Gecko [ herbhavenuk.co.uk ]

Name	Size	Permission	Date
__pycache__	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
collections	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
concurrent	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
config-3.3m	[ DIR ]	drwxr-xr-x	2025-08-15 19:32
ctypes	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
curses	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
dbm	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
distutils	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
email	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
encodings	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
html	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
http	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
idlelib	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
importlib	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
json	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
lib-dynload	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
lib2to3	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
logging	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
multiprocessing	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
plat-linux	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
pydoc_data	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
site-packages	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
sqlite3	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
test	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
unittest	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
urllib	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
venv	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
wsgiref	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
xml	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
xmlrpc	[ DIR ]	drwxr-xr-x	2025-08-15 19:31
__future__.py	4.48 KB	-rw-r--r--	2024-04-17 17:12
__phello__.foo.py	64 B	-rw-r--r--	2024-04-17 17:12
_compat_pickle.py	4.24 KB	-rw-r--r--	2024-04-17 17:12
_dummy_thread.py	4.66 KB	-rw-r--r--	2024-04-17 17:12
_markupbase.py	14.26 KB	-rw-r--r--	2024-04-17 17:12
_osx_support.py	18.41 KB	-rw-r--r--	2024-04-17 17:12
_pyio.py	71.2 KB	-rw-r--r--	2024-04-17 17:12
_strptime.py	21.17 KB	-rw-r--r--	2024-04-17 17:12
_sysconfigdata.py	23.52 KB	-rw-r--r--	2024-04-17 17:12
_threading_local.py	7.24 KB	-rw-r--r--	2024-04-17 17:12
_weakrefset.py	5.57 KB	-rw-r--r--	2024-04-17 17:12
abc.py	7.87 KB	-rw-r--r--	2024-04-17 17:12
aifc.py	30.33 KB	-rw-r--r--	2024-04-17 17:12
antigravity.py	475 B	-rw-r--r--	2024-04-17 17:12
argparse.py	86.98 KB	-rw-r--r--	2024-04-17 17:12
ast.py	11.86 KB	-rw-r--r--	2024-04-17 17:12
asynchat.py	11.32 KB	-rw-r--r--	2024-04-17 17:12
asyncore.py	20.27 KB	-rw-r--r--	2024-04-17 17:12
base64.py	13.66 KB	-rwxr-xr-x	2024-04-17 17:12
bdb.py	21.38 KB	-rw-r--r--	2024-04-17 17:12
binhex.py	13.39 KB	-rw-r--r--	2024-04-17 17:12
bisect.py	2.53 KB	-rw-r--r--	2024-04-17 17:12
bz2.py	18.04 KB	-rw-r--r--	2024-04-17 17:12
cProfile.py	6.21 KB	-rwxr-xr-x	2024-04-17 17:12
