Python code coverage for Lib/filecmp.py

#	count	content
1	n/a	"""Utilities for comparing files and directories.
2	n/a
3	n/a	Classes:
4	n/a	dircmp
5	n/a
6	n/a	Functions:
7	n/a	cmp(f1, f2, shallow=True) -> int
8	n/a	cmpfiles(a, b, common) -> ([], [], [])
9	n/a	clear_cache()
10	n/a
11	n/a	"""
12	n/a
13	n/a	import os
14	n/a	import stat
15	n/a	from itertools import filterfalse
16	n/a
17	n/a	__all__ = ['clear_cache', 'cmp', 'dircmp', 'cmpfiles', 'DEFAULT_IGNORES']
18	n/a
19	n/a	_cache = {}
20	n/a	BUFSIZE = 8*1024
21	n/a
22	n/a	DEFAULT_IGNORES = [
23	n/a	'RCS', 'CVS', 'tags', '.git', '.hg', '.bzr', '_darcs', '__pycache__']
24	n/a
25	n/a	def clear_cache():
26	n/a	"""Clear the filecmp cache."""
27	n/a	_cache.clear()
28	n/a
29	n/a	def cmp(f1, f2, shallow=True):
30	n/a	"""Compare two files.
31	n/a
32	n/a	Arguments:
33	n/a
34	n/a	f1 -- First file name
35	n/a
36	n/a	f2 -- Second file name
37	n/a
38	n/a	shallow -- Just check stat signature (do not read the files).
39	n/a	defaults to True.
40	n/a
41	n/a	Return value:
42	n/a
43	n/a	True if the files are the same, False otherwise.
44	n/a
45	n/a	This function uses a cache for past comparisons and the results,
46	n/a	with cache entries invalidated if their stat information
47	n/a	changes. The cache may be cleared by calling clear_cache().
48	n/a
49	n/a	"""
50	n/a
51	n/a	s1 = _sig(os.stat(f1))
52	n/a	s2 = _sig(os.stat(f2))
53	n/a	if s1[0] != stat.S_IFREG or s2[0] != stat.S_IFREG:
54	n/a	return False
55	n/a	if shallow and s1 == s2:
56	n/a	return True
57	n/a	if s1[1] != s2[1]:
58	n/a	return False
59	n/a
60	n/a	outcome = _cache.get((f1, f2, s1, s2))
61	n/a	if outcome is None:
62	n/a	outcome = _do_cmp(f1, f2)
63	n/a	if len(_cache) > 100: # limit the maximum size of the cache
64	n/a	clear_cache()
65	n/a	_cache[f1, f2, s1, s2] = outcome
66	n/a	return outcome
67	n/a
68	n/a	def _sig(st):
69	n/a	return (stat.S_IFMT(st.st_mode),
70	n/a	st.st_size,
71	n/a	st.st_mtime)
72	n/a
73	n/a	def _do_cmp(f1, f2):
74	n/a	bufsize = BUFSIZE
75	n/a	with open(f1, 'rb') as fp1, open(f2, 'rb') as fp2:
76	n/a	while True:
77	n/a	b1 = fp1.read(bufsize)
78	n/a	b2 = fp2.read(bufsize)
79	n/a	if b1 != b2:
80	n/a	return False
81	n/a	if not b1:
82	n/a	return True
83	n/a
84	n/a	# Directory comparison class.
85	n/a	#
86	n/a	class dircmp:
87	n/a	"""A class that manages the comparison of 2 directories.
88	n/a
89	n/a	dircmp(a, b, ignore=None, hide=None)
90	n/a	A and B are directories.
91	n/a	IGNORE is a list of names to ignore,
92	n/a	defaults to DEFAULT_IGNORES.
93	n/a	HIDE is a list of names to hide,
94	n/a	defaults to [os.curdir, os.pardir].
95	n/a
96	n/a	High level usage:
97	n/a	x = dircmp(dir1, dir2)
98	n/a	x.report() -> prints a report on the differences between dir1 and dir2
99	n/a	or
100	n/a	x.report_partial_closure() -> prints report on differences between dir1
101	n/a	and dir2, and reports on common immediate subdirectories.
102	n/a	x.report_full_closure() -> like report_partial_closure,
103	n/a	but fully recursive.
104	n/a
105	n/a	Attributes:
106	n/a	left_list, right_list: The files in dir1 and dir2,
107	n/a	filtered by hide and ignore.
108	n/a	common: a list of names in both dir1 and dir2.
