ariebovenberg / gentools

Tools for generators, generator functions, and generator-based coroutines

Gentools

Tools for generators, generator functions, and generator-based coroutines.

Key features:

• Create reusable generators
• Compose generators
• Build python 2/3-compatible generators (gentools version <1.2 only)

Installation

pip install gentools

Examples

• Make generator functions reusable:

>>> @reusable
... def countdown(value, step):
...     while value > 0:
...         yield value
...         value -= step

>>> from_3 = countdown(3, step=1)
>>> list(from_3)
[3, 2, 1]
>>> list(from_3)
[3, 2, 1]
>>> isinstance(from_3, countdown)  # generator func is wrapped in a class
True
>>> from_3.step  # attribute access to arguments
1
>>> from_3.replace(value=5)  # create new instance with replaced fields
countdown(value=5, step=1)  # descriptive repr()

• map a generator's yield, send, and return values:

>>> @map_return('final value: {}'.format)
... @map_send(int)
... @map_yield('the current max is: {}'.format)
... def my_max(value):
...     while value < 100:
...         newvalue = yield value
...         if newvalue > value:
...             value = newvalue
...     return value

>>> gen = my_max(5)
>>> next(gen)
'the current max is: 5'
>>> gen.send(11.3)
'the current max is: 11'
>>> gen.send(104)
StopIteration('final value: 104')

• relay a generator's yield/send interactions through another generator:

>>> def try_until_positive(outvalue):
...     value = yield outvalue
...     while value < 0:
...         value = yield 'not positive, try again'
...     return value

>>> @relay(try_until_positive)
... def my_max(value):
...     while value < 100:
...         newvalue = yield value
...         if newvalue > value:
...             value = newvalue
...     return value

>>> gen = my_max(5)
>>> next(gen)
5
>>> gen.send(-4)
'not positive, try again'
>>> gen.send(-1)
'not positive, try again'
>>> gen.send(8)
8
>>> gen.send(104)
StopIteration(104)

• make python 2/3 compatible generators with return. (gentools version <1.2 only)

>>> @py2_compatible
... def my_max(value):
...     while value < 100:
...         newvalue = yield value
...         if newvalue > value:
...             value = newvalue
...     return_(value)