linux - 删除几乎没有差异的重复行

Question

我有一个动态 文本文件，可以自动写入一些行。但是重复的条目有问题:

例如:

1111 2222 3333 4444 <- I want this line
5555 6666 7777 8888 <- And this line too
1111 2222 3333 4444
5555 6666 7777 9999 <- Note : 9999 is only one ward change

预期结果:

1111 2222 3333 4444
5555 6666 7777 8888

---

真实测试示例:

exten => 01272786170,1,Set(CALLERID(num)=821)
    same => n,Dial(SIP/port21/01272786170,60,rt)
    same => n,Set(thereis=yes01272786170)
    same => n,Set(calledid=01272786170)
    same => n,GotoIf("calledid" = "01272786170"?ejoin,01272786170,1)
exten => 01272786170,1,Set(CALLERID(num)=826) <- duplicated here with one number change
    same => n,Dial(SIP/port26/01272786170,60,rt) <-
exten => 01272786170,1,Set(CALLERID(num)=827) <-
    same => n,Dial(SIP/port27/01272786170,60,rt) <-

预期结果:

exten => 01272786170,1,Set(CALLERID(num)=821)
    same => n,Dial(SIP/port21/01272786170,60,rt)
    same => n,Set(thereis=yes01272786170)
    same => n,Set(calledid=01272786170)
    same => n,GotoIf("calledid" = "01272786170"?ejoin,01272786170,1)

注意:我希望使用 Linux Shell 完成它。

非常感谢。

Answer 1

使用 awk 和您的第一个示例数据:

如果您实现 Levenshtein 算法(例如 here)并想出足够的编辑距离(下面的 4)，您可以使用这种简单的方法:

awk '
function levdist(str1, str2 ...)  # see the above link for working implementation
{
    ...
}
{
    for(i in a) {                 # iterate all previous stored strings
        l=levdist($0,a[i])        # compute the edit distance
        if(l<=4)                  # if below threshold
            next                  # skip to next string 
    }
    print $0                      # output where threshold was not met
    a[NR]=$0                      # store
}' file

输出:

1111 2222 3333 4444
5555 6666 7777 8888