2 *NOTE* This file is a copy of doc/DETAILS file that comes
3 with gnupg-1.9.94 distribution.
5 Format of colon listings
6 ========================
9 $ gpg --fixed-list-mode --with-colons --list-keys \
10 --with-fingerprint --with-fingerprint wk@gnupg.org
12 pub:f:1024:17:6C7EE1B8621CC013:899817715:1055898235::m:::scESC:
13 fpr:::::::::ECAF7590EB3443B5C7CF3ACB6C7EE1B8621CC013:
14 uid:f::::::::Werner Koch <wk@g10code.com>:
15 uid:f::::::::Werner Koch <wk@gnupg.org>:
16 sub:f:1536:16:06AD222CADF6A6E1:919537416:1036177416:::::e:
17 fpr:::::::::CF8BCC4B18DE08FCD8A1615906AD222CADF6A6E1:
18 sub:r:1536:20:5CE086B5B5A18FF4:899817788:1025961788:::::esc:
19 fpr:::::::::AB059359A3B81F410FCFF97F5CE086B5B5A18FF4:
21 The double --with-fingerprint prints the fingerprint for the subkeys
22 too, --fixed-list-mode is themodern listing way printing dates in
23 seconds since Epoch and does not merge the first userID with the pub
27 1. Field: Type of record
29 crt = X.509 certificate
30 crs = X.509 certificate and private key available
31 sub = subkey (secondary key)
33 ssb = secret subkey (secondary key)
34 uid = user id (only field 10 is used).
35 uat = user attribute (same as user id except for field 10).
37 rev = revocation signature
38 fpr = fingerprint: (fingerprint is in field 10)
39 pkd = public key data (special field format, see below)
40 grp = reserved for gpgsm
42 tru = trust database information
43 spk = signature subpacket
45 2. Field: A letter describing the calculated trust. This is a single
46 letter, but be prepared that additional information may follow
47 in some future versions. (not used for secret keys)
48 o = Unknown (this key is new to the system)
49 i = The key is invalid (e.g. due to a missing self-signature)
50 d = The key has been disabled
51 (deprecated - use the 'D' in field 12 instead)
52 r = The key has been revoked
53 e = The key has expired
54 - = Unknown trust (i.e. no value assigned)
56 '-' and 'q' may safely be treated as the same
57 value for most purposes
58 n = Don't trust this key at all
59 m = There is marginal trust in this key
60 f = The key is fully trusted
61 u = The key is ultimately trusted. This often means
62 that the secret key is available, but any key may
63 be marked as ultimately trusted.
64 3. Field: length of key in bits.
65 4. Field: Algorithm: 1 = RSA
66 16 = Elgamal (encrypt only)
67 17 = DSA (sometimes called DH, sign only)
68 20 = Elgamal (sign and encrypt - don't use them!)
69 (for other id's see include/cipher.h)
71 6. Field: Creation Date (in UTC). For UID and UAT records, this is the
72 self-signature date. Note that the dae is usally printed
73 in seconds since epoch, however, we are migrating to an ISO
74 8601 format (e.g. "19660205T091500"). This is currently
75 only relevant for X.509, A simple way to detect the format
76 is be scannning for the 'T'.
77 7. Field: Key or user ID/user attribute expiration date or empty if none.
78 8. Field: Used for serial number in crt records (used to be the Local-ID).
79 For UID and UAT records, this is a hash of the user ID contents
80 used to represent that exact user ID. For trust signatures,
81 this is the trust depth seperated by the trust value by a
83 9. Field: Ownertrust (primary public keys only)
84 This is a single letter, but be prepared that additional
85 information may follow in some future versions. For trust
86 signatures with a regular expression, this is the regular
87 expression value, quoted as in field 10.
88 10. Field: User-ID. The value is quoted like a C string to avoid
89 control characters (the colon is quoted "\x3a").
90 This is not used with --fixed-list-mode in gpg.
91 A UAT record puts the attribute subpacket count here, a
92 space, and then the total attribute subpacket size.
93 In gpgsm the issuer name comes here
94 An FPR record stores the fingerprint here.
95 The fingerprint of an revocation key is stored here.
96 11. Field: Signature class. This is a 2 digit hexnumber followed by
97 either the letter 'x' for an exportable signature or the
98 letter 'l' for a local-only signature.
99 The class byte of an revocation key is also given here,
100 'x' and 'l' ist used the same way.
101 12. Field: Key capabilities:
106 A key may have any combination of them in any order. In
107 addition to these letters, the primary key has uppercase
108 versions of the letters to denote the _usable_
109 capabilities of the entire key, and a potential letter 'D'
110 to indicate a disabled key.
111 13. Field: Used in FPR records for S/MIME keys to store the fingerprint of
112 the issuer certificate. This is useful to build the
113 certificate path based on certificates stored in the local
114 keyDB; it is only filled if the issue certificate is
115 available. The advantage of using this value is that it is
116 guaranteed to have been been build by the same lookup
117 algorithm as gpgsm uses.
