mcrypt extermination plan

php.internals

Scott Arciszewski

10 years ago
All, I propose the following timeline to give ext/mcrypt the viking funeral it deserves: PHP 7.1 - all mcrypt functions raise E_DEPRECATED PHP 8.0 - remove ext/mcrypt, only make it available through PECL There will be no compatibility layer in this proposal, but if someone wants to build one that sits atop openssl, feel free to do that separately. Scott Arciszewski Chief Development Officer Paragon Initiative Enterprises <https://paragonie.com>

Remi Collet

10 years ago
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Le 07/01/2016 17:54, Scott Arciszewski a écrit :
> All, > > I propose the following timeline to give ext/mcrypt the viking > funeral it deserves: > > PHP 7.1 - all mcrypt functions raise E_DEPRECATED PHP 8.0 - remove > ext/mcrypt, only make it available through PECL
Big +1. We need to solve this chicken and eggs issue. People use it because it exists. It still exists because people still use it. Remi.
> > There will be no compatibility layer in this proposal, but if > someone wants to build one that sits atop openssl, feel free to do > that separately. > > Scott Arciszewski Chief Development Officer Paragon Initiative > Enterprises <https://paragonie.com> >
-----BEGIN PGP SIGNATURE----- Version: GnuPG v2 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iEYEARECAAYFAlaOmT8ACgkQYUppBSnxahjk4wCg7mbjccTdPu4Z3G8cfTp8/H+k ercAn3YDLTsCl06+a6N6W5ORT70hGEK3 =mt8x -----END PGP SIGNATURE-----

Lior Kaplan

10 years ago
On Jan 7, 2016 6:59 PM, "Remi Collet" <remi@fedoraproject.org> wrote:
> > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Le 07/01/2016 17:54, Scott Arciszewski a écrit : > > All, > > > > I propose the following timeline to give ext/mcrypt the viking > > funeral it deserves: > > > > PHP 7.1 - all mcrypt functions raise E_DEPRECATED PHP 8.0 - remove > > ext/mcrypt, only make it available through PECL > > Big +1. > > We need to solve this chicken and eggs issue. > People use it because it exists. > It still exists because people still use it.
Why not move it to pecl as part on 7.1 ? Kaplan

Scott Arciszewski

10 years ago
On Thu, Jan 7, 2016 at 1:32 PM, Lior Kaplan <lior.k@zend.com> wrote:
> On Jan 7, 2016 6:59 PM, "Remi Collet" <remi@fedoraproject.org> wrote: >> >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> Le 07/01/2016 17:54, Scott Arciszewski a écrit : >> > All, >> > >> > I propose the following timeline to give ext/mcrypt the viking >> > funeral it deserves: >> > >> > PHP 7.1 - all mcrypt functions raise E_DEPRECATED PHP 8.0 - remove >> > ext/mcrypt, only make it available through PECL >> >> Big +1. >> >> We need to solve this chicken and eggs issue. >> People use it because it exists. >> It still exists because people still use it. > > Why not move it to pecl as part on 7.1 ? > > Kaplan
I'd love that, personally. Another idea might be E_DEPRECATED in 7.1 and PECL-only in 7.1+1 (whether that's 7.2 or 8.0). Scott Arciszewski Chief Development Officer Paragon Initiative Enterprises <https://paragonie.com>

Pierre Joye

10 years ago
On Jan 7, 2016 11:54 PM, "Scott Arciszewski" <scott@paragonie.com> wrote:
> > All, > > I propose the following timeline to give ext/mcrypt the viking funeral > it deserves: > > PHP 7.1 - all mcrypt functions raise E_DEPRECATED > PHP 8.0 - remove ext/mcrypt, only make it available through PECL > > There will be no compatibility layer in this proposal, but if someone > wants to build one that sits atop openssl, feel free to do that > separately. >
Sounds good to me :)

Marco Pivetta

10 years ago
While I'd love to see mcrypt die, unless we all forgot how semver works, this isn't how it can be done :-\ If you want to actually drop something, regardless of how bad it is (and I know mcrypt is bad), then the next major version is where this should happen. Note that pushing for an earlier 8.0 is not a problem either. Marco Pivetta http://twitter.com/Ocramius http://ocramius.github.com/ On 7 January 2016 at 20:12, Pierre Joye <pierre.php@gmail.com> wrote:

Marco Pivetta

10 years ago
Edit: never mind - I must have misread somewhere that dropping in 7.2 was a plan. Sorry for the misunderstanding! Marco Pivetta http://twitter.com/Ocramius http://ocramius.github.com/ On 10 January 2016 at 12:56, Marco Pivetta <ocramius@gmail.com> wrote:

Ferenc Kovacs

10 years ago
2016. jan. 10. 12:57 ezt írta ("Marco Pivetta" <ocramius@gmail.com>):
> > While I'd love to see mcrypt die, unless we all forgot how semver works, > this isn't how it can be done :-\ > If you want to actually drop something, regardless of how bad it is (and I > know mcrypt is bad), then the next major version is where this should > happen. > Note that pushing for an earlier 8.0 is not a problem either.
For the record our release process (https://wiki.php.net/rfc/releaseprocess) explicitly states that extension support can be ended/moved to pecl in a minor version.