以下python 2.7示例返回string1和string2之间的匹配块:
import difflib
string1 = "This is a test"
string2 = "This ain't a testament"
s = difflib.SequenceMatcher(lambda x: x == " ", string1, string2)
for block in s.get_matching_blocks():
a,b,size = block
print "string1[%s] and string2[%s] match for %s characters" % block
以下是上述程序的结果:
string1[0] and string2[0] match for 5 characters
string1[5] and string2[6] match for 1 characters
string1[7] and string2[10] match for 7 characters
string1[14] and string2[22] match for 0 characters
我想反转结果并返回string1和string2的不匹配块,如下所示:
string1[6] mismatch for 1 characters
string2[5] mismatch for 1 characters
string2[7] mismatch for 3 characters
string2[17] mismatch for 5 characters
注意:两个字符串的匹配块总数相同,但是不匹配的块将根据字符串而有所不同。
这里是字符串的颜色编码表示,其中black = matched和red = mismatched。
答案 0 :(得分:0)
在我看来,应该可以通过匹配的块来计算不匹配的部分。下面粘贴了一个快速解决方案(称为“仅使用问题中的输入进行了测试”)。看看它是否可以帮助您制定最终的解决方案。
注意:我现在只能访问Python3解释器,但是由于此问题不是特定于版本的,因此我将发布此解决方案。
import difflib
string1 = "This is a test"
string2 = "This ain't a testament"
s = difflib.SequenceMatcher(lambda x: x == " ", string1, string2)
s1_miss = list()
s2_miss = list()
s1_cur_off = 0
s2_cur_off = 0
for block in s.get_matching_blocks():
a,b,size = block
print("string1[%s] and string2[%s] match for %s characters" % block)
if a > s1_cur_off:
s1_miss.append((s1_cur_off, a-1, a-1-s1_cur_off + 1))
s1_cur_off = a + size
if b > s2_cur_off:
s2_miss.append((s2_cur_off, b-1, b-1-s2_cur_off + 1))
s2_cur_off = b + size
print(s1_miss)
print(s2_miss)
输出: 将为每个字符串转储不匹配的列表。列表中的每个元素都有三元组:不匹配的起始偏移量和结束偏移量以及长度(主要用于调试)。
string1[0] and string2[0] match for 5 characters
string1[5] and string2[6] match for 1 characters
string1[7] and string2[10] match for 7 characters
string1[14] and string2[22] match for 0 characters
[(6, 6, 1)]
[(5, 5, 1), (7, 9, 3), (17, 21, 5)]