  If the subpacket type bit 0x40
        is set, then this means that the "critical" bit. revocation information is
        sensitive.  Other bits are for future expansion to other kinds
        of authorizations.  This is found on a self-signature.

        If the "sensitive" flag is set, it implies the keyholder feels this
        subpacket contains private trust information that it describes a
        real-world sensitive relationship.  If this flag is critical that set,
        implementations SHOULD NOT export this signature to other users
        except in cases where the subpacket data needs to be one which available: when the
        signature is understood by being sent to the software.  If designated revoker, or when it
        is accompanied by a revocation signature from that revoker.
        Note that it may be appropriate to isolate this subpacket
        within a separate signature so that it is encountered which is marked critical
but the software does not understand, the handling depends combined with
        other subpackets which need to be exported.

    Notation Data (4 octets of flags, 2 octets of name length,
                   2 octets of value length, M octets of name data,
                   N octets of value data)

        This subpacket describes a "notation" on the
relationship between signature that the issuing key
        issuer wishes to make.  The notation has a name and a value,
        each of which are strings of octets.  There may be more than
        one notation in a signature.  Notations can be used for any
        extension the key that is signed.  If issuer of the signature cares to make.  The
        "flags" field holds four octets of flags.

        All undefined flags MUST be zero.  Defined flags are:
        First octet: 0x80 = human-readable. This note is text, a note
                            from one person to another, and has no
                            meaning to software.
        Other octets: none.

    Key server preferences (N octets of flags)

        This is a valid self-signature (for which list of flags that indicate preferences that the issuer is key
        holder has about how the key that is being signed, either directly or via handled on a username binding),
then the key should not server.  All
        undefined flags MUST be used.  In other cases, the signature
containing zero.

       First octet: 0x80 = No-modify -- the critical subpacket should key holder requests that
                           this key only be ignored.

5.2.2.2 Signature Subpacket Types

Several types of subpackets are currently defined.  Some subpackets
apply to modified or updated by the signature itself and some are attributes
                           key holder or an authorized administrator of
                           the key.
Subpackets that are key server.
       This is found only on a self-signature are placed on self-signature.

    Preferred key server (String)

        This is a user name
certification made by URL of a key server that the key itself. holder prefers be
        used for updates.  Note that keys with multiple user names can
        have a preferred key may have more
than one server for each user name, and thus may have more than one self-signature, and
differing subpackets.

Implementing software should interpret name.  Note also that
        since this is a self-signature's preference
subpackets as narrowly as possible.  For example, suppose URL, the key server can actually be a copy of
        the key has two
usernames, Alice and Bob.  Suppose retrieved by ftp, http, finger, etc.

    Primary user id (1 octet, boolean)

        This is a flag in a user id's self signature that Alice prefers states
        whether this user id is the symmetric
algorithm CAST5, and Bob prefers IDEA or Triple-DES.  If main user id for this key.  It is
        reasonable for an implementation to resolve ambiguities in
        preferences, etc. by referring to the software
locates primary user id.  If this key via Alice's name, then the preferred algorithm
        flag is
CAST5, if software locates the key via Bob's name, then the preferred
algorithm absent, its value is IDEA. zero.  If the more than one user id in
        a key is located by key id, then algorithm of marked as primary, the default user name implementation may resolve the
        ambiguity in any way it sees fit.

    Policy URL (String)

        This subpacket contains a URL of a document that describes the key provides
        policy under which the default symmetric
algorithm.

The descriptions below describe whether a signature was issued.

    Key Flags (Octet string)

        This subpacket contains a list of binary flags that hold
        information about a key.  It is typically found
in the hashed or unhashed subpacket sections.  If a subpacket string of octets, and an
        implementation MUST NOT assume a fixed size.  This is not
hashed, then so it cannot can
        grow over time.  If a list is shorter than an implementation
        expects, the unstated flags are considered to be trusted.

    Signature creation time (4 octet time field) (Hashed) zero.  The time the signature was made.  Always included with new signatures.

    Issuer (8 octet
        defined flags are:

        First octet:
        0x01 - This key ID) (Non-hashed)

The OP may be used to certify other keys.
        0x02 - This key ID may be used to sign data.
        0x04 - This key may be used to encrypt communications.
        0x08 - This key may be used to encrypt storage.
        0x10 - The private component of the this key issuing the signature.

    Key expiration time (4 octet time field) (Hashed) may have been split by
               a secret-sharing mechanism.
        0x80 - The validity period private component of this key may be in the key.  This posession
               of more than one person.

        Usage notes:

        The flags in this packet may appear in self-signatures or in
        certification signatures.  They mean different things depending
        on who is making the number of seconds after
the key creation time statement -- for example, a certification
        signature that the key expires.  If this is not present or has a value of zero, the key never expires. This "sign data" flag is found only on a
self-signature.

    Preferred symmetric algorithms (array of one-octet values) (Hashed)

Symmetric algorithm numbers stating that indicate which algorithms the key
holder prefers to use.  This
        certification is an ordered list of octets with the most
preferred listed first.  It should be assumed for that only algorithms
listed are supported by use.  On the recipient's software.  Algorithm numbers other hand, the
        "communications encryption" flag in
section 6. This a self-signature is only found on stating
        a self-signature.

    Preferred hash algorithms (array of one-octet values) (Hashed)

Message digest algorithm numbers preference that indicate which algorithms the a given key
holder prefers be used for communications.  Note
        however, that it is a thorny issue to receive.  Like the preferred symmetric algorithms,
the list determine what is ordered. Algorithm numbers are in section 6.
        "communications" and what is "storage." This decision is only
found left
        wholly up to the implementation; the authors of this document
        do not claim any special wisdom on a self-signature.

{{Editor's note: the issue, and realize that
        accepted opinion may change.

        The above preference (hash algs) is controversial.  I
included it in for symmetry, because if someone wants to build "split key" (0x10) and "group key" (0x80) flags are placed
        on a
minimal OP implementation, there needs to be self-signature only; they are meaningless on a way to tell someone that
you won't
        certification signature.  They SHOULD be able to verify placed only on a
        direct-key signature unless it's made with some set
of algorithms.  It also permits to prefer DSA with RIPEMD-160, for
example. If you have an opinion, please state it.}}

    Preferred compression algorithms (array of one-octet values)
        (Hashed)

Compression algorithm numbers (type 0x1f) or a subkey signature (type
        0x18), one that indicate which algorithms the key
holder prefers refers to use.  Like the preferred symmetric algorithms, key the
list is ordered.  Algorithm numbers are in section 6.  If this flag applies to.

    Signer's User ID

        This subpacket allows a keyholder to state which user id is not included, ZIP is preferred. A zero denotes that no
compression is preferred;
        responsible for the signing.  Many keyholders use a single key holder's software may not have
compression software.
        for different purposes, such as business communications as well
        as personal communications.  This subpacket allows such a
        keyholder to state which of their roles is only found on making a self-signature. signature.

Implementations SHOULD implement "preferences".

5.2.3 Signature expiration time (4 octet time field) (Hashed)

The validity period Types

There are a number of the possible meanings for a signature, which are
specified in a signature type octet in any given signature.  This is  These
meanings are:

    - 0x00: Signature of a binary document.

Typically, this means the number signer owns it, created it, or certifies that
it has not been modified.

    - 0x01: Signature of seconds
after a canonical text document.

Typically, this means the signature creation time signer owns it, created it, or certifies that
it has not been modified.  The signature will be calculated over the
text data with its line endings converted to <CR><LF>.

    - 0x02: Standalone signature.

This signature expires.  If this is not present or has a value signature of zero, only its own subpacket contents.  It
is calculated identically to a signature over a zero-length binary
document.  Note that it never expires.

    Exportable (1 octet doesn't make sense to have a V3 standalone
signature.

    - 0x10: The certification of exportability, 0 for not, 1 for exportable)
               (Hashed)

Signature's exportability status.  Packet body contains a boolean flag
indicating whether the signature is exportable. Signatures which are
not exportable are ignored during export User ID and import operations.  If Public Key packet.

The issuer of this packet is certification does not present the signature is assumed make any particular assertion
as to be exportable.

    Revocable (1 octet how well the certifier has checked that the owner of revocability, 0 for not, 1 for revocable)
        (Hashed)

Signature's revocability status.  Packet body contains a boolean flag
indicating whether the signature key is revocable.  Signatures which
in fact the person described by the user ID.  Note that all PGP "key
signatures" are this type of certification.

    - 0x11: This is a persona certification of a User ID and
            Public Key packet.

The issuer of this certification has not revocable get done any later revocation signatures ignored.  They
represent a commitment by verification of the signer
claim that he cannot revoke his
signature for the life owner of his key.  If this packet key is not present the
signature user ID specified.

    - 0x12: This is assumed to be revocable.

    Trust signature (1 octet the casual certification of "level" (depth), 1 octet a User ID and
            Public Key packet.

The issuer of this certification has done some casual verification of trust amount)
     (Hashed)
Signer asserts that
the key claim of identity.

    - 0x13: This is not only valid, but also trustworthy, at the specified level.  Level 0 positive certification of a User ID and
            Public Key packet.

