如何正确包装单元测试差异？
我正在运行python 3.6.4，但是有时单元测试差异无法按预期工作。例如，在以下情况下，在不需要换行行为的情况下，强制单元测试错误与预期错误有关。
import unittest
class TestSemanticRules(unittest.TestCase):
maxDiff = None
def test_badWrapping(self):
self.assertEqual(
"1. Duplicated target language name defined in your grammar on: [@-1,63:87='Abstract Machine Language'<__ANON_3>,3:19]\n"
"2. Duplicated master scope name defined in your grammar on: [@-1,138:147='source.sma'<__ANON_3>,5:20]"
,
"1. Duplicated target language name defined in your grammar on: free_input_string\n"
" text_chunk_end Abstract Machine Language"
"\n"
"2. Duplicated master scope name defined in your grammar on: free_input_string\n"
" text_chunk_end source.sma"
)
unittest.main(failfast=True)
使用python3 test.py运行它，您会看到第一个错误差异行没有被包裹：
预期结果将是：
我尝试搜索替代的diff库，然后尝试使用自定义diff库替换unittest diff作为内置difflib，但是diff的结果相同。因此，我假设unittest软件包正在使用difflib。
import unittest
import difflib
class TestSemanticRules(unittest.TestCase):
maxDiff = None
def myAssertEquals(self, expected, actual):
expected = expected.splitlines( 1 )
actual = actual.splitlines( 1 )
if expected != actual:
diff = difflib.context_diff( expected, actual, fromfile='expected input', tofile='actual output', lineterm='\n' )
self.fail( '\n' + ''.join( diff ) )
def test_badWrapping(self):
self.myAssertEquals(
"1. Duplicated target language name defined in your grammar on: [@-1,63:87='Abstract Machine Language'<__ANON_3>,3:19]\n"
"2. Duplicated master scope name defined in your grammar on: [@-1,138:147='source.sma'<__ANON_3>,5:20]"
,
"1. Duplicated target language name defined in your grammar on: free_input_string\n"
" text_chunk_end Abstract Machine Language"
"\n"
"2. Duplicated master scope name defined in your grammar on: free_input_string\n"
" text_chunk_end source.sma"
)
是否可以配置difflib软件包使用的unittest内置库，所以不会发生这种情况吗？还是difflib软件包有可靠的替代方案？
2 个答案:
答案 0 :(得分：1)
搜索difflib的替代项，我得到3个结果：
- https://github.com/Carreau/difflib2.py（4年无更新）
- https://github.com/google/diff-match-patch
- https://github.com/seperman/deepdiff
然后，使用diff-match-patch，我设法构建以下代码：
import re
import unittest
import textwrap
import diff_match_patch
class DiffMatchPatch(diff_match_patch.diff_match_patch):
def diff_prettyText(self, diffs):
"""Convert a diff array into a pretty Text report.
Args:
diffs: Array of diff tuples.
Returns:
Text representation.
"""
results_diff = []
cut_next_new_line = [False]
# print('\ndiffs:\n%s\n' % diffs)
operations = (self.DIFF_INSERT, self.DIFF_DELETE)
def parse(sign):
# print('new1:', text.encode( 'ascii' ))
if text:
new = text
else:
return ''
new = textwrap.indent( "%s" % new, sign, lambda line: True )
# force the diff change to show up on a new line for highlighting
if len(results_diff) > 0:
new = '\n' + new
if new[-1] == '\n':
if op == self.DIFF_INSERT and next_text and new[-1] == '\n' and next_text[0] == '\n':
cut_next_new_line[0] = True;
# Avoids a double plus sign showing up when the diff has the element (1, '\n')
if len(text) > 1: new = new + '%s\n' % sign
elif next_op not in operations and next_text and next_text[0] != '\n':
new = new + '\n'
# print('new2:', new.encode( 'ascii' ))
return new
for index in range(len(diffs)):
op, text = diffs[index]
if index < len(diffs) - 1:
next_op, next_text = diffs[index+1]
else:
next_op, next_text = (0, "")
if op == self.DIFF_INSERT:
results_diff.append( parse( "+ " ) )
elif op == self.DIFF_DELETE:
results_diff.append( parse( "- " ) )
elif op == self.DIFF_EQUAL:
# print('new3:', text.encode( 'ascii' ))
text = textwrap.indent(text, " ")
if cut_next_new_line[0]:
cut_next_new_line[0] = False
text = text[1:]
results_diff.append(text)
# print('new4:', text.encode( 'ascii' ))
return "".join(results_diff)
def diff_linesToWords(self, text1, text2, delimiter=re.compile('\n')):
"""
Split two texts into an array of strings. Reduce the texts to a string
of hashes where each Unicode character represents one line.
