4 ms·
It's so sad that examples like these are not part of every spec (in fact, the very first thing in any spec). I understand the dangers of programming by example
by _pmf_ 8y ago
It's so sad that examples like these are not part of every spec (in fact, the very first thing in any spec). I understand the dangers of programming by example rather than by spec, but a lack of canonical, verified examples has been a problem with every protocol I've had to implement.
- syncsynchalt 8y agoGood news: for cryptographic protocols and methods they are! Many of the cryptographic primitives that I implemented for my test TLS 1.3 implementation (https://github.com/syncsynchalt/tincan-tls https://github.com/syncsynchalt/tincan-tls) come with test vectors in the spec itself, and the rest always had test vectors supplied in a separate doc. For an example see Appendix C of the AES standard, FIPS-197: https://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.197.pdf https://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.197.pdf In the case of TLS 1.3 I was able to use this IETF draft: https://tools.ietf.org/html/draft-ietf-tls-tls13-vectors-07 https://tools.ietf.org/html/draft-ietf-tls-tls13-vectors-07