calendar.py	22.4 KB	-rw-r--r--	2024-04-17 17:12
cgi.py	34.72 KB	-rwxr-xr-x	2024-04-17 17:12
cgitb.py	11.76 KB	-rw-r--r--	2024-04-17 17:12
chunk.py	5.25 KB	-rw-r--r--	2024-04-17 17:12
cmd.py	14.51 KB	-rw-r--r--	2024-04-17 17:12
code.py	9.79 KB	-rw-r--r--	2024-04-17 17:12
codecs.py	35.11 KB	-rw-r--r--	2024-04-17 17:12
codeop.py	5.85 KB	-rw-r--r--	2024-04-17 17:12
colorsys.py	3.6 KB	-rw-r--r--	2024-04-17 17:12
compileall.py	9.51 KB	-rw-r--r--	2024-04-17 17:12
configparser.py	48.28 KB	-rw-r--r--	2024-04-17 17:12
contextlib.py	8.91 KB	-rw-r--r--	2024-04-17 17:12
copy.py	8.78 KB	-rw-r--r--	2024-04-17 17:12
copyreg.py	6.46 KB	-rw-r--r--	2024-04-17 17:12
crypt.py	1.83 KB	-rw-r--r--	2024-04-17 17:12
csv.py	15.81 KB	-rw-r--r--	2024-04-17 17:12
datetime.py	73.2 KB	-rw-r--r--	2024-04-17 17:12
decimal.py	223.2 KB	-rw-r--r--	2024-04-17 17:12
difflib.py	80.58 KB	-rw-r--r--	2024-04-17 17:12
dis.py	9.9 KB	-rw-r--r--	2024-04-17 17:12
doctest.py	100.52 KB	-rw-r--r--	2024-04-17 17:12
dummy_threading.py	2.75 KB	-rw-r--r--	2024-04-17 17:12
filecmp.py	9.37 KB	-rw-r--r--	2024-04-17 17:12
fileinput.py	13.92 KB	-rw-r--r--	2024-04-17 17:12
fnmatch.py	3.09 KB	-rw-r--r--	2024-04-17 17:12
formatter.py	14.58 KB	-rw-r--r--	2024-04-17 17:12
fractions.py	22.49 KB	-rw-r--r--	2024-04-17 17:12
ftplib.py	39.31 KB	-rw-r--r--	2024-04-17 17:12
functools.py	13.28 KB	-rw-r--r--	2024-04-17 17:12
genericpath.py	3.02 KB	-rw-r--r--	2024-04-17 17:12
getopt.py	7.31 KB	-rw-r--r--	2024-04-17 17:12
getpass.py	5.66 KB	-rw-r--r--	2024-04-17 17:12
gettext.py	20.15 KB	-rw-r--r--	2024-04-17 17:12
glob.py	2.77 KB	-rw-r--r--	2024-04-17 17:12
gzip.py	23.83 KB	-rw-r--r--	2024-04-17 17:12
hashlib.py	6.05 KB	-rw-r--r--	2024-04-17 17:12
heapq.py	17.58 KB	-rw-r--r--	2024-04-17 17:12
hmac.py	4.34 KB	-rw-r--r--	2024-04-17 17:12
imaplib.py	48.94 KB	-rw-r--r--	2024-04-17 17:12
imghdr.py	3.45 KB	-rw-r--r--	2024-04-17 17:12
imp.py	9.5 KB	-rw-r--r--	2024-04-17 17:12
inspect.py	77.11 KB	-rw-r--r--	2024-04-17 17:12
io.py	3.2 KB	-rw-r--r--	2024-04-17 17:12
ipaddress.py	68.66 KB	-rw-r--r--	2024-04-17 17:12
keyword.py	2.01 KB	-rwxr-xr-x	2024-04-17 17:12
linecache.py	3.77 KB	-rw-r--r--	2024-04-17 17:12
locale.py	91.03 KB	-rw-r--r--	2024-04-17 17:12
lzma.py	17.04 KB	-rw-r--r--	2024-04-17 17:12
macpath.py	5.49 KB	-rw-r--r--	2024-04-17 17:12
macurl2path.py	2.67 KB	-rw-r--r--	2024-04-17 17:12
mailbox.py	77.24 KB	-rw-r--r--	2024-04-17 17:12
mailcap.py	7.26 KB	-rw-r--r--	2024-04-17 17:12
mimetypes.py	20.25 KB	-rw-r--r--	2024-04-17 17:12
modulefinder.py	22.65 KB	-rw-r--r--	2024-04-17 17:12
netrc.py	5.61 KB	-rw-r--r--	2024-04-17 17:12
nntplib.py	41.78 KB	-rw-r--r--	2024-04-17 17:12
ntpath.py	19.96 KB	-rw-r--r--	2024-04-17 17:12
nturl2path.py	2.34 KB	-rw-r--r--	2024-04-17 17:12
numbers.py	10.15 KB	-rw-r--r--	2024-04-17 17:12
opcode.py	4.98 KB	-rw-r--r--	2024-04-17 17:12
optparse.py	58.93 KB	-rw-r--r--	2024-04-17 17:12
os.py	33.96 KB	-rw-r--r--	2024-04-17 17:12
os2emxpath.py	4.55 KB	-rw-r--r--	2024-04-17 17:12
pdb.py	59.23 KB	-rwxr-xr-x	2024-04-17 17:12
pickle.py	46.74 KB	-rw-r--r--	2024-04-17 17:12
pickletools.py	79.44 KB	-rw-r--r--	2024-04-17 17:12