118 For "uid" recods this lists the preferences n the sameway the
120 For "sig" records, this is the fingerprint of the key that
121 issued the signature. Note that this is only filled in if
122 the signature verified correctly. Note also that for
123 various technical reasons, this fingerprint is only
124 available if --no-sig-cache is used.
126 14. Field Flag field used in the --edit menu output:
128 15. Field Used in sec/sbb to print the serial number of a token
129 (internal protect mode 1002) or a '#' if that key is a
130 simple stub (internal protect mode 1001)
132 All dates are displayed in the format yyyy-mm-dd unless you use the
133 option --fixed-list-mode in which case they are displayed as seconds
134 since Epoch. More fields may be added later, so parsers should be
135 prepared for this. When parsing a number the parser should stop at the
136 first non-number character so that additional information can later be
139 If field 1 has the tag "pkd", a listing looks like this:
140 pkd:0:1024:B665B1435F4C2 .... FF26ABB:
142 ! !------ for information number of bits in the value
143 !--------- index (eg. DSA goes from 0 to 3: p,q,g,y)
146 The "tru" trust database records have the fields:
148 2: Reason for staleness of trust. If this field is empty, then the
149 trustdb is not stale. This field may have multiple flags in it:
152 t: Trustdb was built with a different trust model than the one we
156 0: Classic trust model, as used in PGP 2.x.
157 1: PGP trust model, as used in PGP 6 and later. This is the same
158 as the classic trust model, except for the addition of trust
161 GnuPG before version 1.4 used the classic trust model by default.
162 GnuPG 1.4 and later uses the PGP trust model by default.
164 4: Date trustdb was created in seconds since 1/1/1970.
165 5: Date trustdb will expire in seconds since 1/1/1970.
167 The "spk" signature subpacket records have the fields:
169 2: Subpacket number as per RFC-2440 and later.
170 3: Flags in hex. Currently the only two bits assigned are 1, to
171 indicate that the subpacket came from the hashed part of the
172 signature, and 2, to indicate the subpacket was marked critical.
173 4: Length of the subpacket. Note that this is the length of the
174 subpacket, and not the length of field 5 below. Due to the need
175 for %-encoding, the length of field 5 may be up to 3x this value.
176 5: The subpacket data. Printable ASCII is shown as ASCII, but other
177 values are rendered as %XX where XX is the hex value for the byte.
180 Format of the "--status-fd" output
181 ==================================
182 Every line is prefixed with "[GNUPG:] ", followed by a keyword with
183 the type of the status line and a some arguments depending on the
184 type (maybe none); an application should always be prepared to see
185 more arguments in future versions.
189 May be issued right before a signature verification starts. This
190 is useful to define a context for parsing ERROR status
191 messages. No arguments are currently defined.
193 GOODSIG <long keyid> <username>
194 The signature with the keyid is good. For each signature only
195 one of the three codes GOODSIG, BADSIG or ERRSIG will be
196 emitted and they may be used as a marker for a new signature.
197 The username is the primary one encoded in UTF-8 and %XX
200 EXPSIG <long keyid> <username>
201 The signature with the keyid is good, but the signature is
202 expired. The username is the primary one encoded in UTF-8 and
205 EXPKEYSIG <long keyid> <username>
206 The signature with the keyid is good, but the signature was
207 made by an expired key. The username is the primary one
208 encoded in UTF-8 and %XX escaped.
210 REVKEYSIG <long keyid> <username>
211 The signature with the keyid is good, but the signature was
212 made by a revoked key. The username is the primary one
213 encoded in UTF-8 and %XX escaped.
215 BADSIG <long keyid> <username>
216 The signature with the keyid has not been verified okay.
217 The username is the primary one encoded in UTF-8 and %XX
220 ERRSIG <long keyid> <pubkey_algo> <hash_algo> \
221 <sig_class> <timestamp> <rc>
222 It was not possible to check the signature. This may be
223 caused by a missing public key or an unsupported algorithm.
224 A RC of 4 indicates unknown algorithm, a 9 indicates a missing
225 public key. The other fields give more information about
226 this signature. sig_class is a 2 byte hex-value.
228 Note, that TIMESTAMP may either be a number with seconds since
229 epoch or an ISO 8601 string which can be detected by the
230 presence of the letter 'T' inside.
232 VALIDSIG <fingerprint in hex> <sig_creation_date> <sig-timestamp>
233 <expire-timestamp> <sig-version> <reserved> <pubkey-algo>
234 <hash-algo> <sig-class> <primary-key-fpr>
236 The signature with the keyid is good. This is the same as
237 GOODSIG but has the fingerprint as the argument. Both status
238 lines are emitted for a good signature. All arguments here
239 are on one long line. sig-timestamp is the signature creation
240 time in seconds after the epoch. expire-timestamp is the
241 signature expiration time in seconds after the epoch (zero
242 means "does not expire"). sig-version, pubkey-algo, hash-algo,
243 and sig-class (a 2-byte hex value) are all straight from the
244 signature packet. PRIMARY-KEY-FPR is the fingerprint of the
245 primary key or identical to the first argument. This is
246 useful to get back to the primary key without running gpg
247 again for this purpose.
249 Note, that *-TIMESTAMP may either be a number with seconds
250 since epoch or an ISO 8601 string which can be detected by the
251 presence of the letter 'T' inside.
