In message <20161206.223057.237264374331072901.levi...@openssl.org> on Tue, 06 Dec 2016 22:30:57 +0100 (CET), Richard Levitte <levi...@openssl.org> said:
levitte> That being said, it should certainly be easy enough to change the levitte> appropriate places to make sure headers are available as well, and I levitte> have zero issues adding a header parameter to the try_decode levitte> prototype and associated functions. Done. levitte> One thing I didn't think of earlier is that PEM_bytes_read_bio() levitte> checks the pem name against a known set, *or* in the private key case, levitte> that the pem name ends with " PRIVATE KEY" (which "TSS KEY BLOB" does levitte> not), so some kind of refactoring is needed to accomodate the levitte> store_file_load() call either way. levitte> (quite frankly, I'm slowly realising that the STORE_FILE_HANDLER code levitte> can replace quite a lot of the discovery code in the PEM module, so levitte> refactoring could be in order either way) The easiest was actually to rewrite PEM_read_bio_PrivateKey() entirely, so it solely uses the internal store_file functions I've provided. I wonder what kind of impact this would have on the community at large. Cheers, Richard -- Richard Levitte levi...@openssl.org OpenSSL Project http://www.openssl.org/~levitte/ -- openssl-dev mailing list To unsubscribe: https://mta.openssl.org/mailman/listinfo/openssl-dev