109	n/a	left_only, right_only: names only in dir1, dir2.
110	n/a	common_dirs: subdirectories in both dir1 and dir2.
111	n/a	common_files: files in both dir1 and dir2.
112	n/a	common_funny: names in both dir1 and dir2 where the type differs between
113	n/a	dir1 and dir2, or the name is not stat-able.
114	n/a	same_files: list of identical files.
115	n/a	diff_files: list of filenames which differ.
116	n/a	funny_files: list of files which could not be compared.
117	n/a	subdirs: a dictionary of dircmp objects, keyed by names in common_dirs.
118	n/a	"""
119	n/a
120	n/a	def __init__(self, a, b, ignore=None, hide=None): # Initialize
121	n/a	self.left = a
122	n/a	self.right = b
123	n/a	if hide is None:
124	n/a	self.hide = [os.curdir, os.pardir] # Names never to be shown
125	n/a	else:
126	n/a	self.hide = hide
127	n/a	if ignore is None:
128	n/a	self.ignore = DEFAULT_IGNORES
129	n/a	else:
130	n/a	self.ignore = ignore
131	n/a
132	n/a	def phase0(self): # Compare everything except common subdirectories
133	n/a	self.left_list = _filter(os.listdir(self.left),
134	n/a	self.hide+self.ignore)
135	n/a	self.right_list = _filter(os.listdir(self.right),
136	n/a	self.hide+self.ignore)
137	n/a	self.left_list.sort()
138	n/a	self.right_list.sort()
139	n/a
140	n/a	def phase1(self): # Compute common names
141	n/a	a = dict(zip(map(os.path.normcase, self.left_list), self.left_list))
142	n/a	b = dict(zip(map(os.path.normcase, self.right_list), self.right_list))
143	n/a	self.common = list(map(a.__getitem__, filter(b.__contains__, a)))
144	n/a	self.left_only = list(map(a.__getitem__, filterfalse(b.__contains__, a)))
145	n/a	self.right_only = list(map(b.__getitem__, filterfalse(a.__contains__, b)))
146	n/a
147	n/a	def phase2(self): # Distinguish files, directories, funnies
148	n/a	self.common_dirs = []
149	n/a	self.common_files = []
150	n/a	self.common_funny = []
151	n/a
152	n/a	for x in self.common:
153	n/a	a_path = os.path.join(self.left, x)
154	n/a	b_path = os.path.join(self.right, x)
155	n/a
156	n/a	ok = 1
157	n/a	try:
158	n/a	a_stat = os.stat(a_path)
159	n/a	except OSError as why:
160	n/a	# print('Can\'t stat', a_path, ':', why.args[1])
161	n/a	ok = 0
162	n/a	try:
163	n/a	b_stat = os.stat(b_path)
164	n/a	except OSError as why:
165	n/a	# print('Can\'t stat', b_path, ':', why.args[1])
166	n/a	ok = 0
167	n/a
168	n/a	if ok:
169	n/a	a_type = stat.S_IFMT(a_stat.st_mode)
170	n/a	b_type = stat.S_IFMT(b_stat.st_mode)
171	n/a	if a_type != b_type:
172	n/a	self.common_funny.append(x)
173	n/a	elif stat.S_ISDIR(a_type):
174	n/a	self.common_dirs.append(x)
175	n/a	elif stat.S_ISREG(a_type):
176	n/a	self.common_files.append(x)
177	n/a	else:
178	n/a	self.common_funny.append(x)
179	n/a	else:
180	n/a	self.common_funny.append(x)
181	n/a
182	n/a	def phase3(self): # Find out differences between common files
183	n/a	xx = cmpfiles(self.left, self.right, self.common_files)
184	n/a	self.same_files, self.diff_files, self.funny_files = xx
185	n/a
186	n/a	def phase4(self): # Find out differences between common subdirectories
187	n/a	# A new dircmp object is created for each common subdirectory,
188	n/a	# these are stored in a dictionary indexed by filename.