253 SIG_ID <radix64_string> <sig_creation_date> <sig-timestamp>
254 This is emitted only for signatures of class 0 or 1 which
255 have been verified okay. The string is a signature id
256 and may be used in applications to detect replay attacks
257 of signed messages. Note that only DLP algorithms give
258 unique ids - others may yield duplicated ones when they
259 have been created in the same second.
261 Note, that SIG-TIMESTAMP may either be a number with seconds
262 since epoch or an ISO 8601 string which can be detected by the
263 presence of the letter 'T' inside.
266 ENC_TO <long keyid> <keytype> <keylength>
267 The message is encrypted to this keyid.
268 keytype is the numerical value of the public key algorithm,
269 keylength is the length of the key or 0 if it is not known
270 (which is currently always the case).
273 No data has been found. Codes for what are:
275 2 - Expected a packet but did not found one.
276 3 - Invalid packet found, this may indicate a non OpenPGP
278 4 - signature expected but not found
279 You may see more than one of these status lines.
282 Unexpected data has been encountered
283 0 - not further specified 1
286 TRUST_UNDEFINED <error token>
287 TRUST_NEVER <error token>
291 For good signatures one of these status lines are emitted
292 to indicate how trustworthy the signature is. The error token
293 values are currently only emiited by gpgsm.
295 PKA_TRUST_GOOD <mailbox>
296 PKA_TRUST_BAD <mailbox>
297 Depending on the outcome of the PKA check one of the above
298 status codes is emitted in addition to a TRUST_* status.
299 Without PKA info available or
302 This is deprecated in favor of KEYEXPIRED.
304 KEYEXPIRED <expire-timestamp>
305 The key has expired. expire-timestamp is the expiration time
306 in seconds after the epoch.
308 Note, that TIMESTAMP may either be a number with seconds since
309 epoch or an ISO 8601 string which can be detected by the
310 presence of the letter 'T' inside.
313 The used key has been revoked by its owner. No arguments yet.
316 The ASCII armor is corrupted. No arguments yet.
319 The IDEA algorithms has been used in the data. A
320 program might want to fallback to another program to handle
321 the data if GnuPG failed. This status message used to be emitted
322 also for RSA but this has been dropped after the RSA patent expired.
323 However we can't change the name of the message.
335 NEED_PASSPHRASE <long main keyid> <long keyid> <keytype> <keylength>
336 Issued whenever a passphrase is needed.
337 keytype is the numerical value of the public key algorithm
338 or 0 if this is not applicable, keylength is the length
339 of the key or 0 if it is not known (this is currently always the case).
341 NEED_PASSPHRASE_SYM <cipher_algo> <s2k_mode> <s2k_hash>
342 Issued whenever a passphrase for symmetric encryption is needed.
344 NEED_PASSPHRASE_PIN <card_type> <chvno> [<serialno>]
345 Issued whenever a PIN is requested to unlock a card.
348 No passphrase was supplied. An application which encounters this
349 message may want to stop parsing immediately because the next message
350 will probably be a BAD_PASSPHRASE. However, if the application
351 is a wrapper around the key edit menu functionality it might not
352 make sense to stop parsing but simply ignoring the following
355 BAD_PASSPHRASE <long keyid>
356 The supplied passphrase was wrong or not given. In the latter case
357 you may have seen a MISSING_PASSPHRASE.
360 The supplied passphrase was good and the secret key material
364 The symmetric decryption failed - one reason could be a wrong
365 passphrase for a symmetrical encrypted message.
368 The decryption process succeeded. This means, that either the
369 correct secret key has been used or the correct passphrase
370 for a conventional encrypted message was given. The program
371 itself may return an errorcode because it may not be possible to
372 verify a signature for some reasons.
374 NO_PUBKEY <long keyid>
375 NO_SECKEY <long keyid>
376 The key is not available
378 IMPORT_CHECK <long keyid> <fingerprint> <user ID>
379 This status is emitted in interactive mode right before
380 the "import.okay" prompt.
382 IMPORTED <long keyid> <username>
383 The keyid and name of the signature just imported
385 IMPORT_OK <reason> [<fingerprint>]
386 The key with the primary key's FINGERPRINT has been imported.
388 0 := Not actually changed
389 1 := Entirely new key.
393 16 := Contains private key.
394 The flags may be ORed.
396 IMPORT_PROBLEM <reason> [<fingerprint>]
397 Issued for each import failure. Reason codes are:
398 0 := "No specific reason given".
399 1 := "Invalid Certificate".
400 2 := "Issuer Certificate missing".
401 3 := "Certificate Chain too long".
402 4 := "Error storing certificate".
404 IMPORT_RES <count> <no_user_id> <imported> <imported_rsa> <unchanged>
405 <n_uids> <n_subk> <n_sigs> <n_revoc> <sec_read> <sec_imported> <sec_dups> <not_imported>
406 Final statistics on import process (this is one long line)
408 FILE_START <what> <filename>
409 Start processing a file <filename>. <what> indicates the performed
416 Marks the end of a file processing which has been started
421 Mark the start and end of the actual decryption process. These
422 are also emitted when in --list-only mode.
