src/keymanagement.c
Mon, 08 May 2017 08:18:47 +0200 merge "default" into "ENGINE-199"
Sun, 07 May 2017 13:45:04 +0200 ENGINE-183 remove session.use_only_own_private_keys, and add existing priv keys to own_keys table at initial startup. Also added fpr from own_keys to own_key_is_listed, influencing result of ellect_own_key.
Thu, 04 May 2017 15:16:47 +0200 Merged default into diphoton, enforcing 2 devices and 1 account limit
Fri, 28 Apr 2017 12:02:57 +0200 update_identity() now takes PEP_ct_key_expired as a special case, and replaces it with actual key rating if not expired enymore. Otherwise, partner's rating with expired key was never comming back to normal, even if key was renewed.
Fri, 21 Apr 2017 21:38:20 +0200 add more ERROR() and GOTO() to find out why the call failed. it is in keymanagement.c :-D
Mon, 27 Feb 2017 17:43:10 +0100 ENGINE-174: stowing fixes to ratings while attempting to reactivate branch
Thu, 26 Jan 2017 19:07:45 +0100 avoid unitialized local pointer variables to reduce hard-to-find bugs
Fri, 13 Jan 2017 15:37:35 +0100 Removed assert, because an empty username is totally valid input from an email, and we deal with it directly in the code that follows where the assert was.
Mon, 09 Jan 2017 10:28:02 +0100 merged in default
Tue, 20 Dec 2016 22:51:41 +0100 ENGINE-137 (fixing side effect of optimization) update_identity: added key re-election when a key has been deleted from keyring. fixed test blacklist_accept_new_key_test using that feature
Tue, 27 Dec 2016 21:13:41 +0100 adding license info to each file
Tue, 27 Dec 2016 21:13:41 +0100 adding license info to each file
Wed, 21 Dec 2016 10:47:27 +0100 ENGINE-137 make update_identity question stored fpr if there may be a more trustworthy key associated with that user_id
Tue, 20 Dec 2016 22:51:41 +0100 ENGINE-137 (fixing side effect of optimization) update_identity: added key re-election when a key has been deleted from keyring. fixed test blacklist_accept_new_key_test using that feature
Tue, 20 Dec 2016 14:07:45 +0100 ENGINE-137 fix optimization of update_identity, elect_pubkey already does blacklisted key exclusion, no need to redo it here.
Tue, 20 Dec 2016 12:06:56 +0100 ENGINE-137 3x optimization in update_identity, though it breaks blacklist test. To be continued
Thu, 15 Dec 2016 17:34:04 +0100 Fixed additional null username issue in update_identity
Wed, 14 Dec 2016 17:08:28 +0100 Prevent assertion when username is not given for own identity, and doing keygen
Mon, 28 Nov 2016 17:48:28 +0100 merged in default
Mon, 28 Nov 2016 17:48:05 +0100 ENGINE-112: updated get_trust in update_identity and stopped set_identity from setting empty fprs as keypairs and trust entries
Fri, 25 Nov 2016 15:59:58 +0100 Copy-paste is bad.
Mon, 21 Nov 2016 22:27:25 +0100 ENGINE-140 clean'up own_identities_retrieve so that available pEpEngine API returns all own identities, not only those that can be synced
Fri, 18 Nov 2016 00:13:40 +0100 ENGINE-140 changed identities flags update policy
Tue, 15 Nov 2016 15:56:18 +0100 ENGINE-140 #comment Added own_keys table, linking pgp_keypair to identity in case user_id = PEP_OWN_USERID.
Sat, 12 Nov 2016 00:24:14 +0100 No more _update_indentiy() and with_myself parameter. Rather, added a _myself(), with a do_keygen parameter. Now, only explicit call to myself(), key_mistrusted(), encrypt_message(), encrypt_message_to_self() may trigger keygen, not outgoing_message_rating() or update_identity()
Fri, 11 Nov 2016 14:45:42 +0100 ENGINE-84: fixed merge conflicts
Fri, 11 Nov 2016 14:10:30 +0100 ENGINE-84: merge with default
Tue, 08 Nov 2016 17:57:06 +0100 ENGINE-84: fixed status TODOs.
Thu, 10 Nov 2016 21:10:24 +0100 ENGINE-140: merged
Thu, 10 Nov 2016 01:35:59 +0100 fixed memory leak in myself
Wed, 09 Nov 2016 15:28:45 +0100 Backed out changeset 2ba0f96f6c0e. Broke KeySync, but for some reason last test run didn't catch it. Will graft those changes in a branch.
