diff options
Diffstat (limited to 'g10/ChangeLog')
| -rw-r--r-- | g10/ChangeLog | 3663 |
1 files changed, 3345 insertions, 318 deletions
diff --git a/g10/ChangeLog b/g10/ChangeLog index 9c5acd8a1..21c44df4e 100644 --- a/g10/ChangeLog +++ b/g10/ChangeLog @@ -1,429 +1,3446 @@ -2000-11-13 Werner Koch <wk@gnupg.org> +2002-06-21 Stefan Bellon <sbellon@sbellon.de> - * Makefile.am: Do not install any program + * g10.c [__riscos__]: Moved RISC OS specific stuff to util/riscos.c + and include/util.h. -2000-10-12 Werner Koch <wk@gnupg.org> + * gpgv.c [__riscos__]: Likewise. - * keygen.c (keygen_add_std_prefs): Add Rijndael to the prefs. +2002-06-20 David Shaw <dshaw@jabberwocky.com> + + * keydb.h, pkclist.c (select_algo_from_prefs): Allow passing a + suggested algorithm which will be used if available. + + * encode.c (encode_crypt, encrypt_filter), sign.c (sign_file): Use + new select_algo_from_prefs feature to check if forcing an + algorithm would violate the recipient preferences. + + * photoid.c (get_default_photo_command, show_photos): Use + different default viewers on different platforms. Currently we + have Win 9x, Win NT (2k, xp), Mac OSX, RISC OS, and "everybody + else". These are #ifdefs as much as possible to avoid clutter. + + * g10.c (strusage, build_list), keyedit.c (show_prefs), main.h, + misc.c (compress_algo_to_string, check_compress_algo), pkclist.c + (algo_available), keygen.c (keygen_set_std_prefs): New + algo_to_string and check functions for compress algorithms. + +2002-06-20 Werner Koch <wk@gnupg.org> + + * misc.c (setsysinfo): Removed a #warning for Alpha's uniligedn + trap disabling - it is quite possible that this is a debug relict. + +2002-06-20 Stefan Bellon <sbellon@sbellon.de> + + * g10.c [__riscos__]: Added image file system feature. + + * gpgv.c [__riscos__]: Added image file system feature. + + * photoid.c (show_photos) [__riscos__]: Set RISC OS filetype of + photo id according to MIME type. + +2002-06-19 David Shaw <dshaw@jabberwocky.com> + + * hkp.c (parse_hkp_index): Don't leak memory when failing out of a + bad HKP keyserver. + + * g10.c (add_notation_data): Relax slightly the rules as to what + can go into a notation name - 2440 allows "@", for example. + +2002-06-17 David Shaw <dshaw@jabberwocky.com> + + * import.c (clean_subkeys, import_one): Only allow at most 1 + binding sig and at most 1 revocation sig on a subkey, as per + 2440:11.1. + + * hkp.c (parse_hkp_index, hkp_search): Error if the keyserver + returns an unparseable HKP response. + +2002-06-15 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (show_key_with_all_names), keylist.c + (list_keyblock_print): Show "[expired]" before expired uids. + + * keyedit.c (show_key_with_all_names_colon), mainproc.c + (list_node), keylist.c (list_keyblock_colon): Show flag 'e' for + expired user ids. Use "uat" for user attribute packets instead of + "uid". Also use '<count> <length>' rather than the fake user id + string on attributes. + + * keygen.c (keygen_add_revkey): Remove unused code. + + * misc.c (check_permissions): Check directory permissions + properly - they are not special files. + + * pkclist.c (expand_id, expand_group, build_pk_list): When + expanding groups before building a pk list, inherit flags from the + original pre-expanded string. + + * pubkey-enc.c (is_algo_in_prefs): Don't use prefs from expired + uids. + +2002-06-14 David Shaw <dshaw@jabberwocky.com> + + * free-packet.c (copy_signature): Properly copy a signature that + carries a revocation key on it. + + * pkclist.c (expand_id, expand_group, build_pk_list): Groups now + work properly when used in the "Enter the user ID" prompt. + +2002-06-14 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (show_key_with_all_names): Display warning if a user + tries to show prefs on a v3 key with a v3 selfsig. + + * kbnode.c (dump_kbnode): Show if a uid is expired. + + * import.c (merge_blocks, import_revoke_cert): Show user ID + receiving a revocation certificate. + + * free-packet.c (cmp_user_ids): Properly compare attribute ids. + + * pkclist.c (expand_groups): Maintain the strlist flags while + expanding. Members of an expansion inherit their flags from the + expansion key. + + * options.h, cipher.c (write_header), g10.c (main), keygen.c + (keygen_set_std_prefs): remove the personal_mdc flag. It no + longer serves a purpose now that the personal preference lists are + split into cipher/digest/zip. + +2002-06-14 Timo Schulz <ts@winpt.org> + + * skclist.c (is_insecure): Implemented. + +2002-06-12 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (keyserver_spawn): Properly handle PROGRAM responses + when they have a CRLF ending. Noted by Keith Ray. + + * keyserver.c (keyserver_spawn): Handle CRLF endings from + keyserver helpers. Also don't leak the last line worth of memory + from the keyserver response. + + * main.h, misc.c (deprecated_warning): New function to warn about + deprecated options and commands. + + * g10.c (main), keyserver-internal.h, keyserver.c + (parse_keyserver_uri): Use new deprecated function to warn about + honor-http-proxy, auto-key-retrieve, and x-broken-hkp. + +2002-06-11 David Shaw <dshaw@jabberwocky.com> + + * Makefile.am: link gpg with NETLIBS for the built-in HKP access. + +2002-06-10 David Shaw <dshaw@jabberwocky.com> + + * options.h, keyserver.c (keyserver_opts), g10.c (main): New + keyserver option "include-subkeys". This feature already existed, + but now can be turned off. It defaults to on. + + * options.h, keyserver.c (parse_keyserver_options, + keyserver_spawn): There are now enough options to justify making a + structure for the keyserver options rather than a page of + if-then-else-if-then-etc. + + * getkey.c (merge_keys_and_selfsig, merge_selfsigs_main): Fix bug + in calculating key expiration dates. + +2002-06-09 David Shaw <dshaw@jabberwocky.com> + + * keydb.h, getkey.c (get_user_id_native), import.c (import_one): + Display user ID while importing a key. Note this applies to both + --import and keyserver --recv-keys. + + * exec.c (exec_finish): Log unnatural exit (core dump, killed + manually, etc) for fork/exec/pipe child processes. + +2002-06-08 Timo Schulz <ts@winpt.org> + + * encode.c (encode_symmetric): Disable the compat flag + when the expert mode is enabled. + +2002-06-07 David Shaw <dshaw@jabberwocky.com> + + * options.skel, options.h, main.h, keydb.h, pkclist.c + (build_pk_list, expand_groups), g10.c (main, add_group): Add new + "group" command to allow one name to expand into multiple keys. + For simplicity, and to avoid potential loops, we only expand once + - you can't make an alias that points to an alias. + + * main.h, g10.c (main), keygen.c (build_personal_digest_list): + Simplify the default digest list - there is really no need for the + other hashes since they will never be used after SHA-1 in the + list. + + * options.skel, options.h, g10.c (main), hkp.c (hkp_ask_import, + hkp_export, hkp_search), keyserver.c (parse_keyserver_options, + parse_keyserver_uri, keyserver_work, keyserver_refresh): Make the + "x-broken-hkp" keyserver scheme into keyserver-option + "broken-http-proxy". Move honor_http_proxy into + keyserver_options. Canonicalize the three variations of "hkp", + "x-hkp", and "x-broken-hkp" into "hkp". + +2002-06-07 Stefan Bellon <sbellon@sbellon.de> + + * g10.c [__riscos__]: Added --attribute-file to do the same as + --attribute-fd, but with a filename not a fd as argument. + Added magic symbol for RISC OS to use different memory management. + + * gpgv.c [__riscos__]: Added magic symbol for RISC OS to use + different memory management. + +2002-06-06 David Shaw <dshaw@jabberwocky.com> + + * main.h, g10.c (main), keygen.c (build_personal_digest_list): Put + in a default digest preference list consisting of SHA-1, followed + by every other installed digest except MD5. Note this is the same + as having no digest preference at all except for SHA-1 being + favored. + + * options.h, g10.c (main), keygen.c (keygen_set_std_prefs), + pkclist.c (select_algo_from_prefs): Split + --personal-preference-list into three: + --personal-{cipher|digest|compress}-preferences. This allows a + user to set one without affecting another (i.e. setting only a + digest pref doesn't imply an empty cipher pref). + + * exec.c (exec_read): This is a safer way of guessing the return + value of system(). Noted by Stefan Bellon. + +2002-06-05 David Shaw <dshaw@jabberwocky.com> + + * hkp.c (parse_hkp_index): Be more robust with keyservers + returning very unparseable responses. + + * exec.c (exec_read): Catch and display an error when the remote + process exits unnaturally (i.e. segfault) so the user knows what + happened. Also fix exec_write stub which has a different number + of arguments now. + +2002-06-05 Timo Schulz <ts@winpt.org> + + * encode.c (encode_simple): Ignore the new mode for RFC1991. + * mainproc.c (symkey_decrypt_sesskey): Better check for weird + keysizes. + +2002-06-05 Timo Schulz <ts@winpt.org> + + * encode.c (encode_sesskey): New. + (encode_simple): Use it here. But by default we use the compat + mode which supress to generate encrypted session keys. + +2002-06-05 Timo Schulz <ts@winpt.org> + + * mainproc.c (symkey_decrypt_sesskey): New. + (proc_symkey_enc): Support for encrypted session keys. + +2002-06-04 David Shaw <dshaw@jabberwocky.com> + + * sign.c (hash_for, sign_file): When encrypting and signing at the + same time, consult the various hash prefs to pick a hash algorithm + to use. Pass in a 160-bit hint if any of the signing keys are + DSA. + + * keydb.h, pkclist.c (select_algo_from_prefs, algo_available): + Pass a "hints" opaque pointer in to let the caller give hints as + to what algorithms would be acceptable. The only current hint is + for PREFTYPE_HASH to require a 160-bit hash for DSA. Change all + callers in encode.c (encode_crypt, encrypt_filter) and sign.c + (sign_file). If we settle on MD5 as the best algorithm based + solely on recepient keys and SHA1 is also a possibility, use SHA1 + unless the user intentionally chose MD5. This is as per 2440:13. + + * exec.c (make_tempdir): Fix duplicated filename problem. + +2002-06-03 David Shaw <dshaw@jabberwocky.com> + + * packet.h, parse-packet.c (enum_sig_subpkt): Report back from + enum_sig_subpkt when a subpacket is critical and change all + callers in keylist.c (show_policy_url, show_notation), mainproc.c + (print_notation_data), and pkclist.c (do_show_revocation_reason). + + * keylist.c (show_policy_url, show_notation): Display if the + policy or notation is critical. + +2002-06-03 David Shaw <dshaw@jabberwocky.com> + + * main.h, g10.c (main), keylist.c (dump_attribs, set_attrib_fd, + list_keyblock_print, list_keyblock_colon), status.h, status.c + (get_status_string): New --attribute-fd feature to dump the + contents of attribute subpackets for frontends. If --status-fd is + also used, then a new status tag ATTRIBUTE is provided for each + subpacket. + + * packet.h, getkey.c (fixup_uidnode, merge_selfsigs_main, + merge_selfsigs_subkey), parse-packet.c (setup_user_id): Keep track + of the expiration time of a user ID, and while we're at it, use + the expired flag from the selfsig rather than reparsing the + SIG_EXPIRE subpacket. + + * photoid.c (generate_photo_id): When adding a new photo ID, + showing the photo for confirmation is not safe when noninteractive + since the "user" may not be able to dismiss a viewer window. + Noted by Timo Schulz. + +2002-06-03 David Shaw <dshaw@jabberwocky.com> + + * options.skel: Sample photo viewers for Win32. + + * misc.c (pct_expando): Use the seckey for %k/%K if the pubkey is + not available. + + * photoid.h, photoid.c (show_photos): Include the seckey in case a + user tries to view a photo on a secret key, and change all callers + in keyedit.c (menu_showphoto), keylist.c (list_keyblock_print), + and photoid.c (generate_photo_id). + +2002-06-02 David Shaw <dshaw@jabberwocky.com> + + * photoid.c (show_photos): Work properly when not called with a + public key. + +2002-05-31 David Shaw <dshaw@jabberwocky.com> + + * sign.c (mk_notation_and_policy): Free unneeded buffer. + + * hkp.c (parse_hkp_index): Properly handle the '&' character + (i.e. "&") in HKP responses. + + * getkey.c (merge_selfsigs_main): Fix reversed expiration time + check with self-sigs. + + * keyedit.c (sign_uids): When making a new self-sig on a v3 key, + make a v3 self-sig unless it is currently a v3 self-sig being + promoted to v4. + +2002-05-31 Timo Schulz <ts@winpt.org> + + * pkclist.c (do_show_revocation_reason): Don't use capital + letters for non-interactive output. + (show_revocation_reason): Now it is global. + * pubkey-enc.c (get_it): Show if the key has been revoked. + +2002-05-30 David Shaw <dshaw@jabberwocky.com> + + * sign.c (write_signature_packets, sign_file, clearsign_file, + sign_symencrypt_file): Make a v4 signature if a policy URL or + notation is set, unless v3 sigs are forced via rfc1991 or + force-v3-sigs. Also remove some doubled code and clarify an error + message (we don't sign in PGP2 mode - just detach-sign). + + * parse-packet.c (parse_one_sig_subpkt): Add KS_FLAGS to the "any + size" section. + +2002-05-29 David Shaw <dshaw@jabberwocky.com> + + * keygen.c (keygen_set_std_prefs, add_feature_mdc): Use "mdc" and + "no-mdc" in the prefs string to allow switching on and off the MDC + feature. This is needed to properly export a key from GnuPG for + use on PGP which does not support MDC - without this, MDC-capable + implementations will still try and generate MDCs which will break + PGP. + + * keygen.c (keygen_get_std_prefs): Show "[mdc]" in prefs string if + it is enabled. + + * options.h, g10.c (main), cipher.c (write_header), keygen.c + (keygen_set_std_prefs): For consistency, allow the user to specify + mdc/no-mdc in the --personal-preference-list. If disabled, it + acts just like --disable-mdc. + +2002-05-29 David Shaw <dshaw@jabberwocky.com> + + * options.h, exec.c: Add some debugging info, using the 1024 debug + flag. + + * exec.c (win_system): New system()-like function for win32 that + does not return until the child process terminates. Of course, + this doesn't help if the process itself exits before it is + finished. + +2002-05-29 Werner Koch <wk@gnupg.org> + + * encode.c (encode_simple): Intialize PKT when --no-literal is used. + + * keyedit.c (show_key_with_all_names_colon): Renamed the record + for revocation keys to "rvk". + +2002-05-27 Werner Koch <wk@gnupg.org> + + * keyedit.c (show_key_with_all_names_colon): New. + (show_key_with_all_names): Divert to new function when required. + Sanitize printing of revoker name. + +2002-05-27 David Shaw <dshaw@jabberwocky.com> + + * build-packet.c (build_sig_subpkt): Handle setting sig flags for + certain subpacket types (notation, policy url, exportable, + revocable). keyedit.c (sign_mk_attrib): Flags no longer need to + be set here. + + * packet.h, parse-packet.c (parse_one_sig_subpkt), build-packet.c + (build_sig_subpkt): Call parse_one_sig_subpkt to sanity check + buffer lengths before building a sig subpacket. + +2002-05-26 David Shaw <dshaw@jabberwocky.com> + + * sign.c (mk_notation_and_policy): Include secret key to enable %s + expandos, and pass notations through pct_expando as well. + + * main.h, misc.c (pct_expando): Add %s and %S expandos for + signer's keyid. + +2002-05-25 David Shaw <dshaw@jabberwocky.com> + + * g10.c (strusage, build_list): Add compress algorithms to + --version list. Show algorithm numbers when --verbose --version + is done. + +2002-05-22 David Shaw <dshaw@jabberwocky.com> + + * options.h, main.h, keygen.c (keygen_set_set_prefs, + keygen_get_std_prefs, keygen_upd_std_prefs), keyedit.c + (keyedit_menu), g10.c (main), pkclist.c (select_algo_from_prefs): + Add --personal-preference-list which allows the user to factor in + their own preferred algorithms when the preference lists are + consulted. Obviously, this does not let the user violate a + recepient's preferences (and the RFC) - this only influences the + ranking of the agreed-on (and available) algorithms from the + recepients. Suggested by David Hollenberg. + + * options.h, keygen.c (keygen_set_std_prefs), g10.c (main): Rename + --preference-list to --default-preference-list (as that is what it + really is), and make it a true default in that if the user selects + "default" they get this list and not the compiled-in list. + +2002-05-22 Werner Koch <wk@gnupg.org> + + * g10.c (main): Add missing LF in a info printout and made it + translatable. Noted by Michael Tokarev. + +2002-05-21 Werner Koch <wk@gnupg.org> + + * g10.c (main): Removed the undef of USE_SHM_COPROCESSING which + was erroneously introduced on 2002-01-09. + + * signal.c (got_fatal_signal): Don't write the Nul to stderr. + Reported by David Hollenberg. + +2002-05-18 David Shaw <dshaw@jabberwocky.com> + + * main.h, g10.c (main), revoke.c (gen_desig_revoke): Generate a + designated revocation via --desig-revoke + + * keyedit.c (keyedit_menu, menu_addrevoker): New "addrevoker" + command to add a designated revoker to a key. + +2002-05-17 David Shaw <dshaw@jabberwocky.com> + + * gpgv.c: Add stub for get_ownertrust(). + + * g10.c (main): --allow-freeform-uid should be implied by + OpenPGP. Add --no-allow-freeform-uid. + + * keyedit.c (sign_uids): Issue a warning when signing a + non-selfsigned uid. + + * getkey.c (merge_selfsigs_main): If a key has no selfsigs, and + allow-non-selfsigned-uid is not set, still try and make the key + valid by checking all uids for a signature from an ultimately + trusted key. + +2002-05-16 David Shaw <dshaw@jabberwocky.com> + + * main.h, keygen.c (keygen_add_revkey): Add revocation key + subpackets to a signature (callable by + make_keysig_packet). (write_direct_sig): Write a 1F direct key + signature. (parse_revocation_key): Parse a string in + algo:fpr:sensitive format into a revocation + key. (get_parameter_revkey, do_generate_keypair): Call above + functions when prompted from a batch key