424 BEGIN_ENCRYPTION <mdc_method> <sym_algo>
426 Mark the start and end of the actual encryption process.
429 Mark the start of the actual signing process. This may be used
430 as an indication that all requested secret keys are ready for
433 DELETE_PROBLEM reason_code
434 Deleting a key failed. Reason codes are:
436 2 - Must delete secret key first
437 3 - Ambigious specification
439 PROGRESS what char cur total
440 Used by the primegen and Public key functions to indicate progress.
441 "char" is the character displayed with no --status-fd enabled, with
442 the linefeed replaced by an 'X'. "cur" is the current amount
443 done and "total" is amount to be done; a "total" of 0 indicates that
444 the total amount is not known. 100/100 may be used to detect the
446 Well known values for WHAT:
447 "pk_dsa" - DSA key generation
448 "pk_elg" - Elgamal key generation
449 "primegen" - Prime generation
450 "need_entropy" - Waiting for new entropy in the RNG
451 "file:XXX" - processing file XXX
452 (note that current gpg versions leave out the
454 "tick" - generic tick without any special meaning - useful
455 for letting clients know that the server is
457 "starting_agent" - A gpg-agent was started because it is not
461 SIG_CREATED <type> <pubkey algo> <hash algo> <class> <timestamp> <key fpr>
462 A signature has been created using these parameters.
466 (only the first character should be checked)
467 class: 2 hex digits with the signature class
469 Note, that TIMESTAMP may either be a number with seconds since
470 epoch or an ISO 8601 string which can be detected by the
471 presence of the letter 'T' inside.
473 KEY_CREATED <type> <fingerprint> [<handle>]
474 A key has been created
475 type: 'B' = primary and subkey
478 The fingerprint is one of the primary key for type B and P and
479 the one of the subkey for S. Handle is an arbitrary
480 non-whitespace string used to match key parameters from batch
483 KEY_NOT_CREATED [<handle>]
484 The key from batch run has not been created due to errors.
487 SESSION_KEY <algo>:<hexdigits>
488 The session key used to decrypt the message. This message will
489 only be emitted when the special option --show-session-key
490 is used. The format is suitable to be passed to the option
491 --override-session-key
494 NOTATION_DATA <string>
495 name and string are %XX escaped; the data may be splitted
496 among several notation_data lines.
498 USERID_HINT <long main keyid> <string>
499 Give a hint about the user ID for a certain keyID.
502 string is %XX escaped
508 INV_RECP <reason> <requested_recipient>
509 Issued for each unusable recipient. The reasons codes
510 currently in use are:
511 0 := "No specific reason given".
513 2 := "Ambigious specification"
514 3 := "Wrong key usage"
519 8 := "Policy mismatch"
520 9 := "Not a secret key"
521 10 := "Key not trusted"
523 Note that this status is also used for gpgsm's SIGNER command
524 where it relates to signer's of course.
527 Issued when no recipients are usable.
529 ALREADY_SIGNED <long-keyid>
530 Warning: This is experimental and might be removed at any time.
533 The output was truncated to MAXNO items. This status code is issued
534 for certain external requests
536 ERROR <error location> <error code>
538 This is a generic error status message, it might be followed
539 by error location specific data. <error token> and
540 <error_location> should not contain a space. The error code
541 is a either a string commencing with a letter or such string
542 prefix with a numerical error code and an underscore; e.g.:
545 ATTRIBUTE <fpr> <octets> <type> <index> <count>
546 <timestamp> <expiredate> <flags>
547 This is one long line issued for each attribute subpacket when
548 an attribute packet is seen during key listing. <fpr> is the
549 fingerprint of the key. <octets> is the length of the
550 attribute subpacket. <type> is the attribute type
551 (1==image). <index>/<count> indicates that this is the Nth
552 indexed subpacket of count total subpackets in this attribute
553 packet. <timestamp> and <expiredate> are from the
554 self-signature on the attribute packet. If the attribute
555 packet does not have a valid self-signature, then the
556 timestamp is 0. <flags> are a bitwise OR of:
557 0x01 = this attribute packet is a primary uid
558 0x02 = this attribute packet is revoked
559 0x04 = this attribute packet is expired
561 CARDCTRL <what> [<serialno>]
562 This is used to control smartcard operations.
563 Defined values for WHAT are:
564 1 = Request insertion of a card. Serialnumber may be given
565 to request a specific card.
566 2 = Request removal of a card.
567 3 = Card with serialnumber detected
568 4 = No card available.
569 5 = No card reader available
572 PLAINTEXT <format> <timestamp> <filename>
573 This indicates the format of the plaintext that is about to be
574 written. The format is a 1 byte hex code that shows the
575 format of the plaintext: 62 ('b') is binary data, 74 ('t') is
576 text data with no character set specified, and 75 ('u') is
577 text data encoded in the UTF-8 character set. The timestamp
578 is in seconds since the epoch. If a filename is available it
579 gets printed as the third argument, percent-escaped as usual.
581 PLAINTEXT_LENGTH <length>
582 This indicates the length of the plaintext that is about to be
583 written. Note that if the plaintext packet has partial length
584 encoding it is not possible to know the length ahead of time.
