This reverts commit 6944aefa3c2ef79cf3f14306ed384d22de36ba7f.
--
The fd-passing works well on Windows with new libassuan (to be 3.0),
and it doesn't require ASSUAN_SOCKET_SERVER_FDPASSING actually.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* kbx/kbxserver.c (prepare_outstream): Use es_sysopen_nc
and avoid the use of translate_sys2libc_fd.
--
On Windows, it's better directly use the system HANDLE.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* kbx/kbxserver.c (kbxd_start_command_handler): No fd-passing udner
Windows.
--
file descriptor passing does not work reliable in libassuan for
Windows and we actually don't need it here. It is not even used by
gpg or gpgsm. As soon as we enable fd-passing in gpgme for Windows
and see that it is robust enough we should back out this patch.
* kbx/keyboxd.h (kbxd_status_printf): New.
* kbx/backend-support.c (be_return_pubkey): Use kbxd_status_printf.
* kbx/kbxserver.c (kbxd_status_printf): New.
* kbxd_start_command_handler (kbxd_start_command_handler): Don't
use set_assuan_context_func, because the function pointer is
shared by multiple threads.
--
GnuPG-bug-id: 5948
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* kbx/keybox-search-desc.h (struct keydb_search_desc): New flag
name_used.
* common/userids.c (classify_user_id): Set flag.
* kbx/kbxserver.c (struct search_backing_store_s): New.
(cmd_search): use a backing store for the const pointers.
(kbxd_start_command_handler): Release the backing store.
--
Well, the search object partly uses buffers but also const
pointers (for strings and the serial number). This when assigning
such objects to an another one we should really take a deep copy and
not just copy the pointer. The more clean solution would have been to
provide a storage option the search object but that needs checking the
code at too many places so that I decided to use a separate backing
store array here.
Signed-off-by: Werner Koch <wk@gnupg.org>
* kbx/kbxserver.c (cmd_search): Use the openpgp hack for calling
classify_user_id.
* kbx/backend-sqlite.c (run_select_statement): Remove angle brackets
in exact addrspec mode.
* g10/call-keyboxd.c (keydb_search): Do not duplicate the left angle
bracket.
* sm/keydb.c (keydb_search): Ditto.
--
Note that the openpgp hack flag of classify_user_id is actually a
misnomer because we actually hack a round a problem in gpgsm. And it
is only over there that we don't set it there. In keyboxd the flag
should be set. And we need to remove the angle brackets of course
because that is how we create the addrspec column values.
Signed-off-by: Werner Koch <wk@gnupg.org>
* kbx/backend-sqlite.c (be_sqlite_rollback): New.
(be_sqlite_commit): New.
(be_sqlite_search): Take care of global transactions.
(be_sqlite_store): Ditto.
(be_sqlite_delete): Ditto.
* kbx/frontend.c (kbxd_rollback, kbxd_commit): New.
* kbx/keyboxd.h (opt): Add vars for transactions.
* kbx/kbxserver.c (struct server_local_s): Add fields next_session and
client_pid.
(session_list): New var.
(cmd_transaction): New.
(register_commands): Register command.
(kbxd_start_command_handler): Store pids and track sessions. Do a
final rollback.
--
This command is currently an experiment to allow a client to run
everything in one session.
Signed-off-by: Werner Koch <wk@gnupg.org>
* kbx/keyboxd.h (struct server_control_s): Replace the two request
objects by just one. Add filter flags.
* kbx/kbxserver.c (cmd_search): Add options --openpgp and --x509.
(cmd_killkeyboxd): Do not return GPG_ERR_EOF.
* kbx/frontend.c (kbxd_release_session_info): Adjust for the new
request object.
(kbxd_search, kbxd_store, kbxd_delete): Ditto.
* kbx/backend-sqlite.c (struct be_sqlite_local_s): Add filter flags.
(run_sql_prepare): Add optional arg 'extra'. Change callers.
(run_sql_bind_ntext): New.
(run_sql_bind_text): Just call run_sql_bind_ntext.
(run_select_statement): Add ctrl arg. Implement the filter flags.
* g10/call-keyboxd.c (keydb_search): Use the --openpgp option.
--
As soon as we implement X.509 we need to have a way to return only
openpgp or x.509 certificates. Gpg/gpgsm will then use the respective
flag.
