# doc/DETAILS                                                -*- org -*-
#+TITLE: GnuPG Details
# Globally disable superscripts and subscripts:
#+OPTIONS: ^:{}
#

# Note: This file uses org-mode; it should be easy to read as plain
# text but be aware of some markup peculiarities: Verbatim code is
# enclosed in #+begin-example, #+end-example blocks or marked by a
# colon as the first non-white-space character, words bracketed with
# equal signs indicate a monospace font, and the usual /italics/,
# *bold*, and _underline_ conventions are recognized.

This is the DETAILS file for GnuPG which specifies some internals and
parts of the external API for GPG and GPGSM.

* Format of the colon listings
  The format is a based on colon separated record, each recods starts
  with a tag string and extends to the end of the line.  Here is an
  example:
#+begin_example
$ gpg --with-colons --list-keys \
      --with-fingerprint --with-fingerprint wk@gnupg.org
pub:f:1024:17:6C7EE1B8621CC013:899817715:1055898235::m:::scESC:
fpr:::::::::ECAF7590EB3443B5C7CF3ACB6C7EE1B8621CC013:
uid:f::::::::Werner Koch <wk@g10code.com>:
uid:f::::::::Werner Koch <wk@gnupg.org>:
sub:f:1536:16:06AD222CADF6A6E1:919537416:1036177416:::::e:
fpr:::::::::CF8BCC4B18DE08FCD8A1615906AD222CADF6A6E1:
sub:r:1536:20:5CE086B5B5A18FF4:899817788:1025961788:::::esc:
fpr:::::::::AB059359A3B81F410FCFF97F5CE086B5B5A18FF4:
#+end_example

Note that new version of GnuPG or the use of certain options may add
new fields to the output.  Parsers should not assume a limit on the
number of fields per line.  Some fields are not yet used or only used
with certain record types; parsers should ignore fields they are not
aware of.

The double =--with-fingerprint= prints the fingerprint for the subkeys
too.  Old versions of gpg used a slightly different format and required
the use of the option =--fixed-list-mode= to conform to the format
described here.


** Description of the fields
*** Field 1 - Type of record

    - pub :: Public key
    - crt :: X.509 certificate
    - crs :: X.509 certificate and private key available
    - sub :: Subkey (secondary key)
    - sec :: Secret key
    - ssb :: Secret subkey (secondary key)
    - uid :: User id (only field 10 is used).
    - uat :: User attribute (same as user id except for field 10).
    - sig :: Signature
    - rev :: Revocation signature
    - fpr :: Fingerprint (fingerprint is in field 10)
    - pkd :: Public key data [*]
    - grp :: Keygrip
    - rvk :: Revocation key
    - tru :: Trust database information [*]
    - spk :: Signature subpacket [*]
    - cfg :: Configuration data [*]

    Records marked with an asterisk are described at [[*Special%20field%20formats][*Special fields]].

*** Field 2 - Validity

    This is a letter describing the computed validity of a key.
    Currently this is a single letter, but be prepared that additional
    information may follow in some future versions. Note that GnuPG <
    2.1 does not set this field for secret key listings.

    - o :: Unknown (this key is new to the system)
    - i :: The key is invalid (e.g. due to a missing self-signature)
    - d :: The key has been disabled
	   (deprecated - use the 'D' in field 12 instead)
    - r :: The key has been revoked
    - e :: The key has expired
    - - :: Unknown validity (i.e. no value assigned)
    - q :: Undefined validity.  '-' and 'q' may safely be treated as
           the same value for most purposes
    - n :: The key is not valid
    - m :: The key is marginal valid.
    - f :: The key is fully valid
    - u :: The key is ultimately valid.  This often means that the
           secret key is available, but any key may be marked as
           ultimately valid.
    - w :: The key has a well known private part.
    - s :: The key has special validity.  This means that it might be
           self-signed and expected to be used in the STEED sytem.

    If the validity information is given for a UID or UAT record, it
    describes the validity calculated based on this user ID.  If given
    for a key record it describes the validity taken from the best
    rated user ID.

    For X.509 certificates a 'u' is used for a trusted root
    certificate (i.e. for the trust anchor) and an 'f' for all other
    valid certificates.

*** Field 3 - Key length

    The length of key in bits.

*** Field 4 - Public key algorithm

    The values here are those from the OpenPGP specs or if they are
    greather than 255 the algorithm ids as used by Libgcrypt.

*** Field 5 - KeyID

    This is the 64 bit keyid as specified by OpenPGP and the last 64
    bit of the SHA-1 fingerprint of an X.509 certifciate.

*** Field 6 - Creation date

    The creation date of the key is given in UTC.  For UID and UAT
    records, this is used for the self-signature date.  Note that the
    date is usally printed in seconds since epoch, however, we are
    migrating to an ISO 8601 format (e.g. "19660205T091500").  This is
    currently only relevant for X.509.  A simple way to detect the new
    format is to scan for the 'T'.  Note that old versions of gpg
    without using the =--fixed-list-mode= option used a "yyyy-mm-tt"
    format.

*** Field 7 - Expiration date

    Key or UID/UAT expiration date or empty if it does not expire.

*** Field 8 - Certificate S/N, UID hash, trust signature info

    Used for serial number in crt records.  For UID and UAT records,
    this is a hash of the user ID contents used to represent that
    exact user ID.  For trust signatures, this is the trust depth
    seperated by the trust value by a space.

*** Field 9 -  Ownertrust

    This is only used on primary keys.  This is a single letter, but
    be prepared that additional information may follow in future
    versions.  For trust signatures with a regular expression, this is
    the regular expression value, quoted as in field 10.

*** Field 10 - User-ID
    The value is quoted like a C string to avoid control characters
    (the colon is quoted =\x3a=).  For a "pub" record this field is
    not used on --fixed-list-mode.  A UAT record puts the attribute
    subpacket count here, a space, and then the total attribute
    subpacket size.  In gpgsm the issuer name comes here.  A FPR
    record stores the fingerprint here.  The fingerprint of a
    revocation key is stored here.
*** Field 11 - Signature class

    Signature class as per RFC-4880.  This is a 2 digit hexnumber
    followed by either the letter 'x' for an exportable signature or
    the letter 'l' for a local-only signature.  The class byte of an
    revocation key is also given here, 'x' and 'l' is used the same
    way.  This field if not used for X.509.

*** Field 12 - Key capabilities

    The defined capabilities are:

    - e :: Encrypt
    - s :: Sign
    - c :: Certify
    - a :: Authentication
    - ? :: Unknown capability

    A key may have any combination of them in any order.  In addition
    to these letters, the primary key has uppercase versions of the
    letters to denote the _usable_ capabilities of the entire key, and
    a potential letter 'D' to indicate a disabled key.

*** Field 13 - Issuer certificate fingerprint or other info

    Used in FPR records for S/MIME keys to store the fingerprint of
    the issuer certificate.  This is useful to build the certificate
    path based on certificates stored in the local key database it is
    only filled if the issuer certificate is available. The root has
    been reached if this is the same string as the fingerprint. The
    advantage of using this value is that it is guaranteed to have
    been been build by the same lookup algorithm as gpgsm uses.

    For "uid" records this field lists the preferences in the same way
    gpg's --edit-key menu does.

    For "sig" records, this is the fingerprint of the key that issued
    the signature.  Note that this is only filled in if the signature
    verified correctly.  Note also that for various technical reasons,
    this fingerprint is only available if --no-sig-cache is used.

*** Field 14 - Flag field

    Flag field used in the --edit menu output

*** Field 15 - S/N of a token

    Used in sec/ssb to print the serial number of a token (internal
    protect mode 1002) or a '#' if that key is a simple stub (internal
    protect mode 1001).  If the option --with-secret is used and a
    secret key is available for the public key, a '+' indicates this.

*** Field 16 - Hash algorithm

    For sig records, this is the used hash algorithm.  For example:
    2 = SHA-1, 8 = SHA-256.

*** Field 17 - Curve name

    For pub, sub, sec, and ssb records this field is used for the ECC
    curve name.
*** Field 18 - TOFU Policy

    This is the TOFU policy.  It is either good, bad, unknown, ask or
    auto.  This is only shows for uid records.

** Special fields

*** PKD - Public key data

    If field 1 has the tag "pkd", a listing looks like this:
#+begin_example
pkd:0:1024:B665B1435F4C2 .... FF26ABB:
    !  !   !-- the value
    !  !------ for information number of bits in the value
    !--------- index (eg. DSA goes from 0 to 3: p,q,g,y)
#+end_example

*** TRU - Trust database information
    Example for a "tru" trust base record:
#+begin_example
    tru:o:0:1166697654:1:3:1:5
#+end_example

    - Field 2 :: Reason for staleness of trust.  If this field is
                 empty, then the trustdb is not stale.  This field may
                 have multiple flags in it:

                 - o :: Trustdb is old
                 - t :: Trustdb was built with a different trust model
                        than the one we are using now.

    - Field 3 :: Trust model

                 - 0 :: Classic trust model, as used in PGP 2.x.
                 - 1 :: PGP trust model, as used in PGP 6 and later.
                        This is the same as the classic trust model,
                        except for the addition of trust signatures.

                 GnuPG before version 1.4 used the classic trust model
                 by default. GnuPG 1.4 and later uses the PGP trust
                 model by default.

    - Field 4 :: Date trustdb was created in seconds since Epoch.
    - Field 5 :: Date trustdb will expire in seconds since Epoch.
    - Field 6 :: Number of marginally trusted users to introduce a new
                 key signer (gpg's option --marginals-needed).
    - Field 7 :: Number of completely trusted users to introduce a new
                 key signer.  (gpg's option --completes-needed)

    - Field 8 :: Maximum depth of a certification chain. (gpg's option
                 --max-cert-depth)

*** SPK - Signature subpacket records

    - Field 2 :: Subpacket number as per RFC-4880 and later.
    - Field 3 :: Flags in hex.  Currently the only two bits assigned
                 are 1, to indicate that the subpacket came from the
                 hashed part of the signature, and 2, to indicate the
                 subpacket was marked critical.
    - Field 4 :: Length of the subpacket.  Note that this is the
                 length of the subpacket, and not the length of field
                 5 below.  Due to the need for %-encoding, the length
                 of field 5 may be up to 3x this value.
    - Field 5 :: The subpacket data.  Printable ASCII is shown as
                 ASCII, but other values are rendered as %XX where XX
                 is the hex value for the byte.

*** CFG - Configuration data

    --list-config outputs information about the GnuPG configuration
    for the benefit of frontends or other programs that call GnuPG.
    There are several list-config items, all colon delimited like the
    rest of the --with-colons output.  The first field is always "cfg"
    to indicate configuration information.  The second field is one of
    (with examples):

    - version :: The third field contains the version of GnuPG.

                 : cfg:version:1.3.5

    - pubkey :: The third field contains the public key algorithms
                this version of GnuPG supports, separated by
                semicolons.  The algorithm numbers are as specified in
                RFC-4880.  Note that in contrast to the --status-fd
                interface these are _not_ the Libgcrypt identifiers.
                Using =pubkeyname= prints names instead of numbers.

                 : cfg:pubkey:1;2;3;16;17

    - cipher :: The third field contains the symmetric ciphers this
                version of GnuPG supports, separated by semicolons.
                The cipher numbers are as specified in RFC-4880.
                Using =ciphername= prints names instead of numbers.

                 : cfg:cipher:2;3;4;7;8;9;10

    - digest :: The third field contains the digest (hash) algorithms
                this version of GnuPG supports, separated by
                semicolons.  The digest numbers are as specified in
                RFC-4880.  Using =digestname= prints names instead of
                numbers.

                 : cfg:digest:1;2;3;8;9;10

    - compress :: The third field contains the compression algorithms
                  this version of GnuPG supports, separated by
                  semicolons.  The algorithm numbers are as specified
                  in RFC-4880.

                 : cfg:compress:0;1;2;3

    - group :: The third field contains the name of the group, and the
               fourth field contains the values that the group expands
               to, separated by semicolons.

               For example, a group of:
                 : group mynames = paige 0x12345678 joe patti
               would result in:
                 : cfg:group:mynames:patti;joe;0x12345678;paige

    - curve :: The third field contains the curve names this version
               of GnuPG supports, separated by semicolons. Using
               =curveoid= prints OIDs instead of numbers.

