IRPy: A Python repository from TApplencourt

IRPy (pronounce /kəˈθuːluː/) extend the python property function in order to support lazy evaluation and mutability of nested properties (property that use properties in their declaration and so forth). Lazy evaluation of properties are quite common (Werkzeug, cached-property...), but coherence problems can arise if you have inference rule between mutable properties. The aim of this library is to solve this problem.

Install

With pip:

pip install irpy

With conda:

conda install -c https://conda.anaconda.org/tapplencourt irpy

You can download irpy.pyx rename it into irpy.py and add it to PYTHONPATH.

API

lazy_property: a simple lazy property;
lazy_property_mutable: a lazy property that can change. When doing so, all of these ancestors are invalided and will be recomputed when needed. Furthermore, all of these descendants are now unattainable;
lazy_property_leaf(mutable,immutable): this function allow node creation's from values defined in the __init__ class method.

But why i need all of this? Or What is a scientific code?

A program is a function of its input data:

output = program (input)

A program can be represented as a production tree where

The root is the output;
The leaves are the input data;
The nodes are the intermediate variables;
The edges represent the relation needs/needed by.

In python, this way of thinking is encouraged by the usage of, special object, property. Indeed each property are one nodes of the production tree. When you nest them, you define implicitly the edges of this tree.

Illustration

This code snippet:

class WeightFactory(object):
    '''Compute the weight of a rectangle'''

    def __init__(self, l1, l2, l3):
        '''Dimension of the box (m)'''
        self.l1 = l1
        self.l2 = l2
        self.l3 = l3

    @property
    def volume(self):
        " V = l1 * l2 * l3 (m^3)"
        return self.l1 * self.l2 * self.l3 

    @property
    def density(self):
        "Volumetric mass density of Iron (kg/m^3)"
        return 7.87 * 10**3

    @property
    def mass(self):
        "m = V * d (kg)"
        return self.volume * self.density

    @property
    def g(self):
        "g-force (m/s2) in equator"
        return 9.7803

    @property
    def weight(self):
        "(Newton)"
        return self.mass*self.g

can be represend as:

In this example l[1,3], volume, density, mass, g, weight are the node of the production tree.

Remarks

In the source code of the snipper, one can see that properties have no explicit parameter (in fact, IRPy mean Implicit Reference Parameter for Python).

And this simplify dramatically sotfware developments. Indeed:

The global production tree is not known by the programmer, the programmer doesn’t handle the execution sequence. Just ask a property, it will be computed on the fly:

f = WeightFactory(1,1,1)
assert ( abs(f.weight - 76970.961) < 1.e-4)

The program is easy to write (adding a new property only require to know about the name of theses implicit parameters);

Lazy evaluation

But, the same data will be recomputed multiple times. A simple solution is to use lazy evaluation of these nodes/properties. Just replace property into lazy_property and you are good to go. (all these example can be found in the examples folder).

Draw back

In IRP paradigm, a Russian nesting lazy property, nodes are by default immutable. This mean that you cannot set these node by hand; the only way to compute a node is by using the function who have be decorated.

For example:

f = WeightFactory(1,1,1)
try:
    f.g = 1.622
except AttributeError:
    print "Node are immutable!"

Mutability

In IRPy, the node who can be changed by hand are decorated with the lazy_property_mutable function.

@lazy_property_mutable
def g(self):
    "g-force (m/s2) in equator"
    return 9.7803

Ancestor are invalided

When this kind of node node (for example g) are set, all her ancestors (wieght in your case) will be recomputed the nest time somebody ask for them.

f = WeightFactory(1,1,1)
assert ( abs(f.weight - 76970.961) < 1.e-4)
f.g = 1.622
assert ( abs(f.weight - 12765.14) < 1.e-4)

Descendant are removed from the tree

Is we change an node (the mass for example), all her descendants are removed from the tree and are, for now, unreachable.

f = WeightFactory(1,1,1)
f.mass = 80
try:
    print f.volume
except AttributeError:
    print "Node have been removed"

TApplencourt/IRPy