> [[email protected] - Fri Feb 18 13:34:58 2011]: > > Hi, > > A collegue has an issue with SMIME_write_PKCS7() using Python and the > M2Crypto library for a multipart (MIME) signature. He uses a memory BIO > as output. He told me that the script works with OpenSSL 0.9.8g, but not > with OpenSSL 0.9.8o. In CVS, I saw that PKCS7 and ASN.1 code for SMIME > changed a lot, it is maybe a regression introduced by a code > factorization. > > > If data is not NULL, int_smime_write_ASN1() calls data_fn > (pk7_output_data() in my case) twice, and write into the output BIO > after the first call to data_fn. The problem is that PKCS7_dataFinal() > sets the output buffer to read only (I don't understood exactly in which > case). > > M2Crypto adds PKCS7_DETACHED flag is data is not NULL. The real call > looks like: > SMIME_write_PKCS7(out, pkcs7, info, PKCS7_TEXT | PKCS7_DETACHED) > > I am sorry, but I don't have a C script to reproduce the issue: only a > Python script (attached to this email, smime.py) using > flags=PKCS7_DETACHED. Output with error:
It isn't clear what the flags to PKCS7_sign() are. If you haven't set PKCS7_DETACHED you shouldn't set the content parameter to SMIME_write_PKCS7(): that is only used for detached content. Steve. -- Dr Stephen N. Henson. OpenSSL project core developer. Commercial tech support now available see: http://www.openssl.org ______________________________________________________________________ OpenSSL Project http://www.openssl.org Development Mailing List [email protected] Automated List Manager [email protected]