                 : cfg:curve:ed25519;nistp256;nistp384;nistp521


* Format of the --status-fd output

  Every line is prefixed with "[GNUPG:] ", followed by a keyword with
  the type of the status line and some arguments depending on the type
  (maybe none); an application should always be prepared to see new
  keyworkds or more arguments in future versions.

** General status codes
*** NEWSIG [<signers_uid>]
    Is issued right before a signature verification starts.  This is
    useful to define a context for parsing ERROR status messages.
    arguments are currently defined.  If SIGNERS_UID is given and is
    not "-" this is the percent escape value of the OpenPGP Signer's
    User ID signature sub-packet.

*** GOODSIG  <long_keyid_or_fpr>  <username>
    The signature with the keyid is good.  For each signature only one
    of the codes GOODSIG, BADSIG, EXPSIG, EXPKEYSIG, REVKEYSIG or
    ERRSIG will be emitted.  In the past they were used as a marker
    for a new signature; new code should use the NEWSIG status
    instead.  The username is the primary one encoded in UTF-8 and %XX
    escaped. The fingerprint may be used instead of the long keyid if
    it is available.  This is the case with CMS and might eventually
    also be available for OpenPGP.

*** EXPSIG  <long_keyid_or_fpr>  <username>
    The signature with the keyid is good, but the signature is
    expired. The username is the primary one encoded in UTF-8 and %XX
    escaped. The fingerprint may be used instead of the long keyid if
    it is available.  This is the case with CMS and might eventually
    also be available for OpenPGP.

*** EXPKEYSIG  <long_keyid_or_fpr> <username>
    The signature with the keyid is good, but the signature was made
    by an expired key. The username is the primary one encoded in
    UTF-8 and %XX escaped.  The fingerprint may be used instead of the
    long keyid if it is available.  This is the case with CMS and
    might eventually also be available for OpenPGP.

*** REVKEYSIG  <long_keyid_or_fpr>  <username>
    The signature with the keyid is good, but the signature was made
    by a revoked key. The username is the primary one encoded in UTF-8
    and %XX escaped. The fingerprint may be used instead of the long
    keyid if it is available.  This is the case with CMS and might
    eventually also beñ available for OpenPGP.

*** BADSIG  <long_keyid_or_fpr>  <username>
    The signature with the keyid has not been verified okay.  The
    username is the primary one encoded in UTF-8 and %XX escaped. The
    fingerprint may be used instead of the long keyid if it is
    available.  This is the case with CMS and might eventually also be
    available for OpenPGP.

*** ERRSIG  <keyid>  <pkalgo> <hashalgo> <sig_class> <time> <rc>
    It was not possible to check the signature.  This may be caused by
    a missing public key or an unsupported algorithm.  A RC of 4
    indicates unknown algorithm, a 9 indicates a missing public
    key. The other fields give more information about this signature.
    sig_class is a 2 byte hex-value.  The fingerprint may be used
    instead of the keyid if it is available.  This is the case with
    gpgsm and might eventually also be available for OpenPGP.

    Note, that TIME may either be the number of seconds since Epoch or
    an ISO 8601 string.  The latter can be detected by the presence of
    the letter 'T'.

*** VALIDSIG <args>

    The args are:

    - <fingerprint_in_hex>
    - <sig_creation_date>
    - <sig-timestamp>
    - <expire-timestamp>
    - <sig-version>
    - <reserved>
    - <pubkey-algo>
    - <hash-algo>
    - <sig-class>
    - [ <primary-key-fpr> ]

    This status indicates that the signature is cryptographically
    valid. This is similar to GOODSIG, EXPSIG, EXPKEYSIG, or REVKEYSIG
    (depending on the date and the state of the signature and signing
    key) but has the fingerprint as the argument. Multiple status
    lines (VALIDSIG and the other appropriate *SIG status) are emitted
    for a valid signature.  All arguments here are on one long line.
    sig-timestamp is the signature creation time in seconds after the
    epoch. expire-timestamp is the signature expiration time in
    seconds after the epoch (zero means "does not
    expire"). sig-version, pubkey-algo, hash-algo, and sig-class (a
    2-byte hex value) are all straight from the signature packet.
    PRIMARY-KEY-FPR is the fingerprint of the primary key or identical
    to the first argument.  This is useful to get back to the primary
    key without running gpg again for this purpose.

    The primary-key-fpr parameter is used for OpenPGP and not
    available for CMS signatures.  The sig-version as well as the sig
    class is not defined for CMS and currently set to 0 and 00.

    Note, that *-TIMESTAMP may either be a number of seconds since
    Epoch or an ISO 8601 string which can be detected by the presence
    of the letter 'T'.

*** SIG_ID  <radix64_string>  <sig_creation_date>  <sig-timestamp>
    This is emitted only for signatures of class 0 or 1 which have
    been verified okay.  The string is a signature id and may be used
    in applications to detect replay attacks of signed messages.  Note
    that only DLP algorithms give unique ids - others may yield
    duplicated ones when they have been created in the same second.

    Note, that SIG-TIMESTAMP may either be a number of seconds since
    Epoch or an ISO 8601 string which can be detected by the presence
    of the letter 'T'.

*** ENC_TO  <long_keyid>  <keytype>  <keylength>
    The message is encrypted to this LONG_KEYID.  KEYTYPE is the
    numerical value of the public key algorithm or 0 if it is not
    known, KEYLENGTH is the length of the key or 0 if it is not known
    (which is currently always the case).  Gpg prints this line
    always; Gpgsm only if it knows the certificate.

*** BEGIN_DECRYPTION
    Mark the start of the actual decryption process.  This is also
    emitted when in --list-only mode.
*** END_DECRYPTION
    Mark the end of the actual decryption process.  This are also
    emitted when in --list-only mode.
*** DECRYPTION_INFO <mdc_method> <sym_algo>
    Print information about the symmetric encryption algorithm and the
    MDC method.  This will be emitted even if the decryption fails.

*** DECRYPTION_FAILED
    The symmetric decryption failed - one reason could be a wrong
    passphrase for a symmetrical encrypted message.

*** DECRYPTION_OKAY
    The decryption process succeeded.  This means, that either the
    correct secret key has been used or the correct passphrase for a
    symmetric encrypted message was given.  The program itself may
    return an errorcode because it may not be possible to verify a
    signature for some reasons.