95% of this function code is copied from `diff_linesToChars` on:
https://github.com/google/diff-match-patch/blob/895a9512bbcee0ac5a8ffcee36062c8a79f5dcda/python3/diff_match_patch.py#L381
Copyright 2018 The diff-match-patch Authors.
https://github.com/google/diff-match-patch
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Args:
text1: First string.
text2: Second string.
delimiter: a re.compile() expression for the word delimiter type
Returns:
Three element tuple, containing the encoded text1, the encoded text2 and
the array of unique strings. The zeroth element of the array of unique
strings is intentionally blank.
"""
lineArray = [] # e.g. lineArray[4] == "Hello\n"
lineHash = {} # e.g. lineHash["Hello\n"] == 4
# "\x00" is a valid character, but various debuggers don't like it.
# So we'll insert a junk entry to avoid generating a null character.
lineArray.append('')
def diff_linesToCharsMunge(text):
"""Split a text into an array of strings. Reduce the texts to a string
of hashes where each Unicode character represents one line.
Modifies linearray and linehash through being a closure.
Args:
text: String to encode.
Returns:
Encoded string.
"""
chars = []
# Walk the text, pulling out a substring for each line.
# text.split('\n') would would temporarily double our memory footprint.
# Modifying text would create many large strings to garbage collect.
lineStart = 0
lineEnd = -1
while lineEnd < len(text) - 1:
lineEnd = delimiter.search(text, lineStart)
if lineEnd:
lineEnd = lineEnd.start()
else:
lineEnd = len(text) - 1
line = text[lineStart:lineEnd + 1]
if line in lineHash:
chars.append(chr(lineHash[line]))
else:
if len(lineArray) == maxLines:
# Bail out at 1114111 because chr(1114112) throws.
line = text[lineStart:]
lineEnd = len(text)
lineArray.append(line)
lineHash[line] = len(lineArray) - 1
chars.append(chr(len(lineArray) - 1))
lineStart = lineEnd + 1
return "".join(chars)
# Allocate 2/3rds of the space for text1, the rest for text2.
maxLines = 666666
chars1 = diff_linesToCharsMunge(text1)
maxLines = 1114111
chars2 = diff_linesToCharsMunge(text2)
return (chars1, chars2, lineArray)
class TestRules(unittest.TestCase):
## Set the maximum size of the assertion error message when Unit Test fail
maxDiff = None
## Whether `characters diff=0`, `words diff=1` or `lines diff=2` will be used
diffMode = 1
def __init__(self, *args, **kwargs):
diffMode = kwargs.pop('diffMode', -1)
if diffMode > -1: self.diffMode = diffMode
super(TestRules, self).__init__(*args, **kwargs)
def setUp(self):
if diff_match_patch: self.addTypeEqualityFunc(str, self.myAssertEqual)
def myAssertEqual(self, expected, actual, msg=""):
"""
How to wrap correctly the unit testing diff?
https://stackoverflow.com/questions/52682351/how-to-wrap-correctly-the-unit-testing-diff
"""
# print( '\n\nexpected\n%s' % expected )
# print( '\n\nactual\n%s' % actual )
if expected != actual:
diff_match = DiffMatchPatch()
if self.diffMode == 0:
diffs = diff_match.diff_main(expected, actual)
else:
diff_struct = diff_match.diff_linesToWords(expected, actual,
re.compile(r'\b') if self.diffMode == 1 else re.compile(r'\n') )
lineText1 = diff_struct[0] # .chars1;
lineText2 = diff_struct[1] # .chars2;
lineArray = diff_struct[2] # .lineArray;
diffs = diff_match.diff_main(lineText1, lineText2, False);
diff_match.diff_charsToLines(diffs, lineArray);
diff_match.diff_cleanupSemantic(diffs)
if msg:
msg += '\n'
else:
msg = "The strings does not match...\n"
self.fail( msg + diff_match.diff_prettyText(diffs) )
def test_characthersDiffModeExample1(self):
self.diffMode = 0
expected = "1. Duplicated target language name defined in your grammar on: [@-1,63:87='Abstract Machine Language'<__ANON_3>,3:19]\n" \
"2. Duplicated master scope name defined in your grammar on: [@-1,138:147='source.sma'<__ANON_3>,5:20]"
actual = "1. Duplicated target language name defined in your grammar on: free_input_string\n" \
" text_chunk_end Abstract Machine Language\n" \
"\n" \
"2. Duplicated master scope name defined in your grammar on: free_input_string\n" \
