compare_projects.py

gistfile1.txt

#!/usr/bin/env python

import os
import sys
import stat
import os.path
import getpass


def walktree(top, callback):
    """
    recursively descend the directory tree rooted at top,
    calling the callback function for each regular file

    From: https://docs.python.org/2/library/stat.html
    """
    for f in os.listdir(top):
        pathname = os.path.join(top, f)
        mode = os.stat(pathname).st_mode
        if stat.S_ISDIR(mode):
            # It's a directory, recurse into it
            walktree(pathname, callback)
        elif stat.S_ISREG(mode):
            # It's a file, call the callback function
            callback(pathname)
        else:
            # Unknown file type, print a message
            print "Skipping %s" % (pathname,)


def _create_new_path(file_, orig_base_path, new_base_path):
    """
    Return file path re-located in new directory
    orig: /projects/user1/project1/blah
    new:  /projects/user1/project2/blah
    """

    relative_path = file_[len(orig_base_path):]
    new_path = os.path.join(new_base_path, relative_path)
    return new_path


def _update_results(results, name):
    """Dictionary counter"""
    results[name] = results.get(name, 0) + 1
    return results


# closure for callback
def callback(path, compare, results):
    """
    path: base path to look at
    compare: base path to compare against
    results: dictionary for holding results
    """

    def visitfile(file_):
        """
        Check file_ to see if it is hardlinked to the same inode in path2
        """

        # update count
        _update_results(results, "count")

        s1 = os.stat(file_)

        new_path = _create_new_path(file_, path, compare)
        if os.path.exists(new_path):
            s2 = os.stat(new_path)

            # compare inode and device number
            if (s1[stat.ST_INO], s1[stat.ST_DEV]) == \
               (s2[stat.ST_INO], s2[stat.ST_DEV]):
                # print "%s == %s" % (file, new_path)
                _update_results(results, "matches")
            else:
                # print "!!!! %s != %s" % (file, new_path)
                _update_results(results, "misses")
        else:
            # print "### Doesn't exist! %s" % (file, new_path)
            _update_results(results, "misses")

    # return function closure
    return visitfile


def main():
    # I prefer a dictionary to collections.Counter
    results = {}

    if len(sys.argv) != 3:
        print "Error:  missing args!"
        print "Usage: compare_env.py project1 project2"
        sys.exit(1)

    # fails without a controlling terminal
    # user = os.getlogin()

    user = getpass.getuser()
    print "User: %s" % (user,)

    path = "/projects/%s/%s/" % (user, sys.argv[1])
    compare = "/projects/%s/%s/" % (user, sys.argv[2])

    if not os.path.exists(path):
        print "Error: project %s doesn't exist" % (sys.argv[1],)
        sys.exit(1)
    if not os.path.exists(compare):
        print "Error: project %s doesn't exist" % (sys.argv[2],)
        sys.exit(1)

    print "Comparing:\n  %s\n  %s" % (path, compare)

    walktree(path, callback(path, compare, results))

    p_match = float(results.get("matches", 0)) / results.get("count", 1) * 100
    p_miss  = float(results.get("misses" , 0)) / results.get("count", 1) * 100

    print
    print "  Links:  %-10d (%7.4f%%)" % (results.get("matches", 0), p_match)
    print "  Unique: %-10d (%7.4f%%)" % (results.get("misses", 0), p_miss)
    print "--------------------------------"
    print "  Total:  %-10d" % (results.get("count", 0),)
    print

if __name__ == '__main__':
    main()