generation file. + + * build-packet.c (build_sig_subpkt): Allow multiple revocation key + subpackets in a single sig. + + * keydb.h, getkey.c (get_seckey_byfprint): Same as + get_pubkey_byfprint, except for secret keys. We only know the + fingerprint of a revocation key, so this is needed to retrieve the + secret key needed to issue a revokation. + + * packet.h, parse-packet.c (parse_signature, parse_revkeys): Split + revkey parsing off into a new function that can be used to reparse + after manipulating the revkey list. + + * sign.c (make_keysig_packet): Ability to make 1F direct key + signatures. + +2002-05-15 David Shaw <dshaw@jabberwocky.com> + + * options.skel: keyserver.pgp.com is gone, so list pgp.surfnet.nl + as a sample LDAP server instead. + + * getkey.c (merge_selfsigs_main): Properly handle multiple + revocation keys in a single packet. Properly handle revocation + keys that are in out-of-order packets. Remove duplicates in + revocation key list. + +2002-05-14 Timo Schulz <ts@winpt.org> + + * exec.c (make_tempdir) [MINGW32]: Added missing '\'. + +2002-05-14 Stefan Bellon <sbellon@sbellon.de> + + * exec.c (make_tempdir): Make use of EXTSEP_S instead of hardcoded + dot as extension separator. + +2002-05-13 David Shaw <dshaw@jabberwocky.com> + + * photoid.c (show_photos): Use the long keyid as the filename for + the photo. Use the short keyid as the filename on 8.3 systems. + + * exec.h, exec.c (make_tempdir, exec_write, exec_finish): Allow + caller to specify filename. This should make things easier on + windows and macs where the file extension is required, but a whole + filename is even better. + + * keyedit.c (show_key_with_all_names, show_prefs): Show proper + prefs for a v4 key uid with no selfsig at all. + + * misc.c (check_permissions): Don't check permissions on + non-normal files (pipes, character devices, etc.) + +2002-05-11 Werner Koch <wk@gnupg.org> + + * mainproc.c (proc_symkey_enc): Avoid segv in case the parser + encountered an invalid packet. + + * keyserver.c (keyserver_export): Get confirmation before sending + all keys. + +2002-05-10 Stefan Bellon <sbellon@sbellon.de> + + * g10.c, hkp.c, keyedit.c, keyserver.c: Replaced all occurrances + of strcasecmp with ascii_strcasecmp and all occurrances of + strncasecmp with ascii_memcasecmp. + +2002-05-10 David Shaw <dshaw@jabberwocky.com> + + * packet.h, getkey.c (fixup_uidnode), keyedit.c (show_prefs): Show + assumed prefs for hash and compression as well as the cipher pref. + Show assumed prefs if there are no prefs at all on a v4 + self-signed key. + + * options.h, g10.c (main), sign.c (make_keysig_packet): New + --cert-digest-algo function to override the default key signing + hash algorithm. + +2002-05-09 David Shaw <dshaw@jabberwocky.com> + + * getkey.c (merge_selfsigs_main): Make sure the revocation key + list starts clean as this function may be called more than once + (e.g. from functions in --edit). + + * g10.c, encode.c (encode_crypt), sign.c (sign_file, + sign_symencrypt_file): Make --compress-algo work like the + documentation says. It should be like --cipher-algo and + --digest-algo in that it can override the preferences calculation + and impose the setting the user wants. No --compress-algo setting + allows the usual preferences calculation to take place. + + * main.h, compress.c (compress_filter): use new + DEFAULT_COMPRESS_ALGO define, and add a sanity check for compress + algo value. + +2002-05-08 David Shaw <dshaw@jabberwocky.com> + + * pkclist.c (select_algo_from_prefs): There is an assumed + compression preference for uncompressed data. + +2002-05-07 David Shaw <dshaw@jabberwocky.com> + + * options.h, g10.c (main), getkey.c (finish_lookup), pkclist.c + (algo_available): --pgp7, identical to --pgp6 except that it + permits a few algorithms that PGP 7 added: AES128, AES192, AES256, + and TWOFISH. Any more of these --pgpX flags, and it'll be time to + start looking at a generic --emulate-pgp X option. + + * export.c (do_export_stream): Warn the user when exporting a + secret key if it or any of its secret subkeys are protected with + SHA1 while simple_sk_checksum is set. + + * parse-packet.c (parse_key): Show when the SHA1 protection is + used in --list-packets. + + * options.h, build-packet.c (do_comment), g10.c (main): Rename + --no-comment as --sk-comments/--no-sk-comments (--no-comment still + works) and make the default be --no-sk-comments. + +2002-05-07 Werner Koch <wk@gnupg.org> + + * keygen.c (get_parameter_algo): Never allow generation of the + deprecated RSA-E or RSA-S flavors of PGP RSA. + (ask_algo): Allow generation of RSA sign and encrypt in expert + mode. Don't allow ElGamal S+E unless in expert mode. + * helptext.c: Added entry keygen.algo.rsa_se. + +2002-05-07 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (sign_uids): If --expert is set, allow re-signing a + uid to promote a v3 self-sig to a v4 one. This essentially + deletes the old v3 self-sig and replaces it with a v4 one. + + * packet.h, parse-packet.c (parse_key), getkey.c + (merge_keys_and_selfsig, merge_selfsigs_main): a v3 key with a v4 + self-sig must never let the v4 self-sig express a key expiration + time that extends beyond the original v3 expiration time. + +2002-05-06 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (sign_uids): When making a self-signature via "sign" + don't ask about sig level or expiration, and include the usual + preferences and such for v4 self-sigs. (menu_set_preferences): + Convert uids from UTF8 to native before printing. + + * keyedit.c (sign_uids): Convert uids from UTF8 to native before + printing. (menu_set_primary_uid): Show error if the user tries to + make a uid with a v3 self-sig primary. + +2002-05-05 David Shaw <dshaw@jabberwocky.com> + + * import.c (import_one): When merging with a key we already have, + don't let a key conflict (same keyid but different key) stop the + import: just skip the bad key and continue. + + * exec.c (make_tempdir): Under Win32, don't try environment + variables for temp directories - GetTempDir tries environment + variables internally, and it's better not to second-guess it in + case MS adds some sort of temp dir handling to Windows at some + point. + +2002-05-05 Timo Schulz <ts@winpt.org> + + * mainproc.c (proc_symkey_enc): Don't ask for a passphrase + in the list only mode. + +2002-05-05 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (keyserver_refresh): --refresh-keys implies + --merge-only so as not to import keys with keyids that match the + ones being refreshed. Noted by Florian Weimer. + +2002-05-04 Stefan Bellon <sbellon@sbellon.de> + + * free-packet.c (copy_public_key): Don't call m_alloc(0), therefore + added consistency check for revkey and numrefkeys. + + * getkey.c (check_revocation_keys): Added consistency check for + revkey and numrefkeys. + + * keyedit.c (show_key_with_all_names): Likewise. + +2002-05-03 David Shaw <dshaw@jabberwocky.com> + + * photoid.c: Provide default image viewer for Win32. + + * misc.c (pct_expando): %t means extension, not name ("jpg", not + "jpeg"). + + * keyserver.c (keyserver_spawn), photoid.c (show_photos), exec.h, + exec.c: Allow the caller to determine the temp file extension when + starting an exec_write and change all callers. + + * keyedit.c (sign_uids): Nonrevocable key signatures cause an + automatic promotion to v4. + + * exec.c: Provide stubs for exec_ functions when NO_EXEC is + defined. + +2002-05-02 David Shaw <dshaw@jabberwocky.com> + + * photoid.h, photoid.c (parse_image_header, image_type_to_string): + Useful functions to return data about an image. + + * packet.h, parse-packet.c (make_attribute_uidname, + parse_attribute_subpkts, parse_attribute), photoid.h, photoid.c + (show_photos): Handle multiple images in a single attribute + packet. + + * main.h, misc.c (pct_expando), sign.c (mk_notation_and_policy), + photoid.c (show_photos): Simpler expando code that does not + require using compile-time string sizes. Call + image_type_to_string to get image strings (i.e. "jpg", + "image/jpeg"). Change all callers. + + * keyedit.c (menu_showphoto), keylist.c (list_keyblock_print): + Allow viewing multiple images within a single attribute packet. + + * gpgv.c: Various stubs for link happiness. + +2002-05-02 David Shaw <dshaw@jabberwocky.com> + + * build-packet.c (build_sig_subpkt), keyedit.c (sign_uids), + options.h, sign.c (mk_notation_and_policy), g10.c (main, + add_notation_data, add_policy_url (new), check_policy_url + (removed)): Allow multiple policy URLs on a given signature. + Split "--notation-data" into "--cert-notation" and + "--sig-notation" so the user can set different policies for key + and data signing. For backwards compatibility, "--notation-data" + sets both, as before. + +2002-05-02 Werner Koch <wk@gnupg.org> + + * options.skel: Removed the comment on trusted-keys because this + option is now deprecated. + +2002-05-01 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (menu_adduid): 2440bis04 says that multiple attribute + packets on a given key are legal. + + * keyserver.c (keyserver_refresh): the fake v3 keyid hack applies + to "mailto" URLs as well since they are also served by pksd. + +2002-04-29 Werner Koch <wk@gnupg.org> + + Added a copyright year for files changed this year. + +2002-04-25 Werner Koch <wk@gnupg.org> + + * g10.c, options.h: New options --display, --ttyname, --ttytype, + --lc-ctype, --lc-messages to be used with future versions of the + gpg-agent. + * passphrase.c (agent_send_option,agent_send_all_options): New. + (agent_open): Send options to the agent. + + * trustdb.c (update_ownertrust, clear_ownertrust): Do an explicit + do_sync because revalidation_mark does it only if when the + timestamp actually changes. + +2002-04-23 David Shaw <dshaw@jabberwocky.com> + + * main.h, keygen.c (do_generate_keypair), keylist.c + (print_signature_stats, list_all, list_one, list_keyblock, + list_keyblock_print, list_keyblock_colon): After generating a new + key, show the key information (name, keyid, fingerprint, etc.) + Also do not print uncheckable signatures (missing key..) in + --check-sigs. Print statistics (N missing keys, etc.) after + --check-sigs. + + * keyedit.c (sign_uids): When signing a key with an expiration + date on it, the "Do you want your signature to expire at the same + time?" question should default to YES. + +2002-04-22 David Shaw <dshaw@jabberwocky.com> + + * parse-packet.c (parse_plaintext), packet.h, plaintext.c + (handle_plaintext): Fix bug in handling literal packets with + zero-length data (no data was being confused with partial body + length). + + * misc.c (pct_expando), options.skel: %t means extension ("jpg"). + %T means MIME type ("image/jpeg"). + + * import.c (import_one): Only trigger trust update if the keyring + is actually changed. + + * export.c (do_export_stream): Missing a m_free. + +2002-04-22 Stefan Bellon <sbellon@sbellon.de> + + * keyid.c (expirestr_from_sk, expirestr_from_sig): Added _() to + string constant. + + * exec.c (make_tempdir) [__riscos__]: Better placement of + temporary file. + +2002-04-20 David Shaw <dshaw@jabberwocky.com> + + * keygen.c (generate_subkeypair): 2440bis04 adds that creating + subkeys on v3 keys is a MUST NOT. + + * getkey.c (finish_lookup): The --pgp6 "use the primary key" + behavior should only apply while data signing and not encryption. + Noted by Roger Sondermann. + +2002-04-19 Werner Koch <wk@gnupg.org> -2000-10-11 Werner Koch <wk@gnupg.org> + * keygen.c (keygen_set_std_prefs): Put back 3DES because the RFC + says it is good form to do so. - * packet.h: Add features sig subpacket definition. - * parse-packet.c (dump_sig_subpkt,parse_one_sig_subpkt, - can_handle_critical): Implemented it here. - * build-packet.c (build_sig_subpkt): and here. - * keygen.c (keygen_add_std_prefs): Generate a features packet. +2002-04-19 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (menu_deluid): Only cause a trust update if we delete + a non-revoked user id. + + * hkp.c (hkp_ask_import), keyserver.c (parse_keyserver_options, + keyserver_spawn), options.h: Remove fast-import keyserver option + (no longer meaningful). + + * g10.c (main), keyedit.c (sign_uids), options.h: Change + --default-check-level to --default-cert-check-level as it makes + clear what it operates on. + + * g10.c (main): --pgp6 also implies --no-ask-sig-expire. + + * delkey.c (do_delete_key): Comment. + + * keyedit.c (sign_uids, keyedit_menu, menu_deluid, menu_delsig, + menu_expire, menu_revsig, menu_revkey): Only force a trustdb check + if we did something that changes it. + + * g10.c: add "--auto-check-trustdb" to override a + "--no-auto-check-trustdb" + +2002-04-19 Werner Koch <wk@gnupg.org> + + * tdbio.c (tdbio_write_nextcheck): Return a status whether the + stamp was actually changed. + * trustdb.c (revalidation_mark): Sync the changes. Removed the + sync operation done by its callers. + (get_validity): Add logic for maintaining a pending_check flag. + (clear_ownertrust): New. + + * keyedit.c (sign_uids): Don't call revalidation_mark depending on + primary_pk. + (keyedit_menu): Call revalidation_mark after "trust". + (show_key_with_all_names): Print a warning on the wrong listed key + validity. + + * delkey.c (do_delete_key): Clear the owenertrust information when + deleting a public key. -2000-10-09 Werner Koch <wk@gnupg.org> +2002-04-18 Werner Koch <wk@gnupg.org> + + * seskey.c (encode_md_value): Print an error message if a wrong + digest algorithm is used with DSA. Changed all callers to cope + with a NULL return. Problem noted by Imad R. Faiad. + +2002-04-18 David Shaw <dshaw@jabberwocky.com> + + * trustdb.c (mark_usable_uid_certs): Properly handle nonrevocable + signatures that can expire. In short, the only thing that can + override an unexpired nonrevocable signature is another unexpired + nonrevocable signature. + + * getkey.c (finish_lookup): Always use primary signing key for + signatures when --pgp6 is on since pgp6 and 7 do not understand + signatures made by signing subkeys. + +2002-04-18 Werner Koch <wk@gnupg.org> + + * trustdb.c (validate_keys): Never schedule a nextcheck into the + past. + (validate_key_list): New arg curtime use it to set next_expire. + (validate_one_keyblock): Take the current time from the caller. + (clear_validity, reset_unconnected_keys): New. + (validate_keys): Reset all unconnected keys. + + * getkey.c (premerge_public_with_secret): Fixed 0x12345678! syntax + for use with secret keys. + (lookup): Advance the searchmode after a search FIRST. + + * seckey-cert.c (do_check): Always calculate the old checksum for + use after unprotection. + + * g10.c, options.skel: New option --no-escape-from. Made + --escape-from and --force-v3-sigs the default and removed them + from the options skeleton. + +2002-04-16 Werner Koch <wk@gnupg.org> + + * parse-packet.c (parse_key): Support a SHA1 checksum as per + draft-rfc2440-bis04. + * packet.h (PKT_secret_key): Add field sha1chk. + * seckey-cert.c (do_check): Check the SHA1 checksum + (protect_secret_key): And create it. + * build-packet.c (do_secret_key): Mark it as sha-1 protected. + * g10.c, options.h: New option --simple-sk-checksum. + +2002-04-13 David Shaw <dshaw@jabberwocky.com> + + * parse-packet.c (parse_signature): Minor fix - signatures should + expire at their expiration time and not one second later. + + * keygen.c (proc_parameter_file): Allow specifying preferences + string (i.e. "s5 s2 z1 z2", etc) in a batchmode key generation + file. + + * keyedit.c (keyedit_menu): Print standard error message when + signing a revoked key (no new translation). + + * getkey.c (merge_selfsigs): Get the default set of key prefs from + the real (not attribute) primary uid. + +2002-04-12 David Shaw <dshaw@jabberwocky.com> + + * pkclist.c (build_pk_list): Fix bug that allowed a key to be + selected twice in batch mode if one instance was the default + recipient and the other was an encrypt-to. Noted by Stefan + Bellon. + + * parse-packet.c (dump_sig_subpkt): Show data in trust and regexp + sig subpackets. + + * keyedit.c (keyedit_menu): Use new function real_uids_left to + prevent deleting the last real (i.e. non-attribute) uid. Again, + according to the attribute draft. (menu_showphoto): Make another + string translatable. + +2002-04-11 David Shaw <dshaw@jabberwocky.com> + + * build-packet.c (build_sig_subpkt): Delete subpackets from both + hashed and unhashed area on update. (find_subpkt): No longer + needed. + + * keyedit.c (sign_uids): With --pgp2 on, refuse to sign a v3 key + with a v4 signature. As usual, --expert overrides. Try to tweak + some strings to a closer match so they can all be translated in + one place. Use different helptext keys to allow different help + text for different questions. + + * keygen.c (keygen_upd_std_prefs): Remove preferences from both + hashed and unhashed areas if they are not going to be used. + +2002-04-10 David Shaw <dshaw@jabberwocky.com> + + * misc.c (pct_expando), options.skel: Use %t to indicate type of a + photo ID (in this version, it's always "jpeg"). Also tweak string + expansion loop to minimize reallocs. + + * mainproc.c (do_check_sig): Variable type fix. + + * keyedit.c (menu_set_primary_uid): Differentiate between true + user IDs and attribute user IDs when making one of them primary. + That is, if we are making a user ID primary, we alter user IDs. + If we are making an attribute packet primary, we alter attribute + packets. This matches the language in the latest attribute packet + draft. + + * keyedit.c (sign_uids): No need for the empty string hack. + + * getkey.c (fixup_uidnode): Only accept preferences from the + hashed segment of the self-sig. + +2002-04-10 Werner Koch <wk@gnupg.org> + + * tdbio.c (migrate_from_v2): Fixed the offset to read the old + ownertrust value and only add entries to the table if we really + have a value. + +2002-04-08 David Shaw <dshaw@jabberwocky.com> + + * status.h, status.c (get_status_string): Add