Tue, 08 Nov 2016 23:04:26 +0100 ENGINE-140: now send all keys that have been at some point used as own key _on that device_
Tue, 08 Nov 2016 14:10:46 +0100 missing install files fixed
Tue, 08 Nov 2016 00:34:53 +0100 ENGINE-84: #comment now generating new keys when no usable private key is available, and not electing/selecting/attempting to use own keys as private keys where private half of the keypair is blacklisted, revoked, or missing. TODO: fix status TODOs, test more before merging back. But it's working, in general.
Tue, 08 Nov 2016 00:12:36 +0100 ENGINE-84: merged in default
Tue, 08 Nov 2016 00:12:17 +0100 ENGINE-84: working on ensuring keypairs missing private keys are not selected for encryption candidates
Sun, 06 Nov 2016 23:32:23 +0100 Fixed problem with null usernames coming from prepackaged mime messages in 'From' for MIME_encrypt_message
Sun, 06 Nov 2016 22:55:37 +0100 stowing changes
Thu, 27 Oct 2016 21:43:59 +0200 ENGINE-129: update_identity now properly returns an ID with an empty fpr string when no key can be elected because none are in the keyring
Tue, 18 Oct 2016 01:32:05 +0200 sync: decrypt_message and myself shouldn't return sync errors except if out of memory
Mon, 17 Oct 2016 23:42:47 +0200 Deleted 'useless extraneous code that got pulled over from an earlier version', according to darthmama
Thu, 13 Oct 2016 16:22:09 +0200 ENGINE-112: #comment Fixed empty fpr problems in stored identities which occurred due to blacklisting. When adding other keys, we now check to see if there is a matching identity entry which lacks an fpr and, if so, we update it instead of adding a new one. Also fixed elect_pubkey NOT to return blacklisted pubkeys.
Wed, 12 Oct 2016 17:22:30 +0200 deleted unused _same_fpr() causing a warning in keymanagement.c
Fri, 07 Oct 2016 11:29:56 +0200 Actually, the flag was correct. Sigh.
Fri, 07 Oct 2016 09:27:29 +0200 don't_use_fpr was always set to true in update_identity. Fixed.
Thu, 06 Oct 2016 22:08:09 +0200 include in keymanagement.c for blacklist_is_listed
Wed, 05 Oct 2016 14:56:54 +0200 ENGINE-107: added ct and rating states for having no key
Mon, 03 Oct 2016 17:16:33 +0200 Engine-97 Merge with default
Thu, 29 Sep 2016 13:07:41 +0200 ENGINE-27: set_identity can be called with a null/empty fpr now per Volker. ensured it does not call blacklist_is_listed on null/empty fprs
Thu, 29 Sep 2016 13:03:56 +0200 ENGINE-27: set_identity can be called with a null/empty fpr now per Volker. ensured it does not call blacklist_is_listed on null/empty fprs
Thu, 29 Sep 2016 12:16:12 +0200 ENGINE-27: ensure set_identity doesn't get called in update_identity if the fpr is null
Thu, 29 Sep 2016 11:08:50 +0200 ENGINE-27: blacklist_is_listed will no longer be called with empty/null strings in update_identity (caused assert failure)
Wed, 28 Sep 2016 13:49:12 +0200 ENGINE-96: #comment all calls renamed. Roker is checking out the semantics of the references to 'compromized' to see if they also need to be renamed to 'mistrusted' for consistency within the engine. THIS IS AN API CHANGE AND WILL BREAK ADAPTERS.
Tue, 27 Sep 2016 00:32:44 +0200 ENGINE-27: #comment I forgot one possible blacklist case. If you still get an assertion fail, I'd appreciate capturing the error type if you can so I can follow the logic. Please pull and try again...
Mon, 26 Sep 2016 16:55:23 +0200 Sync : asynchronously inject KeyGen and CannotDecrypt to fsm
Mon, 26 Sep 2016 12:48:28 +0200 ENGINE-27: type
Mon, 26 Sep 2016 12:40:26 +0200 ENGINE-27: setting fpr to empty string, not NULL, so set_identity doesn't crash.
Sun, 25 Sep 2016 17:46:27 +0200 ENGINE-27: intermittent commit, update_identity should now never try to set or return a blacklisted key
Sat, 24 Sep 2016 18:09:18 +0200 ENGINE-27: update_identity no longer pays attention to the input identity's fpr, but pulls one from elect_pubkey.
Tue, 20 Sep 2016 16:14:41 +0200 merge "default" into "keysync" branch