Clean a directory (e.g. cache) by removing rarely used files.

clean_directory.py

#!/usr/bin/env python

"""
Clean a directory (e.g. cache) by removing rarely used files.

Usage:
  ./clean_directory directory [max_size]

The max_size argument is expected in megabytes. File removals are reported to
standard output and afterwards empty directories are (silently) removed.

Jeroen Laros <j.f.j.laros@lumc.nl>
Martijn Vermaat <m.vermaat.hg@lumc.nl>

This code is in the public domain; it can be used for whatever purpose with
absolutely no restrictions.
"""


import sys
import os


# Default maximum directory size (in megabytes).
DEFAULT_MAX_SIZE = 50


def directory_size(directory):
    """
    Return the size of a directory in bytes.

    @arg directory: Directory under scrutiny.
    @type directory: string

    @return: The total size of {directory} in bytes.
    @rtype: int
    """
    size = 0
    for (path, dirs, files) in os.walk(directory):
        for file in files:
            size += os.path.getsize(os.path.join(path, file))

    return size


def clean_directory(directory, max_size=DEFAULT_MAX_SIZE):
    """
    Keep removing files until the total size of the directory is less than the
    maximum size. Empty directories are removed.

    @arg directory: Directory under scrutiny.
    @type directory: string
    @kwarg max_size: Maximum size of the directory in megabytes.
    @type max_size: int

    First, {directory} is checked for its size. If it exceeds {max_size}, the
    'oldest' files are deleted. Note that accessing a file makes it 'new'.
    """
    max_size = max_size * pow(2, 20)

    if directory_size(directory) < max_size:
        return

    # Build a list of files sorted by access time.
    entries = []
    for (path, dirs, files) in os.walk(directory):
        for file in files:
            file_path = os.path.join(path, file)
            entries.append( (os.stat(file_path).st_atime, file_path) )
    entries.sort()

    # Now start removing pairs of files until the size of the directory is
    # small enough (or until the list is exhausted).
    for entry in entries:
        os.remove(entry[1])
        print 'Removed: %s' % entry[1]
        if directory_size(directory) < max_size:
            break

    # Remove any empty directories.
    for (path, dirs, files) in os.walk(directory, topdown=False):
        for dir in dirs:
            os.rmdir(os.path.join(path, dir))


if __name__ == '__main__':
    if not 1 < len(sys.argv) < 4:
        print 'Usage:\n  ./clean_directory directory [max_size]'
        sys.exit(1)
    try:
        clean_directory(sys.argv[1], int(sys.argv[2]))
    except IndexError:
        clean_directory(sys.argv[1])
    except ValueError:
        print 'Second argument must be an integer'
        sys.exit(1)