KEYEXPIRED, EXPSIG, + and EXPKEYSIG. Add "deprecated-use-keyexpired-instead" to + SIGEXPIRED. + + * sig-check.c (do_check): Start transition from SIGEXPIRED to + KEYEXPIRED, since the actual event is signature verification by an + expired key and not an expired signature. (do_signature_check, + packet.h): Rename as signature_check2, make public, and change all + callers. + + * mainproc.c (check_sig_and_print, do_check_sig): Use status + EXPSIG for an expired, but good, signature. Add the expiration + time (or 0) to the VALIDSIG status line. Use status KEYEXPSIG for + a good signature from an expired key. + + * g10.c (main): remove checks for no arguments now that argparse + does it. + +2002-04-06 Werner Koch <wk@gnupg.org> + + * keyring.c (keyring_get_keyblock): Disable the keylist mode here. + + * encode.c (encode_simple, encode_crypt): Only test on compressed + files if a compress level was not explicity set. + + * keygen.c (keygen_set_std_prefs): Removed Blowfish and Twofish + from the list of default preferences, swapped the preferences of + RMD160 and SHA1. Don't include a preference to 3DES unless the + IDEA kludge gets used. + + * free-packet.c (free_packet): call free_encrypted also for + PKT_ENCRYPTED_MDC. + + * compress.c (release_context): New. + (handle_compressed): Allocate the context and setup a closure to + release the context. This is required because there is no + guarabntee that the filter gets popped from the chain at the end + of the function. Problem noted by Timo and probably also the + cause for a couple of other reports. + (compress_filter): Use the release function if set. + + * tdbio.c [__CYGWIN32__]: Don't rename ftruncate. Noted by + Disastry. + + * parse-packet.c (parse_signature): Put parens around a bit test. + + * exec.c (make_tempdir): Double backslash for TMP directory + creation under Windows. Better strlen the DIRSEP_S constants for + allocation measurements. + + * decrypt.c (decrypt_messages): Release the passphrase aquired + by get_last_passphrase. - * keygen.c (do_generate_keypair): Removed the keyblock locking. +2002-04-02 Werner Koch <wk@gnupg.org> - * ringedit.c (enum_keyblocks): Replaced by ... - (enum_keyblocks_begin): New. - (enum_keyblocks_next): New. - (enum_keyblocks_end): New. And changed all callers. + * Makefile.am (EXTRA_DIST): Removed OPTIONS an pubring.asc - they + are no longer of any use. + +2002-04-03 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (parse_keyserver_options): fix auto-key-retrieve to + actually work as a keyserver-option (noted by Roger Sondermann). + + * keylist.c (reorder_keyblock): do not reorder the primary + attribute packet - the first user ID must be a genuine one. + +2002-03-31 David Shaw <dshaw@jabberwocky.com> + + * keylist.c (list_keyblock_colon): Fix ownertrust display with + --with-colons. + + * keygen.c (generate_user_id), photoid.c (generate_photo_id): + Properly initialize the user ID refcount. A few more "y/n" -> + "y/N" in photoid.c. + + * keyedit.c (ask_revoke_sig): Warn the user if they are about to + revoke an expired sig (not a problem, but they should know). Also + tweak a few prompts to change "y/n" to "y/N", which is how most + other prompts are written. + + * keyserver.c (keyserver_search_prompt): Control-d escapes the + keyserver search prompt. + + * pkclist.c (show_revocation_reason & callers): If a subkey is + considered revoked solely because the parent key is revoked, print + the revocation reason from the parent key. + + * trustdb.c (get_validity): Allow revocation/expiration to apply + to a uid/key with no entry in the trustdb. + +2002-03-29 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (printunquoted): unquote backslashes from keyserver + searches + + * hkp.c (write_quoted): quote backslashes from keyserver searches + +2002-03-26 Werner Koch <wk@gnupg.org> + + * keygen.c (ask_keysize): Removed the warning for key sizes > 1536. + +2002-03-25 Werner Koch <wk@gnupg.org> + + * keyedit.c (sign_uids): Use 2 strings and not a %s so that + translations can be done the right way. + * helptext.c: Fixed small typo. + +2002-03-23 David Shaw <dshaw@jabberwocky.com> + + * import.c (append_uid, merge_sigs): it is okay to import + completely non-signed uids now (with --allow-non-selfsigned-uid). + + * getkey.c (get_primary_uid, merge_selfsigs_main): do not choose + an attribute packet (i.e. photo) as primary uid. This prevents + oddities like "Good signature from [image of size 2671]". This is + still not perfect (one can still select an attribute packet as + primary in --edit), but is closer to the way the draft is going. + + * g10.c (build_list): algorithms should include 110. + + * g10.c (main): --pgp2 implies --no-ask-sig-expire and + --no-ask-cert-expire as those would cause a v4 sig/cert. + + * armor.c (is_armor_header): be more lenient in what constitutes a + valid armor header (i.e. -----BEGIN blah blah-----) as some + Windows programs seem to add spaces at the end. --openpgp makes + it strict again. + +2002-03-18 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (keyserver_search_prompt): Properly handle a "no + keys found" case from the internal HKP code (external HKP is ok). + Also, make a COUNT -1 (i.e. streamed) keyserver response a little + more efficient. + + * g10.c (main): Add --no-allow-non-selfsigned-uid + +2002-03-17 David Shaw <dshaw@jabberwocky.com> + + * g10.c (main): --openpgp implies --allow-non-selfsigned-uid. + + * getkey.c (merge_selfsigs_main): If none of the uids are primary + (because none are valid) then pick the first to be primary (but + still invalid). This is for cosmetics in case some display needs + to print a user ID from a non-selfsigned key. Also use + --allow-non-selfsigned-uid to make such a key valid and not + --always-trust. The key is *not* automatically trusted via + --allow-non-selfsigned-uid. + + * mainproc.c (check_sig_and_print): Make sure non-selfsigned uids + print [uncertain] on verification even though one is primary now. + + * getkey.c (merge_selfsigs): If the main key is not valid, then + neither are the subkeys. + + * import.c (import_one): Allow --allow-non-selfsigned-uid to work + on completely unsigned keys. Print the uids in UTF8. Remove + mark_non_selfsigned_uids_valid(). + + * keyedit.c (show_key_with_all_names): Show revocation key as + UTF8. + + * sign.c (clearsign_file): Allow --not-dash-escaped to work with + v3 keys. + +2002-03-14 Werner Koch <wk@gnupg.org> + + * main.h: Changed the default algorithms to CAST5 and SHA1. + +2002-03-13 David Shaw <dshaw@jabberwocky.com> + + * import.c (chk_self_sigs): Show which user ID a bad self-sig + (invald sig or unsupported public key algorithm) resides on. + + * import.c (chk_self_sigs): any valid self-sig should mark a user + ID or subkey as valid - otherwise, an attacker could DoS the user + by inventing a bogus invalid self-signature. + +2002-03-07 David Shaw <dshaw@jabberwocky.com> + + * g10.c (main): make a few more strings translatable. + + * options.h, options.skel, g10.c (main), gpgv.c, mainproc.c + (check_sig_and_print), keyserver.c (parse_keyserver_options): + --auto-key-retrieve should really be a keyserver-option variable. + + * import.c (revocation_present): new function to print a warning + if a key is imported that has been revoked by designated revoker, + but the designated revoker is not present to verify the + revocation. If keyserver-options auto-key-retrieve is set, try + and fetch the designated revoker from the keyserver. + + * import.c (import_one): call revocation_present after importing a + new key. Note that this applies to --import, --recv-keys, and + --search-keys. + + * keyserver-internal.h, keyserver.c (keyserver_import_fprint): + import via fingerprint (for revocation keys). + + * keyserver.c (keyserver_import_keyid): much simpler + implementation now that we're using KEYDB_SEARCH_DESC internally. + +2002-03-04 David Shaw <dshaw@jabberwocky.com> + + * revoke.c (gen_revoke): do not prompt for revocation reason for + v3 revocations (unless force-v4-certs is on) since they wouldn't + be used anyway. + + * keyedit.c (menu_revsig): show the status of the sigs + (exportable? revocable?) to the user before prompting for which + sig to revoke. Also, make sure that local signatures get local + revocations. + + * keyedit.c (ask_revoke_sig): remind the user which sigs are + local. + + * g10.c (main): Add "exec-path" variable to override PATH for + execing programs. + + * export.c (do_export_stream): properly check return code from + classify_user_id to catch unclassifiable keys. + +2002-03-03 David Shaw <dshaw@jabberwocky.com> + + * parse-packet.c (parse_signature): variable type tweak for RISC + OS (from Stefan) + +2002-02-28 David Shaw <dshaw@jabberwocky.com> + + * getkey.c (check_revocation_keys): New function to check a + revocation against a list of potential revocation keys. Note the + loop-breaking code here. This is to prevent blowing up if A is + B's revocation key, while B is also A's. Note also that this is + written so that a revoked revoker can still issue revocations: + i.e. If A revokes B, but A is revoked, B is still revoked. I'm + not completely convinced this is the proper behavior, but it + matches how PGP does it. It does at least have the advantage of + much simpler code - my first version of this had lots of loop + maintaining code so you could chain revokers many levels deep and + if D was revoked, C was not, which meant that B was, and so on. + It was sort of scary, actually. + + * getkey.c (merge_selfsigs_main): Add any revocation keys onto the + pk. This is particularly interesting since we normally only get + data from the most recent 1F signature, but you need multiple 1F + sigs to properly handle revocation keys (PGP does it this way, and + a revocation key could be marked "sensitive" and hence in a + different signature). Also, if a pk has a revocation key set, + check for revocation sigs that were not made by us - if made by a + valid revocation key, mark the pk revoked. + + * packet.h, getkey.c (cache_public_key): do not cache key if + "dont_cache" is set. This allows the revocation key code to look + up a key and return information that may be inaccurate to prevent + loops without caching the fake data. + + * packet.h, sig-check.c (do_signature_check): Record if a + signature was made by a revoked pk. + + * packet.h, parse-packet.c (parse_one_sig_subpkt, + can_handle_critical, parse_signature): Get revocation key + information out of direct sigs. - * import.c (import_one): Removed keyblock locking becuase this is - now done inside of insert_keyblock(). Removed get_keyblock_handle - because insert_keyblock() now decides what is the default keyring. + * keylist.c (list_keyblock_print): don't assume that the presence + of a 0x20 signature means the key is revoked. With revocation + keys, this may not be true if the revocation key is not around to + verify it or if verification failed. Also, 0x1F should get listed + as "sig", and not "unexpected signature class". + + * keyedit.c (show_key_with_all_names): Add a flag for printing + revoker information and change all callers. + + * import.c (merge_blocks): merge in any new direct key (0x1F) + sigs. + + * import.c (import_revoke_cert): don't keep processing after a + revocation is rejected. + + * import.c (delete_inv_parts): Allow importing a revocation + signature even if it was not issued by the key. This allows a + revocation key to issue it. Of course, the sig still needs to be + checked before we trust it. + + * free-packet.c (copy_public_key): Include a new copy of the + revocation keys when duping a pk. + + * free-packet.c (free_seckey_enc, release_public_key_parts): Free + any revocation keys that are attached to a sig or pk. + + * export.c (do_export_stream): Do not export signatures with + "sensitive" revocation keys in them. + +2002-02-27 David Shaw <dshaw@jabberwocky.com> + + * export.c (do_export_stream): Do not include v3 keys in a + --export-secret-subkeys export. + + * getkey.c (merge_selfsigs_main): If a key isn't valid (say, + because of no self-signature), allow --always-trust to force it + valid so it can be trusted. + +2002-02-25 David Shaw <dshaw@jabberwocky.com> + + * hkp.c (hkp_ask_import), hkp.h, keyserver.c (all): treat key + lists internally as fingerprints when possible. All this is via + KEYDB_SEARCH_DESC - no point in reinventing the wheel. This allows + the helper program to search the keyserver by fingerprint if + desired (and the keyserver supports it). Note that automatic + fingerprint promotion during refresh only applies to v4 keys as a + v4 fingerprint can be easily changed into a long or short key id, + and a v3 cannot. + + * pubkey-enc.c, getkey.c, misc.c, main.h: Take two copies of + hextobyte() from pubkey-enc.c and getkey.c and make them into one + copy in misc.c. + +2002-02-22 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (keyserver_search_prompt): Detect a "no keys found" + case even if the helper program does not explicitly say how many + keys were found. + + * hkp.c (parse_hkp_index): Bug fix - don't report non-revoked keys + as revoked in HKP key searches. + +2002-02-19 Werner Koch <wk@gnupg.org> + + * parse-packet.c (parse_trust): Made parsing more robust. + +2002-02-19 David Shaw <dshaw@jabberwocky.com> + + * hkp.c (parse_hkp_index): Catch corruption in HKP index lines + (can be caused by broken or malicious keyservers). + + * keyserver.c (keyserver_work): Add KEYSERVER_NOT_SUPPORTED for + unsupported actions (say, a keyserver that has no way to search, + or a readonly keyserver that has no way to add). Also add a + USE_EXTERNAL_HKP define to disable the internal HKP keyserver + code. + +2002-02-14 Werner Koch <wk@gnupg.org> + + * g10.c: New option --no-use-agent. + + * pkclist.c (check_signatures_trust): Always print the warning for + unknown and undefined trust. Removed the did_add cruft. Reported + by Janusz A. Urbanowicz. + +2002-02-11 David Shaw <dshaw@jabberwocky.com> + + * hkp.c (parse_hkp_index): Bug fix - properly handle user IDs with + colons (":") in them while HKP searching. + +2002-02-09 David Shaw <dshaw@jabberwocky.com> + + * misc.c (pct_expando): More comments. + + * keydb.h, sign.c (mk_notation_and_policy): Clarify what is a sig + and what is a cert. A sig has sigclass 0x00, 0x01, 0x02, or 0x40, + and everything else is a cert. + + * g10.c (main), keyedit.c (keyedit_menu): Add a "nrlsign" for + nonrevocable and local key signatures. + + * g10.c (main): Add a --no-force-mdc to undo --force-mdc. + + * options.h, g10.c (main), cipher.c (write_header): Add a knob to + --disable-mdc/--no-disable-mdc. Off by default, of course, but is + used in --pgp2 and --pgp6 modes. + + * pkclist.c (build_pk_list): Allow specifying multiple users in + the "Enter the user ID" loop. Enter a blank line to stop. Show + each key+id as it is added. + + * keylist.c (show_policy_url), mainproc.c (print_notation_data): + It is not illegal (though possibly silly) to have multiple policy + URLs in a given signature, so print all that are present. + + * hkp.c (hkp_search): More efficient implementation of URL-ifying + code. + +2002-02-04 David Shaw <dshaw@jabberwocky.com> + + * main.h, misc.c (pct_expando): New function to generalize + %-expando processing in any arbitrary string. + + * photoid.c (show_photo): Call the new pct_expando function rather + than expand strings internally. + + * sign.c (mk_notation_and_policy): Show policy URLs and notations + when making a signature if show-policy/show-notation is on. + %-expand policy URLs during generation. This lets the user have + policy URLs of the form "http://notary.jabberwocky.com/keysign/%K" + which will generate a per-signature policy URL. + + * main.h, keylist.c (show_policy_url, show_notation): Add amount + to indent so the same function can be used in key listings as well + as during sig generation. Change all callers. + +2002-02-04 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c, options.h (parse_keyserver_options, keyidlist): + Workaround for the pksd and OKS keyserver bug that calculates v4 + RSA keyids as if they were v3. The workaround/hack is to fetch + both the v4 (e.g. 99242560) and v3 (e.g. 68FDDBC7) keyids. This + only happens for key refresh while using the HKP scheme and the + refresh-add-fake-v3-keyids keyserver option must be set. This + should stay off by default. + +2002-02-03 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (keyserver_spawn): Bug fix - do not append keys to + each other when --sending more than one. + +2002-02-02 David Shaw <dshaw@jabberwocky.com> + + * options.h, g10.c (main), keyedit.c (sign_uids), sign.c + (mk_notation_and_policy): Split "--set-policy-url" into + "--cert-policy-url" and "--sig-policy-url" so the user can set + different policies for key and data signing. For backwards + compatibility, "--set-policy-url" sets both, as before. + +2002-01-30 Werner Koch <wk@gnupg.org> + + * g10.c (main): --gen-random --armor does now output a base64 + encoded string. + +2002-01-28 David Shaw <dshaw@jabberwocky.com> + + * g10.c (main), options.h, pkclist.c (algo_available): --pgp6 + flag. This is not nearly as involved as --pgp2. In short, it + turns off force_mdc, turns on no_comment, escape_from, and + force_v3_sigs, and sets compression to 1. It also restricts the + user to IDEA (if present), 3DES, CAST5, MD5, SHA1, and RIPEMD160. + See the comments above algo_available() for lots of discussion on + why you would want to do this. + +2002-01-27 David Shaw <dshaw@jabberwocky.com> + + * keygen.c (keygen_set_std_prefs): Comment + + * keyedit.c (sign_uids): Bug fix - when signing with multiple + secret keys at the same time, make sure each key gets the sigclass + prompt. + + * exec.c (exec_finish): Close the iobuf and FILE before trying to + waitpid, so the remote process will get a SIGPIPE and exit. This + is only a factor when using a pipe to