def diff(filename1, filename2):
if not os.path.exists(filename1):
msg = NOT_EXISTS.format(filename1)
print(msg, file=sys.stderr)
sys.exit(1)
if not os.path.exists(filename2):
msg = NOT_EXISTS.format(filename2)
print(msg, file=sys.stderr)
sys.exit(1)
with open(filename1, 'r') as file_obj:
lines1 = file_obj.readlines()
with open(filename2, 'r') as file_obj:
lines2 = file_obj.readlines()
if lines1 == lines2:
msg = '{} and {} are identical'.format(filename1, filename2)
print(msg)
else:
diff_lines = difflib.context_diff(
lines1, lines2, fromfile=filename1, tofile=filename2)
print(''.join(diff_lines))
sys.exit(1)
python类context_diff()的实例源码
def get_diff(value1, value2, name1, name2):
"""Get a diff between two strings.
Args:
value1 (str): First string to be compared.
value2 (str): Second string to be compared.
name1 (str): Name of the first string.
name2 (str): Name of the second string.
Returns:
str: The full diff.
"""
lines1 = [line + '\n' for line in value1.splitlines()]
lines2 = [line + '\n' for line in value2.splitlines()]
diff_lines = difflib.context_diff(
lines1, lines2, fromfile=name1, tofile=name2)
return ''.join(diff_lines)
def compare(self, old: str, old_date: str, new: str, new_date: str,
ctx: common.Context, meta: dict) \
-> ty.Tuple[bool, ty.Optional[str], ty.Optional[dict]]:
# pylint: disable=invalid-sequence-index
old_lines = old.split('\n')
res = list(difflib.context_diff(
old_lines, new.split('\n'),
fromfiledate=old_date, tofiledate=new_date,
lineterm='\n'))
changed_lines = sum(1 for line in res[2:]
if line and line[0] != ' ' and line[0] != '*')
if not _check_changes(ctx, changed_lines, len(old_lines),
self.conf.get("changes_threshold"),
self.conf.get("min_changed")):
return False, None, None
return True, "\n".join(res), self.opts
def main():
parser = argparse.ArgumentParser()
parser.add_argument("file1")
parser.add_argument("file2")
args = parser.parse_args()
lines1 = load_jsonl_file(args.file1)
lines2 = load_jsonl_file(args.file2)
pretty_lines1 = prettify(lines1)
pretty_lines2 = prettify(lines2)
for line in difflib.context_diff(pretty_lines1.splitlines(), pretty_lines2.splitlines(),
fromfile=args.file1, tofile=args.file2):
print(line)
def simple_diff(a, b, fromfile='', tofile='',
fromfiledate='', tofiledate='', n=3, lineterm='\n'):
"""
A function with the same calling signature as difflib.context_diff
(diff -c) and difflib.unified_diff (diff -u) but which prints
output like the simple, unadorned 'diff" command.
"""
sm = difflib.SequenceMatcher(None, a, b)
def comma(x1, x2):
return x1+1 == x2 and str(x2) or '%s,%s' % (x1+1, x2)
result = []
for op, a1, a2, b1, b2 in sm.get_opcodes():
if op == 'delete':
result.append("%sd%d" % (comma(a1, a2), b1))
result.extend(map(lambda l: '< ' + l, a[a1:a2]))
elif op == 'insert':
result.append("%da%s" % (a1, comma(b1, b2)))
result.extend(map(lambda l: '> ' + l, b[b1:b2]))
elif op == 'replace':
result.append("%sc%s" % (comma(a1, a2), comma(b1, b2)))
result.extend(map(lambda l: '< ' + l, a[a1:a2]))
result.append('---')
result.extend(map(lambda l: '> ' + l, b[b1:b2]))
return result
def print_diff_strings(orig, new):
for line in difflib.context_diff(orig.splitlines(), new.splitlines()):
print(line)
def calculate_diff(self, new, old, zone):
return "\n".join(context_diff(
a=old.splitlines(),
b=new.splitlines(),
fromfile=str(zone.domain_active_serial) + '-' + zone.domain_name,
tofile=str(zone.domain_serial) + '-' + zone.domain_name,
lineterm=''))
def calculate_diff(self, config, new, old):
return "\n".join(context_diff(
a=old.splitlines(),
b=new.splitlines(),
fromfile=str(config.active_serial),
tofile=str(config.serial),
lineterm=''))
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def main():
usage = "usage: %prog [options] fromfile tofile"
parser = optparse.OptionParser(usage)
parser.add_option("-c", action="store_true", default=False, help='Produce a context format diff (default)')
parser.add_option("-u", action="store_true", default=False, help='Produce a unified format diff')
parser.add_option("-m", action="store_true", default=False, help='Produce HTML side by side diff (can use -c and -l in conjunction)')
parser.add_option("-n", action="store_true", default=False, help='Produce a ndiff format diff')
parser.add_option("-l", "--lines", type="int", default=3, help='Set number of context lines (default 3)')
(options, args) = parser.parse_args()
if len(args) == 0:
parser.print_help()
sys.exit(1)
if len(args) != 2:
parser.error("need to specify both a fromfile and tofile")
n = options.lines
fromfile, tofile = args
fromdate = time.ctime(os.stat(fromfile).st_mtime)
todate = time.ctime(os.stat(tofile).st_mtime)
fromlines = open(fromfile, 'U').readlines()
tolines = open(tofile, 'U').readlines()
if options.u:
diff = difflib.unified_diff(fromlines, tolines, fromfile, tofile, fromdate, todate, n=n)
elif options.n:
diff = difflib.ndiff(fromlines, tolines)
elif options.m:
diff = difflib.HtmlDiff().make_file(fromlines,tolines,fromfile,tofile,context=options.c,numlines=n)
else:
diff = difflib.context_diff(fromlines, tolines, fromfile, tofile, fromdate, todate, n=n)
sys.stdout.writelines(diff)
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def main():
usage = "usage: %prog [options] fromfile tofile"
parser = optparse.OptionParser(usage)
parser.add_option("-c", action="store_true", default=False, help='Produce a context format diff (default)')
parser.add_option("-u", action="store_true", default=False, help='Produce a unified format diff')
parser.add_option("-m", action="store_true", default=False, help='Produce HTML side by side diff (can use -c and -l in conjunction)')
parser.add_option("-n", action="store_true", default=False, help='Produce a ndiff format diff')
parser.add_option("-l", "--lines", type="int", default=3, help='Set number of context lines (default 3)')
(options, args) = parser.parse_args()
if len(args) == 0:
parser.print_help()
sys.exit(1)
if len(args) != 2:
parser.error("need to specify both a fromfile and tofile")
n = options.lines
fromfile, tofile = args
fromdate = time.ctime(os.stat(fromfile).st_mtime)
todate = time.ctime(os.stat(tofile).st_mtime)
fromlines = open(fromfile, 'U').readlines()
tolines = open(tofile, 'U').readlines()
if options.u:
diff = difflib.unified_diff(fromlines, tolines, fromfile, tofile, fromdate, todate, n=n)
elif options.n:
diff = difflib.ndiff(fromlines, tolines)
elif options.m:
diff = difflib.HtmlDiff().make_file(fromlines,tolines,fromfile,tofile,context=options.c,numlines=n)
else:
diff = difflib.context_diff(fromlines, tolines, fromfile, tofile, fromdate, todate, n=n)
sys.stdout.writelines(diff)
def step(self, output, should_not=False, exactly=False):
child = self.get_scenario_context().command_response['child']
ensure_command_finished(child)
output = 'stdout' if output == 'output' else output
# todo: separate stdout and stderr
# todo: test replace
data = child.logfile_read.getvalue().replace('\r\n', '\n')
data_lines = data.splitlines()
if data.endswith('\n'):
data_lines.append('')
expected = self.get_text().encode('utf-8') # todo: test encode
expected_lines = expected.splitlines()
if expected.endswith('\n'):
expected_lines.append('')
bool_matcher = is_not if should_not else is_
comparison_matcher = equal_to if exactly else contains_string
try:
assert_that(
data,
bool_matcher(
comparison_matcher(expected)
)
)
except AssertionError:
if comparison_matcher == equal_to and bool_matcher == is_:
diff = '\n'.join(
difflib.context_diff(
data_lines,
expected_lines
)
)
raise AssertionError('Comparison error. Diff:\n' + diff)
else:
raise
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def diff_strings(_a, _b):
result = "---------- String A-----------\n"
result += _a + "\n"
result += "---------- String B-----------\n"
result += _b + "\n"
result += "---------- Diff between A and B-----------\n" + "\n"
for line in difflib.context_diff(_a, _b):
result += line
return result
def context_diff(self, lines=3):
return difflib.context_diff(
self.fromlines, self.tolines, self._fromfile,
self._tofile, self.fromdate, self.todate, n=lines)
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def test_tab_delimiter(self):
args = ['one', 'two', 'Original', 'Current',
'2005-01-26 23:30:50', '2010-04-02 10:20:52']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], [
"--- Original\t2005-01-26 23:30:50",
"+++ Current\t2010-04-02 10:20:52"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], [
"*** Original\t2005-01-26 23:30:50",
"--- Current\t2010-04-02 10:20:52"])
def test_no_trailing_tab_on_empty_filedate(self):
args = ['one', 'two', 'Original', 'Current']
ud = difflib.unified_diff(*args, lineterm='')
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"])
cd = difflib.context_diff(*args, lineterm='')
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"])
def testSimplify(self):
inputFile = os.path.join(wikidataDir, "data/1000.json")
linesOut = open(os.path.join(wikidataDir, "test/1000.out")).read().split()
linesNew = subprocess.check_output([wdExtractPy, "-f", "-F", "-l", "en", "-s", "", "-t", "all", inputFile]).split()
diff = list(difflib.context_diff(linesOut, linesNew))
self.assertEqual(len(diff), 0, "Unexpected differences in output of wd-extract.py -f -F -l en -s '' -t all data/1000.json\n"
+ "".join(diff))
