Search specified directory and convert the adjacent images equal to each other into hardlink.

hardlinkimage.py

#!/usr/bin/python

import os
import os.path
import sys

import Image

class ImageFile(object):

    def __init__(self, path):
        self.path = path
        self._data = None

    def empty(self):
        return self.path is None

    def inode(self):
        return os.stat(self.path).st_ino

    def data(self):
        if not self._data is None: return self._data
        if self.empty(): return None
        try:
            im = Image.open(self.path, "r")
        except IOError as e:
            # continue if f is not an image file.
            print >>sys.stderr, self.path, "cannot be opened."
            return None
        self._data = im.tostring()
        return self._data

    def __eq__(self, other):
        return (not self.empty() and not other.empty()
                and (self.path == other.path
                     or self.inode() == other.inode()))

    def __ne__(self, other):
        return not self == other

    def equal_image(self, other):
        return self.data() == other.data()


# check only adjacents
def adjacent_image_scan(path_list):
    lastfile = ImageFile(None)
    for curpath in path_list:
        curfile = ImageFile(curpath)
        if curfile == lastfile: continue
        if curfile.equal_image(lastfile):
            print "rm", curfile.path, "; ln", lastfile.path, curfile.path
            os.unlink(curfile.path)
            os.link(lastfile.path, curfile.path)
        else:
            lastfile = curfile


def main():
    if len(sys.argv) == 1:
        print >>sys.stderr, "Usage:", sys.argv[0], "dir | files..."
    elif len(sys.argv) == 2:
        for root, dirs, files in os.walk(sys.argv[1]):
            adjacent_image_scan(os.path.join(root, f) for f in sorted(files))
    else:
        adjacent_image_scan(sys.argv[1:])


if __name__ == "__main__":
    main()