Question

我想使用grep删除以]:结尾的文件中的所有行，以便下一行以I开头或达到EOF。

即：删除所有以]: 结尾的行，然后后跟一行以文件结尾I开头。

鉴于以下内容：

I am an invalid line ]:
I am a valid line ]:
some text
some text
some text
I am a valid line ]:
some text
I am an invalid line ]:
I am a valid line ]:
some text
some text
some text

输出应为

I am a valid line ]:
some text
some text
some text
I am a valid line ]:
some text
I am a valid line ]:
some text
some text
some text

请注意第1行和第8行I am in invalid line ]:已删除，因为它以]:结尾，下一行以I开头

Answer 1

grep -v 'invalid' file.txt

应该这样做。使用-v，您要求选择与该条件不匹配的行。

示例：

$ cat ~/test.txt
I am an invalid line ]:
I am a valid line ]:
some text
some text
some text
I am a valid line ]:
some text
I am an invalid line ]:
I am a valid line ]:
some text
some text
some text

$ grep -v 'invalid' ~/test.txt
I am a valid line ]:
some text
some text
some text
I am a valid line ]:
some text
I am a valid line ]:
some text
some text
some text

根据问题更改进行更新

sed ':a;N;$!ba;s/]:\nI/]:xxx\nI/g' ~/test.txt | grep -v 'xxx'

使用此方法，我将]:\nI替换为]:xxx\nI。看看我如何在以]:结尾的行后面注入xxx - 换行符-I？然后我grep并删除那些xxx

的行

Answer 2

这个awk应该适合你的要求：

awk '/\]:$/{pno=NR;prec=$0;next} pno && !(/^I/ && NR==pno+1){print prec; pno=0} 1' file

<强>输出：

I am a valid line ]:
some text
some text
some text
I am a valid line ]:
some text
I am a valid line ]:
some text
some text
some text

<强>解体：

/\]:$/{pno=NR;prec=$0;next}  # if ]: is at line end store line # and line and move to next
pno && !(/^I/ && NR==pno+1)  # if next row starts with I and we have prev vars set
{print prec; pno=0}          # print previous line and rest prev line # to 0
1                            # default action to print current record

Answer 3

所有其他答案对我来说似乎有些过分，所以也许我错过了一些东西，但是：

sed 'N;/]:\nI/D'

如何根据条件删除grep行？

3 个答案: