Expose LibCrypt's PKCS5_PBKDF2_HMAC #7264
Conversation
|
This is perfect, thanks and sorry for the confusion! |
src/openssl/algorithm.cr
Outdated
| SHA512 | ||
|
|
||
| def to_evp | ||
| # The internal bindings to the *LibCrypto* digest operations sometimes require a hash algorithm |
There was a problem hiding this comment.
Ditto. Needs to be put in front of to_evp.
But I would only use the bottom text:
Returns the appropriate equivalent hash algorithm that corresponds to the
current enum value.
There was a problem hiding this comment.
"equivalent hash algorithm" should be more precise. This enum already represents hash algorithms. This method returns the corresponding identifier in libcrypto.
There was a problem hiding this comment.
The @r00ster91's suggestion seems very concise to me. Thanks for the suggestion.
But really have to agree that this phrasing adds some ambiguity like stated by @straight-shoota.
There was a problem hiding this comment.
The OpenSSL documentation mentions the expression "message digest function" when referring to this type.
Also, the documentation for the SSLeay library (from where OpenSSL was forked) makes it clear that the acronym EVP stands for Envelope.
There was a problem hiding this comment.
The type of the equivalent hash algorithms in the LibCrypto API is EVP_MD.
I found that this MD stands for Message Digest, and in some point the ancient documentation mentions the expressions and "digest method" and "message digest algorithm" related to this type.
That pretty much confuses me. I thought that the algorithm was called hash and the ouput was called digest. (source).
It seems that back then digest algorithm meant what now we call hash algorithm.
Well... I think this is no help at all. I just wanted to add some value, as I can't by means of language, since I do not write very well in english.
There was a problem hiding this comment.
About this review, can I mark as resolved?
I'm not sure what to do
|
Hmm, why the darwin CI builds are constantly failing...? |
|
@straight-shoota I reckon that, I was just wondering... Thanks for creating an issue for that! |
|
needs a format with |
|
Please rebase on master. This will allow |
|
@mniak I would need to ask you for another rebase on master. I'm sorry this is coming 1 month after the last request. Thanks! |
src/openssl/lib_crypto.cr
Outdated
| @@ -236,6 +236,7 @@ lib LibCrypto | |||
| fun md5 = MD5(data : UInt8*, lengh : LibC::SizeT, md : UInt8*) : UInt8* | |||
|
|
|||
| fun pkcs5_pbkdf2_hmac_sha1 = PKCS5_PBKDF2_HMAC_SHA1(pass : LibC::Char*, passlen : LibC::Int, salt : UInt8*, saltlen : LibC::Int, iter : LibC::Int, keylen : LibC::Int, out : UInt8*) : LibC::Int | |||
| fun pkcs5_pbkdf2_hmac = PKCS5_PBKDF2_HMAC(pass : LibC::Char*, passlen : LibC::Int, salt : UInt8*, saltlen : LibC::Int, iter : LibC::Int, digest : EVP_MD, keylen : LibC::Int, out : UInt8*) : LibC::Int | |||
There was a problem hiding this comment.
Here a {% if compare_versions(LibSSL::OPENSSL_VERSION, "1.0.0") >= 0 %} is missing to avoid declaring the fun if it is not available.
Agreed. Thanks! Co-Authored-By: mniak <[email protected]>
Co-Authored-By: mniak <[email protected]>
Co-Authored-By: mniak <[email protected]>
Co-Authored-By: mniak <[email protected]>
Co-Authored-By: mniak <[email protected]>
Co-Authored-By: mniak <[email protected]>
|
@mniak I will handle the rebase and minor changes I asked and push the changes. Thanks for the patience with this change. |
|
Thanks @mniak and all reviewers |
|
I'm sorry for the delay. Theses days I'm not having mucha spare time as I started in a new position. |
Fixes #6973
Replaces pull request #6975