diff --git a/Lib/filecmp.py b/Lib/filecmp.py new file mode 100644 index 000000000..70a4b23c9 --- /dev/null +++ b/Lib/filecmp.py @@ -0,0 +1,313 @@ +"""Utilities for comparing files and directories. + +Classes: + dircmp + +Functions: + cmp(f1, f2, shallow=True) -> int + cmpfiles(a, b, common) -> ([], [], []) + clear_cache() + +""" + +import os +import stat +from itertools import filterfalse +from types import GenericAlias + +__all__ = ['clear_cache', 'cmp', 'dircmp', 'cmpfiles', 'DEFAULT_IGNORES'] + +_cache = {} +BUFSIZE = 8*1024 + +DEFAULT_IGNORES = [ + 'RCS', 'CVS', 'tags', '.git', '.hg', '.bzr', '_darcs', '__pycache__'] + +def clear_cache(): + """Clear the filecmp cache.""" + _cache.clear() + +def cmp(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)) + 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 + clear_cache() + _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 DEFAULT_IGNORES. + 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 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): # 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 = DEFAULT_IGNORES + 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 OSError: + # print('Can\'t stat', a_path, ':', why.args[1]) + ok = 0 + try: + b_stat = os.stat(b_path) + except OSError: + # 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 (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 = {} + 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] = self.__class__(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) + + __class_getitem__ = classmethod(GenericAlias) + + +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 OSError: + 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() diff --git a/Lib/test/test_filecmp.py b/Lib/test/test_filecmp.py new file mode 100644 index 000000000..9b5ac12bc --- /dev/null +++ b/Lib/test/test_filecmp.py @@ -0,0 +1,250 @@ +import filecmp +import os +import shutil +import tempfile +import unittest + +from test import support +from test.support import os_helper + + +class FileCompareTestCase(unittest.TestCase): + def setUp(self): + self.name = os_helper.TESTFN + self.name_same = os_helper.TESTFN + '-same' + self.name_diff = os_helper.TESTFN + '-diff' + data = 'Contents of file go here.\n' + for name in [self.name, self.name_same, self.name_diff]: + with open(name, 'w', encoding="utf-8") as output: + output.write(data) + + with open(self.name_diff, 'a+', encoding="utf-8") as output: + output.write('An extra line.\n') + self.dir = tempfile.gettempdir() + + def tearDown(self): + os.unlink(self.name) + os.unlink(self.name_same) + os.unlink(self.name_diff) + + def test_matching(self): + self.assertTrue(filecmp.cmp(self.name, self.name), + "Comparing file to itself fails") + self.assertTrue(filecmp.cmp(self.name, self.name, shallow=False), + "Comparing file to itself fails") + self.assertTrue(filecmp.cmp(self.name, self.name_same), + "Comparing file to identical file fails") + self.assertTrue(filecmp.cmp(self.name, self.name_same, shallow=False), + "Comparing file to identical file fails") + + def test_different(self): + self.assertFalse(filecmp.cmp(self.name, self.name_diff), + "Mismatched files compare as equal") + self.assertFalse(filecmp.cmp(self.name, self.dir), + "File and directory compare as equal") + + def test_cache_clear(self): + first_compare = filecmp.cmp(self.name, self.name_same, shallow=False) + second_compare = filecmp.cmp(self.name, self.name_diff, shallow=False) + filecmp.clear_cache() + self.assertTrue(len(filecmp._cache) == 0, + "Cache not cleared after calling clear_cache") + +class DirCompareTestCase(unittest.TestCase): + def setUp(self): + tmpdir = tempfile.gettempdir() + self.dir = os.path.join(tmpdir, 'dir') + self.dir_same = os.path.join(tmpdir, 'dir-same') + self.dir_diff = os.path.join(tmpdir, 'dir-diff') + + # Another dir is created under dir_same, but it has a name from the + # ignored list so it should not affect testing results. + self.dir_ignored = os.path.join(self.dir_same, '.hg') + + self.caseinsensitive = os.path.normcase('A') == os.path.normcase('a') + data = 'Contents of file go here.