脚本提取字符串的一部分

Question

ns/APAC_BankTransfers_Publish/CMB/services/svcPublishBankTransfers/flow.xml

在上面的这一行和类似的行中，我想提取services和flow.xml之间存在的任何内容，并将其保存到变量DIST。

输出应为svcPublishBankTransfers。

Answer 1

使用POSIX中可用的参数扩展机制：

s=ns/APAC_BankTransfers_Publish/CMB/services/svcPublishBankTransfers/flow.xml
s=${s%/flow.xml}    # remove "/flow.xml"
s=${s##*/services/} # remove everything before "services"
echo "$s"

这样做的好处在于纯粹在进行中（比需要通过外部工具管道的方法更快），并且与所有POSIX外壳（灰，破折号，ksh等）兼容。

参考文献：

• BashFAQ #100 ("How do I do string manipulation in bash?")
• BashFAQ #73 ("How can I use parameter expansion? How can I get substrings?")

Answer 2

使用BASH正则表达式：

s='ns/APAC_BankTransfers_Publish/CMB/services/svcPublishBankTransfers/flow.xml'
[[ "$s" =~ /([^/]+)/[^/]*$ ]] && echo "${BASH_REMATCH[1]}"
svcPublishBankTransfers

或者其他：

[[ "$s" =~ /services/([^/]+)/flow\.xml ]] && echo "${BASH_REMATCH[1]}"
svcPublishBankTransfers

Answer 3

bash$ echo "ns/APAC_BankTransfers_Publish/CMB/services/svcPublishBankTransfers/flow.xml" | cut -d '/' -f 5
svcPublishBankTransfers
bash$