New Case Study:See how Anthropic automated 95% of dependency reviews with Socket.Learn More
Socket
Sign inDemoInstall
Socket

lazy-property

Package Overview
Dependencies
Maintainers
1
Alerts
File Explorer

Advanced tools

Socket logo

Install Socket

Detect and block malicious and high-risk dependencies

Install

lazy-property

Makes properties lazy (ie evaluated only when called)

  • 0.0.1
  • PyPI
  • Socket score

Maintainers
1

lazy_property - A package for making properties lazy

Properties <https://docs.python.org/3.4/library/functions.html#property>_ are a very useful feature of Python, effectively allowing an attribute to masquerade as a method (with no arguments other than self). However, sometimes we want to store the results of an expensive computation in an attribute. The straightforward way to do it would be something like the following:

::

import time

def do_some_big_calculation():
    """Simulating some ginormous calculation going on somewhere..."""
    print("{}: Calculations started...".format(time.time))
    time.sleep(5)
    print("{}: Calculations complete!".format(time.time))
    return 42

class SomeClass(object):

    def __init__(self):
        self.calculation_value = do_some_big_calculation()

And that certainly works. However, this means that whenever you instantiate an object from SomeClass, you'll perform this calculation each time. That could be problematic...

So, what we really want is for this calculation to only be done when it's needed. In other words, what we want is for this attribute to be lazy.

To do that, you can create a property, and a "private" attribute, named _calculation_value, used to cache the result, like so:

::

class SomeOtherClass(object):

@property
def calculation_value(self):

    if not hasattr(self, "_calculation_value"):
        self._calculation_value = do_some_big_calculation()

    return self._calculation_value

This package essentially reduces this down to a decorator:

::

import lazy_property

class YetAnotherClass(object):

    @lazy_property.LazyProperty
    def calculation_value(self):

        return do_some_big_calculation()

And when called, the "calculation" is only done once:

::

In [5]:yac = YetAnotherClass()

In [6]: yac.calculation_value
1440798443.228256: Calculations started...
1440798448.229179: Calculations complete!
Out[6]: 42

In [7]: yac.calculation_value
Out[7]: 42

Note, however, that this property is not writable:

::

In [8]: yac.calculation_value = "something_else"
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-8-0bbfd01ac59a> in <module>()
----> 1 yac.calculation_value = "something_else"

AttributeError: can't set attribute

That's what the LazyWritableProperty is for.

::

class SomethingElse(object):

@lazy_property.LazyWritableProperty
def overwritable_calculation_value(self):

    return do_some_big_calculation()

::

In [9]: se = SomethingElse()

In [10]: se.overwritable_calculation_value
1440798779.27305: Calculations started...
1440798784.274711: Calculations complete!
Out[10]: 42

In [11]: se.overwritable_calculation_value
Out[11]: 42

In [12]: se.overwritable_calculation_value = "foo"

In [13]: se.overwritable_calculation_value
Out[13]: 'foo'

Installation

It's up on PyPI:

::

pip install lazy-property

Or, to do it the hard way, clone this repo, enter the directory into which you cloned the repo, and do a

::

python setup.py install

Wait...isn't this a solved problem?

Well, yes, but I couldn't find a lazy attribute implementation that clearly implemented laziness in the (fairly simple) way discussed above.

FAQs


Did you know?

Socket

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Install

Related posts

SocketSocket SOC 2 Logo

Product

  • Package Alerts
  • Integrations
  • Docs
  • Pricing
  • FAQ
  • Roadmap
  • Changelog

Packages

npm

Stay in touch

Get open source security insights delivered straight into your inbox.


  • Terms
  • Privacy
  • Security

Made with ⚡️ by Socket Inc