\n' + for dir in (self.dir, self.dir_same, self.dir_diff, self.dir_ignored): + shutil.rmtree(dir, True) + os.mkdir(dir) + subdir_path = os.path.join(dir, 'subdir') + os.mkdir(subdir_path) + if self.caseinsensitive and dir is self.dir_same: + fn = 'FiLe' # Verify case-insensitive comparison + else: + fn = 'file' + with open(os.path.join(dir, fn), 'w', encoding="utf-8") as output: + output.write(data) + + with open(os.path.join(self.dir_diff, 'file2'), 'w', encoding="utf-8") as output: + output.write('An extra file.\n') + + def tearDown(self): + for dir in (self.dir, self.dir_same, self.dir_diff): + shutil.rmtree(dir) + + def test_default_ignores(self): + self.assertIn('.hg', filecmp.DEFAULT_IGNORES) + + def test_cmpfiles(self): + self.assertTrue(filecmp.cmpfiles(self.dir, self.dir, ['file']) == + (['file'], [], []), + "Comparing directory to itself fails") + self.assertTrue(filecmp.cmpfiles(self.dir, self.dir_same, ['file']) == + (['file'], [], []), + "Comparing directory to same fails") + + # Try it with shallow=False + self.assertTrue(filecmp.cmpfiles(self.dir, self.dir, ['file'], + shallow=False) == + (['file'], [], []), + "Comparing directory to itself fails") + self.assertTrue(filecmp.cmpfiles(self.dir, self.dir_same, ['file'], + shallow=False), + "Comparing directory to same fails") + + # Add different file2 + with open(os.path.join(self.dir, 'file2'), 'w', encoding="utf-8") as output: + output.write('Different contents.\n') + + self.assertFalse(filecmp.cmpfiles(self.dir, self.dir_same, + ['file', 'file2']) == + (['file'], ['file2'], []), + "Comparing mismatched directories fails") + + + def _assert_lists(self, actual, expected): + """Assert that two lists are equal, up to ordering.""" + self.assertEqual(sorted(actual), sorted(expected)) + + + def test_dircmp(self): + # Check attributes for comparison of two identical directories + left_dir, right_dir = self.dir, self.dir_same + d = filecmp.dircmp(left_dir, right_dir) + self.assertEqual(d.left, left_dir) + self.assertEqual(d.right, right_dir) + if self.caseinsensitive: + self._assert_lists(d.left_list, ['file', 'subdir']) + self._assert_lists(d.right_list, ['FiLe', 'subdir']) + else: + self._assert_lists(d.left_list, ['file', 'subdir']) + self._assert_lists(d.right_list, ['file', 'subdir']) + self._assert_lists(d.common, ['file', 'subdir']) + self._assert_lists(d.common_dirs, ['subdir']) + self.assertEqual(d.left_only, []) + self.assertEqual(d.right_only, []) + self.assertEqual(d.same_files, ['file']) + self.assertEqual(d.diff_files, []) + expected_report = [ + "diff {} {}".format(self.dir, self.dir_same), + "Identical files : ['file']", + "Common subdirectories : ['subdir']", + ] + self._assert_report(d.report, expected_report) + + # Check attributes for comparison of two different directories (right) + left_dir, right_dir = self.dir, self.dir_diff + d = filecmp.dircmp(left_dir, right_dir) + self.assertEqual(d.left, left_dir) + self.assertEqual(d.right, right_dir) + self._assert_lists(d.left_list, ['file', 'subdir']) + self._assert_lists(d.right_list, ['file', 