gnupg/OBUGS

		   List of fixed bugs
		  --------------------

(format: severity: [  *] to [***], no, first reported, by, version)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[  *] #1
    pgp263in works fine even with a source file with CR,LF but GnuPG
    and pgp263in has problems if the clearsign has been created by
    pgp263ia.  The reason for this problem is that pgp2 sometimes
    converts CR,LF to CR,CR,LF	and to fix for this it hashes both
    versions.  I was able to reproduce such a problem, that PGP263in
    was not able to verify it's own signature.
    FIX: 1999-05-19  (Most cases are now handled)

[***] #2  1999-02-21
    Problem while importing or deleting public keys in 0.9.3 - 0.9.2
    worked fine.  Error message:
    gpg:[stdin]: key A6A59DB9: secret key not found: public key not found
    FIX: 1999-02-22 wk

[  *] #5
    /home/jam/.gnupg/pubring.gpg: can't open gdbm file: Can't be writer
    keyblock resource `/home/jam/.gnupg/pubring.gpg': file open error
    OOPS in close enum_keyblocks - ignored
    [gdbm is experimental and will be replaced by the new keybox code]
    FIX: 1999-07-22 (Fixed the second error, there will be no fix for
    the first one, because GDBM is to be replaced)

[  *] #7 1999-02-22 <dwpalmer@dwpalm.jf.intel.com> 0.9.3
    Conventional encryption incompatibility:
     $ gpg -c --cipher-algo cast5 --compress-algo 1 --no-comment secrets.txt
    Creates a file that gpg can decrypt, but PGP 5.5 has problems with it.
    PGP decrypts 6416k out of 6424k, then complains with "PGP Warning",
    "The keyring contains a bad (corrupted) PGP packet".  The resulting
    file is missing information from the front.
    FIX: 1999-02-26 temporary fix in encrypt_simple()


[  *] #8 1999-02-25 <kazu@iijlab.net> 0.9.3
    %gpg --encrypt -r kazu@iijlab.net --batch foo
    gpg: Warning: using insecure memory!
    gpg: 11C23F61: no info to calculate a trust probability
    This creates a symmetrically encrypted message WITHOUT a session key
    encrypted with public cryptographic(i.e. foo.gpg). This is probably
    FIX: 199-02-26 wk

[ **] #9 1999-02-25
    Misalignment in md5.c#md5_write.
    FIX: 1999-02-26 wk

[ **] #10 1999-03-01
    Armor detection code is broken.  Direct import of keyrings is not possible.
    FIX: 1999-03-02 wk

[***] #11 1999-02-25
    "cipher algo 10 not found".
    FIX: 1999-02-25 wk

[ **] #12 1999-03-10
    gpg --list-secret-keys --with-colon SEGVs
    FIX: 1999-03-10

[  *] #13 1999-04-05
    Trying to generate very large keys fails with a BUG in read_pool()
    FIX: 1999-04-06


[  *] #14 1999-04-05 <anonymous>
    If you use --s2k-cipher-algo twofish, the the program crashes with
    a BUG at line 226 of passphrase.c.
    FIX: 1999-04-06


[ **] #15 1999-04-05
    Hash calculation for subkey bindings is not according to rfc2440 if
    a 4 byte length header is used for the subkey.
    FIX: 1999-04-06

[***] #16 1999-03-23 <jafo@tummy.com>
    Verifying detached signatures with an empty file yields a rc of 0.
    FIX: 1999-05-06

[ **] #17 1999-05-18 <Bodo_Moeller@public.uni-hamburg.de> 0.9.6
    Import does not detect identical user IDs.
    FIX: 1999-05-22

[ **] #19 1999-06-11
    "trustdb transaction too large" with about 500 signatures on a key
    FAEBD5FC.
    FIX: 1999-07-12  (less memory requirement and increased the limit)

[ **] #20 1999-06-16 <jashley@yorktown.designlab.ukans.edu>  0.9.7
    Using "addkey" in the edit menu with more than 1 subkey leads to
    "out of secure memory" in some cases.
    FIX: 1999-06-17  (Twofish uses too much memory and the memory
    becomes fragmented - workaround is using CAST5 to protect passphrases)

[  *] #21 1999-06-17
    Ctrl-D does not work correct for messages entered at the tty.
    FIX: 1999-06-18  (Better EOF detection on terminals)

[  *] #22 1999-07-22
    Solaris make has problems with the generated POTFILES - seems to be a
    gettext bug.  Use GNU make as a workaround.
    FIX: 1999-12-03 (meanwhile fixed in gettext)

[  *] #23 1999-09-03 <todd.brooks@yale.edu> 0.9.11
    Only the first signature of a cleartext sig seems to be verified.
    Can't fix it in 1.0 because the code is semi-frozen.
    HMMM: Can't reprodude the bug here - it just works.

[ **] #25 1999-10-04  <wk@gnupg.org> 1.0
    Validity check based on user name is only done when listing a key;
    the key ID based checking (selecting the user ID with the most
    validity) is used in all other cases.  The Edit menu does not have
    a way to disable user ID based validity.
    FIX: 1999-12-03 (1.0.x only)

[  *] #26 1999-11-11
    gpg still does not take UTF8 strings to select a UID.
    FIX: 1999-11-12

[ **] #27 1999-11-12 <oren@hishome.net>
    Unknown packets (type 17 - photo ID?) mess up the checking
    of self-signature because they are simply ignored.
    FIX: 1999-11-12 (1.0.x only)

[ **] #29 1999-12-01 <karney@pppl.gov>
    Using a --gen-key on a new user account (w/o ~./gnupg) doesn't pay
    attention to the standard options files, which get's installed
    afterwards.
    FIX: 1999-12-02