communicate. + + * exec.c (exec_write): Disable cache-on-close of the fd iobuf (is + this right? Why is a fd iobuf cached at all?) + +2002-01-26 Werner Koch <wk@gnupg.org> + + * g10.c, options.h: New option --gpg-agent-info + * passphrase.c (agent_open): Let it override the environment info. + * seckey-cert.c (check_secret_key): Always try 3 times when the + agent is enabled. + * options.skel: Describe --use-agent. + +2002-01-24 David Shaw <dshaw@jabberwocky.com> + + * pubkey-enc.c (is_algo_in_prefs, get_it): Only check preferences + against keys with v4 self sigs - there is really little point in + warning for every single non-IDEA message encrypted to an old key. + + * pkclist.c (select_algo_from_prefs): Only put in the fake IDEA + preference if --pgp2 is on. + + * mainproc.c (check_sig_and_print): Print "Expired" for expired + but good signatures (this still prints "BAD" for expired but bad + signatures). + +2002-01-23 David Shaw <dshaw@jabberwocky.com> + + * keygen.c (ask_keysize): Cosmetic: don't present a RSA signing + key as a "keypair" which can be 768 bits long (as RSA minimum is + 1024). + + * pubkey-enc.c (is_algo_in_prefs): Allow IDEA as a fake preference + for v3 keys with v3 selfsigs. + +2002-01-22 David Shaw <dshaw@jabberwocky.com> + + * packet.h, getkey.c (merge_selfsigs_main), pkclist.c + (select_algo_from_prefs): Implement the fake IDEA preference as + per RFC2440:12.1. This doesn't mean that IDEA will be used (the + plugin may not be present), but it does mean that a v3 key with a + v3 selfsig has an implicit IDEA preference instead of 3DES. v3 + keys with v4 selfsigs use preferences as normal. + + * encode.c (encode_crypt): if select_algo_from_prefs fails, this + means that we could not find a cipher that both keys like. Since + all v4 keys have an implicit 3DES preference, this means there is + a v3 key with a v3 selfsig in the list. Use 3DES in this case as + it is the safest option (we know the v4 key can handle it, and + we'll just hope the v3 key is being used in an implementation that + can handle it). If --pgp2 is on, warn the user what we're doing + since it'll probably break PGP2 compatibility. + + * g10.c (main): Do not force using IDEA for encrypted files in + --pgp2 mode - let the fake IDEA preference choose this for us for + better compatibility when encrypting to multiple keys, only some + of which are v3. + + * keygen.c (keygen_set_std_prefs): Put 3DES on the end of the + default cipher pref list (RFC2440: "...it is good form to place it + there explicitly."). If the user has the IDEA plugin installed, + put a preference for IDEA *after* 3DES to effectively disable its + use for everything except encrypting along with v3 keys. + + * encode.c, g10.c, sign.c: Change the PGP2 warning line from + "... will not be usable ..." to "... may not be usable ..." as the + user could be using one of the enhanced PGP2 variations. + + * helptext.c: Revise the sign_uid.class help text as suggested by + Stefan. + +2002-01-20 Werner Koch <wk@gnupg.org> + + * passphrase.c (passphrase_to_dek): Add tryagain_text arg to be + used with the agent. Changed all callers. + (agent_get_passphrase): Likewise and send it to the agent + * seckey-cert.c (do_check): New arg tryagain_text. + (check_secret_key): Pass the string to do_check. + * keygen.c (ask_passphrase): Set the error text is required. + * keyedit.c (change_passphrase): Ditto. + + * passphrase.c (agent_open): Disable opt.use_agent in case of a + problem with the agent. + (agent_get_passphrase): Ditto. + (passphrase_clear_cache): Ditto. + +2002-01-19 Werner Koch <wk@gnupg.org> + + * passphrase.c (agent_open): Add support for the new Assuan based + gpg-agent. New arg to return the used protocol version. + (agent_get_passphrase): Implemented new protocol here. + (passphrase_clear_cache): Ditto. + (readline): New. + +2002-01-15 Timo Schulz <ts@winpt.org> + + * encode.c (encode_crypt_files): Fail if --output is used. + + * g10.c: New command --decrypt-files. + + * decrypt.c (decrypt_messages): New. + +2002-01-09 David Shaw <dshaw@jabberwocky.com> + + * g10.c, misc.c, gpgv.c: move idea_cipher_warn to misc.c so gpgv.c + doesn't need a stub for it any longer. + + * g10.c (get_temp_dir), main.h: no longer used (it's in exec.c now) + + * g10.c (main), delkey.c (delete_keys), main.h : Allow + --delete-key (now --delete-keys, though --delete-key still works, + of course) to delete multiple keys in one go. This applies to + --delete-secret-key(s) and --delete-secret-and-public-key(s) as + well. + +2002-01-09 Timo Schulz <ts@winpt.org> + + * encode.c (encode_crypt_files): Now it behaves like verify_files. + + * g10.c (main): We don't need to check argc for encode_crypt_files + any longer. + +2002-01-09 Timo Schulz <ts@winpt.org> + + * exec.c: Include windows.h for dosish systems. + +2002-01-08 Timo Schulz <ts@winpt.org> + + * g10.c (main): New description for --encrypt-files. + +2002-01-08 Werner Koch <wk@gnupg.org> + + * g10.c (main): Must register the secring for encryption because + it is needed to figure out the default recipient. Reported by + Roger Sondermann. + +2002-01-05 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (menu_adduid): Require --expert before adding a photo + ID to a v3 key, and before adding a second photo ID to any key. + + * keyedit.c (keyedit_menu): Don't allow adding photo IDs in + rfc1991 or pgp2 mode. + + * getkey.c (merge_selfsigs_subkey): Permit v3 subkeys. Believe it + or not, this is allowed by rfc 2440, and both PGP 6 and PGP 7 work + fine with them. + + * g10.c, options.h, keyedit.c, sign.c: Move the "ask for + expiration" switch off of --expert, which was getting quite + overloaded, and onto ask-sig-expire and ask-cert-expire. Both + default to off. + + * g10.c (main): Change the default compression algo to 1, to be + more OpenPGP compliant (PGP also uses this, so it'll help with + interoperability problems as well). + + * encode.c (encode_crypt): Handle compression algo 2, since the + default is now 1. + + * build-packet.c (build_attribute_subpkt): Fix off-by-one error. + +2002-01-05 Werner Koch <wk@gnupg.org> + + * g10.c (main): Do not register the secret keyrings for certain + commands. + + * keydb.c (keydb_add_resource): Use access to test for keyring + existence. This avoids cached opened files which are bad under + RISC OS. + +2002-01-04 David Shaw <dshaw@jabberwocky.com> + + * sign.c (sign_file, sign_symencrypt_file): always use one-pass + packets unless rfc1991 is enabled. This allows a signature made + with a v3 key to work in PGP 6 and 7. Signatures made with v4 + keys are unchanged. + + * g10.c (main): Disallow non-detached signatures in PGP2 mode. + Move the "you must use files and not pipes" PGP2 warning up so all + the PGP2 stuff is together. + + * encode.c (encode_simple): Use the actual filesize instead of + partial length packets in the internal literal packet from a + symmetric message. This breaks PGP5(?), but fixes PGP2, 6, and 7. + It's a decent tradeoff. Note there was only an issue with + old-style RFC1991 symmetric messages. 2440-style messages in 6 + and 7 work with or without partial length packets. + +2002-01-03 David Shaw <dshaw@jabberwocky.com> + + * g10.c (main): Removed --no-default-check-level option, as it is + not consistent with other "default" options. Plus, it is the same + as saying --default-check-level 0. + + * exec.c (exec_read): Disallow caching tempfile from child + process, as this keeps the file handle open and can cause unlink + problems on some platforms. + + * keyserver.c (keyserver_search_prompt): Minor tweak - don't + bother to transform keyids into textual form if they're just going + to be transformed back to numbers. + +2002-01-03 Timo Schulz <ts@winpt.org> + + * g10.c: New command --encrypt-files. + + * verify.c (print_file_status): Removed the static because + encode_crypt_files also uses this function. + + * main.h (print_files_status): New. + (encode_crypt_files): New. + + * encode.c (encode_crypt_files): New. + +2002-01-02 Stefan Bellon <sbellon@sbellon.de> + + * keyserver.c: Moved util.h include down in order to avoid + redefinition problems on RISC OS. + + * keyring.c (keyring_lock): Only lock keyrings that are writable. + + * keyring.c (keyring_update_keyblock): Close unused iobuf. + + * hkp.c (parse_hkp_index, hkp_search) [__riscos__]: Changed + unsigned char* to char* because of compiler issues. + + * exec.c (exec_finish) [__riscos__]: Invalidate close cache so + that file can be unlinked. + +2001-12-28 David Shaw <dshaw@jabberwocky.com> + + * g10.c (main): Use a different strlist to check extensions since + they need to be handled seperately now. + + * misc.c,main.h (check_permissions): Properly handle permission + and ownership checks on files in the lib directory + (e.g. /usr/local/lib/gnupg), which are owned by root and are + world-readable, and change all callers to specify extension or + per-user file. + + * photoid.c (show_photo), keyserver.c (keyserver_spawn): Bug fix - + don't call exec_finish if exec_write fails. + + * keyserver.c (keyserver_spawn): Look for OPTIONS from the + keyserver helper - specifically, a "OUTOFBAND" option for the + email keyserver. + + * mainproc.c (list_node), keylist.c (list_keyblock_colon), + import.c (delete_inv_parts), export.c (do_export_stream): Use + signature flags for exportability check rather than re-parsing the + subpacket. + + * keyid.c, keydb.h (get_lsign_letter): No longer needed. + +2001-12-27 David Shaw <dshaw@jabberwocky.com> + + * exec.c (exec_finish): Show errors when temp files cannot be + deleted for whatever reason. + + * exec.c (exec_read): Don't rely on WEXITSTATUS being present. + + * exec.c (make_tempdir): Add temp file creator for win32. Don't + create an incoming temp file if the exec is write-only. + + * keyserver.c (keyserver_spawn): Clean up error handling, for when + the spawn fails. + + * photoid.c (show_photo): Clean up error handling. + + * misc.c (check_permissions): Neaten. + +2001-12-25 David Shaw <dshaw@jabberwocky.com> + + * mkdtemp.c (mkdtemp): Add copyleft info and tweak the 'X' counter + to be a bit simpler. + + * keyserver.c, photoid.c: Remove unused headers left over from + when the exec functions lived there. + +2001-12-23 Timo Schulz <ts@winpt.org> + + * misc.c (check_permissions): Do not use it for W32 systems. + + * tdbio.c (migrate_from_v2): Define ftruncate as chsize() for W32. + + * mkdtemp.c: W32 support. + + * photoid.c: Ditto. + + * exec.c: Ditto. + +2001-12-22 David Shaw <dshaw@jabberwocky.com> + + * exec.c (make_tempdir): avoid compiler warning with const + + * mkdtemp.c (mkdtemp): catch the empty ("") string case in case + someone repurposes mkdtemp at some point. + + * photoid.c (generate_photo_id, show_photo): some type changes + from Stefan Bellon. + + * exec.c (make_tempdir): handle Win32 systems, suggested by Timo + Schulz. + +2001-12-22 Werner Koch <wk@gnupg.org> + + * encode.c (encode_simple, encode_crypt): i18n 2 strings. + +2001-12-22 Timo Schulz <ts@winpt.org> + + * encode.c (encode_simple, encode_crypt): Use is_file_compressed + to avoid to compress compressed files. + +2001-12-22 Werner Koch <wk@gnupg.org> + + * keyserver.c (keyserver_spawn): Removed some variables + declaration due to shadowing warnings. + + * build-packet.c (build_attribute_subpkt): s/index/idx/ to avoid + compiler warnig due to index(3). + + * getkey.c (get_ctx_handle): Use KEYDB_HANDLE as return value. + * keylist.c (list_one): Made resname const. + + * keyedit.c (keyedit_menu): Allow "addphoto" only when --openpgp is + not used. + + * options.skel: Changed one example photo viewer to qiv. + +2001-12-21 David Shaw <dshaw@jabberwocky.com> + + * Makefile.am: add exec.c, exec.h, photoid.c, and photoid.h + + * build-packet.c (build_attribute_subpkt): new function to build + the raw attribute subpacket. Note that attribute subpackets have + the same format as signature subpackets. + + * exec.c: new file with generic exec-a-program functionality. + Used by both photo IDs and keyserver helpers. This is pretty much + the same code that used to be keyserver specific, with some + changes to be usable generically. + + * free-packet.c (free_attributes (new)): function to free an + attribute packet. + + * gpgv.c: added stub show_photo + + * keyedit.c (keyedit_menu, menu_adduid, menu_showphoto): can add a + photo (calls generate_photo_id), or display a photo (calls + show_photo) from the --edit menu. New commands are "addphoto", + and "delphoto" (same as "deluid"). + + * keylist.c (list_keyblock_print): show photos during key list if + --show-photos enabled. + + * keyserver.c (keyserver_spawn): use the generic exec_xxx + functions to call keyserver helper. + + * g10.c, options.h: three new options - --{no-}show-photos, and + --photo-viewer to give the command line to display a picture. + + * options.skel: instructions for the photo viewer + + * parse-packet.c (parse_user_id, setup_user_id (new)): common code + for both user IDs and attribute IDs moved to setup_user_id. + + * parse-packet.c (make_attribute_uidname (new)): constructs a fake + "name" for attribute packets (e.g. "[image of size ...]") + + * parse-packet.c (parse_attribute (replaces parse_photo_id), + parse_attribute_subpkts): Builds an array of individual + attributes. Currently only handles attribute image / type jpeg + subpackets. + + * sign.c (hash_uid): Fix bug in signing attribute (formerly + photo_id) packets. + + * packet.h, and callers: globally change "photo_id" to "attribute" + and add structures for attributes. The packet format is generic + attributes, even though the only attribute type thus far defined + is jpeg. + +2001-12-21 David Shaw <dshaw@jabberwocky.com> + + * parse-packet.c (can_handle_critical): Can handle critical + revocation subpackets now. + + * trustdb.c (mark_usable_uid_certs): Disregard revocations for + nonrevocable sigs. Note that this allows a newer revocable + signature to override an older nonrevocable signature. + + * sign.c (make_keysig_packet): add a duration field and change all + callers. This makes make_keysig_packet closer to + write_signature_packets and removes some duplicated expiration + code. + + * keyedit.c (keyedit_menu, menu_revsig, sign_uids, + sign_mk_attrib): Add nrsign command, don't allow revoking a + nonrevocable signature, + + * g10.c (main): Add --nrsign option to nonrevocably sign a key + from the command line. + + * build-packet.c (build_sig_subpkt_from_sig): Comment to explain + the use of CRITICAL. + +2001-12-21 Werner Koch <wk@gnupg.org> + + * g10.c. options.h : New option --show-keyring + * getkey.c (get_ctx_handle): New. + * keylist.c (list_one): Implement option here. By David Champion. + +2001-12-20 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (keyserver_spawn): Use mkdtemp() to make temp + directory. + + * mkdtemp.c: replacement function for those platforms that don't + have mkdtemp (make a temp directory securely). + +2001-12-19 David Shaw <dshaw@jabberwocky.com> + + * misc.c (check_permissions): New function to stat() and ensure + the permissions of GNUPGHOME and the files have safe permissions. + + * keydb.c (keydb_add_resource): Check keyring permissions. + + * tdbio.c (tdbio_set_dbname): Check permissions of trustdb.gpg + + * keyserver.c (keyserver_spawn): Disable keyserver schemes that + involve running external programs if the options file has unsafe + permissions or ownership. + + * g10.c, options.h: New option --no-permission-warning to disable + the permission warning message(s). This also permits use of the + keyserver if it had been disabled (see above). Also check the + permissions/ownership of random_seed. + + * keyserver.c (keyserver_spawn): The new glibc prints a warning + when using mktemp() (the code was already secure, but the warning + was bound to cause confusion). Use a different implementation + based on get_random_bits() instead. Also try a few times to get + the temp dir before giving up. + +2001-12-19 Werner Koch <wk@gnupg.org> + + * g10.c, passphrase.c [CYGWIN32]: Allow this as an alias for MINGW32. + +2001-12-18 David Shaw <dshaw@jabberwocky.com> + + * g10.c (idea_cipher_warn): Add a flag to show the warning always + or once per session and change all callers (show always except for + the secret key protection and unknown cipher from an encrypted + message errors). Also make the strings translatable. + + * pubkey-enc.c (get_it): Add the IDEA cipher warning if the user + tries to decrypt an IDEA encrypted message without the IDEA + plugin. + + * keyserver.c (parse_keyserver_uri): More strict checking of the + keyserver URI. Specifically, fail if the ":port" section is + anything except a number between 1 and 65535. + +2001-12-17 David Shaw <dshaw@jabberwocky.com> + + * keyserver.c (print_keyinfo): No need to check for + control/illegal characters, as utf8_to_native does this for us. + + * mainproc.c (proc_encrypted): Use generic IDEA warning. + + * gpgv.c: add stub for idea_cipher_warn + + * g10.c, hkp.c, keyserver.c: Fix capitalization and plural issues. + + * encode.c (encode_crypt), sign.c (sign_file, clearsign_file): + disable pgp2 mode after the message is no longer pgp2 compatible. + + * g10.c (main): Tweak the PGP2.x IDEA warning to use the generic + warning, and not merely fail if the IDEA plugin isn't there. + + * g10.c (main, idea_cipher_warn), keygen.c (set_one_pref), + seckey-cert.c (do_check): Add a generic IDEA warning for when the + IDEA plugin is not present. This pops up when the user uses + "--cipher-algo idea", when setpref is used to set a "S1" + preference, and when a secret key protected with IDEA is used. + +2001-12-15 Werner Koch <wk@gnupg.org> + + * keyserver.c (keyserver_spawn): Assert that we have dropped privs. + +2001-12-13 Werner Koch <wk@gnupg.org> + + * pubkey-enc.c (get_session_key): Check that the public key + algorithm is indeed usable for en/decryption. This avoid a + strange error message from pubkey_decrypt