The issuer of this certification has done substantial verification of
the same meaning as an ordinary
validity signature.  Level 1 means claim of identity.

Please note that the signed key vagueness of these certification claims is asserted to
be not a
flaw, but a valid trusted introducer, with the 2nd octet feature of the body
specifying system.  Because PGP places final authority
for validity upon the degree receiver of trust. Level 2 means a certification, it may be that the signed key is
asserted to one
authority's casual certification might be trusted more rigorous than some other
authority's positive certification.  These classifications allow a
certification authority to issue level 1 trust signatures, i.e. that it fine-grained claims.

    - 0x18: This is used for a "meta introducer".  Generally, a level n trust signature asserts
that by a signature key is trusted to issue level n-1 trust signatures.  The trust
amount is in bind a range from 0-255, interpreted such that values less than
120 indicate partial trust and values of 120 or greater indicate
complete trust.  Implementations SHOULD emit values of 60 for partial
trust and 120
      subkey which will be used for complete trust.

    Regular expression (null-terminated regular expression) (Hashed)

Used in conjunction with trust encryption.

The signature packets (of level > 0) to
limit the scope of trust which is extended.  Only signatures by calculated directly on the
target subkey itself, not on any
User ID or other packets.

    - 0x1f: Signature directly on a key

This signature is calculated directly on user IDs which match a key.  It binds the regular expression
information in the body
of this packet have trust extended by signature subpackets to the trust packet.

    Additional recipient request (1 octet of class, 1 octet of algid,
                                  20 octets of fingerprint) (Hashed)

Key holder requests encryption key, and is appropriate
to additional recipient when data be used for subpackets which provide information about the key, such
as the revocation key subpacket.  It is
encrypted also appropriate for statements
that non-self certifiers want to this username.  If the class octet contains 0x80, then make about the key holder strongly requests that itself, rather than
the additional recipient be added to
an encryption.  Implementing software may treat this subpacket in any
way it sees fit. binding between a key and a name.

    - 0x20: This signature is found only on used to revoke a self-signature.

    Revocation key (1 octet of class, 1 octet of algid, 20 octets of
	              fingerprint) (Hashed)

Authorizes key.

The signature is calculated directly on the specified key being revoked.  A
revoked key is not to issue be used.  Only revocation self-signatures on
this key.  Class octet must have bit 0x80 set, other bits are for

future expansion to other kinds of signature authorizations. signatures by the key
being revoked, or by an authorized revocation key, should be
considered.

    - 0x28: This is
found on a self-signature.

    Notation Data (4 octets of flags, 2 octets of name length,
                   2 octets of value length, M octets of name data,
                   N octets of value data) (Hashed)

This subpacket describes used to revoke a "notation" on the subkey.

The signature that is calculated directly on the issuer
wishes subkey being revoked.  A
revoked subkey is not to make.  The notation has a name and a value, each of be used.  Only revocation signatures by the
top-level signature key which are
strings of octets.  There may is bound to this subkey, or by an
authorized revocation key, should be more than one notation in considered.

    - 0x30: This signature revokes an earlier user ID certification
      signature (signature class 0x10 through 0x13).

It should be issued by the same key which issued the revoked signature,
and should have a later creation date than the signature it revokes.

    - 0x40: Timestamp signature.
Notations can be used

This signature is only meaningful for any extension the issuer of timestamp contained in it.

5.2.4 Computing Signatures

All signatures are formed by producing a hash over the signature
cares to make. data,
and then using the resulting hash in the signature algorithm.

The "flags" field holds four octets of flags. All
undefined flags MUST be zero.  Defined flags are:
        First octet: 0x80 = human-readable. This note signature data is text, a note
                            from one person simple to another, compute for document signatures (types
0x00 and has no
                            meaning to software.
        Other octets: none.

    Key server preferences (N octets of flags) (Hashed)

This 0x01), for which the document itself is the data.  For
standalone signatures, this is a list of flags that indicate preferences that null string.

When a signature is made over a key, the key holder
has about how hash data starts with the key is handled on
octet 0x99, followed by a key server.  All undefined flags
MUST be zero.

       First octet: 0x80 = No-modify -- two-octet length of the key, and then body of
the key holder requests packet. (Note that this is an old-style packet header for a key
packet with two-octet length.) A subkey signature (type 0x18) then
hashes the subkey, using the same format as the main key.  Key
revocation signatures (types 0x20 and 0x28) hash only be modified or updated by the key holder or being
revoked.

A certification signature (type 0x10 through 0x13) then hashes the user
name being bound to the key.  A V3 certification hashes the contents of
the name packet, without any header.  A V4 certification hashes the
constant 0xd4 (which is an authorized administrator old-style CTB with the length-of-length set
to zero), a four-octet number giving the length of the key server.
This username, and
then the username data.

Once the data body is found only on hashed, then a self-signature.

    Preferred key server (String) (Hashed) trailer is hashed.  A V3 signature
hashes five octets of the packet body, starting from the signature type
field.  This data is a URL the signature type, followed by the four-octet
signature time.  A V4 signature hashes the packet body starting from
its first field, the version number, through the end of a key server that the hashed

subpacket data.  Thus, the fields hashed are the signature version, the
signature type, the public key holder prefers be used for
updates.  Note that keys with multiple user names can have a preferred
key server for each user name. This algorithm, the hash algorithm, the
hashed subpacket length, and the hashed subpacket body.

After all this has been hashed, the resulting hash field is found only on a self-signature.

Implementations SHOULD implement a "preference" used in the
signature algorithm, and MAY implement a
"request."

{{Editor's note:  None placed at the end of the preferences have a way to specify a
negative preference (for example, I like Triple-DES, don't use
algorithm X).  Tacitly, signature packet.

5.3 Symmetric-Key Encrypted Session-Key Packets (Tag 3)

The Symmetric-Key Encrypted Session Key packet holds the absence
conventional-cipher encryption of an algorithm from a set is a
negative preference, but should there be an explicit way session key used to give encrypt a
negative preference? -jdcc}}

{{Editor's note:  A missing feature is to invalidate (or revoke)
message.  Zero or more Encrypted Session Key packets and/or
Conventional Encrypted Session Key packets may precede a user
id, rather than the entire key.  Lots of people want this, and many
people have keys cluttered with old work email addresses.  There is
another related issue, that Symmetrically
Encrypted Data Packet that holds an encrypted message.  The message is
encrypted with key rollover -- suppose I'm
retiring an old key, but I don't want to have to lose all my
certification signatures.  It would be nice if there were a way for a session key, and the session key to transfer is itself to a new one.  Lastly, if either (or both) of
these encrypted
and stored in the Encrypted Session Key packet or the Conventional
Encrypted Session Key packet.

If the Symmetrically Encrypted Data Packet is desirable, do we handle them with a new signature type, preceded by one or

with notations, more
Symmetric-Key Encrypted Session Key packets, each specifies a
passphrase which are an extension mechanism.  I think that it
makes sense may be used to make decrypt the message.  This allows a revocation type (because it's analogous
message to the
other forms of revocation), but rollover might be best implemented as
an extension. --jdcc}}

{{Editor's note:  PGP 3 designed, but never implemented encrypted to a number of
other subpacket types.  They were: public keys, and also to one or
more pass phrases.  This packet type is new, and is not generated by
PGP 2.x or PGP 5.0.

The body of this packet consists of:
    - A signature one-octet version number; number. The only currently defined version is
      4.
    - A set
of one-octet number describing the symmetric algorithm used.
    - A string-to-key (S2K) specifier, length as defined above.
    - Optionally, the encrypted session key usage flags (signing key, encryption itself, which is decrypted
          with the string-to-key object.

If the encrypted session key for communication, is not present (which can be detected on
the basis of packet length and
encryption S2K specifier size), then the S2K
algorithm applied to the passphrase produces the session key for storage); User ID of the signer; Policy URL; net
location of
decrypting the key.

Some of these options are things file, using the WG has talked about as being a
Good Thing -- like flags denoting if a symmetric cipher algorithm from the
Symmetric-Key Encrypted Session Key packet.

If the encrypted session key is a comm key or a storage
key.  My design present, the result of such a feature would be different than applying the other
one, though. I think it would be a great idea to have a URL that's a
location S2K
algorithm to find the key, so people who prefer passphrase is used to have a web, ftp, or
finger location can use those.  However, some of them (like a URL) are
also perfect for designing in decrypt just that encrypted
session key field, using CFB mode with extensions.  After all, we only have
128 subpacket constants.

--jdcc}}

5.2.3 Signature Types

There are a number of possible meanings for a signature, which are
specified in a signature type octet in any given signature.  These
meanings are:

	- 0x00: Signature an IV of a binary document.
Typically, this means the signer owns it, created it, or certifies that
it has not been modified.

	- 0x01: Signature all zeros.  The
decryption result consists of a canonical text document.
Typically, this means the signer owns it, created it, or certifies one-octet algorithm identifier that
it has not been modified.  The signature will be calculated over
specifies the
textual data with its line endings converted conventional encryption algorithm used to <CR><LF>.

    - 0x02: Standalone signature.
This signature is a signature of only its own subpacket contents.  It encrypt the
following Symmetrically Encrypted Data Packet, followed by the session
key octets themselves.