189	n/a	# The hide and ignore properties are inherited from the parent
190	n/a	self.subdirs = {}
191	n/a	for x in self.common_dirs:
192	n/a	a_x = os.path.join(self.left, x)
193	n/a	b_x = os.path.join(self.right, x)
194	n/a	self.subdirs[x] = dircmp(a_x, b_x, self.ignore, self.hide)
195	n/a
196	n/a	def phase4_closure(self): # Recursively call phase4() on subdirectories
197	n/a	self.phase4()
198	n/a	for sd in self.subdirs.values():
199	n/a	sd.phase4_closure()
200	n/a
201	n/a	def report(self): # Print a report on the differences between a and b
202	n/a	# Output format is purposely lousy
203	n/a	print('diff', self.left, self.right)
204	n/a	if self.left_only:
205	n/a	self.left_only.sort()
206	n/a	print('Only in', self.left, ':', self.left_only)
207	n/a	if self.right_only:
208	n/a	self.right_only.sort()
209	n/a	print('Only in', self.right, ':', self.right_only)
210	n/a	if self.same_files:
211	n/a	self.same_files.sort()
212	n/a	print('Identical files :', self.same_files)
213	n/a	if self.diff_files:
214	n/a	self.diff_files.sort()
215	n/a	print('Differing files :', self.diff_files)
216	n/a	if self.funny_files:
217	n/a	self.funny_files.sort()
218	n/a	print('Trouble with common files :', self.funny_files)
219	n/a	if self.common_dirs:
220	n/a	self.common_dirs.sort()
221	n/a	print('Common subdirectories :', self.common_dirs)
222	n/a	if self.common_funny:
223	n/a	self.common_funny.sort()
224	n/a	print('Common funny cases :', self.common_funny)
225	n/a
226	n/a	def report_partial_closure(self): # Print reports on self and on subdirs
227	n/a	self.report()
228	n/a	for sd in self.subdirs.values():
229	n/a	print()
230	n/a	sd.report()
231	n/a
232	n/a	def report_full_closure(self): # Report on self and subdirs recursively
233	n/a	self.report()
234	n/a	for sd in self.subdirs.values():
235	n/a	print()
236	n/a	sd.report_full_closure()
237	n/a
238	n/a	methodmap = dict(subdirs=phase4,
239	n/a	same_files=phase3, diff_files=phase3, funny_files=phase3,
240	n/a	common_dirs = phase2, common_files=phase2, common_funny=phase2,
241	n/a	common=phase1, left_only=phase1, right_only=phase1,
242	n/a	left_list=phase0, right_list=phase0)
243	n/a
244	n/a	def __getattr__(self, attr):
245	n/a	if attr not in self.methodmap:
246	n/a	raise AttributeError(attr)
247	n/a	self.methodmap[attr](self)
248	n/a	return getattr(self, attr)
249	n/a
250	n/a	def cmpfiles(a, b, common, shallow=True):
251	n/a	"""Compare common files in two directories.
252	n/a
253	n/a	a, b -- directory names
254	n/a	common -- list of file names found in both directories
255	n/a	shallow -- if true, do comparison based solely on stat() information
256	n/a
257	n/a	Returns a tuple of three lists:
258	n/a	files that compare equal
259	n/a	files that are different
260	n/a	filenames that aren't regular files.
261	n/a
262	n/a	"""
263	n/a	res = ([], [], [])
264	n/a	for x in common:
265	n/a	ax = os.path.join(a, x)
266	n/a	bx = os.path.join(b, x)
267	n/a	res[_cmp(ax, bx, shallow)].append(x)
268	n/a	return res
269	n/a
270	n/a
271	n/a	# Compare two files.
272	n/a	# Return:
273	n/a	# 0 for equal
274	n/a	# 1 for different
275	n/a	# 2 for funny cases (can't stat, etc.)
276	n/a	#
277	n/a	def _cmp(a, b, sh, abs=abs, cmp=cmp):
278	n/a	try:
279	n/a	return not abs(cmp(a, b, sh))
280	n/a	except OSError:
281	n/a	return 2
282	n/a
283	n/a
284	n/a	# Return a copy with items that occur in skip removed.
285	n/a	#
286	n/a	def _filter(flist, skip):
287	n/a	return list(filterfalse(skip.__contains__, flist))
288	n/a
289	n/a
290	n/a	# Demonstration and testing.
291	n/a	#
292	n/a	def demo():
293	n/a	import sys
294	n/a	import getopt
295	n/a	options, args = getopt.getopt(sys.argv[1:], 'r')
296	n/a	if len(args) != 2:
297	n/a	raise getopt.GetoptError('need exactly two args', None)
298	n/a	dd = dircmp(args[0], args[1])
299	n/a	if ('-r', '') in options:
300	n/a	dd.report_full_closure()
301	n/a	else:
302	n/a	dd.report()
303	n/a
304	n/a	if __name__ == '__main__':
305	n/a	demo()