On Mon, Jul 05, 2004, Stephen Soltesz wrote: > Hello, > > I apologize if my last question was too involved to test easily. > > I've attached a new version of the test code that is much more direct in > its use of a message digest. > > The program opens /dev/zero and reads four bytes, one byte at a time, and > after each byte retrieves the digest using BIO_gets(). > > As you can see from the output, the first digest is not equal to the > following three. > > Why is this? Am I misusing the BIO_gets() api? Is there an initial value > injected into the BIO stream for digests that must be cleared before a > first call to BIO_gets can be made? > > Any help you can provide would be appreciated. >
The intended use of a digest BIO is to only retrieve the digest once. The current implementations behaviour when BIO_gets() is called more than once is undefined: the underlying digest will attempt to interpret a finalized context as an unfinalized one with unpredictable results. You should be able to work around this by calling BIO_reset() on the BIO but this will also send BIO_reset() down the chain which may have undesirable consequences. Alternatively calling BIO_set_md() again with a NULL digest value should work. Steve. -- Dr Stephen N. Henson. Email, S/MIME and PGP keys: see homepage OpenSSL project core developer and freelance consultant. Funding needed! Details on homepage. Homepage: http://www.drh-consultancy.demon.co.uk ______________________________________________________________________ OpenSSL Project http://www.openssl.org User Support Mailing List [EMAIL PROTECTED] Automated List Manager [EMAIL PROTECTED]
