filecmp.py

"""Utilities for comparing files and directories.

Classes:
 dircmp

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

"""

importos
importstat
fromitertoolsimportfilterfalse
fromtypesimportGenericAlias

__all__= ['clear_cache', 'cmp', 'dircmp', 'cmpfiles', 'DEFAULT_IGNORES']

_cache= {}
BUFSIZE=8*1024

DEFAULT_IGNORES= [
'RCS', 'CVS', 'tags', '.git', '.hg', '.bzr', '_darcs', '__pycache__']

defclear_cache():
"""Clear the filecmp cache."""
_cache.clear()

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

 Arguments:

 f1 -- First file name

 f2 -- Second file name

 shallow -- treat files as identical if their stat signatures (type, size,
 mtime) are identical. Otherwise, files are considered different
 if their sizes or contents differ. [default: True]

 Return value:

 True if the files are the same, False otherwise.

 This function uses a cache for past comparisons and the results,
 with cache entries invalidated if their stat information
 changes. The cache may be cleared by calling clear_cache().

 """

s1=_sig(os.stat(f1))
s2=_sig(os.stat(f2))
ifs1[0] !=stat.S_IFREGors2[0] !=stat.S_IFREG:
returnFalse
ifshallowands1==s2:
returnTrue
ifs1[1] !=s2[1]:
returnFalse

outcome=_cache.get((f1, f2, s1, s2))
ifoutcomeisNone:
outcome=_do_cmp(f1, f2)
iflen(_cache) >100: # limit the maximum size of the cache
clear_cache()
_cache[f1, f2, s1, s2] =outcome
returnoutcome

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

def_do_cmp(f1, f2):
bufsize=BUFSIZE
withopen(f1, 'rb') asfp1, open(f2, 'rb') asfp2:
whileTrue:
b1=fp1.read(bufsize)
b2=fp2.read(bufsize)
ifb1!=b2:
returnFalse
ifnotb1:
returnTrue

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

 dircmp(a, b, ignore=None, hide=None, *, shallow=True)
 A and B are directories.
 IGNORE is a list of names to ignore,
 defaults to DEFAULT_IGNORES.
 HIDE is a list of names to hide,
 defaults to [os.curdir, os.pardir].
 SHALLOW specifies whether to just check the stat signature (do not read
 the files).
 defaults to True.

 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 instances (or MyDirCmp instances if this
 object is of type MyDirCmp, a subclass of dircmp), keyed by names
 in common_dirs.
 """

def__init__(self, a, b, ignore=None, hide=None, *, shallow=True): # Initialize
self.left=a
self.right=b
ifhideisNone:
self.hide= [os.curdir, os.pardir] # Names never to be shown
else:
self.hide=hide
ifignoreisNone:
self.ignore=DEFAULT_IGNORES
else:
self.ignore=ignore
self.shallow=shallow

defphase0(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()

defphase1(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)))

defphase2(self): # Distinguish files, directories, funnies
self.common_dirs= []
self.common_files= []
self.common_funny= []

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

ok=True
try:
a_stat=os.stat(a_path)
except (OSError, ValueError):
# See https://github.com/python/cpython/issues/122400
# for the rationale for protecting against ValueError.
# print('Can\'t stat', a_path, ':', why.args[1])
ok=False
try:
b_stat=os.stat(b_path)
except (OSError, ValueError):
# print('Can\'t stat', b_path, ':', why.args[1])
ok=False

ifok:
a_type=stat.S_IFMT(a_stat.st_mode)
b_type=stat.S_IFMT(b_stat.st_mode)
ifa_type!=b_type:
self.common_funny.append(x)
elifstat.S_ISDIR(a_type):
self.common_dirs.append(x)
elifstat.S_ISREG(a_type):
self.common_files.append(x)
else:
self.common_funny.append(x)
else:
self.common_funny.append(x)

defphase3(self): # Find out differences between common files
xx=cmpfiles(self.left, self.right, self.common_files, self.shallow)
self.same_files, self.diff_files, self.funny_files=xx

defphase4(self): # Find out differences between common subdirectories
# A new dircmp (or MyDirCmp if dircmp was subclassed) 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= {}
forxinself.common_dirs:
a_x=os.path.join(self.left, x)
b_x=os.path.join(self.right, x)
self.subdirs[x] =self.__class__(a_x, b_x, self.ignore, self.hide,
shallow=self.shallow)

defphase4_closure(self): # Recursively call phase4() on subdirectories
self.phase4()
forsdinself.subdirs.values():
sd.phase4_closure()

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

defreport_partial_closure(self): # Print reports on self and on subdirs
self.report()
forsdinself.subdirs.values():
print()
sd.report()

defreport_full_closure(self): # Report on self and subdirs recursively
self.report()
forsdinself.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):
ifattrnotinself.methodmap:
raiseAttributeError(attr)
self.methodmap[attr](self)
returngetattr(self, attr)

__class_getitem__=classmethod(GenericAlias)


defcmpfiles(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= ([], [], [])
forxincommon:
ax=os.path.join(a, x)
bx=os.path.join(b, x)
res[_cmp(ax, bx, shallow)].append(x)
returnres


# Compare two files.
# Return:
# 0 for equal
# 1 for different
# 2 for funny cases (can't stat, NUL bytes, etc.)
#
def_cmp(a, b, sh, abs=abs, cmp=cmp):
try:
returnnotabs(cmp(a, b, sh))
except (OSError, ValueError):
return2


# Return a copy with items that occur in skip removed.
#
def_filter(flist, skip):
returnlist(filterfalse(skip.__contains__, flist))


# Demonstration and testing.
#
defdemo():
importsys
importgetopt
options, args=getopt.getopt(sys.argv[1:], 'r')
iflen(args) !=2:
raisegetopt.GetoptError('need exactly two args', None)
dd=dircmp(args[0], args[1])
if ('-r', '') inoptions:
dd.report_full_closure()
else:
dd.report()

if__name__=='__main__':
demo()