Super-fast and clean conversions to numbers.
Super-fast and clean conversions to numbers.
fastnumbers is a module with the following three objectives:
- Provide drop-in replacements for the Python built-in int and float that on average are up to 2x faster. These functions should behave identically to the Python built-ins except for a few specific corner-cases as mentioned in the API documentation.
- Provide a set of convenience functions that wrap the above int and float replacements and provides easy, concise, powerful, fast and flexible error handling.
- Provide a set of functions that can be used to rapidly identify if an input could be converted to int or float.
NOTICE: The first major version release after Jan 1, 2020 will drop support for Python 2.7.
The below examples showcase the fast_float function, which is a fast conversion function with error-handling. Please see the API Documentation for other functions that are available from fastnumbers.
>>> from fastnumbers import fast_float, float as fnfloat >>> # Convert string to a float >>> fast_float('56.07') 56.07 >>> # Unconvertable string returned as-is by default >>> fast_float('bad input') 'bad input' >>> # Unconvertable strings can trigger a default value >>> fast_float('bad input', default=0) 0 >>> # 'default' is also the first optional positional arg >>> fast_float('bad input', 0) 0 >>> # Integers are converted to floats >>> fast_float(54) 54.0 >>> # One can ask inf or nan to be substituted with another value >>> fast_float('nan') nan >>> fast_float('nan', nan=0.0) 0.0 >>> fast_float(float('nan'), nan=0.0) 0.0 >>> fast_float('56.07', nan=0.0) 56.07 >>> # The default built-in float behavior can be triggered with >>> # "raise_on_invalid" set to True. >>> fast_float('bad input', raise_on_invalid=True) #doctest: +IGNORE_EXCEPTION_DETAIL Traceback (most recent call last): ... ValueError: invalid literal for float(): bad input >>> # A key function can be used to return an alternate value for invalid input >>> fast_float('bad input', key=len) 9 >>> fast_float(54, key=len) 54.0 >>> # Single unicode characters can be converted. >>> fast_float(u'\u2164') # Roman numeral 5 (V) 5.0 >>> fast_float(u'\u2466') # 7 enclosed in a circle 7.0
NOTE: If you need locale-dependent conversions, supply the fastnumbers function of your choice to locale.atof.
import locale locale.setlocale(locale.LC_ALL, 'de_DE.UTF-8') print(atof('468,5', func=fast_float)) # Prints 468.5
Just how much faster is fastnumbers than a pure python implementation? Please see the following Jupyter notebooks for timing information on various Python versions.
How Is fastnumbers So Fast?
CPython goes to great lengths to ensure that your string input is converted to a number correctly (you can prove this to yourself by examining the source code for integer conversions and for float conversions), but this extra effort is only needed for very large integers or for floats with many digits or large exponents. For integers, if the result could fit into a C long then a naive algorithm of < 10 lines of C code is sufficient. For floats, if the number does not require high precision or does not have a large exponent (such as “-123.45e6”) then a short naive algorithm is also possible.
These naive algorithms are quite fast, but the performance improvement comes at the expense of being unsafe (no protection against overflow or round-off errors). fastnumbers uses a heuristic to determine if the input can be safely converted with the much faster naive algorithm. These heuristics are extremely conservative - if there is any chance that the naive result would not give exactly the same result as the built-in functions then it will fall back on CPython’s conversion function. For this reason, fastnumbers is aways at least as fast as CPython’s built-in float and int functions, and oftentimes is significantly faster because most real-world numbers pass the heuristic.
$ pip install fastnumbers
How to Run Tests
Please note that fastnumbers is NOT set-up to support python setup.py test.
The recommended way to run tests is with tox. Suppose you want to run tests for Python 3.6 - you can run tests by simply executing the following:
$ tox -e py36
tox will create virtual a virtual environment for your tests and install all the needed testing requirements for you.
If you want to run testing on all of Python 2.7, 3.4, 3.5, 3.6, and 3.7 you can simply execute
If you do not wish to use tox, you can install the testing dependencies and run the tests manually using pytest - fastnumbers contains a Pipfile for use with pipenv that makes it easy for you to install the testing dependencies:
$ pipenv install --skip-lock --dev $ pipenv install --skip-lock -e . $ pipenv run pytest
Please visit the changelog.
Release history Release notifications
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.