pipes.py	8.71 KB	-rw-r--r--	2024-04-17 17:12
pkgutil.py	21.03 KB	-rw-r--r--	2024-04-17 17:12
platform.py	49.55 KB	-rwxr-xr-x	2024-04-17 17:12
plistlib.py	14.43 KB	-rw-r--r--	2024-04-17 17:12
poplib.py	11.11 KB	-rw-r--r--	2024-04-17 17:12
posixpath.py	13.92 KB	-rw-r--r--	2024-04-17 17:12
pprint.py	12.4 KB	-rw-r--r--	2024-04-17 17:12
profile.py	20.95 KB	-rwxr-xr-x	2024-04-17 17:12
pstats.py	25.75 KB	-rw-r--r--	2024-04-17 17:12
pty.py	4.94 KB	-rw-r--r--	2024-04-17 17:12
py_compile.py	6.56 KB	-rw-r--r--	2024-04-17 17:12
pyclbr.py	13.12 KB	-rw-r--r--	2024-04-17 17:12
pydoc.py	99.26 KB	-rwxr-xr-x	2024-04-17 17:12
queue.py	8.63 KB	-rw-r--r--	2024-04-17 17:12
quopri.py	7.14 KB	-rwxr-xr-x	2024-04-17 17:12
random.py	25.06 KB	-rw-r--r--	2024-04-17 17:12
re.py	14.62 KB	-rw-r--r--	2024-04-17 17:12
reprlib.py	4.99 KB	-rw-r--r--	2024-04-17 17:12
rlcompleter.py	5.4 KB	-rw-r--r--	2024-04-17 17:12
runpy.py	10.17 KB	-rw-r--r--	2024-04-17 17:12
sched.py	6.25 KB	-rw-r--r--	2024-04-17 17:12
shelve.py	8.05 KB	-rw-r--r--	2024-04-17 17:12
shlex.py	11.23 KB	-rw-r--r--	2024-04-17 17:12
shutil.py	38.23 KB	-rw-r--r--	2024-04-17 17:12
site.py	21.46 KB	-rw-r--r--	2024-04-17 17:12
smtpd.py	29.5 KB	-rwxr-xr-x	2024-04-17 17:12
smtplib.py	37.13 KB	-rwxr-xr-x	2024-04-17 17:12
sndhdr.py	6.07 KB	-rw-r--r--	2024-04-17 17:12
socket.py	14.56 KB	-rw-r--r--	2024-04-17 17:12
socketserver.py	23.63 KB	-rw-r--r--	2024-04-17 17:12
sre_compile.py	15.96 KB	-rw-r--r--	2024-04-17 17:12
sre_constants.py	7.06 KB	-rw-r--r--	2024-04-17 17:12
sre_parse.py	29.5 KB	-rw-r--r--	2024-04-17 17:12
ssl.py	23.9 KB	-rw-r--r--	2024-04-17 17:12
stat.py	4.2 KB	-rw-r--r--	2024-04-17 17:12
string.py	9.19 KB	-rw-r--r--	2024-04-17 17:12
stringprep.py	12.61 KB	-rw-r--r--	2024-04-17 17:12
struct.py	238 B	-rw-r--r--	2024-04-17 17:12
subprocess.py	65.99 KB	-rw-r--r--	2024-04-17 17:12
sunau.py	17.11 KB	-rw-r--r--	2024-04-17 17:12
symbol.py	2 KB	-rwxr-xr-x	2024-04-17 17:12
symtable.py	7.21 KB	-rw-r--r--	2024-04-17 17:12
sysconfig.py	24.58 KB	-rw-r--r--	2024-04-17 17:12
tabnanny.py	11.14 KB	-rwxr-xr-x	2024-04-17 17:12
tarfile.py	86.78 KB	-rwxr-xr-x	2024-04-17 17:12
telnetlib.py	26.71 KB	-rw-r--r--	2024-04-17 17:12
tempfile.py	22.47 KB	-rw-r--r--	2024-04-17 17:12
textwrap.py	16.1 KB	-rw-r--r--	2024-04-17 17:12
this.py	1003 B	-rw-r--r--	2024-04-17 17:12
threading.py	44.57 KB	-rw-r--r--	2024-04-17 17:12
timeit.py	12.1 KB	-rwxr-xr-x	2024-04-17 17:12
token.py	2.96 KB	-rw-r--r--	2024-04-17 17:12
tokenize.py	24.29 KB	-rw-r--r--	2024-04-17 17:12
trace.py	30.75 KB	-rwxr-xr-x	2024-04-17 17:12
traceback.py	11.7 KB	-rw-r--r--	2024-04-17 17:12
tty.py	879 B	-rw-r--r--	2024-04-17 17:12
types.py	3.09 KB	-rw-r--r--	2024-04-17 17:12
uu.py	6.61 KB	-rwxr-xr-x	2024-04-17 17:12
uuid.py	21.83 KB	-rw-r--r--	2024-04-17 17:12
warnings.py	13.5 KB	-rw-r--r--	2024-04-17 17:12
wave.py	18.14 KB	-rw-r--r--	2024-04-17 17:12
weakref.py	11.23 KB	-rw-r--r--	2024-04-17 17:12
webbrowser.py	22.38 KB	-rwxr-xr-x	2024-04-17 17:12
xdrlib.py	5.25 KB	-rw-r--r--	2024-04-17 17:12
zipfile.py	64.87 KB	-rw-r--r--	2024-04-17 17:12

Rename

"""Utilities for comparing files and directories.

Classes:
    dircmp

Functions:
    cmp(f1, f2, shallow=True) -> int
    cmpfiles(a, b, common) -> ([], [], [])

"""

import os
import stat
from itertools import filterfalse

__all__ = ["cmp", "dircmp", "cmpfiles"]

_cache = {}
BUFSIZE = 8*1024

def cmp(f1, f2, shallow=True):
    """Compare two files.

Arguments:

f1 -- First file name

f2 -- Second file name

shallow -- Just check stat signature (do not read the files).
               defaults to 1.

Return value:

True if the files are the same, False otherwise.

This function uses a cache for past comparisons and the results,
    with a cache invalidation mechanism relying on stale signatures.

"""

s1 = _sig(os.stat(f1))
    s2 = _sig(os.stat(f2))
    if s1[0] != stat.S_IFREG or s2[0] != stat.S_IFREG:
        return False
    if shallow and s1 == s2:
        return True
    if s1[1] != s2[1]:
        return False

outcome = _cache.get((f1, f2, s1, s2))
    if outcome is None:
        outcome = _do_cmp(f1, f2)
        if len(_cache) > 100:      # limit the maximum size of the cache
            _cache.clear()
        _cache[f1, f2, s1, s2] = outcome
    return outcome

def _sig(st):
    return (stat.S_IFMT(st.st_mode),
            st.st_size,
            st.st_mtime)

def _do_cmp(f1, f2):
    bufsize = BUFSIZE
    with open(f1, 'rb') as fp1, open(f2, 'rb') as fp2:
        while True:
            b1 = fp1.read(bufsize)
            b2 = fp2.read(bufsize)
            if b1 != b2:
                return False
            if not b1:
                return True

# Directory comparison class.
#
class dircmp:
    """A class that manages the comparison of 2 directories.

dircmp(a, b, ignore=None, hide=None)
      A and B are directories.
      IGNORE is a list of names to ignore,
        defaults to ['RCS', 'CVS', 'tags'].
      HIDE is a list of names to hide,
        defaults to [os.curdir, os.pardir].

High level usage:
      x = dircmp(dir1, dir2)
      x.report() -> prints a report on the differences between dir1 and dir2
       or
      x.report_partial_closure() -> prints report on differences between dir1
            and dir2, and reports on common immediate subdirectories.
      x.report_full_closure() -> like report_partial_closure,
            but fully recursive.

Attributes:
     left_list, right_list: The files in dir1 and dir2,
        filtered by hide and ignore.
     common: a list of names in both dir1 and dir2.
     left_only, right_only: names only in dir1, dir2.
     common_dirs: subdirectories in both dir1 and dir2.
     common_files: files in both dir1 and dir2.
     common_funny: names in both dir1 and dir2 where the type differs between
        dir1 and dir2, or the name is not stat-able.
     same_files: list of identical files.
     diff_files: list of filenames which differ.
     funny_files: list of files which could not be compared.
     subdirs: a dictionary of dircmp objects, keyed by names in common_dirs.
     """

def __init__(self, a, b, ignore=None, hide=None): # Initialize
        self.left = a
        self.right = b
        if hide is None:
            self.hide = [os.curdir, os.pardir] # Names never to be shown
        else:
            self.hide = hide
        if ignore is None:
            self.ignore = ['RCS', 'CVS', 'tags'] # Names ignored in comparison
        else:
            self.ignore = ignore

def phase0(self): # Compare everything except common subdirectories
        self.left_list = _filter(os.listdir(self.left),
                                 self.hide+self.ignore)
        self.right_list = _filter(os.listdir(self.right),
                                  self.hide+self.ignore)
        self.left_list.sort()
        self.right_list.sort()

def phase1(self): # Compute common names
        a = dict(zip(map(os.path.normcase, self.left_list), self.left_list))
        b = dict(zip(map(os.path.normcase, self.right_list), self.right_list))
        self.common = list(map(a.__getitem__, filter(b.__contains__, a)))
        self.left_only = list(map(a.__getitem__, filterfalse(b.__contains__, a)))
        self.right_only = list(map(b.__getitem__, filterfalse(a.__contains__, b)))

def phase2(self): # Distinguish files, directories, funnies
        self.common_dirs = []
        self.common_files = []
        self.common_funny = []