*** SESSION_KEY <algo>:<hexdigits>
    The session key used to decrypt the message.  This message will
    only be emitted if the option --show-session-key is used.  The
    format is suitable to be passed as value for the option
    --override-session-key.  It is not an indication that the
    decryption will or has succeeded.

*** BEGIN_ENCRYPTION  <mdc_method> <sym_algo>
    Mark the start of the actual encryption process.

*** END_ENCRYPTION
    Mark the end of the actual encryption process.

*** FILE_START <what> <filename>
    Start processing a file <filename>.  <what> indicates the performed
    operation:
    - 1 :: verify
    - 2 :: encrypt
    - 3 :: decrypt

*** FILE_DONE
    Marks the end of a file processing which has been started
    by FILE_START.

*** BEGIN_SIGNING
    Mark the start of the actual signing process. This may be used as
    an indication that all requested secret keys are ready for use.

*** ALREADY_SIGNED <long-keyid>
    Warning: This is experimental and might be removed at any time.

*** SIG_CREATED <type> <pk_algo> <hash_algo> <class> <timestamp> <keyfpr>
    A signature has been created using these parameters.
    Values for type <type> are:
      - D :: detached
      - C :: cleartext
      - S :: standard
    (only the first character should be checked)

    <class> are 2 hex digits with the OpenPGP signature class.

    Note, that TIMESTAMP may either be a number of seconds since Epoch
    or an ISO 8601 string which can be detected by the presence of the
    letter 'T'.

*** NOTATION_
    There are actually three related status codes to convey notation
    data:

    - NOTATION_NAME <name>
    - NOTATION_FLAGS <critical> <human_readable>
    - NOTATION_DATA <string>

    <name> and <string> are %XX escaped.  The data may be split among
    several NOTATION_DATA lines.  NOTATION_FLAGS is emitted after
    NOTATION_NAME and gives the critical and human readable flags;
    the flag values are either 0 or 1.

*** POLICY_URL <string>
    Note that URL in <string> is %XX escaped.

*** PLAINTEXT <format> <timestamp> <filename>
    This indicates the format of the plaintext that is about to be
    written.  The format is a 1 byte hex code that shows the format of
    the plaintext: 62 ('b') is binary data, 74 ('t') is text data with
    no character set specified, and 75 ('u') is text data encoded in
    the UTF-8 character set.  The timestamp is in seconds since the
    epoch.  If a filename is available it gets printed as the third
    argument, percent-escaped as usual.

*** PLAINTEXT_LENGTH <length>
    This indicates the length of the plaintext that is about to be
    written.  Note that if the plaintext packet has partial length
    encoding it is not possible to know the length ahead of time.  In
    that case, this status tag does not appear.

*** ATTRIBUTE <arguments>
    The list or arguments are:
    - <fpr>
    - <octets>
    - <type>
    - <index>
    - <count>
    - <timestamp>
    - <expiredate>
    - <flags>

    This is one long line issued for each attribute subpacket when an
    attribute packet is seen during key listing.  <fpr> is the
    fingerprint of the key.  <octets> is the length of the attribute
    subpacket.  <type> is the attribute type (e.g. 1 for an image).
    <index> and <count> indicate that this is the N-th indexed
    subpacket of count total subpackets in this attribute packet.
    <timestamp> and <expiredate> are from the self-signature on the
    attribute packet.  If the attribute packet does not have a valid
    self-signature, then the timestamp is 0.  <flags> are a bitwise OR
    of:
    - 0x01 :: this attribute packet is a primary uid
    - 0x02 :: this attribute packet is revoked
    - 0x04 :: this attribute packet is expired

*** SIG_SUBPACKET <type> <flags> <len> <data>
    This indicates that a signature subpacket was seen.  The format is
    the same as the "spk" record above.

** Key related
*** INV_RECP, INV_SGNR
    The two similar status codes:

    - INV_RECP <reason> <requested_recipient>
    - INV_SGNR <reason> <requested_sender>

    are issued for each unusable recipient/sender. The reasons codes
    currently in use are:

       -  0 :: No specific reason given
       -  1 :: Not Found
       -  2 :: Ambigious specification
       -  3 :: Wrong key usage
       -  4 :: Key revoked
       -  5 :: Key expired
       -  6 :: No CRL known
       -  7 :: CRL too old
       -  8 :: Policy mismatch
       -  9 :: Not a secret key
       - 10 :: Key not trusted
       - 11 :: Missing certificate
       - 12 :: Missing issuer certificate
       - 13 :: Key disabled
       - 14 :: Syntax error in specification

    If no specific reason was given a previously emitted status code
    KEY_CONSIDERED may be used to analyzed the problem.

    Note that for historical reasons the INV_RECP status is also used
    for gpgsm's SIGNER command where it relates to signer's of course.
    Newer GnuPG versions are using INV_SGNR; applications should
    ignore the INV_RECP during the sender's command processing once
    they have seen an INV_SGNR.  Different codes are used so that they
    can be distinguish while doing an encrypt+sign operation.

*** NO_RECP <reserved>
    Issued if no recipients are usable.

*** NO_SGNR <reserved>
    Issued if no senders are usable.

*** KEY_CONSIDERED <fpr> <flags>
    Issued to explian the lookup of a key.  FPR is the hexified
    fingerprint of the primary key.  The bit values for FLAGS are:

    - 1 :: The key has not been selected.
    - 2 :: All subkeys of the key are expired or have been revoked.

*** KEYEXPIRED <expire-timestamp>
    The key has expired.  expire-timestamp is the expiration time in
    seconds since Epoch.  This status line is not very useful because
    it will also be emitted for expired subkeys even if this subkey is
    not used.  To check whether a key used to sign a message has
    expired, the EXPKEYSIG status line is to be used.

    Note, that the TIMESTAMP may either be a number of seconds since
    Epoch or an ISO 8601 string which can be detected by the presence
    of the letter 'T'.

