如何从包含许多相似模式的行中获取每个模式？(how to grep each pattern from a line consisting many alike patterns?)

如何从包含许多相似模式的行中获取每个模式？(how to grep each pattern from a line consisting many alike patterns?)

假设我有一行如下：

blah blah. Jon abc efg Don. blah blah. Jon afjfa kadkca Don.blah blah.

现在我想得到的模式是“Jon”作为第一个单词，“Don”作为最后一个单词，中间没有“Jon”或“Don”。这意味着我想：

Jon abc efg Don Jon afjfa kadkca Don

首先，我试过了

egrep -o 'Jon.*Don'

但它表明

Jon abc efg Don. blah blah. Jon afjfa kadkca Don

所以我试图从模式的中间排除“Don”

egrep -o 'Jon.*[^(Don)]Don'

但它也没有用。

suppose I have a line like the following one:

blah blah. Jon abc efg Don. blah blah. Jon afjfa kadkca Don.blah blah.

Now I want to get patterns with "Jon" as the first word and "Don" as the last one, with no "Jon" or "Don" in between.That means I want:

Jon abc efg Don Jon afjfa kadkca Don

Firstly, I tried

egrep -o 'Jon.*Don'

But it shows

Jon abc efg Don. blah blah. Jon afjfa kadkca Don

So I tried to exclude "Don" from middle of the pattern

egrep -o 'Jon.*[^(Don)]Don'

But it didn't work either.

最满意答案

你可以试试这个，

grep -oP 'Jon.*?Don' file

egrep是折旧的，与grep -E相同。 这不会支持非贪心量词。 您需要使用-P perl-regexp参数来实现非贪婪的形式。

You may try this,

grep -oP 'Jon.*?Don' file

egrep is depreciated which is same as grep -E . This won't support non-greedy quantifier. You need to use -P perl-regexp parameter to achieve the non-greedy form.