" text_chunk_end source.sma" \
with self.assertRaises( AssertionError ) as error:
self.myAssertEqual( expected, actual )
print( '\nerror.exception\n%s\n' % str(error.exception) )
self.assertEqual(
"The strings does not match...\n"
" 1. Duplicated target language name defined in your grammar on: \n"
"- [@-1,63:87='\n"
"+ free_input_string\n"
"+ text_chunk_end \n"
" Abstract Machine Language\n"
"- '<__ANON_3>,3:19]\n"
"+ \n"
" 2. Duplicated master scope name defined in your grammar on: \n"
"- [@-1,138:147='\n"
"+ free_input_string\n"
"+ text_chunk_end \n"
" source.sma\n"
"- '<__ANON_3>,5:20]"
, str(error.exception) )
def test_wordsDiffModeExample1(self):
self.diffMode = 1
expected = "1. Duplicated target language name defined in your grammar on: [@-1,63:87='Abstract Machine Language'<__ANON_3>,3:19]\n" \
"2. Duplicated master scope name defined in your grammar on: [@-1,138:147='source.sma'<__ANON_3>,5:20]"
actual = "1. Duplicated target language name defined in your grammar on: free_input_string\n" \
" text_chunk_end Abstract Machine Language\n" \
"\n" \
"2. Duplicated master scope name defined in your grammar on: free_input_string\n" \
" text_chunk_end source.sma" \
with self.assertRaises( AssertionError ) as error:
self.myAssertEqual( expected, actual )
print( '\nerror.exception\n%s\n' % str(error.exception) )
self.assertEqual(
"The strings does not match...\n"
" 1. Duplicated target language name defined in your grammar on: \n"
"- [@-1,63:87='Abstract Machine Language'<__ANON_3>,3:19]\n"
"+ free_input_string\n"
"+ text_chunk_end Abstract Machine Language\n"
"+ \n"
" 2. Duplicated master scope name defined in your grammar on: \n"
"- [@-1,138:147='source.sma'<__ANON_3>,5:20]\n"
"+ free_input_string\n"
"+ text_chunk_end source.sma"
, str(error.exception) )
def test_linesDiffModeExample1(self):
self.diffMode = 2
expected = "1. Duplicated target language name defined in your grammar on: [@-1,63:87='Abstract Machine Language'<__ANON_3>,3:19]\n" \
"2. Duplicated master scope name defined in your grammar on: [@-1,138:147='source.sma'<__ANON_3>,5:20]"
actual = "1. Duplicated target language name defined in your grammar on: free_input_string\n" \
" text_chunk_end Abstract Machine Language\n" \
"\n" \
"2. Duplicated master scope name defined in your grammar on: free_input_string\n" \
" text_chunk_end source.sma" \
with self.assertRaises( AssertionError ) as error:
self.myAssertEqual( expected, actual )
print( '\nerror.exception\n%s\n' % str(error.exception) )
self.assertEqual(
"The strings does not match...\n"
"- 1. Duplicated target language name defined in your grammar on: [@-1,63:87='Abstract Machine Language'<__ANON_3>,3:19]\n"
"- 2. Duplicated master scope name defined in your grammar on: [@-1,138:147='source.sma'<__ANON_3>,5:20]\n"
"+ 1. Duplicated target language name defined in your grammar on: free_input_string\n"
"+ text_chunk_end Abstract Machine Language\n"
"+ \n"
"+ 2. Duplicated master scope name defined in your grammar on: free_input_string\n"
"+ text_chunk_end source.sma"
, str(error.exception) )
unittest.main(failfast=True, verbosity=2)
使用diffMode=0作为字符
使用diffMode=1作为单词
使用diffMode=2作为行
似乎比unittest模块中的内置行为要好。新的diff_prettyText()仍将如何改进？
参考
- Comparison of multi-line strings in Python unit test
-
PyCharm show full diff when unittest fails for multiline string?
-
How to print the comparison of two multiline strings in unified diff format?
- python difflib comparing files
- How to print the 2 full objects instead of show diff on a python unit test error?
答案 1 :(得分：0)
这是python上的错误，可以通过应用此修补程序进行修复：
- unittest assertEqual difference output foiled by newlines
- [{https://bugs.python.org/issue35687](The]单元测试模块diff缺少/忘记了/没有在+和？之前放置换行符。对于某些输入）
相关问题
最新问题
- 我写了这段代码，但我无法理解我的错误
- 我无法从一个代码实例的列表中删除 None 值，但我可以在另一个实例中。为什么它适用于一个细分市场而不适用于另一个细分市场？
- 是否有可能使 loadstring 不可能等于打印？卢阿
- java中的random.expovariate()
- Appscript 通过会议在 Google 日历中发送电子邮件和创建活动
- 为什么我的 Onclick 箭头功能在 React 中不起作用？
- 在此代码中是否有使用“this”的替代方法？
- 在 SQL Server 和 PostgreSQL 上查询，我如何从第一个表获得第二个表的可视化
- 每千个数字得到
- 更新了城市边界 KML 文件的来源？