if for some reasons a + bad algorithm indentifier is passed. + +2001-12-12 David Shaw <dshaw@jabberwocky.com> + + * Fixed some types for portability. Noted by Stefan Bellon. + +2001-12-11 Werner Koch <wk@gnupg.org> + + * hkp.c (hkp_export): Do not print possible control characters + from a keyserver response. + (parse_hkp_index): Made uid an unsigned char* because it is passed to + isspace(). + (hkp_search): Ditto for the char* vars. + + * g10.c (main): Print the IDEA warning also for -c and -se. + + * g10.c (get_temp_dir): Assert that we have dropped privs + + * encode.c (encode_crypt): Include the first key into the --pgp2 + check. + +2001-12-07 David Shaw <dshaw@jabberwocky.com> + + * g10.c, options.h: New option --pgp2. This is identical to + "--rfc1991 --cipher-algo idea --compress-algo 1 --digest-algo md5 + --force_v3_sigs" with the addition of an warning to advise the + user not to use a pipe (which would break pgp2 compatibility). + + * encode.c (encode_crypt): warn if the user tries to encrypt to + any key that is not RSA and <= 2048 bits when the --pgp2 option is + used. + + * sign.c (sign_file, clearsign_file): When using --pgp2, make a v3 + sig, and warn if the signature is made with a non-v3 key. + +2001-12-05 David Shaw <dshaw@jabberwocky.com> + + * sign.c (sign_file, clearsign_file, sign_symencrypt_file): Prompt + for sig expiration if --expert is set and --force-v3-sigs is not + set (v3 sigs cannot expire). + + * mainproc.c (check_sig_and_print): After checking a sig, print + expiration status. This causes a error return if the sig is + expired. + + * build-packet.c (build_sig_subpkt_from_sig): Include a critical + sig expiration subpacket if the sig is to expire. + + * keyedit.c (sign_uids): Do not sign an expired key unless + --expert is set, in which case prompt. Also, offer to expire a + signature when the key the user is signing expires. + + * keygen.c (ask_expire_interval): Add a value to determine whether + to prompt for a key or sig expiration and change all callers. + + * keyid.c: New functions: expirestr_from_sig and + colon_expirestr_from_sig. + + * keylist.c (list_keyblock_colon): Show sig expiration date in the + --with-colons listing. + + * sign.c (make_keysig_packet, write_signature_packets): Pass in an + optional timestamp for the signature packet, and change all + callers. + + * keyedit.c (sign_mk_attrib): Include a critical expiration + subpacket in the signature if an expiration date is given. + +2001-12-04 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (sign_uids): If the user tries to sign a + locally-signed key, allow the cert to be promoted to a full + exportable signature. This essentially deletes the old + non-exportable sig, and replaces it with a new exportable one. + +2001-12-04 David Shaw <dshaw@jabberwocky.com> + + * keyedit.c (keyedit_menu): Do not allow signing a revoked key + unless --expert is set, and ask even then. + + * keyedit.c (sign_uids): Do not allow signing a revoked UID unless + --expert is set, and ask even then. + + * g10.c, options.h : New option --expert + +2001-11-16 David Shaw <dshaw@jabberwocky.com> + + * Allow the user to select no compression via "--compress-algo 0" + on the command line. + + * keyedit.c (show_prefs): Show compression preferences in the + long-form "showpref" style. + + * keygen.c (set_one_pref): Permit setting a no-compression ("Z0") + preference. + + * getkey.c (fixup_uidnode): Fix compression preference corruption + bug. + +2001-12-02 David Shaw <dshaw@jabberwocky.com> + + * g10.c: Add advisory --for-your-eyes-only option as per section + 5.9 of 2440. + +2001-12-05 David Shaw <dshaw@jabberwocky.com> + + * Force a V4 sig if the user has a notation or policy URL set. + +2001-12-04 David Shaw <dshaw@jabberwocky.com> + + * g10.c: Add options --keyserver-options, --temp-directory, and + auto-key-retrieve (the opposite of no-auto-key-retrieve). + + * hkp.c (hkp_search): New function to handle searching a HKP + keyserver for a key + + * hkp.c (hkp_ask_import, hkp_export): Pretty large changes to make + them communicate via the generic functions in keyserver.c + + * keyserver.c: new file with generic keyserver routines for + getting keys from a keyserver, sending keys to a keyserver, and + searching for keys on a keyserver. Calls the internal HKP stuff + in hkp.c for HKP keyserver functions. Other calls are handled by + an external program which is spawned and written to and read from + via pipes. Platforms that don't have pipes use temp files. + +2001-11-20 David Shaw <dshaw@jabberwocky.com> + + * options.h, g10.c: New options show-notation, no-show-notation, + default-check-level, no-default-check-level, show-policy-url, + no-show-policy-url. + + * packet.h, sign.c (make_keysig_packet), parse-packet.c + (parse_signature), free-packet.c (free_seckey_enc): Fill in + structures for notation, policy, sig class, exportability, etc. + + * keyedit.c, keylist.c (print_and_check_one_sig, + list_keyblock_print): Show flags in signature display for cert + details (class, local, notation, policy, revocable). If selected, + show the notation and policy url. + + * keyedit.c (sign_uids): Prompt for and use different key sig + classes. + + * helptext.c (helptexts): Add help text to explain different + key signature classes + +2001-11-26 David Shaw <dshaw@jabberwocky.com> + + * trustdb.c (mark_usable_uid_certs): Fix segfault from bad + initialization and fix reversed key signature expiration check. + +2001-11-09 Werner Koch <wk@gnupg.org> + + * export.c (do_export_stream): Put all given names into a search + description and change the loop so that all matching names are + returned. + +2001-11-08 Werner Koch <wk@gnupg.org> + + * pubkey-enc.c (get_it): To reduce the number of questions on the + MLs print the the name of cipher algorithm 1 with the error message. + + * mainproc.c: Changed the way old rfc1991 encryption cipher is + selected. Based on a patch by W Lewis. + + * pkclist.c (do_edit_ownertrust): Allow to skip over keys, the non + working "show info" is now assigned to "i" + * trustdb.c (ask_ownertrust, validate_keys): Implement a real quit + here. Both are by David Shaw. + + * trustdb.c (validate_keys): Make sure next_exipire is initialized. + + * sign.c (make_keysig_packet): Use SHA-1 with v4 RSA keys. + + * g10.c, options.h : New option --[no-]froce-v4-certs. + * sign.c (make_keysig_packet): Create v4 sigs on v4 keys even with + a v3 key. Use that new option. By David Shaw + + * revoke.c (ask_revocation_reason): Allow to select "no reason". + By David Shaw. + + * keyid.c (fingerprint_from_sk): Calculation of an v3 fpr was + plain wrong - nearly the same code in fingerprint_from_pk is correct. + + * build-packet.c (do_secret_key): Added a few comments to the code. + +2001-11-07 Werner Koch <wk@gnupg.org> + + * g10.c (main): Print a warning when -r is used w/o encryption. + Suggested by Pascal Scheffers. + +2001-10-23 Werner Koch <wk@gnupg.org> + + * keyedit.c (keyedit_menu): Changed helptext for showpref + command. Suggested by Reinhard Wobst. + + * keyring.c (keyring_search): When marking the offtbl ready, take + into account that we may have more than one keyring. + +2001-10-22 Werner Koch <wk@gnupg.org> + + * Makefile.am: Do not use OMIT_DEPENDENCIES + + * build-packet.c (build_sig_subpkt): Default is now to put all + types of subpackets into the hashed area and only list those which + should go into the unhashed area. + +2001-10-18 Werner Koch <wk@gnupg.org> + + * keydb.c (keydb_add_resource): Rearranged the way we keep track + of the resource. There will now be an entry for each keyring here + and not in keyring.c itself. Store a token to allow creation of a + keyring handle. Changed all functions to utilize this new design. + (keydb_locate_writable): Make a real implementation. + * keyring.c (next_kr): Removed and changed all callers to set the + resource directly from the one given with the handle. + (keyring_is_writable): New. + (keyring_rebuild_cache): Add an arg to pass the token from keydb. + +2001-10-17 Werner Koch <wk@gnupg.org> + + * keyring.c (keyring_search): Enabled word search mode but print a + warning that it is buggy. + +2001-10-11 Werner Koch <wk@gnupg.org> + + * hkp.c (hkp_ask_import): No more need to set the port number for + the x-hkp scheme. + (hkp_export): Ditto. + +2001-10-06 Stefan Bellon <sbellon@sbellon.de> + + * passphrase.c [__riscos__]: Disabled agent specific stuff. + * g10.c: New option --no-force-v3-sigs. + +2001-10-04 Werner Koch <wk@gnupg.org> + + * export.c (do_export_stream): Do not push the compress filter + here because the context would run out of scope due to the + iobuf_close done by the caller. + (do_export): Do it here instead. + +2001-09-28 Werner Koch <wk@gnupg.org> + + * keyedit.c (sign_uids): Always use the primary key to sign keys. + * getkey.c (finish_lookup): Hack to return only the primary key if + a certification key has been requested. + + * trustdb.c (cmp_kid_for_make_key_array): Renamed to + (validate_one_keyblock): this and changed arg for direct calling. + (make_key_array): Renamed to + (validate_one_keyblock): this and changed args for direct calling. + (mark_usable_uid_certs, validate_one_keyblock) + (validate_key_list): Add next_expire arg to keep track of + expiration times. + (validate_keys): Ditto for UTKs and write the stamp. + + * tdbio.c (migrate_from_v2): Check return code of tbdio_sync. + + * tdbdump.c (import_ownertrust): Do a tdbio_sync(). + + * keyring.c: Made the offtbl an global object. + +2001-09-27 Werner Koch <wk@gnupg.org> + + * pkclist.c (do_edit_ownertrust): Allow settin of ultimate trust. + + * trustdb.c (mark_keyblock_seen): New. + (make_key_array): Use it to mark the subkeys too. + (validate_keys): Store validity for ultimatly trusted keys. + +2001-09-26 Werner Koch <wk@gnupg.org> + + * pkclist.c (check_signatures_trust, do_we_trust): Removed the + invocation of add_ownertrust. Minor changes to the wording. + (add_ownertrust, add_ownertrust_cb): Removed. + + * trustdb.c (get_validity): Allow to lookup the validity using a + subkey. + + * trustdb.c (new_key_hash_table): Increased the table size to 1024 + and changed the masks accordingly. + (validate): Changed stats printing. + (mark_usable_uid_certs): New. + (cmp_kid_for_make_key_array): Does now check the signatures and + figures out a usable one. + +2001-09-25 Werner Koch <wk@gnupg.org> + + * keyring.c (new_offset_item,release_offset_items) + (new_offset_hash_table, lookup_offset_hash_table) + (update_offset_hash_table, update_offset_hash_table_from_kb): New. + (keyring_search): Use a offset table to optimize search for + unknown keys. + (keyring_update_keyblock, keyring_insert_keyblock): Insert new + offsets. + * getkey.c (MAX_UNK_CACHE_ENTRIES): Removed the unknown keys + caching code. + + * g10.c, options.h, import.c: Removed the entire + allow-secret-key-import stuff because the validity is now + controlled by other means. + + * g10.c: New command --rebuild-keydb-caches. + * keydb.c (keydb_rebuild_caches): New. + * keyring.c (do_copy): Moved some code to + (create_tmp_file, rename_tmp_file, write_keyblock): new functions. + (keyring_rebuild_cache): New. + + * packet.h (PKT_ring_trust): Add sigcache field. + * parse-packet.c (parse_trust): Parse sigcache. + * keyring.c (do_copy): Always insert a sigcache packet. + (keyring_get_keyblock): Copy the sigcache packet to the signature. + * sig-check.c (cache_sig_result): Renamed from + cache_selfsig_result. Changed implementation to use the flag bits + and changed all callers. + (mdc_kludge_check): Removed this unused code. + (do_check): Do not set the sig flags here. + + * import.c (read_block): Make sure that ring_trust packets are + never imported. + * export.c (do_export_stream): and never export them. + + * trustdb.c (make_key_array): Skip revoked and expired keys. + +2001-09-24 Werner Koch <wk@gnupg.org> + + * g10.c, options.h: New option --no-auto-check-trustdb. + + * keygen.c (do_generate_keypair): Set newly created keys to + ultimately trusted. + + * tdbio.h, tdbio.c: Removed all support for records DIR, KEY, UID, + PREF, SIG, SDIR and CACH. Changed migration function to work + direct on the file. + (tdbio_read_nextcheck): New. + (tdbio_write_nextcheck): New. + +2001-09-21 Werner Koch <wk@gnupg.org> + + Revamped the entire key validation system. + * trustdb.c: Complete rewrite. No more validation on demand, + removed some functions, adjusted to all callers to use the new + and much simpler interface. Does not use the LID anymore. + * tdbio.c, tdbio.h: Add new record types trust and valid. Wrote a + migration function to convert to the new trustdb layout. + * getkey.c (classify_user_id2): Do not allow the use of the "#" + prefix. + * keydb.h: Removed the TDBIDX mode add a skipfnc to the + descriptor. + * keyring.c (keyring_search): Implemented skipfnc. + + * passphrase.c (agent_open): Add missing bracket. Include windows.h. + +2001-09-19 Werner Koch <wk@gnupg.org> + + * keylist.c (print_fingerprint): Renamed from fingerprint, made + global available. Added new arg to control the print style. + * mainproc.c (print_fingerprint): Removed. + * pkclist.c (print_fpr, fpr_info): Removed and changed callers to + use print_fingerprint. + * keyedit.c (show_fingerprint): Ditto. + + * passphrase.c (writen, readn) + (agent_open, agent_close) + (agent_get_passphrase) + (passphrase_clear_cache): Support for W32. Contributed by Timo. + + * import.c (import_one): Release keydb handles at 2 more places. + + * keyring.c (keyring_release): Close the iobuf. + (keyring_get_keyblock): Init ret_kb to NULL and store error contidion. + + * import.c (import_new_stats_handle): New. + (import_release_stats_handle): New. + (import_print_stats): Renamed from static fnc print_stats. + (import_keys, import_keys_stream): Add an optional status handle + arg and changed all callers. + * hkp.c (hkp_ask_import): Add an stats_handle arg and changed all + callers. + + * mainproc.c (print_pkenc_list): Use print_utf8_string2(). + +2001-09-18 Werner Koch <wk@gnupg.org> + + * g10.c: New command --refresh-keys. + * hkp.c (hkp_refresh_keys): New. Contributed by Timo Schulz. + + * parse-packet.c (parse): Stop on impossible packet lengths. + +2001-09-17 Werner Koch <wk@gnupg.org> + + * mainproc.c (print_notation_data): Wrap notation data status lines + after 50 chars. + + * mainproc.c (proc_pubkey_enc): Make option try-all-secrets work. + By disastry@saiknes.lv. + +2001-09-14 Werner Koch <wk@gnupg.org> + + * parse-packet.c (dump_sig_subpkt): List key server preferences + and show the revocable flag correctly. Contributed by David Shaw. + +2001-09-09 Werner Koch <wk@gnupg.org> + + * keyedit.c (keyedit_menu): No need to define another p. + + * keylist.c (print_capabilities): s/used/use/ so that it + does not shadow a global. + * sign.c (sign_file): Renamed arg encrypt to encryptflag + * keygen.c: Replaced all "usage" by "use". + * misc.c (openpgp_pk_algo_usage): Ditto. + + * pubkey-enc.c (get_it): Renamed arg k to enc so that the later + defined k does not shadow it. + + * parse-packet.c (parse_gpg_control): No need to define another i. + + * getkey.c (get_pubkey_byfprint): Must use the enum values and not + the fprint_len. + * keyring.c (keyring_search): Removed a non-sense break. Both + bugs pointed out by Stefan. + +2001-09-07 Werner Koch <wk@gnupg.org> + + * status.c, status.h: Added NO_RECP and ALREADY_SIGNED. + * pkclist.c (build_pk_list): Issue NO_RECP. + * keyedit.c (sign_uids): Added experimental ALREADY_SIGNED + + * hkp.c (hkp_import): Use log_error. Bug reported by Neal H + Walfield. + + * getkey.c (classify_user_id2): Change args to take the desc union + direct. It was a stupid idea to pass the individual fields of an + union to this function. Changed all callers. + (classify_user_id): Ditto and allow to pass NULL as the description. + +2001-09-06 Werner Koch <wk@gnupg.org> + + * getkey.c (fixup_uidnode): Features flag is now a bit vector. + * keygen.c (add_feature_mdc): Ditto. + + Revamped the entire key I/O code to be prepared for other ways of + key storages and to get rid of the existing shit. GDBM support has + gone. + * keydb.c: New + * keyring.c, keyring.h: New. + * ringedit.c: Removed. Moved some stuff to keyring.c + * getkey.c: Changed everything related to the key retrieving + functions which are now using the keydb_ functions. + (prepare_search, word_match_chars, word_match) + (prepare_word_match, compare_name): Moved to keyring.c + (get_pubkey_byname): Removed ctx arg and add ret_kdbhd + arg. Changed all callers. + (key_byname): Use get_pubkey_end to release the context and take + new ret_kbdhd arg. Changed all callers. + (classify_user_id2): Fill the 16 byte fingerprint up with 4 null + bytes not with zero bytes of value 4, tsss. + * import.c (import_one): Updated to use the new keydb interface. (import_secret_one): Ditto. (import_revoke_cert): Ditto. - (import_one): Ditto. - -Fri Oct 6 14:29:16 CEST 2000 Werner Koch <wk@openit.de> - - Started to rework the whole getkey/ringedit stuff to make - it simpler, correcter and faster. - - * parse-packet.c (parse_packet): Add a 3rd arg to return the filepos. - Changed all callers. - * getkey.c (classify_user_id): Add new mode 21. - (find_by_fpr): Find using this new mode. - (get_seckey_byname): New arg to return the context. Changed all - callers. - * keyid.c (unified_fingerprint_from_pk): New. - (unified_fingerprint_from_sk): New. - * ringedit.c (find_keyblock_bypk): Changed to use the unified - fingerprint for lookup. I can't see a reason why we did compare - the entire public key. - (find_keyblock_bysk): Ditto. - (search,cmp_pubkey,cmp_seckey): Removed. - (keyring_search, do_kbxf_search): Removed. - (locate_keyblock_by_fpr,locate_keyblock_by_keyid): Removed. - (find_keyblock_byname): Removed use o search function. - (find_secret_keyblock_byname): Ditto. - (merge_public_with_secret): Fixed removing