Welcome Guest, Not a member yet? Register   Sign In
CodeIgniter 4 Important Features
#11

@ivantcholakov sorry but I can't agree with you. Here are 2-3 points which I can't accept.

1st. Putting CI 3 based app on production with CI 4 system sounds as very bad practise. How ever with good compatibility it may be not to critical problem.
2nd. Upgrading that app part by part means that at your production or local version you will have app which will not be either CI3 or CI4. At any time of this upgrade period you wont have Stable version of your own app if the BC is broken somewhere your app can Die..
3rd. There is no point to hurry with that push, so if you want an app to become CI 4 version based, you can upgrade it with all the time you need for that.

I think that there is point to upgrade only if you want to have a CI 4 based app..

PS:
Long term doesn't means to push live something which is not ok (CI3 based app with CI4 core). You can migrate it to CI 4 and then to push it Live to your clients.
The CI3 third-party libraries migration is too custom moment - are the contributors live, are they going to upgrade it and etc..
Those libs must follow CodeIgniter not the opposite, they have to cover CI 4 standards.
Best VPS Hosting : Digital Ocean
#12

I think that CI4 should break with compatibility and be a brand new branch in the codeigniter series.

If you want LTS (long term support), have CI 3 for the next 10 years. Just keep on doing security maintenance.

New projects will be done with CI 4.
Some folks will migrate to CI 4.
Other people, well they will stick with CI 3. Sorry but you can't please everyone.

This situation is much like PHP 7.

New projects will be done with PHP 7.
Some folks will migrate to PHP 7.
Other people, on shared hosting, well they will stick with PHP 5.x. Sorry but you can't please everyone.

If you don't get this point I'm trying to make. Well... ¯\_(ツ)_/¯
#13

@no1youknowz I don't get it, it is not only LTS the goal. At some point I would be reluctant to pay for CI3-based new features. I've got a fussy technical idea about the scenario I've described. It depends on the future CI4 code, I'll see. Anyway I think at the moment I could make it alone.
#14

(07-28-2015, 12:54 AM)sv3tli0 Wrote: MS is good example but its not correct to compare Windows with CodeIgniter.
Its very different if its made for Clients and if its Developer Tool.

I am against compatibility layers because we don't need them at all.

Most of what I was discussing about Windows is relevant to development, probably in part because I developed software for Windows for 10 years before I became a web developer.

CI is a web framework, and, in that role, supplies many of the same features that an operating system supplies for traditional software developers.

At the top level, MS breaks down the Windows API into the following categories:
- Administration and Management (configuration, directory services, etc.)
- Diagnostics (debugging, logging, error handling, performance counters, etc.)
- Graphics and Multimedia (drawing, images, multimedia, OpenGL, etc.)
- Networking (settings, server APIs, network monitors, etc.)
- Security (Active Directory, Authentication, Authorization, Cryptography, Certificate Enrollment, and Security Management)
- System Services (I/O, File System, Device Management, Time, Registry, Memory Management, etc.)
- Windows User Interface (Strings, Menus, Dialog Boxes, Unicode, Hooks, etc.)

With some time, you could probably find everything that CI supplies for developers somewhere in the Windows API. The real difference is that the Windows API covers a lot more ground, and, amazingly enough, if you do some careful checking to make sure your arithmetic can survive the transition, you can recompile a 32 bit application written for Windows 95 to run as a 64 bit application in Windows 10, despite the fact that Windows 10 shares little more than this API with Windows 95 (because Windows 10 descended from NT, not 9x).

You may or may not remember this, but a little over a decade ago MS seemed ready to bet the farm on something they called the .Net Framework. They even planned to name their next version of Windows "Windows.Net" for a little while. 4+ versions later, they still haven't gotten rid of the lowly Windows API, and it's usually not that difficult to upgrade applications from one version of .Net to the next. It's even called a Framework, so it shouldn't be too difficult to find the equivalences between .Net and CI. There are even numerous ways to call Win32-based DLLs from .Net applications.

