bash 为bash中每行的每个单词添加前缀
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/38263274/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Add prefix to each word of each line in bash
提问by Christopher Loen
I have a variable called deps:
我有一个名为 deps 的变量：
deps='word1 word2'
I want to add a prefix to each word of the variable.
我想为变量的每个单词添加一个前缀。
I tried with:
我试过：
echo $deps | while read word do \ echo "prefix-$word" \ done
but i get:
但我得到：
bash: syntax error near unexpected token `done'
bash：意外标记“完成”附近的语法错误
any help? thanks
有什么帮助吗？谢谢
回答by SLePort
With sed :
使用 sed ：
$ deps='word1 word2'
$ echo "$deps" | sed 's/[^ ]* */prefix-&/g'
prefix-word1 prefix-word2
回答by Heinrich Hartmann
For well behaved strings, the best answer is:
对于表现良好的字符串，最佳答案是：
printf "prefix-%s\n" $deps
as suggested by 123 in the comments to fedorqui's answer.
正如 123 在对 fedorqui 的回答的评论中所建议的那样。
Explanation:
解释：
- Without quoting, bash will split the contents of
$depsaccording to$IFS(which defaults to" \n\t") before callingprintf printfevaluates the pattern for each of the provided arguments and writes the output to stdout.printfis a shell built-in (at least for bash) and does not fork another process, so this is faster than sed-based solutions.
- 如果不引用，bash 将在调用之前
$deps根据$IFS（默认为" \n\t"）拆分内容printf printf评估每个提供的参数的模式并将输出写入标准输出。printf是一个内置的 shell（至少对于 bash）并且不会分叉另一个进程，所以这比基于 sed 的解决方案更快。
回答by codeling
In another questionI just came across the markers for beginning (\<) and end (\>) of words. With those you can shorten the solution of SLePortabove somewhat. The solution also nicely extends to appending a suffix, which I needed in addition to the prefix, but couldn't figure out how to use above solution for it, as the &also includes the possible trailing whitespace after the word.
在另一个问题中，我刚刚遇到了单词开头 ( \<) 和结尾 ( \>) 的标记。有了这些你可以稍微缩短上面的SLePort的解决方案。该解决方案还很好地扩展到附加后缀，除了前缀之外我还需要它，但无法弄清楚如何使用上述解决方案，因为&它还包括单词后面可能的尾随空格。
So my solution is this:
所以我的解决方案是这样的：
$ deps='word1 word2'
# add prefix:
$ echo "$deps" | sed 's/\</prefix-/g'
prefix-word1 prefix-word2
# add suffix:
$ echo "$deps" | sed 's/\>/-suffix/g'
word1-suffix word2-suffix
Explanation:\<matches the beginning of every word, and \>matches the end of each word. You can simply "replace" these by the prefix/suffix, resulting in them being prepended/appended. There is no need to reference them anymore in the replacement, as these are not "real" characters anyway!
说明：\<匹配每个单词的开头，\>匹配每个单词的结尾。您可以简单地用前缀/后缀“替换”它们，从而使它们被前置/附加。无需在替换中再引用它们，因为无论如何这些都不是“真实”字符！
回答by fedorqui 'SO stop harming'
You can read the string into an array and then prepend the string to every item:
您可以将字符串读入数组，然后将字符串添加到每个项目中：
$ IFS=' ' read -r -a myarray <<< "word1 word2"
$ printf "%s\n" "${myarray[@]}"
word1
word2
$ printf "prefix-%s\n" "${myarray[@]}"
prefix-word1
prefix-word2
