Useful Python snippets

_pythonguidelines.md

General tips and guidelines for Python development

Code management

• Always prefer huge modules (i.e. few big source files) instead of Java's "one class, one file" policy. Or you can use Python packages instead (directory with __init__.py file) to organize multiple files.
• This tutorial about directory structure for Python projects.
• When creating "abstract" classes with some default methods, raise NotImplementedError in the base class. You can do so also with data attributes with @property in the base class like this.

Coding standards from Python PEP 8 Coding standards

• Always prefer spaces indentation, Guido told you so !
• Max 79 chars per lines, thanks Guido
• As Guido said: "Code in the core Python distribution should always use UTF-8 (or ASCII in Python 2). Files using ASCII (in Python 2) or UTF-8 (in Python 3) should NOT have an encoding declaration."
• "Imports should usually be on separate lines"
• Very VERY important naming conventions !

General coding tips

• Do not use (object) attributes if the attribute is not used at least 2 times in the object, otherwise use a local method argument.

Tools

• AutoPEP8 Checker

coll.py

# defaultdict (http://docs.python.org/2/library/collections.html#collections.defaultdict)
# lets create default keys as we access them.

import collections

# It uses the default factory str to build empty elements automatically from this class.
# (Any class with 0 argument works here)
l = collections.defaultdict(str)

l[123]  # It will create an empty str object at non-existent key 123
print dict(l)

# Unlimited "clean" defaultdict implementation for tree-like structure (Source: http://ohuiginn.net/mt/2010/07/nested_dictionaries_in_python.html)
class NestedDict(dict):
    def __getitem__(self, key):
        if key in self: return self.get(key)
        return self.setdefault(key, NestedDict())
# More optimized version
class _NestedDict(dict):
    __slots__ = ()
    def __missing__(self, key):
        self[key] = value = _NestedDict()
        return value

crossplatform.py

# Correct way to use path concatenation for cross-platform use
os.listdir(os.path.join('mydir', 'subdir'))

sequence.py

# split seq in length size chunks list (works even if len(seq) is not a multiple of length)
[seq[i:i + length] for i in range(0, len(seq), length)]

# "2-level" nested list comprehension
test = {
    "foo": (
        u"Some",
        u"Text",
    ),
    "bar": (
        u"I want in a PLAIN list",
    )
}

plain_list = [x for y in test.values() for x in y]  # [u'Some', u'Text', u'I want in a PLAIN list']

# Split to two different variables at ' '
(foo, bar) = baz.split(' ', 1)