for x in self.common:
            a_path = os.path.join(self.left, x)
            b_path = os.path.join(self.right, x)

ok = 1
            try:
                a_stat = os.stat(a_path)
            except os.error as why:
                # print('Can\'t stat', a_path, ':', why.args[1])
                ok = 0
            try:
                b_stat = os.stat(b_path)
            except os.error as why:
                # print('Can\'t stat', b_path, ':', why.args[1])
                ok = 0

if ok:
                a_type = stat.S_IFMT(a_stat.st_mode)
                b_type = stat.S_IFMT(b_stat.st_mode)
                if a_type != b_type:
                    self.common_funny.append(x)
                elif stat.S_ISDIR(a_type):
                    self.common_dirs.append(x)
                elif stat.S_ISREG(a_type):
                    self.common_files.append(x)
                else:
                    self.common_funny.append(x)
            else:
                self.common_funny.append(x)

def phase3(self): # Find out differences between common files
        xx = cmpfiles(self.left, self.right, self.common_files)
        self.same_files, self.diff_files, self.funny_files = xx

def phase4(self): # Find out differences between common subdirectories
        # A new dircmp object is created for each common subdirectory,
        # these are stored in a dictionary indexed by filename.
        # The hide and ignore properties are inherited from the parent
        self.subdirs = {}
        for x in self.common_dirs:
            a_x = os.path.join(self.left, x)
            b_x = os.path.join(self.right, x)
            self.subdirs[x]  = dircmp(a_x, b_x, self.ignore, self.hide)

def phase4_closure(self): # Recursively call phase4() on subdirectories
        self.phase4()
        for sd in self.subdirs.values():
            sd.phase4_closure()

def report(self): # Print a report on the differences between a and b
        # Output format is purposely lousy
        print('diff', self.left, self.right)
        if self.left_only:
            self.left_only.sort()
            print('Only in', self.left, ':', self.left_only)
        if self.right_only:
            self.right_only.sort()
            print('Only in', self.right, ':', self.right_only)
        if self.same_files:
            self.same_files.sort()
            print('Identical files :', self.same_files)
        if self.diff_files:
            self.diff_files.sort()
            print('Differing files :', self.diff_files)
        if self.funny_files:
            self.funny_files.sort()
            print('Trouble with common files :', self.funny_files)
        if self.common_dirs:
            self.common_dirs.sort()
            print('Common subdirectories :', self.common_dirs)
        if self.common_funny:
            self.common_funny.sort()
            print('Common funny cases :', self.common_funny)

def report_partial_closure(self): # Print reports on self and on subdirs
        self.report()
        for sd in self.subdirs.values():
            print()
            sd.report()

def report_full_closure(self): # Report on self and subdirs recursively
        self.report()
        for sd in self.subdirs.values():
            print()
            sd.report_full_closure()

methodmap = dict(subdirs=phase4,
                     same_files=phase3, diff_files=phase3, funny_files=phase3,
                     common_dirs = phase2, common_files=phase2, common_funny=phase2,
                     common=phase1, left_only=phase1, right_only=phase1,
                     left_list=phase0, right_list=phase0)

def __getattr__(self, attr):
        if attr not in self.methodmap:
            raise AttributeError(attr)
        self.methodmap[attr](self)
        return getattr(self, attr)

def cmpfiles(a, b, common, shallow=True):
    """Compare common files in two directories.

a, b -- directory names
    common -- list of file names found in both directories
    shallow -- if true, do comparison based solely on stat() information

Returns a tuple of three lists:
      files that compare equal
      files that are different
      filenames that aren't regular files.

"""
    res = ([], [], [])
    for x in common:
        ax = os.path.join(a, x)
        bx = os.path.join(b, x)
        res[_cmp(ax, bx, shallow)].append(x)
    return res

# Compare two files.
# Return:
#       0 for equal
#       1 for different
#       2 for funny cases (can't stat, etc.)
#
def _cmp(a, b, sh, abs=abs, cmp=cmp):
    try:
        return not abs(cmp(a, b, sh))
    except os.error:
        return 2

# Return a copy with items that occur in skip removed.
#
def _filter(flist, skip):
    return list(filterfalse(skip.__contains__, flist))

# Demonstration and testing.
#
def demo():
    import sys
    import getopt
    options, args = getopt.getopt(sys.argv[1:], 'r')
    if len(args) != 2:
        raise getopt.GetoptError('need exactly two args', None)
    dd = dircmp(args[0], args[1])
    if ('-r', '') in options:
        dd.report_full_closure()
    else:
        dd.report()

if __name__ == '__main__':
    demo()