Note: because an all-zero IV is calculated identically to used for this decryption, the S2K
specifier MUST use a signature over salt value, either a zero-length binary
document.

	- 0x10: The generic certification of a User ID and Public Key
	  packet. Salted S2K or an
Iterated-Salted S2K.  The issuer of this certification does salt value will insure that the decryption
key is not make repeated even if the passphrase is reused.

5.4 One-Pass Signature Packets (Tag 4)

The One-Pass Signature packet precedes the signed data and contains
enough information to allow the receiver to begin calculating any particular assertion
as
hashes needed to how well verify the certifier has checked that signature.  It allows the owner Signature Packet
to be placed at the end of the key is
in fact message, so that the person described by signer can compute
the user ID.  Note that all entire signed message in one pass.

A One-Pass Signature does not interoperate with PGP "key
signatures" are this type 2.6.x or earlier.

The body of certification. this packet consists of:
    - 0x11: This A one-octet version number. The current version is a persona certification 3.
    - A one-octet signature type. Signature types are described
          in section 5.2.3.
    - A one-octet number describing the hash algorithm used.
    - A one-octet number describing the public key algorithm used.
    - An eight-octet number holding the key ID of the signing key.
    - A one-octet number holding a User ID and
	  Public Key packet.
It means flag showing whether the signature
is nested.  A zero value indicates that the issuer of this certification has not done any
verification of next packet is
another One-Pass Signature packet which describes another
signature to be applied to the claim that same message data.

5.5 Key Material Packet

A key material packet contains all the owner information about a public or
private key.  There are four variants of this key is the user ID

specified.  Note that no released version of PGP has generated packet type, and two
major versions.  Consequently, this
type of certification.

	- 0x12: This section is the casual certification of a User ID and complex.

5.5.1 Key Packet Variants

5.5.1.1 Public Key packet.
It means that the issuer of this certification has done some casual
verification of the claim Packet (Tag 6)

A Public Key packet starts a series of identity.  Note packets that no version of PGP forms an OP key
(sometimes called an OP certificate).

5.5.1.2 Public Subkey Packet (Tag 14)

A Public Subkey packet (tag 14) has
generated this type of certification, nor is there any definition of
what constitutes a casual certification.

	- 0x13: This is exactly the positive certification of same format as a User ID and Public
Key packet.
It means that packet, but denotes a subkey.  One or more subkeys may be
associated with a top-level key.  By convention, the issuer of this certification has done substantial
verification of top-level key
provides signature services, and the claim of identity.  Note that subkeys provide encryption
services.

Note: in PGP 2.6.X, tag 14 was intended to indicate a comment packet.
This tag was selected for reuse because no previous version of PGP has
generated this type of certification, nor is there any definition of
what constitutes ever
emitted comment packets but they did properly ignore them.  Public
Subkey packets are ignored by PGP 2.6.X and do not cause it to fail,
providing a positive certification.  Please also note that limited degree of backwards compatibility.

5.5.1.3 Secret Key Packet (Tag 5)

A Secret Key packet contains all the
vagueness of these certification systems information that is not found in a flaw,
Public Key packet, including the public key material, but a feature
of also includes
the system.  Because PGP places final authority for validity upon secret key material after all the receiver of a certification, it may be that one authority's casual
certification might be more rigorous than some other authority's
positive certification.

{{Editor's note:  While there public key fields.

5.5.1.4 Secret Subkey Packet (Tag 7)

A Secret Subkey packet (tag 7) is a scale of identification signatures
in the range 0x10 to 0x13, most subkey analog of them have never been implemented or
used.  Current implementations only use 0x10, the "generic
certification." Should Secret Key
packet, and has exactly the others be removed?  RFC 1991 went to some
trouble to explain which ones same format.

5.5.2 Public Key Packet Formats

There are two versions of key-material packets.  Version 3 packets were defined
first generated PGP 2.6.  Version 2 packets are identical in format to
Version 3 packets, but not implemented, are generated by PGP 2.5 or read
but not generated.  I think we should before.  PGP 5.0
introduces version 4 packets, with new fields and semantics.  PGP 2.6.X
will not do that.  If we define them,
they should be accept key-material packets with versions greater than 3.

OP implementations SHOULD create keys with version 4 format.  An
implementation MAY features at generate a V3 key to ensure interoperability with
old software; note, however, that V4 keys correct some security
deficiencies in V3 keys.  These deficiencies are described below.  An
implementation MUST NOT create a V3 key with a public key algorithm
other than RSA.

A version 3 public key or public subkey packet contains:
    - A one-octet version number (3).
    - A four-octet number denoting the very least.  If we're not going to
use them, they shouldn't be in time that the spec. --jdcc}} key was created.
    - 0x18: This is used for a signature by a signature A two-octet number denoting the time in days that this key to bind a
	  subkey which will be used for encryption.
The signature is calculated directly on the subkey itself,
      valid. If this number is zero, then it does not on any
User ID or other packets. expire.
    - 0x20: This signature is used to revoke a key.
The signature is calculated directly on A one-octet number denoting the public key being revoked.  A
revoked algorithm of this key is not to be used.  Only revocation signatures by
    - A series of multi-precision integers comprising the key
being revoked, or by an authorized revocation key, should be
considered.
      material:
    - 0x28: This is used to revoke a subkey. multiprecision integer (MPI) of RSA public modulus n;
    - an MPI of RSA public encryption exponent e.

The signature is calculated directly on fingerprint of the subkey being revoked.  A
revoked subkey key is not to be used.  Only revocation signatures formed by hashing the body (but not the
two-octet length) of the MPIs that form the
top-level signature key which is bound to this subkey, or material (public
modulus n, followed by exponent e) with MD5.

The eight-octet key ID of the key consists of the low 64 bits of the
public modulus of an
authorized revocation key, should be considered.

	- 0x30: This signature revokes an earlier user RSA key.

Since the release of V3 keys, there have been a number of improvements
desired in the key format.  For example, if the key ID certification
	  signature (signature class 0x10 - 0x13).
It should be issued by is a function of
the public modulus, it is easy for a person to create a key that has
the same key which issued ID as some existing key.  Similarly, MD5 is no longer the revoked signature,
preferred hash algorithm, and should have not hashing the length of an MPI with its
body increases the chances of a later creation date.

    - 0x40: Timestamp signature.

{{Editor's note: fingerprint collision.

The timestamp signature version 4 format is left over from RFC 1991,
and has never been fully designed nor implemented.  Is this similar to the sort version 3 format except for the
absence of
thing best handled by notations? --jdcc}}

{{Editor's note:  It would be nice to have a validity period.  This has been moved to the signature
packet.  In addition, fingerprints of version 4 keys are calculated
differently from version 3 keys, as described in section "Enhanced Key
Formats."

A version 4 packet contains:
    - A one-octet version number (4).
    - A four-octet number denoting the time that applied to the key alone, rather than a was created.
    - A one-octet number denoting the public key plus a user name.  Perhaps algorithm of this key
    - A series of multi-precision integers comprising the key
      material.  This algorithm-specific portion is:

    Algorithm Specific Fields for RSA public keys:
    - multiprecision integer (MPI) of RSA public modulus n;
    - MPI of RSA public encryption exponent e.

    Algorithm Specific Fields for DSA public keys:
    - MPI of DSA prime p;
    - MPI of DSA group order q (q is
best done with a notation. --jdcc}}

{{Editor's note:  There prime divisor of p-1);
    - MPI of DSA group generator g;
    - MPI of DSA public key value y (= g**x where x is presently no way secret).

    Algorithm Specific Fields for a key-signer (a.k.a.
certifier) to sign a main key along with a subkey.  There are a number Elgamal public keys:
    - MPI of useful situations for a set Elgamal prime p;
    - MPI of keys (main plus subkeys) to all be
signed together.  How do we solve this? --jdcc}}

5.3 Conventional Encrypted Session-Key Packets (Tag 3)

The Conventional Encrypted Session Key packet holds the
conventional-cipher encryption Elgamal group generator g;
    - MPI of a session Elgamal public key used to encrypt a
message.  Zero or more Encrypted Session Key packets and/or
Conventional Encrypted Session value y (= g**x where x
      is secret).

5.5.3 Secret Key packets may precede a Symmetrically
Encrypted Data Packet that holds an encrypted message. Formats

The message is
encrypted with a session key, and the session key is itself encrypted
and stored in the Encrypted Session Secret Key packet or and Secret Subkey packets contain all the Conventional
Encrypted Session Key packet.

If data of the Symmetrically Encrypted Data Packet is preceded by one or more
Conventional Encrypted Session
Public Key and Public Subkey packets, each specifies a passphrase
which may be used to decrypt the message.  This allows a message to be with additional
algorithm-specific secret key data appended, in encrypted to a number of public keys, and also to one or more pass
phrases.  This packet type is new, and is not generated by PGP 2.x or
PGP 5.0. form.

The body of this packet consists of: contains:
    - A one-octet version number. The only currently defined version Public Key or Public Subkey packet, as described above
    - One octet indicating string-to-key usage conventions.  0 indicates
      that the secret key data is
	  4. not encrypted.  255 indicates that a
      string-to-key specifier is being given.  Any other value
      is a conventional encryption algorithm specifier.
    - A [Optional] If string-to-key usage octet was 255, a one-octet number describing the symmetric algorithm used.
      conventional encryption algorithm.
    - A [Optional] If string-to-key (S2K) specifier, usage octet was 255, a string-to-key
      specifier.  The length of the string-to-key specifier is implied
      by its type, as defined described above.
    - Optionally, [Optional] If secret data is encrypted, eight-octet Initial Vector
      (IV).
    - Encrypted multi-precision integers comprising the encrypted session secret key itself, which is decrypted
          with data.
      These algorithm-specific fields are as described below.

    - Two-octet checksum of the string-to-key object.

If plaintext of the encrypted session key is not present (which can be detected on algorithm-specific
      portion (sum of all octets, mod 65536).

    Algorithm Specific Fields for RSA secret keys:
    - multiprecision integer (MPI) of RSA secret exponent d.
    - MPI of RSA secret prime value p.
    - MPI of RSA secret prime value q (p < q).
    - MPI of u, the basis multiplicative inverse of packet length and S2K p, mod q.

    Algorithm Specific Fields for DSA secret keys:
    - MPI of DSA secret exponent x.

    Algorithm Specific Fields for Elgamal secret keys:
    - MPI of Elgamal secret exponent x.

Secret MPI values can be encrypted using a passphrase.  If a
string-to-key specifier size), then is given, that describes the S2K algorithm applied for
converting the passphrase to a key, else a simple MD5 hash of the
passphrase is used.  Implementations SHOULD use a string-to-key
specifier; the simple hash is for backwards compatibility.  The cipher
for encrypting the passphrase produces MPIs is specified in the session secret key for
decrypting packet.

Encryption/decryption of the file, secret data is done in CFB mode using the symmetric cipher algorithm
key created from the
Conventional Encrypted Session Key packet.

If passphrase and the encrypted session key Initial Vector from the packet.
A different mode is present, used with RSA keys than with other key formats.
With RSA keys, the result of applying MPI bit count prefix (i.e., the S2K
algorithm to first two octets) is
not encrypted.  Only the passphrase MPI non-prefix data is used to decrypt just encrypted.
Furthermore, the CFB state is resynchronized at the beginning of each
new MPI value, so that encrypted
session key field, using the CFB mode block boundary is aligned with an IV of all zeros.  The
decryption result consists the start
of the MPI data.

With non-RSA keys, a one-octet algorithm identifier simpler method is used.  All secret MPI values are
encrypted in CFB mode, including the MPI bitcount prefix.

The 16-bit checksum that
specifies follows the conventional encryption algorithm used to encrypt algorithm-specific portion is the
following Symmetrically Encrypted Data Packet, followed by
algebraic sum, mod 65536, of the session

key plaintext of all the
algorithm-specific octets themselves.

Note: because an all-zero IV (including MPI prefix and data).  With RSA
keys, the checksum is used for this decryption, stored in the S2K
specifier MUST use a salt value, either a a Salted S2K or an
Iterated-Salted S2K.  The salt clear.  With non-RSA keys, the
checksum is encrypted like the algorithm-specific data.  This value will insure is
used to check that the decryption
key passphrase was correct.

5.6 Compressed Data Packet (Tag 8)

The Compressed Data packet contains compressed data.  Typically, this
packet is not repeated even if found as the passphrase is reused.

5.4 One-Pass contents of an encrypted packet, or following a
Signature Packets (Tag 4)

The or One-Pass Signature packet precedes the signed data packet, and contains
enough information to allow the receiver to begin calculating any
hashes needed to verify the signature.  It allows literal data
packets.

The body of this packet consists of:
    - One octet that gives the Signature Packet algorithm used to be placed at compress the end packet.
    - The remainder of the message, so that the signer can compute
the entire signed message in one pass. packet is compressed data.

A One-Pass Signature does not interoperate Compressed Data Packet's body contains an RFC1951 DEFLATE block that
compresses some set of packets.  See section "Packet Composition" for
details on how messages are formed.

5.7 Symmetrically Encrypted Data Packet (Tag 9)

The Symmetrically Encrypted Data packet contains data encrypted with PGP 2.6.x a
conventional (symmetric-key) algorithm.  When it has been decrypted, it
will typically contain other packets (often literal data packets or earlier.
compressed data packets).

The body of this packet consists of:

    - A one-octet version number. Encrypted data, the output of the selected conventional cipher
          operating in PGP's variant of Cipher Feedback (CFB) mode.

The current version is 3.
	- A one-octet signature type. Signature types are described conventional cipher used may be specified in section 5.2.3.
	- A one-octet number describing an Encrypted Session
Key or Conventional Encrypted Session Key packet which precedes the hash algorithm used.
	- A one-octet number describing
Symmetrically Encrypted Data Packet.  In that case, the public key algorithm used.
	- An eight-octet number holding cipher
algorithm octet is prepended to the session key ID before it is encrypted.
If no packets of these types precede the signing key.
	- A one-octet number holding a flag showing whether the signature
is nested.  A zero value indicates that encrypted data, the next packet IDEA
algorithm is
another One-Pass Signature packet which describes another
signature to be applied to used with the same message data.

5.5 Key Material Packet

A session key material packet contains all calculated as the information about a public or
private key.  There are four variants MD5 hash of this packet type, and two
major versions.  Consequently, this section
the passphrase.

The data is complex.

5.5.1 Key Packet Variants

5.5.1.1 Public Key Packet (Tag 6)

A Public Key packet starts encrypted in CFB mode, with a series CFB shift size equal to the
cipher's block size.  The Initial Vector (IV) is specified as all
zeros.  Instead of packets that forms an OP key
(sometimes called using an IV, OP certificate).

5.5.1.2 Public Subkey Packet (Tag 14)

A Public Subkey packet (tag 14) has exactly the same format as a Public
Key packet, but denotes a subkey.  One or more subkeys may be
associated with prefixes a top-level key.  By convention, 10 octet string to the top-level key
provides signature services,
data before it is encrypted.  The first eight octets are random, and
the subkeys provide encryption
services.

Note: in PGP 2.6.X, tag 14 was intended to indicate a comment packet.

This tag was selected for reuse because no previous version 9th and 10th octets are copies of the 7th and 8th octets,
respectivelly. After encrypting the first 10 octets, the CFB state is
resynchronized if the cipher block size is 8 octets or less.  The last
8 octets of PGP ever
emitted comment packets but they did properly ignore them.  Public
Subkey packets ciphertext are ignored by PGP 2.6.X passed through the cipher and do not cause it to fail,
providing a limited degree of backwards compatibility.

5.5.1.3 Secret Key Packet (Tag 5)

A Secret Key packet contains all the information that block
boundary is found reset.

The repetition of 16 bits in a
Public Key packet, including the public key material, but also includes 80 bits of random data prepended to
the secret key material after all message allows the public receiver to immediately check whether the
session key fields.

5.5.1.4 Secret Subkey is correct.

5.8 Marker Packet (Obsolete Literal Packet) (Tag 7)

A Secret Subkey packet (tag 7) is the subkey analog 10)

An experimental version of PGP used this packet as the Secret Key Literal packet, and has exactly the same format.

5.5.2 Public Key Packet Formats

There are two versions
but no released version of key-material packets.  Version 3 packets were
first generated PGP 2.6.  Version 2 packets are identical in format to
Version 3 packets, but are generated by PGP 2.5 or before.  PGP 5.0
introduces version 4 packets, Literal packets with new fields this tag.
With PGP 5.x, this packet has been re-assigned and semantics. is reserved for use
as the Marker packet.

The body of this packet consists of:
    - The three octets 0x60, 0x47, 0x60 (which spell "PGP" in UTF-8).

Such a packet MUST be ignored when received.  It may be placed at the
beginning of a message that uses features not available in PGP 2.6.X
will not accept key-material packets with versions greater than 3.

OP implementations SHOULD create keys with version 4 format.  An
implementation MAY generate a V3 key in
order to ensure interoperability with
old software; note, however, cause that V4 keys correct some security
deficiencies in V3 keys.  These deficiencies are described below.  An
implementation MUST NOT create a V3 key with a public key algorithm
other than RSA.

A version 3 public key or public subkey packet contains:
    - A one-octet version number (3).
    - A four-octet number denoting the time to report that newer software necessary to
process the key was created.
    - message.

5.9 Literal Data Packet (Tag 11)

A two-octet number denoting Literal Data packet contains the time in days body of a message; data that this key is
      valid. If this number is zero, then it does not expire.
    - A one-octet number denoting the public key algorithm
to be further interpreted.

The body of this key packet consists of:
    - A series of multi-precision integers comprising one-octet field that describes how the key
      material:
    - data is formatted.

If it is a multiprecision integer (MPI) of RSA public modulus n;
    - an MPI of RSA public encryption exponent e.

The fingerprint of 'b' (0x62), then the key literal packet contains binary data. If
it is formed a 't' (0x74), then it contains text data, and thus may need line
ends converted to local form, or other text-mode changes.  RFC 1991
also defined a value of 'l' as a 'local' mode for machine-local
conversions.  This use is now deprecated.