*** KEYREVOKED
    The used key has been revoked by its owner.  No arguments yet.

*** NO_PUBKEY  <long keyid>
    The public key is not available

*** NO_SECKEY  <long keyid>
    The secret key is not available

*** KEY_CREATED <type> <fingerprint> [<handle>]
    A key has been created.  Values for <type> are:
      - B :: primary and subkey
      - P :: primary
      - S :: subkey
    The fingerprint is one of the primary key for type B and P and the
    one of the subkey for S.  Handle is an arbitrary non-whitespace
    string used to match key parameters from batch key creation run.

*** KEY_NOT_CREATED [<handle>]
    The key from batch run has not been created due to errors.

*** TRUST_
    These are several similar status codes:

    - TRUST_UNDEFINED <error_token>
    - TRUST_NEVER     <error_token>
    - TRUST_MARGINAL  [0  [<validation_model>]]
    - TRUST_FULLY     [0  [<validation_model>]]
    - TRUST_ULTIMATE  [0  [<validation_model>]]

    For good signatures one of these status lines are emitted to
    indicate the validity of the key used to create the signature.
    The error token values are currently only emitted by gpgsm.

    VALIDATION_MODEL describes the algorithm used to check the
    validity of the key.  The defaults are the standard Web of Trust
    model for gpg and the the standard X.509 model for gpgsm.  The
    defined values are

       - pgp   :: The standard PGP WoT.
       - shell :: The standard X.509 model.
       - chain :: The chain model.
       - steed :: The STEED model.
       - tofu  :: The TOFU model

    Note that the term =TRUST_= in the status names is used for
    historic reasons; we now speak of validity.

*** TOFU_USER <fingerprint_in_hex> <mbox>

    This status identifies the key and the userid for all following
    Tofu information.  The fingerprint is the fingerprint of the
    primary key and the mbox is in general the addr-spec part of the
    userid encoded in UTF-8 and percent escaped.  The fingerprint is
    indentical for all TOFU_USER lines up to a NEWSIG line.

*** TOFU_STATS <validity> <sign-count> 0 [<policy> [<tm1> <tm2>]]

    Statistics for the current user id.

    Values for VALIDITY are:
    - 0 :: conflict
    - 1 :: key without history
    - 2 :: key with too little history
    - 3 :: key with enough history for basic trust
    - 4 :: key with a lot of history

    Values for POLICY are:
    - none    :: No Policy set
    - auto    :: Policy is "auto"
    - good    :: Policy is "good"
    - bad     :: Policy is "bad"
    - ask     :: Policy is "ask"
    - unknown :: Policy is not known.

    TM1 gives the number of seconds since the the first messages was
    verified.  TM2 gives the number of seconds since the most recent
    message was verified.

*** TOFU_STATS_SHORT <long_string>

    Information about the TOFU binding for the signature.
    Example: "15 signatures verified. 10 messages encrypted"

*** TOFU_STATS_LONG <long_string>

    Information about the TOFU binding for the signature in verbose
    format.  The LONG_STRING is percent escaped.
    Example: 'Verified 9 messages signed by "Werner Koch
    (dist sig)" in the past 3 minutes, 40 seconds.  The most
    recent message was verified 4 seconds ago.'

*** PKA_TRUST_
    This is is one:

    - PKA_TRUST_GOOD <addr-spec>
    - PKA_TRUST_BAD  <addr-spec>

    Depending on the outcome of the PKA check one of the above status
    codes is emitted in addition to a =TRUST_*= status.

** Remote control
*** GET_BOOL, GET_LINE, GET_HIDDEN, GOT_IT

    These status line are used with --command-fd for interactive
    control of the process.

*** USERID_HINT <long main keyid> <string>
    Give a hint about the user ID for a certain keyID.

*** NEED_PASSPHRASE <long keyid> <long main keyid> <keytype> <keylength>
    Issued whenever a passphrase is needed.  KEYTYPE is the numerical
    value of the public key algorithm or 0 if this is not applicable,
    KEYLENGTH is the length of the key or 0 if it is not known (this
    is currently always the case).

*** NEED_PASSPHRASE_SYM <cipher_algo> <s2k_mode> <s2k_hash>
    Issued whenever a passphrase for symmetric encryption is needed.

*** NEED_PASSPHRASE_PIN <card_type> <chvno> [<serialno>]
    Issued whenever a PIN is requested to unlock a card.

*** MISSING_PASSPHRASE
    No passphrase was supplied.  An application which encounters this
    message may want to stop parsing immediately because the next
    message will probably be a BAD_PASSPHRASE.  However, if the
    application is a wrapper around the key edit menu functionality it
    might not make sense to stop parsing but simply ignoring the
    following BAD_PASSPHRASE.

*** BAD_PASSPHRASE <long keyid>
    The supplied passphrase was wrong or not given.  In the latter
    case you may have seen a MISSING_PASSPHRASE.

*** GOOD_PASSPHRASE
    The supplied passphrase was good and the secret key material
    is therefore usable.

** Import/Export
*** IMPORT_CHECK <long keyid> <fingerprint> <user ID>
    This status is emitted in interactive mode right before
    the "import.okay" prompt.

*** IMPORTED   <long keyid>  <username>
    The keyid and name of the signature just imported

*** IMPORT_OK  <reason> [<fingerprint>]
    The key with the primary key's FINGERPRINT has been imported.
    REASON flags are:

    - 0 :: Not actually changed
    - 1 :: Entirely new key.
    - 2 :: New user IDs
    - 4 :: New signatures
    - 8 :: New subkeys
    - 16 :: Contains private key.

    The flags may be ORed.

*** IMPORT_PROBLEM <reason> [<fingerprint>]
    Issued for each import failure.  Reason codes are:

    - 0 :: No specific reason given.
    - 1 :: Invalid Certificate.
    - 2 :: Issuer Certificate missing.
    - 3 :: Certificate Chain too long.
    - 4 :: Error storing certificate.

*** IMPORT_RES <args>
    Final statistics on import process (this is one long line). The
    args are a list of unsigned numbers separated by white space:

    - <count>
    - <no_user_id>
    - <imported>
    - always 0 (formerly used for the number of RSA keys)
    - <unchanged>
    - <n_uids>
    - <n_subk>
    - <n_sigs>
    - <n_revoc>
    - <sec_read>
    - <sec_imported>
    - <sec_dups>
    - <skipped_new_keys>
    - <not_imported>
    - <skipped_v3_keys>

*** EXPORTED  <fingerprint>
    The key with <fingerprint> has been exported.  The fingerprint is
    the fingerprint of the primary key even if the primary key has
    been replaced by a stub key during secret key export.