More importantly, I'm not looking for long-term support of my CI3 application. I'm looking for a path to transition my application from CI3 to CI4, much like I just finished transitioning from CI2 to CI3. I may have had to override a bunch of core classes, add some version-dependent functionality, and create some helpers to allow a smoother transition for deprecated features (or to back-port new features to help move the application code forward), but at least the number of code changes to get things up and running on the latest version of the framework was manageable. By the end of the year, I intend to strip out most of the CI2 compatibility in my application, then start leveraging CI3 and Composer to move forward.

I might be willing to rewrite a couple hundred controllers, 150+ models, and a couple dozen libraries, but if I have to start down that road with the knowledge that it's an all-or-nothing project, I'll go looking for a framework with a little more long-term stability.
#15

+1 REST support!
#16

should be Codeigniter 4 have a REST API like larvel ?
#17

I think it's a waste of time focusing in compatibility layers unless it's really easy to do.

Let's say I have 100 CI projects. Why do I need to migrate to CI4 all of them? Even if it's easy it will be one more source of trouble. From the old CI projects only a few will justify to migrate probably. And those are probably important projects that need to keep evolving.

Those who don't need to keep evolving will die sooner or later. If you are not developing it in a consistent way why worry about migrate it?

Basically this is my point of view. Some things don't need to evolve and others do. Whatever makes the evolution quicker and better it's very welcome. If compatibility layers and so on will turn CI4 better so be it, if they are only stalling and serving a very small minority of cases then just forget about it and let evolution take control Smile
#18

(This post was last modified: 08-19-2016, 06:53 AM by sv3tli0.)

(08-19-2016, 03:51 AM)Ivo Miranda Wrote: I think it's a waste of time focusing in compatibility layers unless it's really easy to do.

Let's say I have 100 CI projects. Why do I need to migrate to CI4 all of them? Even if it's easy it will be one more source of trouble. From the old CI projects only a few will justify to migrate probably. And those are probably important projects that need to keep evolving.

Those who don't need to keep evolving will die sooner or later. If you are not developing it in a consistent way why worry about migrate it?

Basically this is my point of view. Some things don't need to evolve and others do. Whatever makes the evolution quicker and better it's very welcome. If compatibility layers and so on will turn CI4 better so be it, if they are only stalling and serving a very small minority of cases then just forget about it and let evolution take control Smile

If you have 100 CI projects, its not 100% needed to move them to CI 4 ASAP. 
You have to separate them to few groups : 
1. Needs to be upgraded. - CI 4 ASAP
2. No point to be upgraded. - No point of CI 4
3. Can wait upgrade.. - CI 4 at good moment in the future

Most important thing will be to start your future projects with CI 4. CI 3 is option too, it wont be killed. Smile
Best VPS Hosting : Digital Ocean
#19

(08-19-2016, 06:52 AM)sv3tli0 Wrote:
(08-19-2016, 03:51 AM)Ivo Miranda Wrote: I think it's a waste of time focusing in compatibility layers unless it's really easy to do.

Let's say I have 100 CI projects. Why do I need to migrate to CI4 all of them? Even if it's easy it will be one more source of trouble. From the old CI projects only a few will justify to migrate probably. And those are probably important projects that need to keep evolving.

Those who don't need to keep evolving will die sooner or later. If you are not developing it in a consistent way why worry about migrate it?

Basically this is my point of view. Some things don't need to evolve and others do. Whatever makes the evolution quicker and better it's very welcome. If compatibility layers and so on will turn CI4 better so be it, if they are only stalling and serving a very small minority of cases then just forget about it and let evolution take control Smile

If you have 100 CI projects, its not 100% needed to move them to CI 4 ASAP. 
You have to separate them to few groups : 
1. Needs to be upgraded. - CI 4 ASAP
2. No point to be upgraded. - No point of CI 4
3. Can wait upgrade.. - CI 4 at good moment in the future

Most important thing will be to start your future projects with CI 4. CI 3 is option too, it wont be killed. Smile

Yes of course. That's what I was trying to say Smile It seems that some people will die if they don't upgrade to CI4 despite getting no benefit with the upgrade. As long as CI3 get's security updates for some more time I don't see the trouble.
#20

What is the future of XSS protection in the CI4?




Theme © iAndrew 2016 - Forum software by © MyBB