Drop PHP 5.2 suppport from the library

[treffynnon]

Remove the requirement for Idiorm to be 5.2 compatible.

See the comments on issue #78 for more information on why this must take place.

This will affect a small number of users hopefully, but if there is a major backlash then I am not averse to leaving this requirement drop til version 3. Please discuss.

[tag]

More than concerns over 5.3-only (which won't bother me, although I've seen the concern raised by a couple of others), I'm curious about the long-term roadmap. What features do you have in mind for 1.4? 1.5? 2.0? Is there timeline for when these versions might appear? Is there intent (or discussion about) PSR-0 organization, PSR-1 method renaming, ... etc.? These seem to be the most common changes in some of the other forks out there.

I've seen some interesting features discussed here and in other forks, but it would be great to have more insight into where you want to go, what you're working on, and what you want help with.

[treffynnon]

I agree that our internal plans could be publicised better, but to be honest we are all so busy that a formal plan is not in place.

V2 will adhere to PSR-1 and PSR-0 though. @j4mie are hoping to meet up over a pint soon so I get a better idea of where he saw it going as well.

My current plan is to close off all pull requests and feature requests for the 1.4.0 release. Then besides some minor back ports and bug fixes end development on v1.* so we can concentrate on v2 of the library.

V2 is likely to be more of a ground up rewrite to be honest. But the important thing for now is to clean up the v1 stuff.

Hopefully I can get a more formal roadmap in the wiki soon.

[hellogerard]

FWIW, I have used Paris/Idiorm in every new project for almost 2 years now (when I haven't used Mongo :o ) and have not run into any major limitations. It provides just enough abstraction to simplify 90% of what I need to do, and with the remaining 10% I am happy to do something more explicit or low-level. If I wanted something more powerful, I would be using something like Doctrine or Propel. Something to keep in mind for V2.

That said, I would love to see a nice, simple, drop-in database migration tool, a la Rails. Laravel's is probably the closest to what I have in mind in terms of simplicity/function, but I don't want to use the framework and it's not on Packagist.

[treffynnon]

I agree and simplicity is key to all Idiorm development. We are all
attracted to it for the same reasons so it would be foolish to lose sight
of that like you say. It is more about doing a similar amount of features
but making use of the latest PHP has to offer to simplify or improve
aspects of the ORM.

This is going to be easier by making a big BC break at v2.

With that mantra of simplicity in mind it is very unlikely that migrations
will ever be handled by Idiorm. There are a number of projects out there
that already cover this ground in PHP.

It is something that a number of people have asked for tho

FWIW, I have used Paris/Idiorm in every new project for almost 2 years now
(when I haven't used Mongo :o ) and have not run into any major
limitations. It provides just enough abstraction to simplify 90% of what I
need to do, and with the remaining 10% I am happy to do something more
explicit or low-level. If I wanted something more powerful, I would be
using something like Doctrine or Propel. Something to keep in mind for V2.

That said, I would love to see a nice, simple, drop-in database migration
tool, a la Rails. Laravel's is probably the closest to what I have in mind
in terms of simplicity/function, but I don't want to use the framework and
it's not on Packagist.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/86#issuecomment-10902357.

[treffynnon]

I have given this more thought and I am going to change the versioning I laid out before.

v2.0.0 will be the PHP version 5.3 break
v3.0.0 will be a rewrite of the library

This is so I can continue to support users of PHP 5.2 with bug fixes and some backports of features where possible to the v1.0.0 branch of the code.