[openssl-project] A proposal for an updated OpenSSL version scheme (v2)

Richard Levitte levitte at openssl.org
Fri Sep 21 13:02:29 UTC 2018

In message <CAHEJ-S4wB1RPqzYeGaQrE28KdUd5ZVYuMeraJL9FQ1QtdViOyg at mail.gmail.com> on Fri, 21 Sep 2018 21:17:12 +1000, Tim Hudson <tjh at cryptsoft.com> said:


So you mean we'd lose the integer form entirely?  Hey, I have zero
issue with that!  However, that's going to make life hard for everyone
that wants to be able to build against different OpenSSL versions and
use this macro in pre-processing their source.  I don't think we'll
get many happy faces with such a move (talk about an API break and
making life hard on people).

I hope I misunderstand what you're going for...

(hey, if we want to drop that integer, or make it something different,
maybe we should follow the POSIX example and make that YYYYMM
(year+month) of the release?  It's incremental and certainly usable
with pre-processing!  ....  and less hard on our users than dropping
the number altogether)

> and OPENSSL_VERSION_TEXT should be "X.Y.Z [-patch][+buildmeta]" and
> that would be a simple, direct, and expected mapping to OpenSSL for
> semantic versioning.

Semantic versioning says MAJOR.MINOR.PATCH, i.e. Z would be the patch
number.  Why you want to have the patch indicator where semantic
versioning has pre-release information (which is exactly what we do
with the -dev and -prex suffixes), I do not understand.

Richard Levitte         levitte at openssl.org
OpenSSL Project         http://www.openssl.org/~levitte/

More information about the openssl-project mailing list