Question

我需要在变量$word中包含<xml> </xml>和$line的所有实例，但不要在字母数字字符出现在之前或之后。例如，给定此行，如果$word设置为“fish”：

fish. The fish is a goldfish. Who likes fish?

这变为：

<xml>fish</xml>. The <xml>fish</xml> is a goldfish. Who likes <xml>fish</xml>?

这是我到目前为止用AWk尝试过的，但它不起作用：

echo "${line}" | awk '{
    gsub(/^|[^[:alnum:]]*($word)*$|[^[:alnum:]]*/, "<xml>&</xml>");
}1'

Answer 1

这个sed应该做的工作：

> s='fish. The fish is a goldfish. Who likes fish?'
> sed -r 's#(^|[^[:alnum:]])(fish)([^[:alnum:]]|$)#\1<xml>\2</xml>\3#g' <<< "$s"
<xml>fish</xml>. The <xml>fish</xml> is a goldfish. Who likes <xml>fish</xml>?

Answer 2

你可能正在寻找单词边界，似乎无法找到一种简单的方法来使用gawk，但这里是一个perl版本：

echo 'fish. The fish is a goldfish. Who likes fish?' | perl -p -e 's/\b(fish)\b/<xml>\1<\/xml>/g'
<xml>fish</xml>. The <xml>fish</xml> is a goldfish. Who likes <xml>fish</xml>?

Answer 3

echo "${line}" | sed s'@fish@<xml>fish</xml>@'g - My solution.

如何查找和替换AWK，但不能在前后字符是字母数字时？

3 个答案: