bash 如何将“watch”命令添加到 shell 脚本？
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/7128554/
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
How do I add `watch' command to a shell script?
提问by Kevin
I have a script that greps and awks and counts for ip addresses with multiple http status codes. It doesn't really matter what the script is though, because my goal is as follows:
我有一个脚本，可以使用多个 http 状态代码对 ip 地址进行 grep 和 awks 计数。不过，脚本是什么并不重要，因为我的目标如下：
I want to invoke the `watch' command inside this script so that it will update the display every few seconds with new data received from the dynamic apache logs.
我想在这个脚本中调用‘watch’命令，这样它就会每隔几秒用从动态 apache 日志接收到的新数据更新显示。
i can call my script fine as follows:
我可以按如下方式调用我的脚本：
$ watch --no-title ./bad_request.sh
But I would much rather have the `watch' command inside the script itself, to make calling the script from the command prompt much cleaner.
但是我更愿意在脚本本身中使用“watch”命令，以使从命令提示符调用脚本更加清晰。
Here is the script:
这是脚本：
#!/bin/bash
#identify repeated offenders through 401 and 403 status codes
for ip in `awk '{print }' /var/log/apache2/* | sort -u`; do echo -n "$ip "; awk '{print " " }' /var/log/apache2/* | egrep "( 401| 403)"| grep -c $ip; done | sort -k2 -r
Now, I have tried just inserting "watch -d --no-title" inside the script, right before the for loop, but it errors out angrily. I think it's because it is only processing until it reaches the end of the first command. I've tried putting backticks, and $() around the entire script, as well. I've also tried making the bulk of the script a bash function, and calling watch on the function. No dice.
现在，我尝试在脚本中插入“watch -d --no-title”，就在 for 循环之前，但它愤怒地出错了。我认为这是因为它只在到达第一个命令的末尾时才进行处理。我也试过在整个脚本周围加上反引号和 $() 。我还尝试将大部分脚本设为 bash 函数，并在该函数上调用 watch。没有骰子。
Any ideas?
有任何想法吗？
By the way, I'm also open to improvements on the script - I do realize it's a bit redundant/inefficient. Of course, that should probably be reserved for a different Stack Overflow question.
顺便说一句，我也愿意改进脚本 - 我确实意识到它有点多余/效率低下。当然，这可能应该保留给不同的 Stack Overflow 问题。
Thanks,
谢谢，
Kevin
凯文
EDIT: And one more thing, I can just call while true; do <bulk of script>; sleep 1; clear;but I hate that. It works, but the screen flickers, and it's just not the right way to do this.
编辑：还有一件事，我可以打电话，while true; do <bulk of script>; sleep 1; clear;但我讨厌那样。它可以工作，但屏幕闪烁，这不是正确的方法。
EDIT 2: Another nasty hack that works, is to simply create two scripts. The first one is:
编辑 2：另一个讨厌的 hack 是简单地创建两个脚本。第一个是：
watch --no-title ./bad_request
And then just call that script. That works fine, but there has to be a better way.
然后只需调用该脚本。这很好用，但必须有更好的方法。
EDIT 3 (sorry...): I took the -d flag off of `watch'. It's not necessary for my script.
编辑 3（抱歉...）：我从“watch”中去掉了 -d 标志。我的脚本不需要。
采纳答案by holygeek
Heed the unix philosophy:
注意 Unix 哲学：
A program should do one and only one thing and do it well.
Applied to your case:
适用于您的案例：
A script should do one and only one thing and do it well.
Your "nasty hack" is actually a step in the right direction.
你的“讨厌的黑客”实际上是朝着正确方向迈出的一步。
You'll never know if one day you'll need to "watch --no-title" on another script. At that point if you follow the unix philosophy you'd already have a script to do that.
您永远不会知道有一天您是否需要在另一个脚本上“观看 --no-title”。到那时，如果您遵循 unix 哲学，那么您已经有一个脚本可以做到这一点。
You already see one complication of trying to make a script do too many things at once - quote hell.
您已经看到尝试让脚本一次做太多事情的一种复杂性 - 引用地狱。
Keep it simple.
把事情简单化。
回答by Thomas Berger
the correct usage would be:
正确的用法是：
watch -d --no-title "/bin/bash for ip in `awk '{print }' /var/log/apache2/* | sort -u`; do echo -n "$ip "; awk '{print " " }' /var/log/apache2/* | egrep '( 401| 403)'| grep -c $ip; done | sort -k2 -r"