    - File name as a string (one-octet length, followed by hashing file name),
      if the body (but not encrypted data should be saved as a file.

If the
two-octet length) of special name "_CONSOLE" is used, the MPIs message is considered to be
"for your eyes only".  This advises that form the key material (public
modulus n, followed by exponent e) with MD5.

The eight-octet key ID of message data is unusually
sensitive, and the key consists of receiving program should process it more carefully,
perhaps avoiding storing the low 64 bits of received data to disk, for example.

    - A four-octet number that indicates the
public modulus modification date of an RSA key.

Since the release of V3 keys, there have been a number of improvements
desired in
file, or the key format.  For example, if creation time of the key ID is packet, or a function zero that indicates the
present time.

    - The remainder of the public modulus, it packet is easy for a person literal data.

Text data is stored with <CR><LF> text endings (i.e. network-normal
line endings).  These should be converted to create a key that has native line endings by the same key ID as some existing key.  Similarly, MD5
receiving software.

5.10 Trust Packet (Tag 12)

The Trust packet is no longer the

preferred hash algorithm, used only within keyrings and is not hashing normally
exported.  Trust packets contain data that record the length user's
specifications of an MPI which key holders are trustworthy introducers, along
with its
body increases the chances other information that implementing software uses for trust
information.

Trust packets SHOULD NOT be emitted to output streams that are
transferred to other users, and they SHOULD be ignored on any input
other than local keyring files.

5.11 User ID Packet (Tag 13)

A User ID packet consists of a fingerprint collision.

The version 4 format data which is similar intended to represent the version 3 format except for the
absence
name and email address of a validity period.  This has been moved to the signature
packet.  In addition, fingerprints of version 4 keys key holder.  By convention, it includes
an RFC822 mail name, but there are calculated
differently from version 3 keys, as described elsewhere.

A version 4 no restrictions on its content.  The
packet contains:
    - A one-octet version number (4).
    - A four-octet number denoting the time that length in the key was created.
    - A one-octet number denoting header specifies the public key algorithm of this key
    - A series length of multi-precision integers comprising the key
      material.  This algorithm-specific portion is:

    Algorithm Specific Fields for RSA public keys:
    - multiprecision integer (MPI) of RSA public modulus n;
    - MPI of RSA public encryption exponent e.

    Algorithm Specific Fields for DSA public keys:
    - MPI of DSA prime p;
    - MPI of DSA group order q (q user name.  If
it is a prime divisor of p-1);
    - MPI of DSA group generator g;
    - MPI of DSA public key value y (= g**x where x text, it is secret).

    Algorithm Specific Fields encoded in UTF-8.

6.  Radix-64 Conversions

As stated in the introduction, OP's underlying native representation
for Elgamal public keys:
    - MPI objects is a stream of Elgamal prime p;
    - MPI arbitrary octets, and some systems desire
these objects to be immune to damage caused by character set
translation, data conversions, etc.

In principle, any printable encoding scheme that met the requirements
of Elgamal group generator g;
    - MPI the unsafe channel would suffice, since it would not change the
underlying binary bit streams of Elgamal public key value y (= g**x where x
      is secret).

5.5.3 Secret Key Packet Formats

The Secret Key and Secret Subkey packets contain all the native OP data structures.  The OP
standard specifies one such printable encoding scheme to ensure
interoperability.

OP's Radix-64 encoding is composed of two parts: a base64 encoding of
the
Public Key binary data, and Public Subkey packets, with additional
algorithm-specific secret key data appended, in encrypted form. a checksum.  The packet contains:
    - A Public Key or Public Subkey packet, as described above
    - One octet indicating string-to-key usage conventions.  0 indicates
	  that base64 encoding is identical to
the secret key data MIME base64 content-transfer-encoding [RFC 2045, Section 6.8].  An
OP implementation MAY use ASCII Armor to protect the raw binary data.

The checksum is not encrypted.  255 indicates that a
	  string-to-key specifier 24-bit CRC converted to four characters of radix-64
encoding by the same MIME base64 transformation, preceded by an equals
sign (=).  The CRC is being given.  Any other value computed by using the generator 0x864CFB and an
initialization of 0xB704CE.  The accumulation is a conventional encryption done on the data
before it is converted to radix-64, rather than on the converted data.
A sample implementation of this algorithm specifier.
	- [Optional] If string-to-key usage octet was 255, a one-octet
	  conventional encryption algorithm.
	- [Optional] If string-to-key usage octet was 255, a string-to-key
	  specifier. is in the next section.

The checksum with its leading equal sign MAY appear on the first line
after the Base64 encoded data.

Rationale for CRC-24:  The length size of 24 bits fits evenly into printable
base64.  The nonzero initialization can detect more errors than a zero
initialization.

6.1 An Implementation of the string-to-key specifier is implied
	  by its type, as described above.
	- [Optional] If secret CRC-24 in "C"

#define CRC24_INIT 0xb704ce
#define CRC24_POLY 0x1864cfb

crc24 crc_bytes(unsigned char *bytes, size_t len)
{
    crc24 crc = CRC_INIT;
    int i;

    while (len--) {
        crc ^= *bytes++;
        for (i = 0; i < 8; i++) {
            crc <<= 1;
            if (crc & 0x1000000)
                crc ^= CRC24_POLY;
        }
    }
    return crc;
}

6.2 Forming ASCII Armor

When OP encodes data is encrypted, eight-octet Initial Vector
      (IV).
	- Encrypted multi-precision integers comprising into ASCII Armor, it puts specific headers around
the secret key data.
	  These algorithm-specific fields are as described below.

	- Two-octet checksum of data, so OP can reconstruct the plaintext data later.  OP informs the user
what kind of data is encoded in the algorithm-specific
	  portion (sum ASCII armor through the use of all octets, mod 65536).

    Algorithm Specific Fields for RSA secret keys: the
headers.

Concatenating the following data creates ASCII Armor:

        - multiprecision integer (MPI) An Armor Header Line, appropriate for the type of RSA secret exponent d. data
        - MPI of RSA secret prime value p. Armor Headers
        - MPI of RSA secret prime value q (p < q). A blank (zero-length, or containing only whitespace) line
        - MPI of u, the multiplicative inverse of p, mod q.

    Algorithm Specific Fields for DSA secret keys: The ASCII-Armored data
        - MPI of DSA secret exponent x.

    Algorithm Specific Fields for Elgamal secret keys: An Armor Checksum
        - MPI of Elgamal secret exponent x.

Secret MPI values can be encrypted using a passphrase.  If a
string-to-key specifier is given, that describes the algorithm for
converting The Armor Tail, which depends on the passphrase to a key, else a simple MD5 hash Armor Header Line.

An Armor Header Line consists of the
passphrase is used.  Implementations SHOULD use a string-to-key
specifier; appropriate header line text
surrounded by five (5) dashes ('-', 0x2D) on either side of the simple hash is for backwards compatibility. header
line text.  The cipher
for encrypting the MPIs header line text is specified in chosen based upon the secret key packet.

Encryption/decryption type of the secret data
that is done being encoded in CFB mode using the
key created from the passphrase Armor, and the Initial Vector from the packet.
A different mode how it is being encoded.  Header
line texts include the following strings:

    BEGIN PGP MESSAGE               used with RSA for signed, encrypted, or
                                    compressed files

    BEGIN PGP PUBLIC KEY BLOCK      used for armoring public keys than with other key formats.
With RSA keys, the MPI bit count prefix (i.e.,

    BEGIN PGP PRIVATE KEY BLOCK     used for armoring private keys

    BEGIN PGP MESSAGE, PART X/Y     used for multi-part messages, where
                                    the first two octets) armor is
not encrypted.  Only the MPI non-prefix data split amongst Y parts,
                                    and this is encrypted.
Furthermore, the CFB state Xth part out of Y.

    BEGIN PGP MESSAGE, PART X       used for multi-part messages, where
                                    this is resynchronized at the beginning Xth part of each
new MPI value, so an
                                    unspecified number of parts.
                                    Requires the MESSAGE-ID Armor
                                    Header to be used.

    BEGIN PGP SIGNATURE             used for detached signatures,
                                    OP/MIME signatures, and signatures
                                    following clearsigned messages

The Armor Headers are pairs of strings that can give the CFB user or the
receiving OP message block boundary is aligned with some information about how to decode or use
the start message.  The Armor Headers are a part of the MPI data.

With non-RSA keys, armor, not a simpler method is used.  All secret MPI values part of
the message, and hence are
encrypted in CFB mode, including not protected by any signatures applied to
the MPI bitcount prefix. message.

The 16-bit checksum that follows the algorithm-specific portion format of an Armor Header is that of a key-value pair.  A colon
(':' 0x38) and a single space (0x20) separate the
algebraic sum, mod 65536, key and value.  OP
should consider improperly formatted Armor Headers to be corruption of
the plaintext ASCII Armor.  Unknown keys should be reported to the user, but OP
should continue to process the message.

Currently defined Armor Header Keys are:

    - "Version", which states the OP Version used to encode the
    message.

    - "Comment", a user-defined comment.