Signed-off-by: Werner Koch <wk@gnupg.org>
No functional changes, just fixing minor spelling issues.
---
Most of these were identified from the command line by running:
codespell \
--ignore-words-list fpr,stati,keyserver,keyservers,asign,cas,iff,ifset \
--skip '*.po,ChangeLog*,help.*.txt,*.jpg,*.eps,*.pdf,*.png,*.gpg,*.asc' \
doc g13 g10 kbx agent artwork scd tests tools am common dirmngr sm \
NEWS README README.maint TODO
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
* common/util.h (UBID_LEN): New. Use it at all places.
* kbx/keybox-blob.c (create_blob_finish): Do not write the UBID item.
* kbx/keybox-dump.c (print_ubib): Remove.
(_keybox_dump_blob): Do not print the now removed ubid flag.
* kbx/keybox-search-desc.h (struct keydb_search_desc): Use constants
for the size of the ubid and grip.
* kbx/keybox-search.c (blob_cmp_ubid): New.
(has_ubid): Make it a simple wrapper around blob_cmp_ubid.
(keybox_get_data): Add arg 'r_ubid'.
* kbx/frontend.h (enum kbxd_store_modes): New.
* kbx/kbxserver.c (cmd_store): Add new option --insert.
* kbx/backend-cache.c (be_cache_initialize): New.
(be_cache_add_resource): Call it here.
* kbx/backend-kbx.c (be_kbx_seek): Remove args 'fpr' and 'fprlen'.
(be_kbx_search): Get the UBID from keybox_get_data.
* kbx/backend-support.c (be_fingerprint_from_blob): Replace by ...
(be_ubid_from_blob): new. Change all callers.
* kbx/frontend.c (kbxd_add_resource): Temporary disable the cache but
use the new cache init function.
(kbxd_store): Replace arg 'only_update' by 'mode'. Seek using the
ubid. Take care of the mode.
--
It turned out that using the hash of the entire blob was not helpful.
Thus we redefine the Unique-Blob-ID (UBID) as the primary fingerprint
of the blob. In case this is a v5 OpenPGP key a left truncated
version of the SHA-256 hash is used; in all other cases the full SHA-1
hash. Using a SHA-256 hash does not make sense because v4 keys are
and will for some time be the majority of keys and thus padding them
with zeroes won't make any difference. Even if fingerprint collisions
can eventually be created we will assume that the keys are bogus and
that it does not make sense to store its twin also in our key storage.
We can also easily extend the update code to detect a collision and
reject the update.
Signed-off-by: Werner Koch <wk@gnupg.org>
* kbx/backend.h (enum database_types): Add DB_TYPE_CACHE.
(struct db_request_part_s): Add seqno fields.
(struct db_request_s): Add infos for the cache backend.
* kbx/backend-support.c (struct backend_handle_s): Add 'backend_id'.
(strdbtype): Support DB_TYPE_CACHE.
(be_generic_release_backend): Ditto.
(be_find_request_part): New.
(be_return_pubkey): New arg UBID and chnage status name.
* kbx/backend-cache.c: New.
* kbx/backend-kbx.c (be_kbx_init_request_part): New.
(be_kbx_search): Factor some code out to a support function.
(be_kbx_seek): New.
* kbx/frontend.c (kbxd_add_resource): Support DB_TYPE_CACHE.
(kbxd_search): Support the NEXR operation with the cache.
* kbx/keybox-search-desc.h (KEYDB_SEARCH_MODE_UBID): New.
(struct keydb_search_desc): Add field u.ubid.
* kbx/keybox-search.c (has_ubid): New.
(keybox_search): Support the UBID search.
--
This adds a caching backend to the keyboxd. This tries to accommodate
for duplicate use of fingerprints and thus be correct in case a
fingerprint is used in several keys. It also turned out that we need
to have a unique identifier (UBID) to identify a keyblock or X.509
certificate. In particular with an OpenPGP keyblob we can't easily
use the primary fingerprint as an identifier because that fingerprint
may also be used as subkey in another key. Thus using a hash of the
entire keyblock is a better identifier to be used to address a
keyblock for restarting a search or for identifying the keyblock to be
updated. Note that this new UBID is not a permanent identifier
because it changes with all keyblock update; it should be viewed as a
handle to the keyblock or X509 cert.