subkeys. - (premerge_public_with_secret): New. - - * ringedit.c: Removed all GDBM support - - * ringedit.c (read_keyblock): Removed. - * ringedit.c (find_keyblock_byname,find_secret_keyblock_byname, - find_keyblock_bypk,find_keyblock_bysk): Moved from here to .... - * getkey.c: ... here. Changed first arg to return a keyblock and - changed all callers to merge the old read_keyblock() with these - functions. - -Wed Oct 4 13:16:18 CEST 2000 Werner Koch <wk@openit.de> - - * getkey.c (merge_selfsigs_main): Fixed for v3 keys. - - * sign.c (hash_for): New arg to take packet version in account. Changed - all callers. - (write_one_sig): New. Moved the shared code from sign_file and - clearsign_file to here. - * skclist.c (build_sk_list): Fixed usage check. - * pkclist.c (build_pk_list): Ditto. - - * encode.c (encode_crypt): Removed duplicated stuff by using - encrypt_filter as sign.c already did. Removed already disabled - comment-packet code. - -Mon Sep 18 16:35:45 CEST 2000 Werner Koch <wk@openit.de> - - * parse-packet.c (dump_sig_subpkt): Dump key flags. - (parse_one_sig_subpkt,can_handle_critical): Add KeyFlags support. - * build-packet.c (build_sig_subpkt): Ditto. + * delkey.c (do_delete_key): Ditto. + * keyedit.c (keyedit_menu): Ditto. + (get_keyblock_byname): Removed. + * revoke.c (gen_revoke): Ditto. + * export.c (do_export_stream): Ditto. + * trustdb.c (update_trustdb): Ditto. + * g10.c, gpgv.c (main): Renamed add_keyblock_resource to + keydb_add_resource. + * Makefile.am: Added and removed files. - * g10.c: New option --allow-freeform-uid. By Jeroen C. van Gelderen. - * keygen.c (ask_user_id): Implemented here. + * keydb.h: Moved KBNODE typedef and MAX_FINGERPRINT_LEN to + * global.h: this new header. + +2001-09-03 Werner Koch <wk@gnupg.org> - * parse-packet.c (dump_sig_subpkt): Print info about the ARR. + * passphrase.c (agent_get_passphrase): Changed nread to size_t. + (passphrase_clear_cache): Ditto. - * openfile.c (overwrite_filep): Always return okay if the file is - called /dev/null. - (make_outfile_name): Add ".sign" to the list of know extensions. - (open_sigfile): Ditto. + * keyid.c (mk_datestr): Avoid trigraphs. + (fingerprint_from_pk): Cache the keyid in the pk. - * getkey.c: Large parts rewritten to have a better sub key selection - and handle some meta information from signatures more correctly. - (get_primary_seckey): Removed. - * seckey_cert.c (do_check): Set main keyid from the data in the sk. - * free-packet.c (copy_public_parts_to_secret_key): New. - * sig-check.c (check_key_signature2): Enabled shortcut for already - checked signatures. - * keydb.h: New macros IS_xxx_SIG, IS_xxx_REV. - * misc.c (openpgp_pk_algo_usage): New. - * packet.h: New field req_uage and do not use pubkey_usage anymore - to request a specific usage. Changed at all places. - * keyid.c (keyid_from_sk): Cache the keyid in the sk + * options.h: Add opt.with_fingerprint so that we know whether the + corresponding options was used. + * g10.c (main): Set it here. + * pkclist.c (check_signatures_trust): Always print fingerprint + when this option is used. Mixed a minor memory leak. - * passphrase.c (hash_passphrase): Removed funny assert. Reported by - David Mathog. + * status.c, status.h: New status INV_RECP. + * pkclist.c (build_pk_list): Issue this status. - * keyedit.c (keyedit_menu): Allow "debug" on secret keys. +2001-08-31 Werner Koch <wk@gnupg.org> - * keygen.c (keygen_add_std_prefs): Changed order of preferences to - twofish, cast5, blowfish. + * parse-packet.c (parse_key,parse_pubkeyenc) + (parse_signature): Return error on reading bad MPIs. + + * mainproc.c (check_sig_and_print): Always print the user ID even + if it is not bound by a signature. Use the primary UID in the + status messages and encode them in UTF-8 + * status.c (write_status_text_and_buffer): New. + +2001-08-30 Werner Koch <wk@gnupg.org> + + * packet.h (sigsubpkttype_t): Add SIGSUBPKT_FEATURES. + (PKT_public_key, PKT_user_id): Add a flag for it. + * parse-packet.c, build-packet.c: Add support for them. + * getkey.c (fixup_uidnode, merge_selfsigs): Set the MDC flags. + * keygen.c (add_feature_mdc): New. + (keygen_upd_std_prefs): Always set the MDC feature. + * keyedit.c (show_prefs): List the MDC flag + * pkclist.c (select_mdc_from_pklist): New. + * encode.c (encode_crypt, encrypt_filter): Test whether MDC + should be used. + * cipher.c (write_header): Set MDC use depending on the above test. + Print more status info. + + * delkey.c (do_delete_key): Kludge to delete a secret key with no + public key available. + + * ringedit.c (find_secret_keyblock_direct): New. + * getkey.c (seckey_available): Simplified. + + * ringedit.c (cmp_seckey): Now compares the secret key against the + public key while ignoring all secret parts. + (keyring_search): Use a public key packet as arg. Allow to search + for subnkeys + (search): Likewise. Changed all callers. + (find_secret_keyblock_bypk): New. + (find_secret_keyblock_byname): First locate the pubkey and then + find the correponding secret key. + * parse-packet.c (parse): Renamed pkttype arg to onlykeypkts and + changed code accordingly. Changed all callers. + (search_packet): Removed pkttype arg. + * keyedit.c (keyedit_menu): First locate the public key and then + try to locate a secret key. + + * ringedit.c (locate_keyblock_by_fpr): Removed. + (locate_keyblock_by_keyid): Removed. + (find_keyblock_bysk): Removed. + + * sig-check.c (check_key_signature2): Print the keyid along with + the wrong sig class errors. + +2001-08-24 Werner Koch <wk@gnupg.org> + + * sign.c (sign_file): Stripped the disabled comment packet code. + (sign_file, sign_symencrypt_file): Moved common code to .. + (write_onepass_sig_packets): .. this new function. + (sign_file, clearsign_file, sign_symencrypt_file): Moved common + code to + (write_signature_packets): this new function. + (write_signature_packets, make_keysig_packet) + (update_keysig_packet): Moved common code to + (hash_uid, hash_sigclass_to_magic): these new functions + (sign_file, sign_symencrypt_file): Moved common code to + (write_plaintext_packet): this new function. + +2001-08-21 Stefan Bellon <sbellon@sbellon.de> + + * trustdb.c (query_trust_info): Changed trustlevel to signed int. + * g10.c [__riscos__]: Fixed handling of --use-agent --lock-multiple. + +2001-08-20 Werner Koch <wk@gnupg.org> + + * encr-data.c (decrypt_data): Keep track on whether we already + printed information about the used algorithm. + * mainproc.c (proc_encrypted): Removed the non-working IDEA hack + and print a message about the assumed algorithm. + * passphrase.c (passphrase_to_dek): Use the same algorithm as above. + (proc_symkey_enc): Print the algorithm, so that the user knows it + before entering the passphrase. + (proc_pubkey_enc, proc_pubkey_enc): Zero the DEK out. + * encode.c (encode_crypt, encrypt_filter): Ditto. + + * g10.c: Allow for --sign --symmetric. + * sign.c (sign_and_symencrypt): New. + + Applied patches from Stefan Bellon <sbellon@sbellon.de> to support + RISC OS. Nearly all of these patches are identified by the + __riscos__ macro. + * compress.c: Added a couple of casts. + * g10.c [__riscos__]: Some patches and new options foo-file similar + to all foo-fd options. + * gpgv.c, openfile.c, ringedit.c, tdbio.c: Minor fixes. Mainly + replaced hardcoded path separators with EXTSEP_S like macros. + * passprase.c [__riscos__]: Disabled agent stuff + * trustdb.c (check_trust): Changed r_trustlevel to signed int to + avoid mismatch problems in pkclist.c + * pkclist.c (add_ownertrust): Ditto. + * plaintext.c (handle_plaintext) [__riscos__]: Print a note when + file can't be created. + * options.h [__riscos__]: Use an extern unless included from the + main module. + * signal.c (got_fatal_signal) [__riscos__]: Close all files. + +2001-08-14 Werner Koch <wk@gnupg.org> + + * keygen.c (ask_algo): New arg r_usage. Allow for RSA keys. + (gen_rsa): Enabled the code. + (do_create): Enabled RSA branch. + (parse_parameter_usage): New. + (proc_parameter_file): Handle usage parameter. + (read_parameter_file): Ditto. + (generate_keypair): Ditto. + (generate_subkeypair): Ditto. + (do_generate_keypair): Ditto. + (do_add_key_flags): New. + (keygen_add_std_prefs): Use the new function. + (keygen_add_key_flags_and_expire): New. + (write_selfsig, write_keybinding): Handle new usage arg. + * build-packet.c (build_sig_subpkt): Make sure that key flags go + into the hashed area. + + * keygen.c (write_uid): Initialize the reference cunter. + + * keyedit.c (keyedit_menu): No more need to update the trustdb for + preferences. Added calls to merge keblock. + + * kbnode.c (dump_kbnode): Print some more flags. + +2001-08-10 Werner Koch <wk@gnupg.org> + + Revamped the preference handling. + + * packet.h (prefitem_t, preftype_t): New. + (PKT_public_key): Added a uid field. + (PKT_user_id): Added field to store preferences and a reference + counter. + * parse-packet.c (parse_user_id,parse_photo_id): Initialize them + * free-packet.c (free_user_id): Free them. + (copy_user_id): Removed. + (scopy_user_id): New. + (cmp_user_ids): Optimized for identical pointers. + (release_public_key_parts): Release the uid. + (copy_public_key_with_new_namehash): Removed. + (copy_prefs): New. + * keyedit.c (menu_adduid): Use the new shallow copy user id. + (show_prefs): Adjusted implementation. + (keyedit_menu): No more need to update the trustdb after changing + preferences. + * getkey.c (fixup_uidnode): Store preferences. + (find_by_name): Return a user id packet and remove namehash stuff. + (lookup): Removed the unused namehash stuff. + (finish_lookup): Added foundu arg. + (pk_from_block): Removed the namehash arg and changed all callers. + (merge_selfsigs): Copy prefs to all keys. + * trustdb.c (get_pref_data): Removed. + (is_algo_in_prefs): Removed. + (make_pref_record): Deleted and removed all class. + * pkclist.c (select_algo_from_prefs): Adjusted for the new + preference implementation. + * pubkey-enc.c (is_algo_in_prefs): New. + (get_it): Use that new function. + +2001-08-09 Werner Koch <wk@gnupg.org> + + * build-packet.c (build_sig_subpkt): Fixed calculation of + newarea->size. + + * g10.c (main): New option "--preference-list" + * keyedit.c (keyedit_menu): New commands "setpref" and "updpref". + (menu_set_preferences): New. + * keygen.c (keygen_set_std_prefs): New. + (set_one_pref): New. + (check_zip_algo): New. + (keygen_get_std_prefs): New. + (keygen_upd_std_prefs): New + (keygen_add_std_prefs): Move the pref setting code into the above fnc. + * build-packet.c (build_sig_subpkt): Updated the list of allowed + to update subpackets. + +2001-08-08 Werner Koch <wk@gnupg.org> + + * packet.h (subpktarea_t): New. + (PKT_signature): Use that type for hashed_data and unhashed_data and + removed the _data prefix from those fields. Changed all users. + * parse-packet.c (parse_signature): Changed allocation for that. + (parse_sig_subpkt): Changed declaration + (enum_sig_subpkt): Ditto and changed implementation accordingly. + * free-packet.c (cp_subpktarea): Renamed from cp_data_block and + adjusted implementation. Changed caller. + * sig-check.c (mdc_kludge_check): Adjusted the hashing. + (do_check): Ditto. + * sign.c (sign_file, clearsign_file, make_keysig_packet, + update_keysig_packet): Ditto. + * build-packet.c (build_sig_subpkt): Partial rewrite. + (find_subpkt): Adjusted and made static. + (delete_sig_subpkt): Adjusted. + (do_signature): Ditto. - * gpg.c: The --trusted-key option is back. - * trustdb.c (verify_own_key): Handle this option. - (add_ultimate_key): Moved stuff from verify_own_key to this new func. - (register_trusted_key): New. + * keygen.c (ask_keysize): Do not print the notes about suggested + key sizes if just a DSA key is generated. - * openfile.c (try_make_homedir): Changes for non-Posix systems. - * gpg.c (main): Take the default homedir from macro. + * trustdb.c (add_ultimate_key): s/log_error/log_info/ for + duplicated inserted trusted keys. - * encode.c (encode_simple, encode_crypt): Fix for large files. - * sign.c (sign_file): Ditto. +2001-08-07 Werner Koch <wk@gnupg.org> - * gpg.c (main): Don't set --quite along with --no-tty. By Frank Tobin. + * sign.c (sleep): Redefine for W32. - * misc.c (disable_core_dump): Don't display a warning here but a return - a status value and ... - * gpg.c (main): ...print warning here. Suggested by Sam Roberts. + * g10.c, options.h: Set new flag opt.no_homedir_creation when + --no-options is given. + * openfile.c (try_make_homedir): Don't create the homedir in that case. - * misc.c (print_pubkey_algo_note): Do not print the RSA notice. - * sig-check.c (do_signature_check): Do not emit the RSA status message. - * pubkey-enc.c (get_session_key): Ditto. +2001-08-03 Werner Koch <wk@gnupg.org> - * ringedit.c (cmp_seckey): Fix for v4 RSA keys. - * seckey-cert.c (do_check): Workaround for PGP 7 bug. + * armor.c (armor_filter): Removed the default comment string + because it could get us in trouble due to translations using non + ascii characters. - * pkclist.c (algo_available): Removed hack to disable Twofish. +2001-08-01 Werner Koch <wk@gnupg.org> - * gpg.c (main): Default S2K algorithms are now SHA1 and CAST5 - this - should solve a lot of compatibility problems with other OpenPGP - apps because those algorithms are SHOULD and not optional. The old - way to force it was by using the --openpgp option whith the drawback - that this would disable a couple of workarounds for PGP. + * keylist.c (list_keyblock_print): Do not list revoked UIDs unless + in verbose mode and we do no signature listing. + + * getkey.c (finish_lookup): Skip subkeys which are not yet valid. + * g10.c, options.h: New option --ignore-valid-from. + + * sign.c (make_keysig_packet): Added new sigversion argument to + allow the caller to force generation of required signature + version. Changed all callers. Suggested by Thomas Roessler. + + * keyedit.c (sign_uids): Force v4 signature generation for local + sigs. Removed the check for local signature and pre-v4 keys. + +2001-07-27 Werner Koch <wk@gnupg.org> + + * keyedit.c (sign_uids): Check that we are not trying to to a + lsign with a pre-v4 key. Bug noticed by Thomas Roessler. + +2001-07-26 Werner Koch <wk@gnupg.org> + + * parse-packet.c (parse_photo_id): Reset all variables. + * getkey.c (merge_selfsigs_main): Removed checks on PHOTO_ID + because this is handled identically to a user ID. + +2001-07-06 Werner Koch <wk@gnupg.org> + + * cipher.c (write_header): Don't use MDC with --rfc1991. Suggested + by disastry@saiknes.lv. + +2001-07-05 Werner Koch <wk@gnupg.org> + + * g10.c, options.h: New option --preserve-permissions. + * ringedit.c (add_keyblock_resource): Use it here + (keyring_copy): and here. + + * trustdb.c (verify_own_keys): Be more silent on --quiet. + Suggested by Thomas Roessler. + * sig-check.c (check_key_signature2): Ditto. + * mainproc.c (proc_encrypted, proc_tree): Ditto + * getkey.c (lookup): Ditto. + +2001-07-04 Werner Koch <wk@gnupg.org> + + * ringedit.c (add_keyblock_resource): Restore filename in case of error. + +2001-06-25 Werner Koch <wk@gnupg.org> + + * kbnode.c (dump_kbnode): Print the signature timestamp. + + * keyedit.c (keyedit_menu): New menu point "primary". + (change_primary_uid_cb): New. + (menu_set_primary_uid): New. + * sign.c (update_keysig_packet): New. + * build-packet.c (build_sig_subpkt): Put the primary UID flag into + the hashed area. Allow update of some more packets. + +2001-06-15 Werner Koch <wk@gnupg.org> + + * getkey.c (merge_selfsigs): Exit gracefully when a secret key is + encountered. May happen if a secret key is in public keyring. + Reported by Francesco Potorti. + +2001-06-12 Werner Koch <wk@gnupg.org> + + * getkey.c (compare_name): Use ascii_memistr(), ascii_memcasecmp() + * keyedit.c (keyedit_menu): Use ascii_strcasecmp(). + * armor.c (radix64_read): Use ascii_toupper(). + * ringedit.c (do_bm_search): Ditto. + * keygen.c (read_parameter_file): Ditto. + * openfile.c (CMP_FILENAME): Ditto. + * g10.c (i18n_init): We can now use just LC_ALL. + +2001-05-29 Werner Koch <wk@gnupg.org> + + * keygen.c (generate_subkeypair): Print a warning if a subkey is + created on a v3 key. Suggested by Brian M. Carlson. + +2001-05-27 Werner Koch <wk@gnupg.org> + + * keyid.c (get_lsign_letter): New. + * keylist.c (list_keyblock_colon): Use it here. + * mainproc.c (list_node): and here. + + * getkey.c, packet.h, free-packet.c: Removed that useless key + created field; I dunno why I introducded this at all - the + creation time is always bound to the key packet and subject to + fingerprint calculation etc. + + * getkey.c (fixup_uidnode): Add keycreated arg and use this + instead of the signature timestamp to calculate the + help_key_expire. Bug reported by David R. Bergstein. + (merge_selfsigs_main): Correct key expiration time calculation. + (merge_selfsigs_subkey): Ditto. + +2001-05-25 Werner Koch <wk@gnupg.org> + + * revoke.c (gen_revoke): Add a cast to a tty_printf arg. + * delkey.c (do_delete_key): Ditto. + * keyedit.c (print_and_check_one_sig): Ditto. + (ask_revoke_sig): Ditto. + (menu_revsig): Ditto. + (check_all_keysigs): Removed unused arg. + +2001-05-23 Werner Koch <wk@gnupg.org> + + * g10.c (opts): Typo fix by Robert C. Ames. + +2001-05-06 Werner Koch <wk@gnupg.org> + + * revoke.c: Small typo fix + +2001-05-04 Werner Koch <wk@gnupg.org> + + * passphrase.c (passphrase_clear_cache): Shortcut if agent usage + is not enabled. + +2001-05-01 Werner Koch <wk@gnupg.org> + + * passphrase.c (writen): Replaced ssize_t by int. Thanks to + to Robert Joop for reporting that SunOS 4.1.4 does not have it. + +2001-04-28 Werner Koch <wk@gnupg.org> + + * getkey.c (merge_public_with_secret): pkttype was not set to subkey. + +2001-04-27 Werner Koch <wk@gnupg.org> + + * skclist.c (build_sk_list): Changed one log_debug to log_info. + +2001-04-25 Werner Koch <wk@gnupg.org> + + * keyedit.c (show_prefs): Add a verbose mode. + (show_key_with_all_names): Pass