Bug 1979713 - Review Request: hash-library - Portable C++ hashing library
Summary: Review Request: hash-library - Portable C++ hashing library
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Fedora
Classification: Fedora
Component: Package Review
Version: rawhide
Hardware: All
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Nobody's working on this, feel free to take it
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2021-07-06 19:34 UTC by Jerry James
Modified: 2021-07-08 02:30 UTC (History)
3 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2021-07-08 02:30:59 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)

Description Jerry James 2021-07-06 19:34:21 UTC
Spec URL: https://jjames.fedorapeople.org/hash-library/hash-library.spec
SRPM URL: https://jjames.fedorapeople.org/hash-library/hash-library-8-1.fc35.src.rpm
Fedora Account System Username: jjames
Description: This package is a new dependency of normaliz.  It contains a library of hash algorithms for C++.  In a nutshell:

- computes CRC32, MD5, SHA1 and SHA256 (most common member of the SHA2 functions), Keccak and its SHA3 sibling
- optional HMAC (keyed-hash message authentication code)
- no external dependencies, small code size
- can work chunk-wise (for example when reading streams block-by-block)
- portable: supports Windows and Linux, tested on Little Endian and Big Endian CPUs
- roughly as fast as Linux core hashing functions

Comment 1 Zbigniew Jędrzejewski-Szmek 2021-07-07 08:55:35 UTC
Hmm, does this make sense? Shouldn't it use openssl instead?

Comment 2 Jerry James 2021-07-07 21:59:56 UTC
That was my first reaction, too.  There already several implementations of these algorithms around.  Why yet another library?

*Shrug*.  The normaliz package wants this library.  I don't know why.  It must carry some advantage for them.

Comment 3 Jerry James 2021-07-08 02:30:59 UTC
I've looked a little deeper, and I can see no reason to prefer this new library to the openssl implementation.  I've created a patch to use libcrypto instead of this library, so I am withdrawing this review request.


Note You need to log in before you can comment on or make changes to this bug.