*** EXPORT_RES <args>

    Final statistics on export process (this is one long line). The
    args are a list of unsigned numbers separated by white space:

    - <count>
    - <secret_count>
    - <exported>


** Smartcard related
*** CARDCTRL <what> [<serialno>]
    This is used to control smartcard operations.  Defined values for
    WHAT are:

      - 1 :: Request insertion of a card.  Serialnumber may be given
             to request a specific card.  Used by gpg 1.4 w/o
             scdaemon
      - 2 :: Request removal of a card.  Used by gpg 1.4 w/o scdaemon.
      - 3 :: Card with serialnumber detected
      - 4 :: No card available
      - 5 :: No card reader available
      - 6 :: No card support available
      - 7 :: Card is in termination state

*** SC_OP_FAILURE [<code>]
    An operation on a smartcard definitely failed.  Currently there is
    no indication of the actual error code, but application should be
    prepared to later accept more arguments.  Defined values for
    <code> are:

      - 0 :: unspecified error (identically to a missing CODE)
      - 1 :: canceled
      - 2 :: bad PIN

*** SC_OP_SUCCESS
    A smart card operaion succeeded.  This status is only printed for
    certain operation and is mostly useful to check whether a PIN
    change really worked.

** Miscellaneous status codes
*** NODATA  <what>
    No data has been found.  Codes for WHAT are:

    - 1 :: No armored data.
    - 2 :: Expected a packet but did not found one.
    - 3 :: Invalid packet found, this may indicate a non OpenPGP
           message.
    - 4 :: Signature expected but not found

    You may see more than one of these status lines.

*** UNEXPECTED <what>
    Unexpected data has been encountered.  Codes for WHAT are:
    - 0 :: Not further specified
    - 1 :: Corrupted message structure

*** TRUNCATED <maxno>
    The output was truncated to MAXNO items.  This status code is
    issued for certain external requests.

*** ERROR <error location> <error code> [<more>]
    This is a generic error status message, it might be followed by
    error location specific data. <error code> and <error_location>
    should not contain spaces.  The error code is a either a string
    commencing with a letter or such a string prefixed with a
    numerical error code and an underscore; e.g.: "151011327_EOF".
*** WARNING <location> <error code> [<text>]
    This is a generic warning status message, it might be followed by
    error location specific data. <error code> and <location>
    should not contain spaces.  The error code is a either a string
    commencing with a letter or such a string prefixed with a
    numerical error code and an underscore; e.g.: "151011327_EOF".
*** SUCCESS [<location>]
    Postive confirmation that an operation succeeded.  It is used
    similar to ISO-C's EXIT_SUCCESS.  <location> is optional but if
    given should not contain spaces.  Used only with a few commands.

*** FAILURE <location> <error_code>
    This is the counterpart to SUCCESS and used to indicate a program
    failure.  It is used similar to ISO-C's EXIT_FAILURE but allows
    conveying more information, in particular a gpg-error error code.
    That numerical error code may optionally have a suffix made of an
    underscore and a string with an error symbol like "151011327_EOF".
    A dash may be used instead of <location>.

*** BADARMOR
    The ASCII armor is corrupted.  No arguments yet.

*** DELETE_PROBLEM <reason_code>
    Deleting a key failed.  Reason codes are:
    - 1 :: No such key
    - 2 :: Must delete secret key first
    - 3 :: Ambigious specification
    - 4 :: Key is stored on a smartcard.

*** PROGRESS <what> <char> <cur> <total> [<units>]
    Used by the primegen and Public key functions to indicate
    progress.  <char> is the character displayed with no --status-fd
    enabled, with the linefeed replaced by an 'X'.  <cur> is the
    current amount done and <total> is amount to be done; a <total> of
    0 indicates that the total amount is not known. The condition
      :       TOTAL && CUR == TOTAL
    may be used to detect the end of an operation.

    Well known values for WHAT are:

           - pk_dsa   :: DSA key generation
           - pk_elg   :: Elgamal key generation
           - primegen :: Prime generation
           - need_entropy :: Waiting for new entropy in the RNG
           - tick :: Generic tick without any special meaning - useful
                     for letting clients know that the server is still
                     working.
           - starting_agent :: A gpg-agent was started because it is not
                                running as a daemon.
           - learncard :: Send by the agent and gpgsm while learing
                          the data of a smartcard.
           - card_busy :: A smartcard is still working

    <units> is sometines used to describe the units for <current> and
    <total>.  For example "B", "KiB", or "MiB".

*** BACKUP_KEY_CREATED <fingerprint> <fname>
    A backup of a key identified by <fingerprint> has been writte to
    the file <fname>; <fname> is percent-escaped.

*** MOUNTPOINT <name>
    <name> is a percent-plus escaped filename describing the
    mountpoint for the current operation (e.g. used by "g13 --mount").
    This may either be the specified mountpoint or one randomly
    choosen by g13.

*** PINENTRY_LAUNCHED <pid>
    This status line is emitted by gpg to notify a client that a
    Pinentry has been launched.  <pid> is the PID of the Pinentry.  It
    may be used to display a hint to the user but can't be used to
    synchronize with Pinentry.  Note that there is also an Assuan
    inquiry line with the same name used internally or, if enabled,
    send to the client instead of this status line.  Such an inquiry
    may be used to sync with Pinentry

** Obsolete status codes
*** SIGEXPIRED
    Removed on 2011-02-04.  This is deprecated in favor of KEYEXPIRED.
*** RSA_OR_IDEA
    Obsolete.  This status message used to be emitted for requests to
    use the IDEA or RSA algorithms.  It has been dropped from GnuPG
    2.1 after the respective patents expired.
*** SHM_INFO, SHM_GET, SHM_GET_BOOL, SHM_GET_HIDDEN
    These were used for the ancient shared memory based co-processing.
*** BEGIN_STREAM, END_STREAM
    Used to issued by the experimental pipemode.


* Format of the --attribute-fd output

  When --attribute-fd is set, during key listings (--list-keys,
  --list-secret-keys) GnuPG dumps each attribute packet to the file
  descriptor specified.  --attribute-fd is intended for use with
  --status-fd as part of the required information is carried on the
  ATTRIBUTE status tag (see above).