    - "MessageID", a 32-character string of printable characters.  The
    string must be the same for all parts of a multi-part message that
    uses the
algorithm-specific octets (including MPI prefix and data).  With RSA
keys, "PART X" Armor Header.  MessageID strings should be unique
    enough that the checksum is stored in recipient of the clear.  With non-RSA keys, mail can associate all the parts
    of a message with each other.  A good checksum or cryptographic
    hash function is encrypted like the algorithm-specific data.  This value sufficent.

    The MessageID should not appear unless it is
used to check that in a multi-part
    message. If it appears at all, it MUST be computed from the passphrase was correct.

5.6 Compressed Data Packet (Tag 8)

The Compressed Data packet contains compressed data.  Typically, this
packet message
    in a deterministic fashion, rather than contain a purely random
    value.  This is found as to allow anyone to determine that the contents of an encrypted packet, or following MessageID
    cannot serve as a
Signature or One-Pass Signature packet, and contains literal data
packets.

The body covert means of this packet consists of:
	- One octet that gives leaking cryptographic key
    information.

The Armor Tail Line is composed in the algorithm used to compress same manner as the packet.
	- The remainder of Armor Header
Line, except the packet string "BEGIN" is compressed data.

A Compressed Data Packet's body contains an RFC1951 DEFLATE block that

compresses some set replaced by the string "END."

6.3 Encoding Binary in Radix-64

The encoding process represents 24-bit groups of packets.  See section 7 for details on how
messages input bits as output
strings of 4 encoded characters.  Proceeding from left to right, a
24-bit input group is formed by concatenating three 8-bit input groups.
These 24 bits are formed.

5.7 Symmetrically Encrypted Data Packet (Tag 9)

The Symmetrically Encrypted Data packet contains data encrypted with then treated as four concatenated 6-bit groups, each
of which is translated into a
conventional (symmetric-key) algorithm. single digit in the Radix-64 alphabet.
When it has been decrypted, it
will typically contain other packets (often literal data packets or
compressed data packets).

The body of this packet consists of:

	- Encrypted data, encoding a bit stream with the output of Radix-64 encoding, the selected conventional cipher
          operating in PGP's variant of Cipher Feedback (CFB) mode.

The conventional cipher used may bit stream
must be specified presumed to be ordered with the most-significant-bit first.
That is, the first bit in an Encrypted Session
Key or Conventional Encrypted Session Key packet which precedes the
Symmetrically Encrypted Data Packet.  In that case, stream will be the cipher
algorithm octet is prepended to high-order bit in the session key before it is encrypted.
If no packets of these types precede
first 8-bit byte, and the encrypted data, eighth bit will be the IDEA
algorithm low-order bit in the
first 8-bit byte, and so on.

         +--first octet--+-second octet--+--third octet--+
         |7 6 5 4 3 2 1 0|7 6 5 4 3 2 1 0|7 6 5 4 3 2 1 0|
         +-----------+---+-------+-------+---+-----------+
         |5 4 3 2 1 0|5 4 3 2 1 0|5 4 3 2 1 0|5 4 3 2 1 0|
         +--1.index--+--2.index--+--3.index--+--4.index--+

Each 6-bit group is used with the session key calculated as the MD5 hash an index into an array of 64 printable
characters from the passphrase. table below.  The character referenced by the index
is placed in the output string.

     Value Encoding  Value Encoding  Value Encoding  Value Encoding
         0 A            17 R            34 i            51 z
         1 B            18 S            35 j            52 0
         2 C            19 T            36 k            53 1
         3 D            20 U            37 l            54 2
         4 E            21 V            38 m            55 3
         5 F            22 W            39 n            56 4
         6 G            23 X            40 o            57 5
         7 H            24 Y            41 p            58 6
         8 I            25 Z            42 q            59 7
         9 J            26 a            43 r            60 8
        10 K            27 b            44 s            61 9
        11 L            28 c            45 t            62 +
        12 M            29 d            46 u            63 /
        13 N            30 e            47 v
        14 O            31 f            48 w         (pad) =
        15 P            32 g            49 x
        16 Q            33 h            50 y

The data is encrypted encoded output stream must be represented in CFB mode, with a CFB shift size equal to the
cipher's block size [Ref].  The Initial Vector (IV) is specified as all
zeros.  Instead lines of using an IV, OP prepends a 10 octet string to the
data before it no more than
76 characters each.

Special processing is encrypted.  The first eight octets performed if fewer than 24 bits are random, and available at
the 9th and 10th octets are copies end of the 7th and 8th octets,
respectivelly. After encrypting the first 10 octets, the CFB state is
resynchronized if the cipher block size is 8 octets or less. data being encoded.  There are three possibilities:

- The last
8 octets of ciphertext are passed through the cipher and the block
boundary data group has 24 bits (3 octets).  No special processing is reset.
needed.

- The repetition of last data group has 16 bits in the 80 bits of random data prepended to
the message allows the receiver to immediately check whether the
session key is correct.

5.8 Marker Packet (Obsolete Literal Packet) (Tag 10)

An experimental version of PGP used this packet (2 octets).  The first two 6-bit
groups are processed as the Literal packet,
but no released version of PGP generated Literal packets with this tag.
With PGP 5.x, this packet above.  The third (incomplete) data group has been re-assigned
two zero-value bits added to it, and is reserved for use processed as above.  A pad
character (=) is added to the Marker packet.

The body of this packet consists of: output.

- The three octets 0x60, 0x47, 0x60 (which spell "PGP" in UTF-8).

Such a packet should be ignored on input.  It may be placed at the
beginning of a message that uses features not available in PGP 2.6.X in
order to cause that version to report that newer software necessary to
process the message.

5.9 Literal Data Packet (Tag 11)

A Literal Data packet contains the body of a message; last data that group has 8 bits (1 octet).  The first 6-bit group is not
to be further interpreted.
processed as above.  The body of this packet consists of:
	- A one-octet field that describes how the second (incomplete) data group has four
zero-value bits added to it, and is formatted.
If it is a 'b' (0x62), then processed as above.  Two pad
characters (=) are added to the literal packet contains binary output.

6.4 Decoding Radix-64

Any characters outside of the base64 alphabet are ignored in Radix-64
data. If
it is a 't' (0x74), then it contains text data, and thus may need  Decoding software must ignore all line
ends converted to local form, breaks or other text-mode changes.  RFC 1991
also defined
characters not found in the table above.

In Radix-64 data, characters other than those in the table, line
breaks, and other white space probably indicate a value of 'l' as transmission error,
about which a 'local' mode for machine-local
conversions.  This use is now deprecated.

	- File name as warning message or even a string (one-octet length, followed by file name),
	  if the encrypted data should message rejection might be saved as a file.
If the special name "_CONSOLE"
appropriate under some circumstances.

Because it is used, used only for padding at the message is considered to end of the data, the
occurrence of any "=" characters may be
"for your eyes only".  This advises taken as evidence that the message end
of the data has been reached (without truncation in transit).  No such
assurance is unusually
sensitive, and the receiving program should process it more carefully,
perhaps avoiding storing possible, however, when the received data to disk, for example.

	- number of octets transmitted
was a multiple of three and no "=" characters are present.

6.5 Examples of Radix-64

Input data:  0x14fb9c03d97e
Hex:     1   4    f   b    9   c     | 0   3    d   9    7   e
8-bit:   00010100 11111011 10011100  | 00000011 11011001 11111110
6-bit:   000101 001111 101110 011100 | 000000 111101 100111 111110
Decimal: 5      15     46     28       0      61     37     63
Output:  F      P      u      c        A      9      l      /

Input data:  0x14fb9c03d9
Hex:     1   4    f   b    9   c     | 0   3    d   9
8-bit:   00010100 11111011 10011100  | 00000011 11011001
                                                pad with 00
6-bit:   000101 001111 101110 011100 | 000000 111101 100100
Decimal: 5      15     46     28       0      61     36
                                                   pad with =
Output:  F      P      u      c        A      9      k      =

Input data:  0x14fb9c03
Hex:     1   4    f   b    9   c     | 0   3
8-bit:   00010100 11111011 10011100  | 00000011
                                       pad with 0000
6-bit:   000101 001111 101110 011100 | 000000 110000
Decimal: 5      15     46     28       0      48
                                            pad with =      =
Output:  F      P      u      c        A four-octet number that indicates the modification date of the
file, or the creation time      w      =      =

6.6 Example of the packet, or a zero an ASCII Armored Message

  -----BEGIN PGP MESSAGE-----
  Version: OP V0.0

  owFbx8DAYFTCWlySkpkHZDKEFCXmFedmFhdn5ucpZKdWFiv4hgaHKPj5hygUpSbn
  l6UWpabo8XIBAA==
  =3m1o
  -----END PGP MESSAGE-----

Note that indicates the
present time.

    - The remainder of the packet is literal data.

Text data this example is stored with <CR><LF> text endings.  This should be
converted to native line endings indented by the receiving software.

5.10 Trust Packet (Tag 12)

The Trust packet is used only within keyrings and two spaces.

7.  Cleartext signature framework

It is not normally
exported.  Trust packets contain data that record the user's
specifications of which key holders are trustworthy introducers, along
with other information that implementing software uses for trust
information.

Trust packets SHOULD NOT be emitted to output streams that are
transferred desirable to other users, and they SHOULD be ignored on any input
other than local keyring files.

