1 Hey Emacs, this is -*- outline -*- mode!
4 ** Some gpg tests fail with gpg 1.3.4-cvs (gpg/t-keylist-sig)
5 The test is currently disabled there and in gpg/t-import.
6 ** Add notation data to key signatures.
9 ** I/O and User Data could be made extensible. But this can be done
10 without breaking the ABI hopefully.
11 ** gpgme_keylist_mode_t should not be an enum. Maybe some other values
12 should also not be an enum. All enums that should be enums need to
13 have a maximum value to ensure a certain minimum width for
15 ** Compatibility interfaces that can be removed in future versions:
16 *** ath compatibility modules.
17 *** gpgme_data_new_from_filepart
18 *** gpgme_data_new_from_file
19 *** gpgme_data_new_with_read_cb
21 *** gpgme_op_import_ext
23 *** gpgme_get_sig_ulong_attr
24 *** gpgme_get_sig_string_attr
26 *** gpgme_get_sig_status
27 *** gpgme_trust_item_release
28 *** gpgme_trust_item_get_string_attr
29 *** gpgme_trust_item_get_ulong_attr
31 *** All Gpgme* typedefs.
34 ** When GNU Pth supports sendmsg/recvmsg, wrap them properly.
35 ** Without timegm (3) support our ISO time parser is not thread safe.
36 There is a configure time warning, though.
39 ** notification system
40 We need a simple notification system, probably a simple callback
41 with a string and some optional arguments. This is for example
42 required to notify an application of a changed smartcard, The
43 application can then do whatever is required. There are other
44 usages too. This notfication system should be independent of any
46 ** --learn-code support
47 This might be integrated with import. we still need to work out how
48 to learn a card when gpg and gpgsm have support for smartcards.
49 ** How to terminate a pending operation? Something like gpgme_op_reset,
50 but where are you allowed to call it (think callback handlers).
51 Then gpgme_op_*list_end can go. Update: The only place where this
52 can go is returning errors from callback handlers, and a function
53 to be called for example from the user event loop code. The user
54 must observe the threading rules! A blocking thread can not be
56 ** Might need a stat() for data objects and use it for length param to gpg.
57 ** Allow to export secret keys.
58 ** Implement support for photo ids.
59 ** New features requested by our dear users, but rejected or left for
61 *** Selecting the key ring, setting the version or comment in output.
62 Rejected because the naive implementation is engine specific, the
63 configuration is part of the engine's configuration or readily
64 worked around in a different way
65 *** Selecting the symmetric cipher.
66 *** Exchanging keys with key servers.
67 ** Allow selection of subkeys
68 ** Allow to return time stamps in ISO format
69 This allows us to handle years later than 2037 properly. With the
70 time_t interface they are all mapped to 2037-12-31
73 ** gpgme.m4 should check --api-version of gpgme-config.
76 ** Document validity and trust issues.
79 ** Do not create/destroy engines, but create engine and then reset it.
80 Internally the reset operation still spawns a new engine process,
81 but this can be replaced with a reset later. Also, be very sure to
82 release everything properly at a reset and at an error. Think hard
83 about where to guarantee what (ie, what happens if start fails, are
84 the fds unregistered immediately - i think so?)
85 ** Optimize the case where a data object has an underlying fd we can pass
86 directly to the engine. This will be automatic with socket I/O and
88 ** Move code common to all engines up from gpg to engine.
89 ** engine operations can return General Error on unknown protocol
90 (it's an internal error, as select_protocol checks already).
91 ** When server mode is implemented properly, more care has to be taken to
92 release all resources on error (for example to free assuan_cmd).
95 ** If an operation failed, make sure that the result functions don't return
96 corrupt partial information. !!!
97 NOTE: The EOF status handler is not called in this case !!!
98 ** If no passphrase cb is installed, status handler is not run even if
99 password is required by crypto engine. !!
100 ** Verify must not fail on NODATA premature if auto-key-retrieval failed.
101 It should not fail silently if it knows there is an error. !!!
102 ** All operations: Better error reporting. !!
103 ** Export status handler need much more work. !!!
104 ** Import should return a useful error when one happened.
105 *** Import does not take notice of NODATA status report.
106 *** When GPGSM does issue IMPORT_OK status reports, make sure to check for
107 them in tests/gpgs m/t-import.c.
108 ** Verify can include info about version/algo/class, but currently
109 this is only available for gpg, not gpgsm.
110 ** Return ENC_TO output in verify result. Again, this is not available
112 ** Genkey should return something more useful than General_Error.
114 On Fri, Jun 27, 2003 at 06:28:23PM +0200, Heiko Abraham wrote:
115 > I have a cipher text and I use 'gpgme_op_decrypt_verify(..)'
116 > for decrypt and get the plaintext. But also I wish a list
117 > of all reciepient, that can also decrypt this file.
119 > If I store the file and check it with 'gpg --list-packets ${filename}'
120 > then I will become also a recipient-list.
121 > It this also possible with gpgme?
123 Currently not, but it is easy to add this to GPGME 0.4.1. At least the key
124 ID and a user ID hint is available from gpg (of course key IDs are not
125 necessarily unique!). I will put it on the TODO list.
126 ** If possible, use --file-setsize to set the file size for proper progress
127 callback handling. Write data interface for file size.
128 ** Optimize the file descriptor list, so the number of open fds is
130 ** Encryption: It should be verified that the behaviour for partially untrusted
131 recipients is correct.
132 ** When GPG issues INV_something for invalid signers, catch them.
135 ** Map ASSUAN/GpgSM ERR error values in a better way than is done now. !!
136 ** Some error values should identify the source more correctly (mostly error
137 values derived from status messages).
140 ** Write a fake gpg-agent so that we can supply known passphrases to
141 gpgsm and setup the configuration files to use the agent. Without
142 this we are testing a currently running gpg-agent which is not a
145 *** Test gpgme_data_release_and_get_mem.
146 *** Test gpgme_data_seek for invalid types.
148 Write a test for ext_keylist.
152 ** Handle malloc and vasprintf errors. But decide first if they should be
153 ignored (and logged with 255?!), or really be assertions. !
156 ** Make sure everything is cleaned correctly (esp. test area).
157 ** Configure test for gpg and gpgsm version (as a warning).
161 ** engine-gpgsm, with-validation
162 Add error checking some time after releasing a new gpgsm.