585 In that case, this status tag does not appear.
587 SIG_SUBPACKET <type> <flags> <len> <data>
588 This indicates that a signature subpacket was seen. The
589 format is the same as the "spk" record above.
591 SC_OP_FAILURE [<code>]
592 An operation on a smartcard definitely failed. Currently
593 there is no indication of the actual error code, but
594 application should be prepared to later accept more arguments.
595 Defined values for CODE are:
596 0 - unspecified error (identically to a missing CODE)
601 A smart card operaion succeeded. This status is only printed
602 for certain operation and is mostly useful to check whether a
603 PIN change really worked.
605 BACKUP_KEY_CREATED fingerprint fname
606 A backup key named FNAME has been created for the key with
610 Format of the "--attribute-fd" output
611 =====================================
613 When --attribute-fd is set, during key listings (--list-keys,
614 --list-secret-keys) GnuPG dumps each attribute packet to the file
615 descriptor specified. --attribute-fd is intended for use with
616 --status-fd as part of the required information is carried on the
617 ATTRIBUTE status tag (see above).
619 The contents of the attribute data is specified by 2440bis, but for
620 convenience, here is the Photo ID format, as it is currently the only
623 Byte 0-1: The length of the image header. Due to a historical
624 accident (i.e. oops!) back in the NAI PGP days, this is
625 a little-endian number. Currently 16 (0x10 0x00).
627 Byte 2: The image header version. Currently 0x01.
629 Byte 3: Encoding format. 0x01 == JPEG.
631 Byte 4-15: Reserved, and currently unused.
633 All other data after this header is raw image (JPEG) data.
636 Format of the "--list-config" output
637 ====================================
639 --list-config outputs information about the GnuPG configuration for
640 the benefit of frontends or other programs that call GnuPG. There are
641 several list-config items, all colon delimited like the rest of the
642 --with-colons output. The first field is always "cfg" to indicate
643 configuration information. The second field is one of (with
646 version: the third field contains the version of GnuPG.
650 pubkey: the third field contains the public key algorithmdcaiphers
651 this version of GnuPG supports, separated by semicolons. The
652 algorithm numbers are as specified in RFC-2440.
654 cfg:pubkey:1;2;3;16;17
656 cipher: the third field contains the symmetric ciphers this version of
657 GnuPG supports, separated by semicolons. The cipher numbers
658 are as specified in RFC-2440.
660 cfg:cipher:2;3;4;7;8;9;10
662 digest: the third field contains the digest (hash) algorithms this
663 version of GnuPG supports, separated by semicolons. The
664 digest numbers are as specified in RFC-2440.
666 cfg:digest:1;2;3;8;9;10
668 compress: the third field contains the compression algorithms this
669 version of GnuPG supports, separated by semicolons. The
670 algorithm numbers are as specified in RFC-2440.
674 group: the third field contains the name of the group, and the fourth
675 field contains the values that the group expands to, separated
678 For example, a group of:
679 group mynames = paige 0x12345678 joe patti
682 cfg:group:mynames:patti;joe;0x12345678;paige
687 Key generation shows progress by printing different characters to
689 "." Last 10 Miller-Rabin tests failed
690 "+" Miller-Rabin test succeeded
691 "!" Reloading the pool with fresh prime numbers
692 "^" Checking a new value for the generator
693 "<" Size of one factor decreased
694 ">" Size of one factor increased
696 The prime number for Elgamal is generated this way:
698 1) Make a prime number q of 160, 200, 240 bits (depending on the keysize)
699 2) Select the length of the other prime factors to be at least the size
700 of q and calculate the number of prime factors needed
701 3) Make a pool of prime numbers, each of the length determined in step 2
702 4) Get a new permutation out of the pool or continue with step 3
703 if we have tested all permutations.
704 5) Calculate a candidate prime p = 2 * q * p[1] * ... * p[n] + 1
705 6) Check that this prime has the correct length (this may change q if
706 it seems not to be possible to make a prime of the desired length)
707 7) Check whether this is a prime using trial divisions and the
709 8) Continue with step 4 if we did not find a prime in step 7.
710 9) Find a generator for that prime.
712 This algorithm is based on Lim and Lee's suggestion from the
713 Crypto '97 proceedings p. 260.
716 Unattended key generation
717 =========================
718 This feature allows unattended generation of keys controlled by a
719 parameter file. To use this feature, you use --gen-key together with
720 --batch and feed the parameters either from stdin or from a file given
723 The format of this file is as follows:
724 o Text only, line length is limited to about 1000 chars.
725 o You must use UTF-8 encoding to specify non-ascii characters.
726 o Empty lines are ignored.
727 o Leading and trailing spaces are ignored.
728 o A hash sign as the first non white space character indicates a comment line.
729 o Control statements are indicated by a leading percent sign, the
730 arguments are separated by white space from the keyword.
731 o Parameters are specified by a keyword, followed by a colon. Arguments
732 are separated by white space.
733 o The first parameter must be "Key-Type", control statements
734 may be placed anywhere.