  The contents of the attribute data is specified by RFC 4880.  For
  convenience, here is the Photo ID format, as it is currently the
  only attribute defined:

  - Byte 0-1 :: The length of the image header.  Due to a historical
                accident (i.e. oops!) back in the NAI PGP days, this
                is a little-endian number.  Currently 16 (0x10 0x00).

  - Byte 2 :: The image header version.  Currently 0x01.

  - Byte 3 :: Encoding format.  0x01 == JPEG.

  - Byte 4-15 :: Reserved, and currently unused.

  All other data after this header is raw image (JPEG) data.


* Layout of the TrustDB

  The TrustDB is built from fixed length records, where the first byte
  describes the record type.  All numeric values are stored in network
  byte order.  The length of each record is 40 bytes.  The first
  record of the DB is always of type 1 and this is the only record of
  this type.

  The record types: directory(2), key(3), uid(4), pref(5), sigrec(6),
  and shadow directory(8) are not anymore used by version 2 of the
  TrustDB.

** Record type 0

   Unused record or deleted, can be reused for any purpose.  Such
   records should in general not exist because deleted records are of
   type 254 and kept in a linked list.

** Version info (RECTYPE_VER, 1)

   Version information for this TrustDB.  This is always the first
   record of the DB and the only one of this type.

   - 1 u8 :: Record type (value: 1).
   - 3 byte :: Magic value ("gpg")
   - 1 u8 :: TrustDB version (value: 2).
   - 1 u8 :: =marginals=. How many marginal trusted keys are required.
   - 1 u8 :: =completes=. How many completely trusted keys are
             required.
   - 1 u8 :: =max_cert_depth=.  How deep is the WoT evaluated.  Along
             with =marginals= and =completes=, this value is used to
             check whether the cached validity value from a [FIXME
             dir] record can be used.
   - 1 u8 :: =trust_model=
   - 1 u8 :: =min_cert_level=
   - 2 byte :: Not used
   - 1 u32 :: =created=. Timestamp of trustdb creation.
   - 1 u32 :: =nextcheck=. Timestamp of last modification which may
              affect the validity of keys in the trustdb.  This value
              is checked against the validity timestamp in the dir
              records.
   - 1 u32 :: =reserved=.  Not used.
   - 1 u32 :: =reserved2=. Not used.
   - 1 u32 :: =firstfree=. Number of the record with the head record
              of the RECTYPE_FREE linked list.
   - 1 u32 :: =reserved3=. Not used.
   - 1 u32 :: =trusthashtbl=. Record number of the trusthashtable.


** Hash table (RECTYPE_HTBL, 10)

   Due to the fact that we use fingerprints to lookup keys, we can
   implement quick access by some simple hash methods, and avoid the
   overhead of gdbm.  A property of fingerprints is that they can be
   used directly as hash values.  What we use is a dynamic multilevel
   architecture, which combines hash tables, record lists, and linked
   lists.

   This record is a hash table of 256 entries with the property that
   all these records are stored consecutively to make one big
   table. The hash value is simple the 1st, 2nd, ... byte of the
   fingerprint (depending on the indirection level).

   - 1 u8 :: Record type (value: 10).
   - 1 u8 :: Reserved
   - n u32 :: =recnum=.  A table with the hash table items fitting into
              this record.  =n= depends on the record length:
              $n=(reclen-2)/4$ which yields 9 for oure current record
              length of 40 bytes.

   The total number of hash table records to form the table is:
   $m=(256+n-1)/n$.  This is 29 for our record length of 40.

   To look up a key we use the first byte of the fingerprint to get
   the recnum from this hash table and then look up the addressed
   record:

   - If that record is another hash table, we use 2nd byte to index
     that hash table and so on;
   - if that record is a hash list, we walk all entries until we find
     a matching one; or
   - if that record is a key record, we compare the fingerprint to
     decide whether it is the requested key;


** Hash list (RECTYPE_HLST, 11)

   See hash table above on how it is used.  It may also be used for
   other purposes.

   - 1 u8 :: Record type (value: 11).
   - 1 u8 :: Reserved.
   - 1 u32 :: =next=.  Record number of the next hash list record or 0
              if none.
   - n u32 :: =rnum=.  Array with record numbers to values.  With
              $n=(reclen-5)/5$ and our record length of 40, n is 7.

** Trust record (RECTYPE_TRUST, 12)

   - 1 u8 :: Record type (value: 12).
   - 1 u8 :: Reserved.
   - 20 byte :: =fingerprint=.
   - 1 u8 :: =ownertrust=.
   - 1 u8 :: =depth=.
   - 1 u8 :: =min_ownertrust=.
   - 1 byte :: Not used.
   - 1 u32 :: =validlist=.
   - 10 byte :: Not used.

** Validity record (RECTYPE_VALID, 13)

   - 1 u8 :: Record type (value: 13).
   - 1 u8 :: Reserved.
   - 20 byte :: =namehash=.
   - 1 u8 :: =validity=
   - 1 u32 :: =next=.
   - 1 u8 :: =full_count=.
   - 1 u8 :: =marginal_count=.
   - 11 byte :: Not used.

** Free record (RECTYPE_FREE, 254)

   All these records form a linked list of unused records in the TrustDB.

   - 1 u8 :: Record type (value: 254)
   - 1 u8 :: Reserved.
   - 1 u32 :: =next=.  Record number of the next rcord of this type.
              The record number to the head of this linked list is
              stored in the version info record.