verbose flag for special value of + with_pref. + (keyedit_menu): New command "showpref" + (show_key_with_all_names): Mark revoked uids and the primary key. + +2001-04-24 Werner Koch <wk@gnupg.org> + + * getkey.c (get_primary_uid): Return a different string in case of + error and made it translatable. + + * build-packet.c (do_secret_key): Ugly, we wrote a zero + instead of the computed ndays. Thanks to M Taylor for complaining + about a secret key import problem. + +2001-04-23 Werner Koch <wk@gnupg.org> + + * hkp.c (hkp_ask_import): Allow to specify a port number for the + keyserver. Add a kudge to set the no_shutdown flag. + (hkp_export): Ditto. + * options.skel: Document the changes + +2001-04-20 Werner Koch <wk@gnupg.org> + + * options.skel: Add some more comments. + +2001-04-19 Werner Koch <wk@gnupg.org> + + * keyid.c (mk_datestr): New. Handles negative times. We must do + this because Windoze segvs on negative times passed to gmtime(). + Changed all datestr_from function to use this one. + + * keyid.c, keyid.h (colon_strtime): New. To implement the + fixed-list-mode. + (colon_datestr_from_pk): New. + (colon_datestr_from_sk): New. + (colon_datestr_from_sig): New. + * keylist.c (list_keyblock_colon): Use these functions here. + * mainproc.c (list_node): Ditto. + +2001-04-18 Werner Koch <wk@gnupg.org> + + * openfile.c (open_sigfile): Fixed the handling of ".sign". + * mainproc.c (proc_tree): Use iobuf_get_real_fname. + Both are by Vincent Broman. + +2001-04-14 Werner Koch <wk@gnupg.org> + + * getkey.c (fixup_uidnode): Removed check for !sig which is + pointless here. Thanks to Jan Niehusmann. + +2001-04-10 Werner Koch <wk@gnupg.org> + + * sig-check.c (check_key_signature2): Use log_info instead of + log_error so that messed up keys do not let gpg return an error. + Suggested by Christian Kurz. + + * getkey.c (merge_selfsigs_main): Do a fixup_uidnode only if we + have both, uid and sig. Thanks to M Taylor. + +2001-04-05 Werner Koch <wk@gnupg.org> + + * armor.c (unarmor_pump_new,unarmor_pump_release): New. + (unarmor_pump): New. + * pipemode.c (pipemode_filter): Use the unarmor_pump to handle + armored or non-armored detached signatures. We can't use the + regular armor_filter becuase this does only chack for armored + signatures the very first time. In pipemode we may have a mix of + armored and binary detached signatures. + * mainproc.c (proc_tree): Do not print the "old style" notice when + this is a pipemode processes detached signature. + (proc_plaintext): Special handling of pipemode detached sigs. + + * packet.h (CTRLPKT_PLAINTEXT_MARK): New. + * parse-packet.c (create_gpg_control): New. + * kbnode.c (dump_kbnode): Support it here. + * mainproc.c (check_sig_and_print): Fixed the check for bad + sequences of multiple signatures. + (proc_plaintext): Add the marker packet. + (proc_tree): We can now check multiple detached signatures. + +2001-04-02 Werner Koch <wk@gnupg.org> + + The length of encrypted packets for blocksizes != 8 was not + correct encoded. I think this is a minor problem, because we + usually use partial length packets. Kudos to Kahil D. Jallad for + pointing this out. + * packet.h: Add extralen to PKT_encrypted. + * cipher.c (write_header): Set extralen. + * build-packet.c (do_encrypted): Use extralen instead of const 10. + (do_encrypted_mdc): Ditto. + * parse-packet.c (parse_encrypted): Set extralen to 0 because we + don't know it here. + +2001-03-30 Werner Koch <wk@gnupg.org> + + * getkey.c (premerge_public_with_secret): Changed wording an add + the keyID to the info message. + +2001-03-29 Werner Koch <wk@gnupg.org> + + * getkey.c (premerge_public_with_secret): Use log_info instead of + log_error when no secret key was found for a public one. + Fix the usage if the secret parts of a key are not available. + + * openfile.c (ask_outfile_name): Trim spaces. + (open_outfile): Allow to enter an alternate filename. Thanks to + Stefan Bellon. + * plaintext.c (handle_plaintext): Ditto. + +2001-03-28 Werner Koch <wk@gnupg.org> + + * mainproc.c (do_check_sig): Allow direct key and subkey + revocation signature. + * sig-check.c (check_key_signature2): Check direct key signatures. + Print the signature class along with an error. + +2001-03-27 Werner Koch <wk@gnupg.org> + + * packet.h: Add a missing typedef to an enum. Thanks to Stefan Bellon. + + * g10.c: New option --no-sig-create-check. + * sign.c (do_sign): Implement it here. + * g10.c: New option --no-sig-cache. + * sig-check.c (check_key_signature2): Implement it here. + (cache_selfsig_result): and here. + + * keylist.c (list_keyblock): Removed debugging stuff. + + * getkey.c (cache_public_key): Made global. + * keygen.c (write_selfsig, write_keybinding): Cache the new key. + + * getkey.c (key_byname): Add new arg secmode and changed all + callers to request explicitly the mode. Deriving this information + from the other supplied parameters does not work if neither pk nor + sk are supplied. + +2001-03-25 Werner Koch <wk@gnupg.org> + + * packet.h (ctrlpkttype_t): New. + * mainproc.c (add_gpg_control,proc_plaintext,proc_tree): Use the + new enum values. + * pipemode.c (make_control): Ditto. + * armor.c (armor_filter): Ditto. + +2001-03-24 Werner Koch <wk@gnupg.org> + + * sign.c (do_sign): Verify the signature right after creation. + +2001-03-23 Werner Koch <wk@gnupg.org> + + * status.c, status.h (STATUS_UNEXPECTED): New. + * mainproc.c (do_proc_packets): And emit it here. + +2001-03-21 Werner Koch <wk@gnupg.org> + + * status.c: Add sys/types.h so that it runs on Ultrix. Reported + by Georg Schwarz.x + + * build-packet.c (build_sig_subpkt): Fixed generaton of packet + length header in case where 2 bytes headers are needed. Thanks to + Piotr Krukowiecki. + +2001-03-19 Werner Koch <wk@gnupg.org> + + * g10.c (main): the default keyring is no always used unless + --no-default-keyring is given. + + * ringedit.c (add_keyblock_resource): invalidate cache after file + creation. + +2001-03-15 Werner Koch <wk@gnupg.org> + + * keygen.c (ask_algo): Changed the warning of the ElGamal S+E Algo. + + * keylist.c (print_capabilities): New. + (list_keyblock_colon): and use it here. + +2001-03-13 Werner Koch <wk@gnupg.org> + + * main.c, options.h: New option --fixed_list_mode. + * keylist.c (list_keyblock_colon): use it here. + + * getkey.c (merge_keys_and_selfsig): Divert merging of public keys + to the function used in key selection.. + * keylist.c (is_uid_valid): Removed. + (list_keyblock): Splitted into .. + (list_keyblock_print, list_keyblock_colon): .. these. + functions. Changed them to use the flags set in the key lookup code. + (reorder_keyblock): New, so that primary user IDs are listed first. + + * ringedit.c (keyring_copy): flush the new iobuf chaces before + rename or remove operations. This is mainly needed for W32. + + * hkp.c [HAVE_DOSISH_SYSTEM]: Removed the disabled code because we + have now W32 socket support in ../util/http.c + + * skclist.c (key_present_in_sk_list): New. + (is_duplicated_entry): New. + (build_sk_list): Check for duplicates and do that before unlocking. + +2001-03-12 Werner Koch <wk@gnupg.org> + + * armor.c (parse_header_line): Removed double empty line check. + (parse_header_line): Replaced trim_trailing_ws with a counting + function so that we can adjust for the next read. + + * options.skel: Fixed 3 typos. By Thomas Klausner. Replaced the + keyserver example by a better working server. + + * parse-packet.c (parse_symkeyenc): Return Invalid_Packet on error. + (parse_pubkeyenc): Ditto. + (parse_onepass_sig): Ditto. + (parse_plaintext): Ditto. + (parse_encrypted): Ditto. + (parse_signature): Return error at other places too. + (parse_key): Ditto. + * g10.c (main): Set opt.list_packets to another value when invoked + with the --list-packets command. + * mainproc.c (do_proc_packets): Don's stop processing when running + under --list-packets command. + + * signal.c (do_sigaction): Removed. + (init_one_signal): New to replace the above. Needed to support + systems without sigactions. Suggested by Dave Dykstra. + (got_fatal_signal,init_signals): Use the above here. + (do_block): Use sigset() if sigprocmask() is not available. + + * armor.c (parse_hash_header): Test on TIGER192, which is the + correct value as per rfc2440. By Edwin Woudt. + +2001-03-08 Werner Koch <wk@gnupg.org> + + * misc.c: Include time.h. By James Troup. + + * getkey.c: Re-enabled the unknown user Id and PK caches and + increased their sizes. + + * getkey.c (merge_selfsigs_main): Set expire date and continue + processing even if we found a revoked key. + (merge_selfsigs_subkeys): Ditto. + + * packet.h: Add an is_revoked flag to the user_id packet. + * getkey.c (fixup_uidnode): Set that flag here. + (merge_selfsigs_main): Fix so that the latest signature is used to + find the self-signature for an UID. + * parse-packet.c (parse_user_id): Zero out all fields. + * mainproc.c (check_sig_and_print): Print the primary user ID + according the the node flag and then all other non-revoked user IDs. + (is_uid_revoked): Removed; it is now handled by the key selection code. + + Changed the year list of all copyright notices. + +2001-03-07 Werner Koch <wk@gnupg.org> + + * getkey.c (finish_lookup): Print an info message only in verbose mode. + +2001-03-05 Werner Koch <wk@gnupg.org> + + * packet.h: Replaced sigsubpkt_t value 101 by PRIV_VERIFY_CACHE. + We have never used the old value, so we can do this without any harm. + * parse-packet.c (dump_sig_subpkt): Ditto. + (parse_one_sig_subpkt): Parse that new sub packet. + * build-packet.c (build_sig_subpkt): Removed the old one from the + hashed area. + (delete_sig_subpkt): New. + (build_sig_subpkt): Allow an update of that new subpkt. + * sig-check.c (check_key_signature2): Add verification caching + (cache_selfsig_result): New. + * export.c (do_export_stream): Delete that sig subpkt before exporting. + * import.c (remove_bad_stuff): New. + (import): Apply that function to all imported data + +2001-03-03 Werner Koch <wk@gnupg.org> + + * getkey.c: Introduced a new lookup context flag "exact" and used + it in all place where we once used primary. + (classify_user_id2): Replaced the old function and add an extra + argument to return whether an exact keyID has been requested. + (key_byname): Removed the unused ctx.primary flag + (get_seckey_byname2): Ditto. + (finish_lookup): Changed debugging output. + +2001-03-02 Werner Koch <wk@gnupg.org> + + * keylist.c (list_one): Remove the merge key calls. + +2001-03-01 Werner Koch <wk@gnupg.org> + + * getkey.c (finish_lookup): Don't use it if we no specific usage + has been requested. + (merge_selfsigs_main): fix UID only if we have an signature. + (lookup): Return UNU_PUBKEY etc. instead of NO_PUBKEY if we found + a key but the requested usage does not allow this key. + * import.c (import_one): Take UNU_PUBKEY into account. + * mainproc.c (list_node): Ditto. + * keylist.c (list_keyblock): Ditto. + * keyedit.c (print_and_check_one_sig): Ditto. + +2001-02-09 Werner Koch <wk@gnupg.org> + + * delkey.c (delete_key): Removed that silly assert which rendered + the whole new stuff meaningless. + +2001-02-08 Werner Koch <wk@gnupg.org> + + * getkey.c (key_byname): It can happen that we have both, sk and pk + NULL, fix for that. + + * parse-packet.c (parse_one_sig_subpkt): Add support for + primary_uid and key_flags. + (can_handle_critical): Ditto + + * parse-packet.c (parse_encrypted): Fixed listing of pktlen for + MDC packets. + + * getkey.c: Backported the version of this file from gpg 1.1. this + involved some changes in other files too. + * parse-packet.c (parse_key): Clear req_usage. + * skclist.c (build_sk_list): Use req_usage to pass the usage + information to the lookup function. + * pkclist.c (build_pk_list): Ditto. + * free-packet.c (copy_public_parts_to_secret_key): New. + * keydb.h: Add IS_* macros to check the sig_class. + * misc.c (openpgp_cipher_test_algo): New. + (openpgp_pk_test_algo): New. + (openpgp_pk_algo_usage): New. + (openpgp_md_test_algo): New. + * packet.h: Add a few fields to PKT_{public,secret}_key and + PKT_user_id. + * seckey-cert.c (do_check): Use the new main_keyid field. + +2001-02-04 Werner Koch <wk@gnupg.org> + + * encr-data.c (decrypt_data): Catch error when we had problems to + parse the encrypted packet. By Timo. + +2001-01-29 Werner Koch <wk@gnupg.org> + + * g10.c (main): --batch does now set nogreeting. + + * delkey.c (do_delete_key): Fixed delete-both functionality. + +2001-01-22 Werner Koch <wk@gnupg.org> + + * g10.c: New command --delete-secret-and-public-key. + * delkey.c (delete_key): Add new arg allow_both. + (do_delete_key): Move most stuff from above to this new function. + +2001-01-12 Werner Koch <wk@gnupg.org> + + * passphrase.c (passphrase_to_dek): Use MD5 when IDEA is installed + and we have no S2K. + * mainproc.c (proc_encrypted): Likewise + +2001-01-11 Werner Koch <wk@gnupg.org> + + * sig-check.c (do_check): Print the signature key expire message + only in verbose mode and added the keyID. + +2001-01-09 Werner Koch <wk@gnupg.org> + + * status.c, status.h: New status USERID_HINT. + (write_status_text): Replace LF and CR int text by C-escape sequence. + + * passphrase.c (passphrase_to_dek): Fixed the NEED_PASSPHRASE + output. It does now always print 2 keyIDs. Emit the new + USERID_HINT. + +2001-01-08 Werner Koch <wk@gnupg.org> + + * g10.c, options.h: New option --no-expensive-trust-checks. + * keylist.c (list_keyblock): Act on this option. + +2001-01-04 Werner Koch <wk@gnupg.org> + + * g10.c (main): Set homedir only in the pre-parsing phase and + replace backslashes in the W32 version. - * gpg.c: New option --merge-only. Suggested by Brendan O'Dea. - * import.c (import_one): Implemented it here. - (import_secret_one): Ditto. - (print_stats): and give some stats. +2001-01-03 Werner Koch <wk@gnupg.org> - * gpg.c: New option --try-all-secrets on suggestion from - Matthias Urlichs. - * pubkey-enc.c (get_session_key): Quite easy to implement here. + * status.c, status.h : New status KEY_CREATED + * keygen.c (do_generate_keypair,generate_subkeypair): Emit it. -Mon Aug 21 17:59:17 CEST 2000 Werner Koch <wk@openit.de> +2000-12-28 Werner Koch <wk@gnupg.org> - * gpg.c: New option --use-agent + * signal.c (got_fatal_signal): Remove lockfiles here because the + atexit stuff does not work due to the use of raise. Suggested by + Peter Fales. + * gpgv.c (remove_lockfiles): New stub. + +2000-12-19 Werner Koch <wk@gnupg.org> + + * status.c, status.h (cpr_get_no_help): New. + * keyedit.c (keyedit_menu): Use it here because we have our own + help list here. + +2000-12-18 Werner Koch <wk@gnupg.org> + + * mainproc.c (print_failed_pkenc): Don't print the sometimes + confusing message about unavailabe secret key. Renamed ... + (print_pkenc_list): ... to this and introduced failed arg. + (proc_encrypted): Print the failed encryption keys and then + the one to be used. + (proc_pubkey_enc): Store also the key we are going to use. + + * mainproc.c (check_sig_and_print): Don't list revoked user IDs. + (is_uid_revoked): New. + +2000-12-08 Werner Koch <wk@gnupg.org> + + * pipemode.c: Made the command work. Currently only for + non-armored detached signatures. + * mainproc.c (release_list): Reset the new pipemode vars. + (add_gpg_control): Handle the control packets for pipemode + * status.c, status.h: New stati {BEGIN,END}_STREAM. + +2000-12-07 Werner Koch <wk@gnupg.org> + + * g10.c: New option --allow-secret-key-import. + * import.c (import_keys,import_keys_stream): Honor this option. + (import): New arg allow_secret and pass that arg down to ... + (import_secret_one): to this and print a warning if secret key + importing is not allowed. + +2000-12-05 Werner Koch <wk@gnupg.org> + + * cipher.c (cipher_filter): Moved the end_encryption status ... + * encode.c (encode_simple,encode_crypt): to here + * sign.c (sign_file): and here. + + * status.c (mywrite): Removed. + (get_status_string): Removed the LFs from the strings. + (set_status_fd,is_status_enabed,write_status_text, + write_status_buffer): Replaced all mywrite by stdio calls and use + fdopen to create a strem. This is needed to make things smoother + in the W32 version. + +2000-12-04 Werner Koch <wk@gnupg.org> + + * import.c (merge_blocks): Increment n_sigs for revocations. + +2000-11-30 Werner Koch <wk@gnupg.org> + + * g10.c (main): Use iobuf_translate_file_handle for all options + with filehandles as arguments. This is function does some magic + for the W32 API. + + * verify.c (verify_signatures): Add a comment rant about the + detached signature problem. + * mainproc.c (proc_tree): Issue an error if a detached signature + is assumed but a standard one was found. + * plaintext.c (hash_datafiles): Don't fall back to read signature + from stdin. + * openfile.c (open_sigfile): Print verbose message only if the + file could be accessed. + +2000-11-24 Werner Koch <wk@gnupg.org> + + * passphrase.c [HAVE_DOSISH_SYSTEM]: Disabled all the agent stuff. + +2000-11-16 Werner Koch <wk@gnupg.org> + + * g10.c: New option --use-agent * passphrase.c (agent_open,agent_close): New. (agent_get_passphrase,agent_clear_passphrase): New. + (passphrase_clear_cache): New. (passphrase_to_dek): Use the agent here. - * seckey-cert.c (do_check): Clear wrong cached passphrases. + * seckey-cert.c (do_check): Clear cached passphrases. -Fri Aug 18 14:27:14 CEST 2000 Werner Koch <wk@openit.de> +2000-11-15 Werner Koch <wk@gnupg.org> - * status.c (do_get_from_fd): Ooops, we used fd instead of opt.command_fd. - Thanks to Michael Tokarev. + * status.c (write_status_text): Moved the big switch to ... + (get_status_string): ... new function. + (write_status_buffer): New. -Mon Jul 31 10:04:47 CEST 2000 Werner Koch <wk@openit.de> + * status.c (mywrite): New and replaced all write() by this. - * encode.c, sign.c, keygen.c, pubkey-enc.c: Replaced all - gcry_sexp_{car,cdr}_{data,mpi} by the new gcry_sexp_nth_{data,mpi} functions. + * status.c, status.h: Add 3 status