735 o Key generation takes place when either the end of the parameter file
736 is reached, the next "Key-Type" parameter is encountered or at the
737 control statement "%commit"
738 o Control statements:
742 Suppress actual key generation (useful for syntax checking).
744 Perform the key generation. An implicit commit is done
745 at the next "Key-Type" parameter.
748 Do not write the key to the default or commandline given
749 keyring but to <filename>. This must be given before the first
750 commit to take place, duplicate specification of the same filename
751 is ignored, the last filename before a commit is used.
752 The filename is used until a new filename is used (at commit points)
753 and all keys are written to that file. If a new filename is given,
754 this file is created (and overwrites an existing one).
755 Both control statements must be given.
756 o The order of the parameters does not matter except for "Key-Type"
757 which must be the first parameter. The parameters are only for the
758 generated keyblock and parameters from previous key generations are not
759 used. Some syntactically checks may be performed.
760 The currently defined parameters are:
761 Key-Type: <algo-number>|<algo-string>
762 Starts a new parameter block by giving the type of the
763 primary key. The algorithm must be capable of signing.
764 This is a required parameter.
765 Key-Length: <length-in-bits>
766 Length of the key in bits. Default is 1024.
767 Key-Usage: <usage-list>
768 Space or comma delimited list of key usage, allowed values are
769 "encrypt", "sign", and "auth". This is used to generate the
770 key flags. Please make sure that the algorithm is capable of
771 this usage. Note that OpenPGP requires that all primary keys
772 are capable of certification, so no matter what usage is given
773 here, the "cert" flag will be on. If no Key-Usage is
774 specified, all the allowed usages for that particular
776 Subkey-Type: <algo-number>|<algo-string>
777 This generates a secondary key. Currently only one subkey
779 Subkey-Length: <length-in-bits>
780 Length of the subkey in bits. Default is 1024.
781 Subkey-Usage: <usage-list>
782 Similar to Key-Usage.
784 If you want to specify a passphrase for the secret key,
785 enter it here. Default is not to use any passphrase.
787 Name-Comment: <string>
789 The 3 parts of a key. Remember to use UTF-8 here.
790 If you don't give any of them, no user ID is created.
791 Expire-Date: <iso-date>|(<number>[d|w|m|y])
792 Set the expiration date for the key (and the subkey). It
793 may either be entered in ISO date format (2000-08-15) or as
794 number of days, weeks, month or years. Without a letter days
796 Preferences: <string>
797 Set the cipher, hash, and compression preference values for
798 this key. This expects the same type of string as "setpref"
800 Revoker: <algo>:<fpr> [sensitive]
801 Add a designated revoker to the generated key. Algo is the
802 public key algorithm of the designated revoker (i.e. RSA=1,
803 DSA=17, etc.) Fpr is the fingerprint of the designated
804 revoker. The optional "sensitive" flag marks the designated
805 revoker as sensitive information. Only v4 keys may be
808 This is an optional parameter only used with the status lines
809 KEY_CREATED and KEY_NOT_CREATED. STRING may be up to 100
810 characters and should not contain spaces. It is useful for
811 batch key generation to associate a key parameter block with a
814 This is an optional parameter that specifies the preferred
815 keyserver URL for the key.
820 %echo Generating a standard key
825 Name-Real: Joe Tester
826 Name-Comment: with stupid passphrase
827 Name-Email: joe@foo.bar
832 # Do a commit here, so that we can later print "done" :-)
836 $ gpg --batch --gen-key foo
838 $ gpg --no-default-keyring --secret-keyring ./foo.sec \
839 --keyring ./foo.pub --list-secret-keys
840 /home/wk/work/gnupg-stable/scratch/foo.sec
841 ------------------------------------------
842 sec 1024D/915A878D 2000-03-09 Joe Tester (with stupid passphrase) <joe@foo.bar>
843 ssb 1024g/8F70E2C0 2000-03-09
847 Layout of the TrustDB
848 =====================
849 The TrustDB is built from fixed length records, where the first byte
850 describes the record type. All numeric values are stored in network
851 byte order. The length of each record is 40 bytes. The first record of
852 the DB is always of type 1 and this is the only record of this type.
854 FIXME: The layout changed, document it here.
858 Unused record, can be reused for any purpose.
862 Version information for this TrustDB. This is always the first
863 record of the DB and the only one with type 1.
865 3 bytes 'gpg' magic value
866 1 byte Version of the TrustDB (2)
867 1 byte marginals needed
868 1 byte completes needed
869 1 byte max_cert_depth
870 The three items are used to check whether the cached
871 validity value from the dir record can be used.
872 1 u32 locked flags [not used]
873 1 u32 timestamp of trustdb creation
874 1 u32 timestamp of last modification which may affect the validity
875 of keys in the trustdb. This value is checked against the
876 validity timestamp in the dir records.
877 1 u32 timestamp of last validation [currently not used]
878 (Used to keep track of the time, when this TrustDB was checked
880 1 u32 record number of keyhashtable [currently not used]
881 1 u32 first free record
882 1 u32 record number of shadow directory hash table [currently not used]
883 It does not make sense to combine this table with the key table
884 because the keyid is not in every case a part of the fingerprint.
