: You can now mark TypedDict items as read-only, preventing modification after creation.
Currently disabled by default, the JIT must be enabled at compile time with the --enable-experimental-jit flag. The immediate performance impact, as the official release notes acknowledge, is modest. Current benchmarks indicate the JIT merely recovers performance lost to the tier-2 optimizer, with many language constructs not yet optimized.
python -q # Quiet mode, then try: >>> def foo(): ... print("Hello") # note syntax highlighting ... >>> [key for key in range(5)] # multiline editing works python 313 release notes verified
: Leading indentation is now stripped from docstrings automatically, reducing memory usage and file size. Python documentation Verified Stability & Support Current Version : The latest maintenance release is
: This is one of the most significant changes in Python's history. It allows users to disable the Global Interpreter Lock (GIL) : You can now mark TypedDict items as
– Allows type narrowing in custom type guards.
Early benchmarks show modest improvements, with larger gains expected in future versions as the JIT matures. 3. A Modern, Powerful REPL >>> [key for key in range(5)] # multiline
class Point(TypedDict): x: ReadOnly[int] y: int
Example (verified):
This remains disabled by default. It must be explicitly configured and built from source using the --enable-experimental-jit flag to realize performance improvements.
: In previous versions, hitting the up arrow recalled single lines of code piecemeal. Now, entire code blocks (such as a full for loop or a multi-line function definition) can be recalled, edited, and rerun as a single cohesive unit.