lcodes for notaions and policy. + * mainproc.c (print_notation_data): Do status output of notations. + +2000-11-13 Werner Koch <wk@gnupg.org> - * keygen.c (gen_dsa,gen_elg): Changed the way the factors are stored. - (factors_from_sexp): Removed. - * comment.c (make_mpi_comment_node): Removed. - (make_comment_node_from_buffer): New. + * sign.c (clearsign_file): Use LF macro to print linefeed. + +2000-11-11 Paul Eggert <eggert@twinsun.com> -Fri Jul 28 18:19:11 CEST 2000 Werner Koch <wk@openit.de> + Clean up the places in the code that incorrectly use "long" or + "unsigned long" for file offsets. The correct type to use is + "off_t". The difference is important on large-file hosts, + where "off_t" is longer than "long". - * sig-check.c (pk_verify): Fixed the S-Exp withe the pkey. + * keydb.h (struct keyblock_pos_struct.offset): + Use off_t, not ulong, for file offsets. + * packet.h (dbg_search_packet, dbg_copy_some_packets, + search_packet, copy_some_packets): Likewise. + * parse-packet.c (parse, dbg_search_packet, search_packet, + dbg_copy_some_packets, copy_some_packets): Likewise. + * ringedit.c (keyring_search): Likewise. - * gpg.c (main): Use setmode(O_BINARY) for MSDOS while generating random bytes - (print_mds): Likewise for stdin. - * plaintext.c (handle_plaintext): Likewise for stdout. + * parse-packet.c (parse): Do not use %lu to report file + offsets in error diagnostics; it's not portable. + * ringedit.c (keyring_search): Likewise. -Tue Jul 25 17:44:15 CEST 2000 Werner Koch <wk@openit.de> +2000-11-09 Werner Koch <wk@gnupg.org> - * keyedit.c (menu_expire): expire date for primary key can be set again. + * g10.c (main): New option --enable-special-filenames. - * keylist.c (is_uid_valid): New. - (list_keyblock): Print validity information for all user IDs. Note, this - has to be done at other places too; for now we have only minimal support. +2000-11-07 Werner Koch <wk@gnupg.org> + + * g10.c (main): New command --pipemode. + * pipemode.c: New. - * sign.c (pk_sign): Changed to use the new S-Exp interface. - * encode.c (pk_encrypt): Ditto. - * sig-check.c (pk_verify): Ditto. - * seckey-cert.c (pk_check_secret_key): Ditto. - * pubkey-enc.c (pk_decrypt): Ditto. - * misc.c (pubkey_nbits): Ditto. - * keygen.c (key_from_sexp,factors_from_sexp,gen_elg,gen_dsa): Ditto. +2000-10-23 Werner Koch <wk@gnupg.org> -Fri Jul 14 19:38:23 CEST 2000 Werner Koch <wk@> + * armor.c (armor_filter): Changed output of hdrlines, so that a CR + is emitted for DOS systems. - Replaced everything with the code from the STABLE-BRANCH-1-0 and - started to backport the changes from the 1.1 development branch - which are dated according to the ChangeLog of the 1.1 from - Sat Sep 18 12:16:08 CEST 1999 to Thu May 25 18:39:11 CEST 2000. - Here are those changes, some of them are duplicates because they - have been done on both branch simultaneously. + * keygen.c (read_parameter_file): Add a cast for isspace(). - * gpg.c (print_mds): Add arg keys as a kludge to print hmacs - (main): New option --print-hmac. + * status.c (myread): Use SIGINT instead of SIGHUP for DOS. - * trustdb.c (verify_own_keys): Do not print warning about unprotected - key when in quiet mode. +2000-10-19 Werner Koch <wk@gnupg.org> - * build-paket.c (do_user_id): Save offset where name has been stored. + * g10.c: New option --ignore-crc-error + * armor.c (invalid_crc): New. + (radix64_read): Act on new option. - * ringedit.c : Add new access method KBXF + * openfile.c (try_make_homedir): Klaus Singvogel fixed a stupid + error introduced on Sep 6th. - * kbxfile.c: New. +2000-10-18 Werner Koch <wk@gnupg.org> - * kbx.h: New. - * kbxblob.c: Started to work on the keybox stuff. + * misc.c (print_cipher_algo_note): Don't print the note for AES. + Changed wording. - * keygen.c (gen_dsa): Modified to work with gcry_pk_genkey. +2000-10-16 Werner Koch <wk@gnupg.org> - * Removed dummy-cipher.h from all files. + * mainproc.c (do_proc_packets): Hack to fix the problem that + signatures are not detected when there is a MDC packet but no + compression packet. + + * g10.c (print_hashline): New. + (print_mds): Use above func with --with-colons. - * keygen.c (gen_elg): Modified to work with gcry_pk_genkey. - (key_from_sexp): New. - (factors_from_sexp): New. + * mainproc.c (check_sig_and_print): Detect multiple signatures + and don't verify them. - * g10.c : Renamed to ... - * gpg.c : ... this - * Makefile.am: And fixed it here. +2000-10-14 Werner Koch <wk@gnupg.org> - * Changed all "g10_"/"GPG_" prefixes to "gpg_"/"GPG_". + * mainproc.c (add_onepass_sig): There is an easier solution to the + error fixed yesterday; just check that we only have onepass + packets. However, the other solution provides an cleaner + interface and opens the path to get access to other information + from the armore headers. + (release_list): Reset some more variables. - * misc.c (mpi_read_opaque): Fixed double counting. +2000-10-13 Werner Koch <wk@gnupg.org> - * seckey-cert.c (do_check): Removed buffer and the unmotivated free - on it. + * mainproc.c (add_gpg_control): New. + (do_proc_packets): use it. + (proc_plaintext): Changed logic to detect clearsigns. + (proc_tree): Check the cleartext sig with some new code. - * pubkey-enc.c (pk_decrypt): New wrapper for the gcry_ function. - * seckey-cert.c (pk_check_secret_key): Likewise. - * encode.c (pk_encrypt): Likewise. + * packet.h: New packet PKT_GPG_CONTROL. + * parse-packet.c (parse_gpg_control): New. + * misc.c (get_session_marker): New. + * armor.c (armor_filter): Replaced the faked 1-pass packet by the + new control packet. - * parse-packet.c (parse_key): Fixed case of unencrypted secret keys. + * keyedit.c (keyedit_menu): Allow batchmode with a command_fd. + * status.c (my_read): New. + (do_get_from_fd): use it. - * misc.c (mpi_print): Use gcry_mpi_aprint. - (pubkey_nbits): Kludge to use the gcry_pk_ API. +2000-10-12 Werner Koch <wk@gnupg.org> - * seskey.c (encode_session_key): Replaced mpi_set_buffer by *_scan. - (do_encode_md): Ditto. - (encode_md_value): Ditto. - * seckey-cert.c (protect_secret_key): Ditto. - * comment.c (make_mpi_comment_node): Replaced mpi_get_buffer by _print. - * pubkey-enc.c (get_it): Ditto. - * sig-check.c (do_signature_check): Ditto. + * keygen.c (keygen_add_std_prefs): Add Rijndael to the prefs. - * keyid.c (do_fingerprint_md): Replaced mpi_get_buffer by gcry_mpi_print. - (v3_keyid): New. - (keyid_from_sk): And use it here. - (keyid_from_pk): Ditto. - (fingerprint_from_sk): Ditto. - (fingerprint_from_pk): Ditto. +2000-10-07 Werner Koch <wk@gnupg.org> - * misc.c (mpi_print): New. + * gpgv.c: Add more stubs for ununsed code to make the binary smaller. - * misc.c (checksum_mpi): Now uses gcry_mpi_print to get the data. +Wed Oct 4 15:50:18 CEST 2000 Werner Koch <wk@openit.de> - * seckey-cert.c (do_check): Replaced mpi_read_from_buffer. + * sign.c (hash_for): New arg to take packet version in account, changed + call callers. - * armor.c (armor_filter): Made the "Comment:" header translatable. + * gpgv.c: New. + * Makefile.am: Rearranged source files so that gpgv can be build with + at least files as possible. - * seckey-cert.c: Removed obsolete mpi_*_protect_flag. - * parse-packet.c: Ditto. +Mon Sep 18 12:13:52 CEST 2000 Werner Koch <wk@openit.de> - * misc.c (mpi_read): Removed the secure argumet becuase it is - never used. Changed all Callers. - (mpi_read_opaque): New. - (mpi_write_opaque): New. - * parse-packet.c (parse_key): Use the opaque method also for - v3 keys. - * build-packet.c (do_secret_key): Likewise. + * hkp.c (not_implemented): Print a notice for W32 - * g10.c (main): Check libgcrypt version. +Fri Sep 15 18:40:36 CEST 2000 Werner Koch <wk@openit.de> - * packet.h: replaced inclusion of mpi.h by a plain typeedef of the - gcry_mpi structure and removed all inclusions of "mpi.h" in all - sources. + * keygen.c (keygen_add_std_prefs): Changed order of preferences to + twofish, cast5, blowfish. - * g10.c: Add --delete-secret-key to the help page. + * pkclist.c (algo_available): Removed hack to disable Twofish. - * g10.c (main): Changed the default homedir to "~/.gnupg-test" so - that we don't mess up with the stable version. +Thu Sep 14 17:45:11 CEST 2000 Werner Koch <wk@openit.de> - * misc.c (mpi_write): New. - (mpi_write): New. + * parse-packet.c (dump_sig_subpkt): Dump key flags. Print special + warning in case of faked ARRs. - * misc.c (checksum_u16_nobug): Removed. - (checksum_mpi_counted_nbits): Renamed to ... - (checksum_mpi): ... this to superseed the old one. Changed all - callers. This is because we do not emulate the old gpg bug anymore. - * g10.c (oEmuChecksumBug): Removed. + * getkey.c (finsih_lookup): Hack so that for v4 RSA keys the subkey + is used for encryption. - * g10.c (register_extension): New... - (main): Use it here instead of register_cipher_extesnion. - (strusage): s/strusage/my_strusage/ . Made static. - (main): Use set_strusage(). +Thu Sep 14 14:20:38 CEST 2000 Werner Koch <wk@openit.de> - * tdbdump.c (HEXTOBIN): Changed the name of the argument, so that - traditional cpp don't mess up the macros. Suggested by Jos Backus. + * g10.c (main): Default S2K algorithms are now SHA1 and CAST5 - this + should solve a lot of compatibility problems with other OpenPGP + apps because those algorithms are SHOULD and not optional. The old + way to force it was by using the --openpgp option whith the drawback + that this would disable a couple of workarounds for PGP. - * armor.c (parse_header_line): Stop parsing on a only WS line too. - Suggested by Aric Cyr. + * g10.c (main): Don't set --quite along with --no-tty. By Frank Tobin. - * misc.c (pull_in_libs): Removed. + * misc.c (disable_core_dump): Don't display a warning here but a return + a status value and ... + * g10.c (main): ...print warnining here. Suggested by Sam Roberts. - * mainproc.c (list_node): Print the PK algo in the --with-colon mode. - * keylist.c (list_keyblock): Ditto. +Wed Sep 13 18:12:34 CEST 2000 Werner Koch <wk@openit.de> - * misc.c (pull_in_libs): Removed pull in of g10c. + * keyedit.c (keyedit_menu): Allow to use "debug" on the secret key. - * misc.c (map_gcry_rc): Removed here and chnaged all users. + * ringedit.c (cmp_seckey): Fix for v4 RSA keys. + * seckey-cert.c (do_check): Workaround for PGP 7 bug. - * getkey.c: Replaced check_pubkey_algo by openpgp_pk_test_algo. - * import.c (delete_inv_parts): Ditto. - * pkclist.c: Ditto. - * skclist.c: Ditto. - * pubkey-enc.c: Ditto. +Wed Sep 6 17:55:47 CEST 2000 Werner Koch <wk@openit.de> - * g10.c (main): Replaced the function to diable PK algos. + * misc.c (print_pubkey_algo_note): Do not print the RSA notice. + * sig-check.c (do_signature_check): Do not emit the RSA status message. + * pubkey-enc.c (get_session_key): Ditto. - * g10.c (main): Replaced get_random_bits by gcry_random_bytes. - * seskey.c (encode_session_key): Likewise. - (make_session_key): Renamed randomize_buffer to gcry_randomize - and use the GCRY_xxx_RANDOM constants. - * cipher.c (write_header): Ditto. - * passphrase.c (hash_passphrase): Ditto. - * seckey-cert.c (protect_secret_key): Ditto. + * encode.c (encode_simple, encode_crypt): Fix for large files. + * sign.c (sign_file): Ditto. - * getkey.c (find_by_name): Replaced rmd160_hash_buffer - by gcry_md_hash_buffer. - * keyedit.c (show_prefs): Ditto. - * keylist.c (list_keyblock): Ditto. - * trustdb.c (print_uid_from_keyblock): Ditto. - (make_uid_records): Ditto. +Wed Sep 6 14:59:09 CEST 2000 Werner Koch <wk@openit.de> - * skclist.c (build_sk_list): Removed the test on faked RNGs. - (is_insecure): Removed. - * g10.c (--quick-random): Removed this option. + * passphrase.c (hash_passphrase): Removed funny assert. Reported by + David Mathog. - * Replaced all PUBKEY_ALGO_xxx by GCRY_PK_xxxx. + * openfile.c (try_make_homedir): Changes for non-Posix systems. + * g10.c (main): Take the default homedir from macro. - * misc.c (pubkey_algo_npkey): New as a wrapper around the gcry fucntion. - (pubkey_algo_nskey): Ditto. - (pubkey_algo_nsig): Ditto. - (pubkey_algo_nenc): Ditto. + * g10.c: The --trusted-key option is back. + * trustdb.c (verify_own_key): Handle this option. + (add_ultimate_key): Moved stuff from verify_own_key to this new func. + (register_trusted_key): New. - * Makefile.am (basicdefs.h): Added. - (install-data-local): Removed the handling for historic gpgm. +Fri Aug 25 16:05:38 CEST 2000 Werner Koch <wk@openit.de> - * misc.c (openpgp_cipher_test_algo): New. - (openpgp_pk_test_algo): New. - (openpgp_md_test_algo): New. + * parse-packet.c (dump_sig_subpkt): Print info about the ARR. + + * openfile.c (overwrite_filep): Always return okay if the file is + called /dev/null. + (make_outfile_name): Add ".sign" to the list of know extensions. + (open_sigfile): Ditto. + +Wed Aug 23 19:52:51 CEST 2000 Werner Koch <wk@openit.de> + + * g10.c: New option --allow-freeform-uid. By Jeroen C. van Gelderen. + * keygen.c (ask_user_id): Implemented here. + +Fri Aug 4 14:23:05 CEST 2000 Werner Koch <wk@openit.de> + + * status.c (do_get_from_fd): Ooops, we used fd instead of opt.command_fd. + Thanks to Michael Tokarev. - * g10.c (build_list): Changed to use the new functions from libgcrypt. +Tue Aug 1 20:06:23 CEST 2000 Werner Koch <wk@openit.de> - * ringedit.c (enum_keyblocks): Set .rt to 0 on open. + * g10.c: New opttion --try-all-secrets on suggestion from Matthias Urlichs. + * pubkey-enc.c (get_session_key): Quite easy to implement here. - * encode.c (encode_simple): Use new CTB when we don't have the - length of the file. This is somewhat strange as the comment above - indicates that this part is actually fixed for PGP 5 - maybe I simply - lost the source line, tsss. +Thu Jul 27 17:33:04 CEST 2000 Werner Koch <wk@openit.de> - * sign.c (clearsign_file): Avoid duplicated Entries in the "Hash:" - line. Those headers are now only _not_ printed when there are - only old-style keys _and_ all hashs are MD5. + * g10.c: New option --merge-only. Suggested by Brendan O'Dea. + * import.c (import_one): Implemented it here + (import_secret_one): Ditto. + (print_stats): and give some stats. - (clearsign_file): Use gcry_md_test_algo() and gcry_md_algo_name(). +Thu Jul 27 12:01:00 CEST 2000 Werner Koch <wk@openit.de> - * openfile.c (make_outfile_name): Use case-insenstive compare for - DOS systems. Add ".pgp" to the list of know extensions. - (open_outfile): For DOS systems try to replace the suffix instead of - appending it. + * g10.c: New options --show-session-key and --override-session-key + * pubkey-enc.c (hextobyte): New. + (get_override_session_key): New. + * mainproc.c (proc_pubkey_enc): Add session-key stuff. + * status.h, status.c (STATUS_SESSION_KEY): New. - * encr-data.c (decrypt_data): Reset error on a weak key. +Thu Jul 27 10:02:38 CEST 2000 Werner Koch <wk@openit.de> - * cipher.c: Replaced the cipher and digest functions by the gcry_ ones. - * seckey-cert.c: Ditto. - * seskey.c: Ditto. - * g10.c (print_mds): Replaced digst functions with the new gcry_ ones. - * keyid.c: Ditto. - * mainproc.c: Ditto. - * passphrase.c: Ditto. - * sig-check.c: Ditto. - * sign.c: Ditto. + * g10.c (main): Use setmode(O_BINARY) for MSDOS while generating random bytes + (print_mds): Likewise for stdin. + * plaintext.c (handle_plaintext): Likewise for stdout. + +Mon Jul 24 10:30:17 CEST 2000 Werner Koch <wk@openit.de> + + * keyedit.c (menu_expire): expire date for primary key can be set again. + +Wed Jul 19 11:26:43 CEST 2000 Werner Koch <wk@openit.de> + + * keylist.c (is_uid_valid): New. + (list_keyblock): Print validity information for all user IDs. Note, this + has to be done at other places too; for now we have only minimal support. + +Wed Jul 12 13:32:06 CEST 2000 Werner Koch <wk@openit.de> - * pkclist.c (do_edit_ownertrust): Made the answer string const. + * helptext.c, pkclist.c: s/superseeded/superseded/ - * basicdefs.h: New. Move some defs and decl to this header. +Mon Jul 10 16:08:57 CEST 2000 Werner Koch <wk@openit.de> - * openfile.c (open_outfile): Fixed the 8dot3 handling. + * parse-packet.c (enum_sig_subpkt): Fixed testing on crtitical bit in case + of a NULL buffer. Reported by Peter Marschall. - * passphrase.c (passphrase_to_dek): Print uid using utf8 func. - * delkey.c (delete_key): Ditto. - * pkclist.c (show_paths,do_edit_ownertrust,do_we_trust): Ditto - (do_we_trust_pre): Ditto. - * trustdb.c (print_user_id,check_uidsigs): Ditto. - * revoke.c (gen_revoke,ask_revoke_sig): Ditto. +Wed Jul 5 13:28:45 CEST 2000 Werner Koch <wk@openit.de> - * filter.h: Changed cipher handle types to the the GCRY_xxx ones. - replaces include cipher by system header include gcrypt.h. - * cipher.c: replaced the cipher functions by the gcry_ ones. - Ditto for the md functions. + * keyedit.c, keyid.c: Add some _() - * misc.c (map_gcry_rc): New. + * argparse.c: Changed the flag to suppress --version handling to also + suppress --help. -Wed Jun 28 11:54:44 CEST 2000 Werner Koch <wk@> +Wed Jun 28 11:54:44 CEST 2000 Werner Koch <wk@openit.de> * armor.c (armor_filter): Set sigclass to 0 in case of non-dash-escaped clearsig. This makes this mode work again. @@ -461,7 +3478,7 @@ Fri Jun 9 10:09:52 CEST 2000 Werner Koch <wk@openit.de> Wed Jun 7 19:19:09 CEST 2000 Werner Koch <wk@openit.de> - * sig-check.c (do_check): Use EMULATE_MDENCODE also on v4 paclets. + * sig-check.c (do_check): Use EMULATE_MDENCODE also on v4 packets. Wed Jun 7 17:25:38 CEST 2000 Werner Koch <wk@openit.de> @@ -3280,3 +6297,13 @@ Thu Feb 12 22:24:42 1998 Werner Koch (wk@frodo) * pubkey-enc.c (get_session_key): rewritten + + Copyright 1998,1999,2000,2001,2002 Free Software Foundation, Inc. + + This file is free software; as a special exception the author gives + unlimited permission to copy and/or distribute it, with or without + modifications, as long as this notice is preserved. + + This file is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY, to the extent permitted by law; without even the + implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. |