After around 18 months of development, Python 3.4 has been released. There were no new language changes for this release, but there were many new features in the standard library and CPython implementation. Python 3.4 may not incorporate any new syntax features but it does have other new features, new library modules, improved library modules, and other enhancements.

Among the highlights of the new version include the Pip-support and new modules for Statistics and asynchronous input-output operations. Pip is a package manager for Python modules and now by default comes with Python. With Pip-support and “pip install”, Python can be extended to useful modules. To this end, the programming language have a module called “ensurepip” integrated into the standard library, which ensures the presence of Pip. The bootstrap process does not require network access, but relies on a built-in standard Pip library version.

The module “statistics.py” provides developers a new library suite of mathematical statistics functions. Whether it comes to averages, medians, variances or standard deviations, Python 3.4 brings all these in the new algorithm module. Also interesting is the use of SIP hash as the default hash algorithm for strings and bytes. The current hash algorithm of CPython is not immune against DoS attacks. The new algorithm promises a secure random number generation for hashes.

The most recent version of Python also presents the implementation of semantic objects and the finalization of a tracemalloc module for tracking the allocated memory, which detect areas of memory that is allocated to objects. The change in the file descriptors is re-designed to prevent default inheritance in sub-processes. Before proceeding with the installation developers will need to keep in mind that the binaries for AMD64 will be compatible with processors that implement the Intel 64 architecture.

Some of the new features that are directly visible to the user are: the tool pip is no longer with Python. It is worth noting that the version 3.3 the same happens with venv; There is object-management of pathways through the file system of the new pathlib module; ability to create counters through the enum; a new module for simple statistical instruments; new debugging tool, which we can now see that binds memory for objects in Python through the tracemalloc; new protocol version Pickle and new module asyncio for asynchronous input/output.

Python is widely used by developers worldwide due to its features like dynamic typing, automatic memory management, full introspection, exception handling, support for multi-threaded computing and comfortable high-level data structures. Python supports structured, object-oriented, functional, imperative, and an aspect-oriented paradigm. Code in Python is organized in classes and functions that can be combined into modules (which in turn can be combined in packages).

You can check in more detail what’s new in Python 3.4 on this page and download from this page.