* Database scheme for the TOFU info

#+begin_src sql
--
-- The VERSION table holds the version of our TOFU data structures.
--
CREATE TABLE version (
  version integer -- As of now this is always 1
);

--
-- The BINDINGS table associates mail addresses with keys.
--
CREATE TABLE bindings (
  oid integer primary key autoincrement,
  fingerprint text, -- The key's fingerprint in hex
  email text,       -- The normalized mail address destilled from user_id
  user_id text,     -- The unmodified user id
  time integer,     -- The time this binding was first observed.
  policy boolean check
       (policy in (1, 2, 3, 4, 5)), -- The trust policy with the values:
                                    --   1 := Auto
                                    --   2 := Good
                                    --   3 := Unknown
                                    --   4 := Bad
                                    --   5 := Ask
  conflict string,  -- NULL or a hex formatted fingerprint.
  unique (fingerprint, email)
);

CREATE INDEX bindings_fingerprint_email on bindings (fingerprint, email);
CREATE INDEX bindings_email on bindings (email);

--
-- The SIGNATURES table records all data signatures we verified
--
CREATE TABLE signatures (
  binding integer not null, -- Link to bindings table,
                            -- references bindings.oid.
  sig_digest text,          -- The digest of the signed message.
  origin text,              -- String describing who initially fed
                            -- the signature to gpg (e.g. "email:claws").
  sig_time integer,         -- Timestamp from the signature.
  time integer,             -- Time this record was created.
  primary key (binding, sig_digest, origin)
);
#+end_src


* GNU extensions to the S2K algorithm

  1 octet  - S2K Usage: either 254 or 255.
  1 octet  - S2K Cipher Algo: 0
  1 octet  - S2K Specifier: 101
  3 octets - "GNU"
  1 octet  - GNU S2K Extension Number.

  If such a GNU extension is used neither an IV nor any kind of
  checksum is used.  The defined GNU S2K Extension Numbers are:

  - 1 :: Do not store the secret part at all.  No specific data
         follows.

  - 2 :: A stub to access smartcards.  This data follows:
         - One octet with the length of the following serial number.
         - The serial number. Regardless of what the length octet
           indicates no more than 16 octets are stored.

  Note that gpg stores the GNU S2K Extension Number internally as an
  S2K Specifier with an offset of 1000.


* Keyserver helper message format

  *This information is obsolete*
  (Keyserver helpers have been replaced by dirmngr)

  The keyserver may be contacted by a Unix Domain socket or via TCP.

  The format of a request is:
#+begin_example
  command-tag
  "Content-length:" digits
  CRLF
#+end_example

  Where command-tag is

#+begin_example
  NOOP
  GET <user-name>
  PUT
  DELETE <user-name>
#+end_example

The format of a response is:

#+begin_example
  "GNUPG/1.0" status-code status-text
  "Content-length:" digits
  CRLF
#+end_example
followed by <digits> bytes of data

Status codes are:

  - 1xx :: Informational - Request received, continuing process

  - 2xx :: Success - The action was successfully received, understood,
           and accepted

  - 4xx :: Client Error - The request contains bad syntax or cannot be
           fulfilled

  - 5xx :: Server Error - The server failed to fulfill an apparently
           valid request


* Object identifiers

  OIDs below the GnuPG arc:

#+begin_example
  1.3.6.1.4.1.11591.2          GnuPG
  1.3.6.1.4.1.11591.2.1          notation
  1.3.6.1.4.1.11591.2.1.1          pkaAddress
  1.3.6.1.4.1.11591.2.2          X.509 extensions
  1.3.6.1.4.1.11591.2.2.1          standaloneCertificate
  1.3.6.1.4.1.11591.2.2.2          wellKnownPrivateKey
  1.3.6.1.4.1.11591.2.12242973   invalid encoded OID
#+end_example



* Miscellaneous notes

** v3 fingerprints
   For packet version 3 we calculate the keyids this way:
    - RSA :: Low 64 bits of n
    - ELGAMAL :: Build a v3 pubkey packet (with CTB 0x99) and
                 calculate a RMD160 hash value from it. This is used
                 as the fingerprint and the low 64 bits are the keyid.

** Simplified revocation certificates
  Revocation certificates consist only of the signature packet;
  "--import" knows how to handle this.  The rationale behind it is to
  keep them small.

** Documentation on HKP (the http keyserver protocol):

   A minimalistic HTTP server on port 11371 recognizes a GET for
   /pks/lookup.  The standard http URL encoded query parameters are
   this (always key=value):

   - op=index (like pgp -kv), op=vindex (like pgp -kvv) and op=get (like
     pgp -kxa)

   - search=<stringlist>. This is a list of words that must occur in the key.
     The words are delimited with space, points, @ and so on. The delimiters
     are not searched for and the order of the words doesn't matter (but see
     next option).

   - exact=on. This switch tells the hkp server to only report exact matching
     keys back. In this case the order and the "delimiters" are important.

   - fingerprint=on. Also reports the fingerprints when used with 'index' or
     'vindex'

   The keyserver also recognizes http-POSTs to /pks/add. Use this to upload
   keys.


   A better way to do this would be a request like:

      /pks/lookup/<gnupg_formatierte_user_id>?op=<operation>

   This can be implemented using Hurd's translator mechanism.
   However, I think the whole keyserver stuff has to be re-thought;
   I have some ideas and probably create a white paper.
** Algorithm names for the "keygen.algo" prompt

  When using a --command-fd controlled key generation or "addkey"
  there is way to know the number to enter on the "keygen.algo"
  prompt.  The displayed numbers are for human reception and may
  change with releases.  To provide a stable way to enter a desired
  algorithm choice the prompt also accepts predefined names for the
  algorithms, which will not change.

   | Name    | No | Description                     |
   |---------+----+---------------------------------|
   | rsa+rsa |  1 | RSA and RSA (default)           |
   | dsa+elg |  2 | DSA and Elgamal                 |
   | dsa     |  3 | DSA (sign only)                 |
   | rsa/s   |  4 | RSA (sign only)                 |
   | elg     |  5 | Elgamal (encrypt only)          |
   | rsa/e   |  6 | RSA (encrypt only)              |
   | dsa/*   |  7 | DSA (set your own capabilities) |
   | rsa/*   |  8 | RSA (set your own capabilities) |
   | ecc+ecc |  9 | ECC and ECC                     |
   | ecc/s   | 10 | ECC (sign only)                 |
   | ecc/*   | 11 | ECC (set your own capabilities) |
   | ecc/e   | 12 | ECC (encrypt only)              |
   | keygrip | 13 | Existing key                    |

   If one of the "foo/*" names are used a "keygen.flags" prompt needs
   to be answered as well.  Instead of toggling the predefined flags,
   it is also possible to set them direct: Use a "=" character
   directly followed by a comination of "a" (for authentication), "s"
   (for signing), or "c" (for certification).