Skip to main content

A package with lisp-like generic functions.

Project description

Overview

gf lets you write generic functions generic functions with multi-methods, that dispatch on all their arguments.

Simple Example

>>> from gf import generic, method
>>> add = generic()
>>> type(add)
<type 'function'>

Lets define add for two integers:

>>> @method(int, int)
... def add(n0, n1):
...     return n0 + n1

Lets test it:

>>> add(1, 2)
3

Calling add with instances of other types fails:

>>> add("Hello ", "World")
Traceback (most recent call last):
...
NotImplementedError: Generic '__main__.add' has no implementation for type(s): __builtin__.str, __builtin__.str

Of course add can also by defined for two strings:

>>> @method(basestring, basestring)
... def add(s0, s1):
...     return s0 + s1

Now our hello world example works:

>>> add("Hello ", "World")
'Hello World'

add can also be defined for a string and an integer:

>>> @method(basestring, int)
... def add(s, n):
...     return s + str(n)

Thus we can add a string and a number:

>>> add("You ", 2)
'You 2'

Python’s Special Methods

gf.Object implements (nearly) all of the special instance methods of a python object as a generic function. The package includes a rational number implementation that makes heavy use of this feature:

@method(object, Rational)
def __add__(a, b):
    """Add an object and a rational number.

    `a` is converted to a `Rational` and then both are added."""
    return Rational(a) + b

@method(Rational, object)
def __add__(a, b):
    """Add a rational number and an object.

    `b` is converted to a `Rational` and then both are added."""
    return a + Rational(b)

gf.Object also has a more Smalltalk means of overwriting object.__str__ and object.__repr__ using a file like object. Again the rational example is instructive about its usage.

@method(Rational, Writer)
def __out__(rational, writer):
    """Write a nice representation of the rational.

    Denominators that equal 1 are not printed."""
    writer("%d", rational.numerator)
    if rational.denominator != 1:
        writer(" / %d", rational.denominator)

@method(Rational, Writer)
def __spy__(rational, writer):
    """Write a debug representation of the rational."""
    writer("%s(", rational.__class__.__name__)
    if rational.numerator != 0:
            writer("%r", rational.numerator)
            if rational.denominator != 1:
                writer(", %r", rational.denominator)
    writer(")")

Changes

A short sketch of the changes done in each release.

Release 0.1.4

The following was fixed in Release 0.1.4:

  • Fixed an issue with variadic methods. Sometimes definitions of variadic methods added after the method was already called where not added.

  • Specified and implemented a precedence rule for overlapping variadic methods of generic functions.

  • Improved generated documentation for variadic methods.

  • Fixed the markup of some notes in the documentation.

Release 0.1.3

The following was changed in Release 0.1.3:

  • Added variadic methods, e.g. multi-methods with a variable number of arguments.

  • Improved the long description text a bit and fixed bug in its markup.

  • Fixed invalid references in the long description.

Release 0.1.2

The following was changed in Release 0.1.2:

  • Added a generic functions for gf.Object.__call__.

  • Added a gf.go.FinalizingMixin.

  • gf.generic now also accepts a type.

  • Improved the exception information for ambiguous calls.

  • Fixed some documentation glitches.

Release 0.1.1

This was the initial release.

Acknowledgements

Guido van Rossum created the core of this package. I just renamed some things and added some convenience stuff. Thank you Guido!

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

gf-0.1.4.tar.gz (22.5 kB view details)

Uploaded Source

Built Distribution

gf-0.1.4-py2.7.egg (33.8 kB view details)

Uploaded Source

File details

Details for the file gf-0.1.4.tar.gz.

File metadata

  • Download URL: gf-0.1.4.tar.gz
  • Upload date:
  • Size: 22.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for gf-0.1.4.tar.gz
Algorithm Hash digest
SHA256 18a0e767952e31c7e86642a3cd76b40e833dbacf5b538c636fbae82ad9e9d31a
MD5 c5c7ac185cd2117ff5d0bf866020d3de
BLAKE2b-256 4fb5ae8e2beceecffd7b77385041f65217c5932333726ef42c7f52ceaec5c665

See more details on using hashes here.

Provenance

File details

Details for the file gf-0.1.4-py2.7.egg.

File metadata

  • Download URL: gf-0.1.4-py2.7.egg
  • Upload date:
  • Size: 33.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for gf-0.1.4-py2.7.egg
Algorithm Hash digest
SHA256 4d128a7f50780857c12be0fd319d203faf7c16227b1f6073d61ff20084b358c6
MD5 9b8afddf93eb1874f64374cd55e62687
BLAKE2b-256 21e76374528f302c4c7fc69b108968528de480076368009b0f7d68a98b3add42

See more details on using hashes here.

Provenance

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page