Redefining PHP's Standard Library: A Focus on Object-Oriented Programming
While developing and helping others develop PHP applications I noticed the trend to use PHP's arrays in nearly every task. Arrays in PHP are useful but they aren't well suited to object-oriented programming since they are not objects. I made this library to make writing object-oriented code easier.
This project requires PHP 5.4 because it uses Traits, the callable type-hint and short-array syntax. Depending on how long this project takes to mature it might require PHP 5.5 for its generators and improved handling of non-scalar keys in Iterators.
This project does not provide an autoloader. This project uses a natural 1-to-1 mapping of the class or interface name to the filesystem. This project supports the tool composer which can generate an autoloader for you.
This project is unstable and subject to significant changes from release to release. Therefore it requires that you exercise patience if you use it in your project.
There is no roadmap for this project. I work on it when I find something in my daily coding that could benefit this library. A few associates of mine also use this (unstable) repository and suggest changes. I intend to release version 1.0.0 when it has been successfully used in multiple projects and can commit to a stable API. Until then, minor versions will be tagged as they are deemed helpful. This project follows semantic versioning.
How can I help?
The best way to help is to use the library and submit issues when you find them. As of right now, every line in the library has been executed in a unit test*. The quality of some of these tests is poor and there are still bugs lurking around.
* Except the Trie and SplayTree classes which are experimental structures.
Can you add X structure?
Maybe. Open an issue and mark it as a feature request and I'll look into it.
Why not use the existing Standard PHP Library?
The current Standard PHP Library (SPL) has many problems, some of which are documented in an unfinished RFC regarding the SPL. I won't go into them here, but I felt I needed to give at least some reasoning for not using it.