{{Editor's note:  I have brushed aside sign a textual octet stream without ASCII armoring
the description of stream itself, so the old PGP
trust packets for signed text is still readable without special
software.  In order to bind a number of reasons.  They are context dependent;
their meaning depends on the packet preceding them in signature to such a keyring.

There cleartext, this
framework is also a security problem with trust packets.  For example,
malicious software can write used. (Note that RFC 2015 defines another way to clear
sign messages for environments that support MIME.)

The cleartext signed message consists of:
  - The cleartext header '-----BEGIN PGP SIGNED MESSAGE-----' on a new public key
    single line,
  - Zero or more "Hash" Armor Headers,
  - Exactly one empty line not included into a user's key ring
with trust packets that make it trusted.

A number of us have discussed this problem, and think the message digest,
  - The dash-escaped cleartext that trust
information should always be self-signed to act as an integrity check,
but other people may have other solutions.

My solution is to make trust packets implementation dependent.  They

are not emitted on export included into the message digest,
  - The ASCII armored signature(s) including the Armor Header and ignored on import.  Because of this, they Armor
    Tail Lines.

If the "Hash" armor header is given, the specified message digest
algorithm is used for the signature.  If there are arguably out no such headers,
SHA-1 is used.  If more than one message digest is used in the
signature, the "Hash" armor header contains a comma-delimited list of scope
used message digests.

Current message digest names are:

    - "SHA1"
    - "MD5"
    - "RIPEMD160"

The cleartext content of this document anyway.  Given that the PGP
implementation of trust packets has security flaws, this seems to message must also be dash-escaped.

Dash escaped cleartext is the best way to deal ordinary cleartext where every line
starting with them.

--jdcc}}

5.11 User ID Packet (Tag 13)

A User ID packet consists of data which a dash '-' (0x2D) is intended to represent prefixed by the
name sequence dash '-'
(0x2D) and email address space ' ' (0x20).  This prevents the parser from recognizing
armor headers of the key holder.  By convention, it includes
an RFC822 mail name, but there are no restrictions cleartext itself.  The message digest is computed
using the cleartext itself, not the dash escaped form.

As with binary signatures on its content. text documents, a cleartext signature is
calculated on the text using canonical <CR><LF> line endings.  The
packet length in line
ending (i.e. the header specifies <CR><LF>) before the length of '-----BEGIN PGP SIGNATURE-----'
line that terminates the user name.  If
it is text, it signed text is encoded in UTF-8.

{{Editor's note:  PRZ thinks there should be more types not considered part of "user ids"
other than the traditional name, such as photos,
signed text.

Also, any trailing whitespace (spaces, and so on.  The above
definition, which assiduously avoids saying that tabs, 0x09) at the content end of
any line is ignored when the
packet cleartext signature is a counted string, calculated.

8.  Regular Expressions

A regular expression is one potential way to handle it.  Another
would be to explicitly state zero or more branches, separated by `|'.  It
matches anything that this packet matches one of the branches.

A branch is zero or more pieces, concatenated.  It matches a string, and
introduce match for
the first, followed by a free-form user identification packet. match for the second, etc.

A related issue with this document is that sometimes it says "user id"
and sometimes "user name." We need some work here.  Present plan piece is to
use "User ID" everywhere. --jdcc}}

{{Editor's note:  Carl Ellison pointed out to me that if we have
non-exportable (local to one's own keyring) usernames that I can assign
to keys I use, then essentially we have SDSI naming in PGP.  This an atom possibly followed by `*', `+', or `?'. An atom
followed by `*' matches a sequence of 0 or more matches of the atom.
An atom followed by `+' matches a sequence of 1 or more matches of the
atom.  An atom fol- lowed by `?' matches a match of the atom, or the
null string.

An atom is a
Good Thing, regular expression in my opinion, but we have to have parentheses (matching a way to define it.
--jdcc}}

5.12 Comment Packet (Tag 16)

A Comment packet is used match for holding data that is not relevant to
software.  Comment packets should be ignored.

{{Editor's note: should?  Must?  What does it mean to ignore them?  For
example, if it's desirable to show
the regular expression), a range (see below), `.' (matching any single
character), `^' (matching the null string at the beginning of the input
string), `$' (matching the null string at the end of the input string),
a comment to `\' followed by a user, then how does single character (matching that interact with should/must and char- acter), or a suitable definition of "ignore." I
believe
single character with no other significance (matching that they MUST be ignored, but displaying them to character).

A range is a user sequence of characters enclosed in `[]'.  It normally
matches any single character from the sequence. If the sequence begins
with `^', it matches any single character not from the rest of the
sequence.  If two char- acters in the sequence are separated by `-',
this is
ignoring them.  Looking inside them shorthand for cryptographic content (like OP
packets) is *not* ignoring them.}}

{{Editor's note: should we put the full list of ASCII characters between them
(e.g. `[0-9]' matches any decimal digit).  To include a literal `]' in an X.509 encapsulation packet type?}}

6.
the sequence, make it the first character (following a possible `^').
To include a literal `-', make it the first or last character.

9.  Constants

This section describes the constants used in OP.

Note that these tables are not exhaustive lists; an implementation MAY
implement an algorithm not on these lists.

6.1

9.1 Public Key Algorithms

1          - RSA (Encrypt or Sign)
2          - RSA Encrypt-Only
3          - RSA Sign-Only
16         - Elgamal Elgamal, see [ELGAMAL]
17         - DSA (Digital Signature Standard)
18         - Elliptic Curve
19         - ECDSA
21         - Diffie-Hellman (X9.42)
100 to 110 - Private/Experimental algorithm.

Implementations MUST implement DSA for signatures, and Elgamal for
encryption.  Implementations SHOULD implement RSA encryption.
Implementations MAY implement any other algorithm.

{{Editor's note: reserve an algorithm for elliptic curve?  Note that
I've left Elgamal signatures completely unmentioned.  I think this is
good. --jdcc}}

6.2

9.2 Symmetric Key Algorithms

0          - Plaintext
1          - IDEA
2          - Triple-DES (DES-EDE, as per spec -
             168 bit key derived from 192)
3          - CAST5 (128 bit key)
4          - Blowfish (128 bit key) key, 16 rounds)
5          - ROT-N (128 bit N)
6          - SAFER-SK128
7          - DES/SK
100 to 110 - Private/Experimental algorithm.

Implementations MUST implement Triple-DES. Implementations SHOULD
implement IDEA and CAST5.Implementations MAY implement any other
algorithm.

6.3

9.3 Compression Algorithms

0          - Uncompressed
1          - ZIP
100 to 110 - Private/Experimental algorithm.

Implementations MUST implement uncompressed data. Implementations
SHOULD implement ZIP.

6.4

9.4 Hash Algorithms

1          - MD5
2          - SHA-1
3          - RIPE-MD/160
4          - HAVAL
100 to 110 - Private/Experimental algorithm.

Implementations MUST implement SHA-1. Implementations SHOULD implement
MD5.

7.

10.  Packet Composition

OP packets may be are assembled into sequences in order to create messages and
to transfer keys.  Not all possible packet sequences are meaningful and
correct.  This describes the rules for how packets should be placed
into sequences.

7.1

10.1 Transferable Public Keys

OP users may transfer public keys.  The essential elements of a
transferable public key are:

    - One Public Key packet
    - Zero or more revocation signatures
    - One or more User ID packets
    - After each User ID packet, zero or more Signature packets
    - Zero or more Subkey packets
    - After each Subkey packet, one or more Signature packets

The Public Key packet occurs first.  Each of the following User ID
packets provides the identity of the owner of this public key.  If
there are multiple User ID packets, this corresponds to multiple means
of identifying the same unique individual user; for example, a user may
enjoy the use of more than one e-mail address, and construct a User ID
packet for each one.

Immediately following each User ID packet, there are zero or more
signature packets.  Each signature packet is calculated on the
immediately preceding User ID packet and the initial Public Key packet.
The signature serves to certify the corresponding public key and user
ID.  In effect, the signer is testifying to his or her belief that this
public key belongs to the user identified by this user ID.

After the User ID packets there may be one or more Subkey packets.
Subkeys In
general, subkeys are used provided in cases where the top-level public key
is a signature-only key.  The  However, any V4 key may have subkeys, and the
subkeys are then may be encryption-only keys that are
bound to the signature key. keys, signature-only keys, or
general-purpose keys.

Each Subkey packet must be followed by at least one Signature packet,
which should be of the subkey binding signature type, and issued by the top
level key.

{{Editor's note:  I think it is a good idea to have signature-only
subkeys, too (or even encrypt-and-sign subkeys), but no implementation
does this.  Should we generalize here? --jdcc}}

Subkey and Key packets may each be followed by a revocation Signature
packet to indicate that the key is revoked.  Revocation signatures are
only accepted if they are issued by the key itself, or by a key which
is authorized to issue revocations via a revocation key subpacket in a
self-signature by the top level key.

Transferable public key packet sequences may be concatenated to allow
transferring multiple public keys in one operation.

7.2

10.2 OP Messages

An OP message is a packet or sequence of packets that corresponds to
the following grammatical rules (comma represents sequential
composition, and vertical bar separates alternatives):

   OP Message :- Encrypted Message | Signed Message | Compressed Message
                                   | Literal Message.

   Compressed Message :- Compressed Data Packet.

   Literal Message :- Literal Data Packet.

   ESK :- Pubic Key Encrypted Session Key Packet |
          Conventionally Encrypted Session Key Packet.

   ESK Sequence :- ESK | ESK Sequence, ESK.

   Encrypted Message :- Symmetrically Encrypted Data Packet |
               ESK Sequence, Symmetrically Encrypted Data Packet.

   One-Pass Signed Message :- One-Pass Signature Packet, OP Message,
               Signature Packet.

   Signed Message :- Signature Packet, OP Message |
               One-Pass Signed Message.

In addition, the decrypting a Symmetrically Encrypted Data packet and
decompressing a Compressed Data packet must yield a valid OP Message.

8.

11.  Enhanced Key Formats

8.1

11.1 Key Structures

The format of V3 OP key using RSA is as follows.  Entries in square
brackets are optional and ellipses indicate repetition.

    RSA Public Key
       [Revocation Self Signature]
        User ID [Signature ...]
       [User ID [Signature ...] ...]

Each signature certifies the RSA public key and the preceding user ID.
The RSA public key can have many user IDs and each user ID can have
many signatures.

The format of an OP V4 key that uses two public keys is very similar
except that the second key is added to the end as a 'subkey' of the
primary key.

    Primary-Key
       [Revocation Self Signature]
       [Direct Key Self Signature...]
        User ID [Signature ...]
       [User ID [Signature ...] ...]

[Subkey Primary-Key-Signature]

The subkey always has a single signature after it that is issued using
the primary key to tie the two keys together.  The new format can use
either the new signature packets or the old signature packets.

In an Elgamal/DSA key, the DSA public key is that has a main key and subkeys, the primary key, the
Elgamal public key is the subkey, MUST be a
key capable of signing.  The subkeys may be keys of any other type, and
either version 3 or 4 of the signature packet can be used.  There may
be other types of V4 keys, too. For example, there may be a single-key
RSA key in V4 format, a DSA primary key with an RSA encryption key,
etc, or RSA primary key with an Elgamal subkey.

It is also possible to have a signature-only subkey.  This permits a
primary key that collects certifications (key signatures) but is used
only used for certifying subkeys that are used for encryption and
signatures.

8.2

11.2 V4 Key IDs and Fingerprints

A V4 fingerprint is the 160-bit SHA-1 hash of the one-octet Packet Tag,
followed by the two-octet packet length, followed by the entire Public
Key packet starting with the version field.  The key ID is either the
low order 32 bits or 64 bits of the fingerprint.  Here are the fields
of the hash material, with the example of a DSA key:

    a.1) 0x99 (1 byte)
    a.2) high order length byte of (b)-(f) (1 byte)
    a.3) low order length byte of (b)-(f) (1 byte)
    b) version number = 4 (1 byte);
    c) time stamp of key creation (4 bytes);
    e) algorithm (1 byte):
         17 = DSA;
    f) Algorithm specific fields.

    Algorithm Specific Fields for DSA keys (example):
    f.1) MPI of DSA prime p;
    f.2) MPI of DSA group order q (q is a prime divisor of p-1);
    f.3) MPI of DSA group generator g;
    f.4) MPI of DSA public key value y (= g**x where x is secret).

9.

12.  Security Considerations

As with any technology involving cryptography, you should check the
current literature to determine if any algorithms used here have been
found to be vulnerable to attack.

This specification uses Public Key Cryptography technologies.
Possession of the private key portion of a public-private key pair is
assumed to be controlled by the proper party or parties.

Certain operations in this specification involve the use of random
numbers.  An appropriate entropy source should be used to generate
these numbers.  See RFC 1750.

The MD5 hash algorithm has been found to have weaknesses
(pseudo-collisions in the compress function) that make some people
deprecate its use.  They consider the SHA-1 algorithm better.

If you are building an authentication system, the recipient may specify
a preferred signing algorithm.  However, the signer would be foolish to
use a weak algorithm simply because the recipient requests it.

Some of the encryption algorithms mentioned in this document have been
analyzed less than others.  For example, although CAST5 is presently
considered strong, it has been analyzed less than Triple-DES.  Other
algorithms may have other controversies surrounding them.

Some technologies mentioned here may be subject to government control
in some countries.

10.

13.  Authors and Working Group Chair

The working group can be contacted via the current chair:

John W. Noerenberg, II
Qualcomm, Inc
6455 Lusk Blvd
San Diego, CA 92131 USA
Email: jwn2@qualcomm.com
Tel: +1 619 658 3510 619-658-3510

The principal authors of this draft are (in alphabetical order):

Jon Callas Pretty Good Privacy,
Network Associates, Inc. 555 Twin Dolphin Drive, #570
Redwood Shores,
4200 Bohannon Drive
Menlo Park, CA 94065, 94025, USA
Email: jon@pgp.com
Tel: +1-650-596-1960 +1-650-473-2860

Lutz Donnerhacke
IKS GmbH
Wildenbruchstr. 15
07745 Jena, Germany
EMail: lutz@iks-jena.de
Tel: +49-3641-675642

Hal Finney Pretty Good Privacy,
Network Associates, Inc. 555 Twin Dolphin Drive, #570
Redwood Shores,
4200 Bohannon Drive

Menlo Park, CA 94065, 94025, USA
Email: hal@pgp.com Tel: +1-650-572-0430

Rodney Thayer
Sable Technology Corporation
246 Walnut Street
Newton, MA  02160 USA
Email: rodney@sabletech.com
Tel: +1-617-332-7292

This draft also draws on much previous work from a number of other
authors who include:  Derek Atkins, Charles Breed, Dave Del Torto, Marc
Dyksterhouse, Gail Haspert, Gene Hoffman, Paul Hoffman, Raph Levine,
Colin Plumb, Will Price, William Stallings, Mark Weaver, and Philip R.
Zimmermann.

11.

14.  References

[CAMPBELL} Campbell, Joe, "C Programmer's Guide to Serial
Communications"

[DONNERHACKE] Donnerhacke, L., et. al, "PGP263in - an improved
international version of PGP",
ftp://ftp.iks-jena.de/mitarb/lutz/crypt/software/pgp/

[ELGAMAL] T.  ElGamal, "A Public-Key Cryptosystem and a Signature
Scheme Based on Discrete Logarithms," IEEE Transactions on Information
Theory, v.  IT-31, n. 4, 1985, pp. 469-472.

[ISO-10646] ISO/IEC 10646-1:1993.  International Standard --
Information technology -- Universal Multiple-Octet Coded Character Set
(UCS) -- Part 1:  Architecture and Basic Multilingual Plane.  UTF-8 is
described in Annex R, adopted but not yet published.  UTF-16 is
described in Annex Q, adopted but not yet published.

[PKCS1] RSA Laboratories, "PKCS #1:  RSA Encryption Standard," version
1.5, November 1993

[RFC822] D.  Crocker, "Standard for the format of ARPA Internet text
messages", RFC 822, August 1982

[RFC1423] D.  Balenson, "Privacy Enhancement for Internet Electronic
Mail:  Part III:  Algorithms, Modes, and Identifiers", RFC 1423,
October 1993

[RFC1641] Goldsmith, D., and M.  Davis, "Using Unicode with MIME", RFC
1641, Taligent inc., July 1994.

[RFC1750] Eastlake, Crocker, & Schiller., Randomness Recommendations
for Security.  December 1994.

[RFC1951] Deutsch, P., DEFLATE Compressed Data Format Specification
version 1.3.  May 1996.

[RFC1983] G.  Malkin., Internet Users' Glossary.  August 1996.

[RFC1991] Atkins, D., Stallings, W., and P.  Zimmermann, "PGP Message
Exchange Formats", RFC 1991, August 1996.

[RFC2015] Elkins, M., "MIME Security with Pretty Good Privacy (PGP)",
RFC 2015, October 1996.

[RFC2044] F.  Yergeau., UTF-8, a transformation format of Unicode and
ISO 10646.  October 1996.

[RFC2045] Borenstein, N., and Freed, N., "Multipurpose Internet Mail
Extensions (MIME) Part One:  Format of Internet Message Bodies.",
November 1996

[RFC2119] Bradner, S., Key words for use in RFCs to Indicate
Requirement Level.  March 1997.

12.

15.  Full Copyright Statement

Copyright 1997 1998 by The Internet Society.  All Rights Reserved.

This document and translations of it may be copied and furnished to
others, and derivative works that comment on or otherwise explain it or
assist in its implementation may be prepared, copied, published and
distributed, in whole or in part, without restriction of any kind,
provided that the above copyright notice and this paragraph are
included on all such copies and derivative works.  However, this
document itself may not be modified in any way, such as by removing the
copyright notice or references to the Internet Society or other
Internet organizations, except as needed for the purpose of developing
Internet standards in which case the procedures for copyrights defined
in the Internet Standards process must be followed, or as required to
translate it into languages other than English.

The limited permissions granted above are perpetual and will not be
revoked by the Internet Society or its successors or assigns.