0.9.7 buglets

[Brian Warner]

Two tiny buglets in 0.9.7:

 It would be handy if the new MISSING_PASSPHRASE status indicator could be
 output when a passphrase is not given in --batch mode. In g10/passphrase.c,
 the code bails out early in batch mode. If it could emit the status just
 before it quits, the mailcrypt parser would be a good bit easier.

 The following message (with armor corrupted by appending BADBAD to it) causes
 a segfault. It's not new, this part of my test setup has caused a gpg segfault
 since the beginning of time. But I thought I'd throw it out there in case it's
 easy to fix.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

This is a sample message
-----BEGIN PGP SIGNATURE-----
Version: GNUPG v0.3.4 (GNU/Linux)
Comment: Get GNUPG from ftp://ftp.guug.de/pub/gcrypt/

iEYEARECAAYFAjXzlxIACgkQtIMGX0t13c8gFgCdEFnPoioXRgs922/1Dtg/tJ6iWywAni+qBADBAD
EsJjb6bh0XjtlsQDl3hMrXJo
=hnQd
-----END PGP SIGNATURE-----

My test setup tells me that 0.9.7 responds with the following:

gpg-0.9.7 --homedir .. --batch --verify CS.s1bad:
stdout:
stderr:
gpg: Warning: using insecure memory!
gpg: CRC error; da0741 - 86741d
gpg: packet(1) with unknown version 215

gpg: Segmentation fault caught ... exiting
status:
rc: 8



cheers,
 -Brian
   warner at lothar.com