From: degski (degski_at_[hidden])
Date: 2020-02-28 15:26:31
On Thu, 27 Feb 2020 at 10:08, Rainer Deyke via Boost <boost_at_[hidden]>
> Looks like you missed the point of my two categories. These are not
> categories of libraries, but categories of needs that users have, which
> an individual library may or may not meet.
No, I understood you well, but collapsed the conclusion towards a
conclusion about libs.
For example: Let's say I want to write a program that reads encrypted
> zip files. I therefore need a library that provides an
> implementation(s) of the specific algorithm(s) used to by encrypted zip
> files, or I need to provide my own.
> It doesn't matter for the purpose of my program that the default
> ZipCrypto used by zip files is a terrible, terrible encryption algorithm
> that should never be used. The file is already encrypted, the damage is
> already done, and I just want to decrypt it.
> Crypto++ doesn't provide a ZipCrypto implementation, but it does provide
> several other algorithms that can also be used to encrypt zip files.
> libsodium, on the other hand, treats encryption as a black box - it only
> provides one secret-key encryption algorithm, and you have to search the
> documentation thoroughly to even find out which algorithm that is.
Hmm, you're implying that you want to use data from one crypto-lib (or any
old way of creating that data), and then use (decrypt) that with whatever
Boost produces. If doing this seriously and the data are (behind the
scenes) encrypted using OpenSSL, one should use OpenSSL to decrypt in my
opinion (common sense), then any problem-fixes get correctly pushed
downstream. This does not seem a very good idea in the first place as the
any old-way-of-creating-that-data might not exactly be what you imagine it
to be. Also the zip example is inherently bad, since this is something
totally different from encryption, but yes for a C++dev, implementing this
functionality using streams, it all looks the same, but I think these
things are substantially different.
-- @realdegski https://brave.com/google-gdpr-workaround/ "We value your privacy, click here!" Sod off! - degski "Anyone who believes that exponential growth can go on forever in a finite world is either a madman or an economist" - Kenneth E. Boulding "Growth for the sake of growth is the ideology of the cancer cell" - Edward P. Abbey
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk