Code Snippets

sed

Sed is very powerful. A single sed statement can turn a cat into cement. Observe:

  echo cat | sed statement

double.py (X11 License, pretty source, API docs)

A Python module for working with IEEE 754 floating-point numbers. While the Python language supports floating-point numbers natively, the built-in language and standard library support for this datatype are inadequate: it's impossible to initialize floating-point variables to special values (like infinity) in a portable way, division by zero is always trapped and raises an exception, and even basic functionality familiar from C or Java is entirely absent from the Python standard library. The double module is intended as a workaround. It supplies basic floating-point constants, floating-point classification methods, sign handling, and division. The new division routine is necessary because Python refuses to evaluate mixed float/long expressions like this one:
```
from __future__ import division
print 1   / (1L << 1024)     # returns 2^(-1024)
print 1.0 / (1L << 1024)     # raises an unnecessary OverflowError
print 1.0 / 0.0              # raises an unnecessary ZeroDivisionError

from double import fdiv
print fdiv(1,   1L << 1024)  # returns 2^(-1024)
print fdiv(1.0, 1L << 1024)  # returns 2^(-1024)
print fdiv(1.0, 0.0)         # returns double.POSITIVE_INFINITY
```