'file2', 'subdir']) + self._assert_lists(d.common, ['file', 'subdir']) + self._assert_lists(d.common_dirs, ['subdir']) + self.assertEqual(d.left_only, []) + self.assertEqual(d.right_only, ['file2']) + self.assertEqual(d.same_files, ['file']) + self.assertEqual(d.diff_files, []) + expected_report = [ + "diff {} {}".format(self.dir, self.dir_diff), + "Only in {} : ['file2']".format(self.dir_diff), + "Identical files : ['file']", + "Common subdirectories : ['subdir']", + ] + self._assert_report(d.report, expected_report) + + # Check attributes for comparison of two different directories (left) + left_dir, right_dir = self.dir, self.dir_diff + shutil.move( + os.path.join(self.dir_diff, 'file2'), + os.path.join(self.dir, 'file2') + ) + d = filecmp.dircmp(left_dir, right_dir) + self.assertEqual(d.left, left_dir) + self.assertEqual(d.right, right_dir) + self._assert_lists(d.left_list, ['file', 'file2', 'subdir']) + self._assert_lists(d.right_list, ['file', 'subdir']) + self._assert_lists(d.common, ['file', 'subdir']) + self.assertEqual(d.left_only, ['file2']) + self.assertEqual(d.right_only, []) + self.assertEqual(d.same_files, ['file']) + self.assertEqual(d.diff_files, []) + expected_report = [ + "diff {} {}".format(self.dir, self.dir_diff), + "Only in {} : ['file2']".format(self.dir), + "Identical files : ['file']", + "Common subdirectories : ['subdir']", + ] + self._assert_report(d.report, expected_report) + + # Add different file2 + with open(os.path.join(self.dir_diff, 'file2'), 'w', encoding="utf-8") as output: + output.write('Different contents.\n') + d = filecmp.dircmp(self.dir, self.dir_diff) + self.assertEqual(d.same_files, ['file']) + self.assertEqual(d.diff_files, ['file2']) + expected_report = [ + "diff {} {}".format(self.dir, self.dir_diff), + "Identical files : ['file']", + "Differing files : ['file2']", + "Common subdirectories : ['subdir']", + ] + self._assert_report(d.report, expected_report) + + def test_dircmp_subdirs_type(self): + """Check that dircmp.subdirs respects subclassing.""" + class MyDirCmp(filecmp.dircmp): + pass + d = MyDirCmp(self.dir, self.dir_diff) + sub_dirs = d.subdirs + self.assertEqual(list(sub_dirs.keys()), ['subdir']) + sub_dcmp = sub_dirs['subdir'] + self.assertEqual(type(sub_dcmp), MyDirCmp) + + def test_report_partial_closure(self): + left_dir, right_dir = self.dir, self.dir_same + d = filecmp.dircmp(left_dir, right_dir) + left_subdir = os.path.join(left_dir, 'subdir') + right_subdir = os.path.join(right_dir, 'subdir') + expected_report = [ + "diff {} {}".format(self.dir, self.dir_same), + "Identical files : ['file']", + "Common subdirectories : ['subdir']", + '', + "diff {} {}".format(left_subdir, right_subdir), + ] + self._assert_report(d.report_partial_closure, expected_report) + + def test_report_full_closure(self): + left_dir, right_dir = self.dir, self.dir_same + d = filecmp.dircmp(left_dir, right_dir) + left_subdir = os.path.join(left_dir, 'subdir') + right_subdir = os.path.join(right_dir, 'subdir') + expected_report = [ + "diff {} {}".format(self.dir, self.dir_same), + "Identical files : ['file']", + "Common subdirectories : ['subdir']", + '', + "diff {} {}".format(left_subdir, right_subdir), + ] + self._assert_report(d.report_full_closure, expected_report) + + def _assert_report(self, dircmp_report, expected_report_lines): + with support.captured_stdout() as stdout: + dircmp_report() + report_lines = stdout.getvalue().strip().split('\n') + self.assertEqual(report_lines, expected_report_lines) + + +if __name__ == "__main__": + unittest.main()