Bash - 计算字符串中的子字符串
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/26212889/
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
Bash - Counting substrings in a string
提问by choroba
I am kinda struggling to write a command which counts how many times a sub-string appears in a string.
我有点努力编写一个命令来计算子字符串在字符串中出现的次数。
Instead of running the code below 10 times, I would rather previously count how many times the sub-string appears and adapt the "for" based on its result:
与其将代码运行 10 次以下，我宁愿先计算子字符串出现的次数并根据其结果调整“for”：
Here you can see the code:
在这里你可以看到代码：
CommandResult="Interface Chipset Driver mon0 Unknown iwlwifi - [phy0]wlan0 Unknown iwlwifi - [phy0]"
for i in `seq 0 9`;
do
InstanceID="mon"$i
if echo "$CommandResult" | grep -q "$InstanceID"; then
echo "found"
fi
done
Any help would be appreciated!
任何帮助，将不胜感激！
Thanks,
谢谢，
回答by Cyrus
Feel free to try this to get the number:
请随意尝试以获取号码：
echo "$CommandResult" | tr " " "\n" | grep -c "$InstanceID"
回答by choroba
I'd use grep -oto extract the desired string from the output:
我会用来grep -o从输出中提取所需的字符串：
#!/bin/bash
CommandResult="Interface Chipset Driver mon0 Unknown iwlwifi - [phy0]wlan0 Unknown iwlwifi - [phy0]
Interface Chipset Driver mon12 Unknown iwlwifi - [phy0]wlan0 Unknown iwlwifi - [phy0]"
for InstanceId in $(grep -o 'mon[0-9]\+' <<< "$CommandResult") ; do
echo "found $InstanceId "$(grep -c "$InstanceId" <<< "$CommandResult")' times'
done
回答by Cyclonecode
You could do like this:
你可以这样做：
#!/bin/bash
CommandResult="Interface Chipset Driver mon0 Unknown iwlwifi - [phy0]wlan0 Unknown iwlwifi - [phy0]"
InstanceId="mon0";
count=`grep -o "$InstanceId" <<< "$CommandResult" | wc -l`
echo "$InstanceId encountered "$count" times";
The above would produce an output like this:
以上将产生这样的输出：
mon0 encountered 1 times
mon0 遇到 1 次
The above could easily be expanded to take take a string as input:
上面的内容可以很容易地扩展为以字符串作为输入：
#!/bin/bash
CommandResult=
InstanceId="mon0";
count=`grep -o "$InstanceId" <<< "$CommandResult" | wc -l`
echo "$InstanceId encountered "$count" times";
Then you could call it like this:
那么你可以这样称呼它：
./script.sh "Interface chipset mon0 mon0 unknown . test"
or perhaps send the output from another command as an argument:
或者可能将另一个命令的输出作为参数发送：
./script.sh `cat file.txt`
of course xargswould also work:
当然xargs也可以：
cat script.txt | xargs ./script.sh
回答by TimoM
Wanted to share another solution. 3,5 years later, but better late than never I guess. ;)
想分享另一个解决方案。3.5 年后，但我猜迟到总比没有好。;)
I was working on a script that had to do the same, that is get the number of substrings in a string, and I was using count=$(echo '$string' | grep -o ... | wc -l)at first.
I got what I wanted, but when looping through ~1500 files with 0...8000 lines in each the performance was just terrible: the script took about 49 minutesto complete.
So, I went and searched for alternative approaches and eventually found this:
我正在编写一个必须执行相同操作的脚本，即获取字符串中子字符串的数量，并且我首先使用count=$(echo '$string' | grep -o ... | wc -l)。
我得到了我想要的东西，但是当循环遍历大约 1500 个文件时，每个文件都有 0...8000 行，性能非常糟糕：脚本花了大约49 分钟才能完成。
所以，我去寻找替代方法，最终找到了这个：
InstanceId="mon0";
tmp="${CommandResult//$InstanceId}"
count=$(((${#CommandResult} - ${#tmp}) / ${#InstanceId}))
Got the same result but waaaay quicker, in 8-9 minutes.
得到了相同的结果，但 waaaay 更快，在 8-9 分钟内。
Explained:
解释：
tmp="${CommandResult//$InstanceId}"
This removes all occurrences of $InstanceIdfrom $CommandResultand places it in tmp.
Actually we're using substring replacement here with replacement string missing. Syntax for substring replacement is ${string//substring/replacement}(this replaces all occurrences of substring with replacement).
这将删除所有出现的$InstanceIdfrom$CommandResult并将其放入tmp.
实际上，我们在这里使用子字符串替换，但缺少替换字符串。子字符串替换的语法是${string//substring/replacement}（这将用替换替换所有出现的子字符串）。
count=$(((${#CommandResult} - ${#tmp}) / ${#InstanceId}))
This gives us the number of occurrences of $InstanceIdin $CommandResult.${#string}gives string length so (${#CommandResult} - ${#tmp])is length of all occurrences of $InstanceId(remember, we removed all occurrences of $InstanceIdfrom $CommandResultand placed the result in $tmp).
Then we just divide the substraction with length of $InstanceIdto get the number of $InstanceIdoccurrences.
这给了我们$InstanceIdin的出现次数$CommandResult。${#string}给出字符串长度，所以(${#CommandResult} - ${#tmp])是所有出现的长度$InstanceId（请记住，我们删除了所有出现的$InstanceIdfrom$CommandResult并将结果放在 中$tmp）。
然后我们只需将减法除以长度$InstanceId即可得到$InstanceId出现次数。
For further info about substring replacement and string length, see e.g. https://www.tldp.org/LDP/abs/html/string-manipulation.html.
有关子字符串替换和字符串长度的更多信息，请参见例如https://www.tldp.org/LDP/abs/html/string-manipulation.html。
回答by jm666
you can use:
您可以使用：
mon_num=$(airmon-ng | grep -Poc '\bmon\d+\b')
echo here are $mon_num mon interfaces
for m in $(airmon-ng | grep -Po '\bmon\d+\b')
do
#do something
echo "this is $m"
done
for the airmon-ngoutput such:
对于这样的airmon-ng输出：
Interface Chipset Driver
mon0 Unknown iwlwifi - [phy0]
mon1 Unknown iwlwifi - [phy0]
mon2 Unknown iwlwifi - [phy0]
wlan0 Unknown iwlwifi - [phy0]
will print
将打印
here are 3 mon interfaces
this is mon0
this is mon1
this is mon2