885 1 u32 record number of the trusthashtbale
888 Record type 2: (directory record)
890 Informations about a public key certificate.
891 These are static values which are never changed without user interaction.
895 1 u32 LID . (This is simply the record number of this record.)
896 1 u32 List of key-records (the first one is the primary key)
897 1 u32 List of uid-records
901 1 byte maximum validity of all the user ids
902 1 u32 time of last validity check.
903 1 u32 Must check when this time has been reached.
904 (0 = no check required)
907 Record type 3: (key record)
909 Informations about a primary public key.
910 (This is mainly used to lookup a trust record)
915 1 u32 next - next key record
918 1 byte pubkey algorithm
919 1 byte length of the fingerprint (in bytes)
920 20 bytes fingerprint of the public key
921 (This is the value we use to identify a key)
923 Record type 4: (uid record)
925 Informations about a userid
926 We do not store the userid but the hash value of the userid because that
931 1 u32 LID points to the directory record.
932 1 u32 next next userid
933 1 u32 pointer to preference record
934 1 u32 siglist list of valid signatures
936 1 byte validity of the key calculated over this user id
937 20 bytes ripemd160 hash of the username.
940 Record type 5: (pref record)
942 This record type is not anymore used.
946 1 u32 LID; points to the directory record (and not to the uid record!).
947 (or 0 for standard preference record)
949 30 byte preference data
951 Record type 6 (sigrec)
953 Used to keep track of key signatures. Self-signatures are not
954 stored. If a public key is not in the DB, the signature points to
955 a shadow dir record, which in turn has a list of records which
956 might be interested in this key (and the signature record here
961 1 u32 LID points back to the dir record
962 1 u32 next next sigrec of this uid or 0 to indicate the
965 1 u32 Local_id of signatures dir or shadow dir record
966 1 byte Flag: Bit 0 = checked: Bit 1 is valid (we have a real
967 directory record for this)
968 1 = valid is set (but may be revoked)
972 Record type 8: (shadow directory record)
974 This record is used to reserve a LID for a public key. We
975 need this to create the sig records of other keys, even if we
976 do not yet have the public key of the signature.
977 This record (the record number to be more precise) will be reused
978 as the dir record when we import the real public key.
982 1 u32 LID (This is simply the record number of this record.)
984 1 byte pubkey algorithm
986 1 u32 hintlist A list of records which have references to
987 this key. This is used for fast access to
988 signature records which are not yet checked.
989 Note, that this is only a hint and the actual records
990 may not anymore hold signature records for that key
991 but that the code cares about this.
996 Record Type 10 (hash table)
998 Due to the fact that we use fingerprints to lookup keys, we can
999 implement quick access by some simple hash methods, and avoid
1000 the overhead of gdbm. A property of fingerprints is that they can be
1001 used directly as hash values. (They can be considered as strong
1003 What we use is a dynamic multilevel architecture, which combines
1004 hashtables, record lists, and linked lists.
1006 This record is a hashtable of 256 entries; a special property
1007 is that all these records are stored consecutively to make one
1008 big table. The hash value is simple the 1st, 2nd, ... byte of
1009 the fingerprint (depending on the indirection level).
1011 When used to hash shadow directory records, a different table is used
1012 and indexed by the keyid.
1016 n u32 recnum; n depends on the record length:
1017 n = (reclen-2)/4 which yields 9 for the current record length
1020 the total number of such record which makes up the table is:
1022 which is 29 for a record length of 40.
1024 To look up a key we use the first byte of the fingerprint to get
1025 the recnum from this hashtable and look up the addressed record:
1026 - If this record is another hashtable, we use 2nd byte
1027 to index this hash table and so on.
1028 - if this record is a hashlist, we walk all entries
1029 until we found one a matching one.
1030 - if this record is a key record, we compare the
1031 fingerprint and to decide whether it is the requested key;
1034 Record type 11 (hash list)
1036 see hash table for an explanation.
1037 This is also used for other purposes.
1041 1 u32 next next hash list record
1042 n times n = (reclen-5)/5
1045 For the current record length of 40, n is 7
1049 Record type 254 (free record)
1051 All these records form a linked list of unused records.
1061 GNUPG uses PGP 2 packet headers and also understands OpenPGP packet header.
1062 There is one enhancement used with the old style packet headers:
1064 CTB bits 10, the "packet-length length bits", have values listed in
1065 the following table:
1067 00 - 1-byte packet-length field
1068 01 - 2-byte packet-length field
1069 10 - 4-byte packet-length field
1070 11 - no packet length supplied, unknown packet length
1072 As indicated in this table, depending on the packet-length length
1073 bits, the remaining 1, 2, 4, or 0 bytes of the packet structure field
1074 are a "packet-length field". The packet-length field is a whole
1075 number field. The value of the packet-length field is defined to be
1076 the value of the whole number field.
1078 A value of 11 is currently used in one place: on compressed data.
1079 That is, a compressed data block currently looks like <A3 01 . . .>,
1080 where <A3>, binary 10 1000 11, is an indefinite-length packet. The
1081 proper interpretation is "until the end of the enclosing structure",
1082 although it should never appear outermost (where the enclosing
1083 structure is a file).
1085 + This will be changed with another version, where the new meaning of
1086 + the value 11 (see below) will also take place.
1088 + A value of 11 for other packets enables a special length encoding,
1089 + which is used in case, where the length of the following packet can
1090 + not be determined prior to writing the packet; especially this will
1091 + be used if large amounts of data are processed in filter mode.
1093 + It works like this: After the CTB (with a length field of 11) a
1094 + marker field is used, which gives the length of the following datablock.
1095 + This is a simple 2 byte field (MSB first) containing the amount of data
1096 + following this field, not including this length field. After this datablock
1097 + another length field follows, which gives the size of the next datablock.
1098 + A value of 0 indicates the end of the packet. The maximum size of a
1099 + data block is limited to 65534, thereby reserving a value of 0xffff for
1100 + future extensions. These length markers must be inserted into the data
1101 + stream just before writing the data out.
1103 + This 2 byte field is large enough, because the application must buffer
1104 + this amount of data to prepend the length marker before writing it out.
1105 + Data block sizes larger than about 32k doesn't make any sense. Note
1106 + that this may also be used for compressed data streams, but we must use
1107 + another packet version to tell the application that it can not assume,
1108 + that this is the last packet.
1111 GNU extensions to the S2K algorithm
1112 ===================================
1113 S2K mode 101 is used to identify these extensions.
1114 After the hash algorithm the 3 bytes "GNU" are used to make
1115 clear that these are extensions for GNU, the next bytes gives the
1116 GNU protection mode - 1000. Defined modes are:
1117 1001 - do not store the secret part at all
1118 1002 - a stub to access smartcards (not used in 1.2.x)
1123 NOTE: This is deprecated and will be removed in future versions.
1125 This mode can be used to perform multiple operations with one call to
1126 gpg. It comes handy in cases where you have to verify a lot of
1127 signatures. Currently we support only detached signatures. This mode
1128 is a kludge to avoid running gpg n daemon mode and using Unix Domain
1129 Sockets to pass the data to it. There is no easy portable way to do
1130 this under Windows, so we use plain old pipes which do work well under
1131 Windows. Because there is no way to signal multiple EOFs in a pipe we
1132 have to embed control commands in the data stream: We distinguish
1133 between a data state and a control state. Initially the system is in
1134 data state but it won't accept any data. Instead it waits for
1135 transition to control state which is done by sending a single '@'
1136 character. While in control state the control command os expected and
1137 this command is just a single byte after which the system falls back
1138 to data state (but does not necesary accept data now). The simplest
1139 control command is a '@' which just inserts this character into the
1142 Here is the format we use for detached signatures:
1143 "@<" - Begin of new stream
1144 "@B" - Detached signature follows.
1145 This emits a control packet (1,'B')
1146 <detached_signature>
1147 "@t" - Signed text follows.
1148 This emits the control packet (2, 'B')
1150 "@." - End of operation. The final control packet forces signature
1152 "@>" - End of stream
1161 * For packet version 3 we calculate the keyids this way:
1162 RSA := low 64 bits of n
1163 ELGAMAL := build a v3 pubkey packet (with CTB 0x99) and calculate
1164 a rmd160 hash value from it. This is used as the
1165 fingerprint and the low 64 bits are the keyid.
1167 * Revocation certificates consist only of the signature packet;
1168 "import" knows how to handle this. The rationale behind it is
1177 Keyserver Message Format
1178 =========================
1180 The keyserver may be contacted by a Unix Domain socket or via TCP.
1182 The format of a request is:
1186 "Content-length:" digits
1190 Where command-tag is
1198 The format of a response is:
1201 "GNUPG/1.0" status-code status-text
1202 "Content-length:" digits
1205 followed by <digits> bytes of data
1210 o 1xx: Informational - Request received, continuing process
1212 o 2xx: Success - The action was successfully received, understood,
1215 o 4xx: Client Error - The request contains bad syntax or cannot be
1218 o 5xx: Server Error - The server failed to fulfill an apparently
1223 Documentation on HKP (the http keyserver protocol):
1225 A minimalistic HTTP server on port 11371 recognizes a GET for /pks/lookup.
1226 The standard http URL encoded query parameters are this (always key=value):
1228 - op=index (like pgp -kv), op=vindex (like pgp -kvv) and op=get (like
1231 - search=<stringlist>. This is a list of words that must occur in the key.
1232 The words are delimited with space, points, @ and so on. The delimiters
1233 are not searched for and the order of the words doesn't matter (but see
1236 - exact=on. This switch tells the hkp server to only report exact matching
1237 keys back. In this case the order and the "delimiters" are important.
1239 - fingerprint=on. Also reports the fingerprints when used with 'index' or
1242 The keyserver also recognizes http-POSTs to /pks/add. Use this to upload
1246 A better way to do this would be a request like:
1248 /pks/lookup/<gnupg_formatierte_user_id>?op=<operation>
1250 This can be implemented using Hurd's translator mechanism.
1251 However, I think the whole key server stuff has to be re-thought;
1252 I have some ideas and probably create a white paper.