工作bash命令在函数或别名中中断

Question

当我以交互方式运行此命令时，它会输出预期的日志名称(上次修改，按服务名称文件(。

# Gets last active service log in a /environmentsDir/serviceName/var/output/logs directory. You need to cd 1st.
ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}'

当我试图为别名表达式或函数转义它时，无法正确执行，由于不同的原因总是失败。

在CLI中，为了使函数或别名正常工作，正确的转义方法是什么？

例如，如果我想跟踪日志文件，函数或别名应该允许我：

lastLog=escaped(ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}')
tail -f $(lastLog)

编辑：

添加尝试功能的输出：

21:55:28-user@host:/environmentsDir/env/serviceName/var/output/logs$ ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}'
serviceName.2022-04-12-21
21:55:33-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog() {
logbash: syntax error near unexpected token `('
21:55:42-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog() { 
logbash: syntax error near unexpected token `('
21:55:46-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog { 
> ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}'
head: invalid option -- 't'
Try 'head --help' for more information.
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.
21:56:10-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog() { ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}'; }
logbash: syntax error near unexpected token `('
21:56:41-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog() { ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}'; }^C
21:56:56-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog() {ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}' ; }
logbash: syntax error near unexpected token `('
21:57:14-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog { ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}' ;  }
logbash: syntax error near unexpected token `}'
# This one doesn't blow up but still fails when executed:
21:57:35-user@host:/environmentsDir/env/serviceName/var/output/logs$ function lastLog { ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}' ;  }
21:57:47-user@host:/environmentsDir/env/serviceName/var/output/logs$ lastLog 
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.

第二次编辑：

@约翰·库格尔曼的答案是正确的，我已经试过了，但没有成功。正如@gordon davisson在下面的评论中指出的那样，添加的问题是，我添加了一个与函数同名的别名，该别名的优先级越来越高，也被破坏了。

Answer 1

函数的好处在于它们不需要任何特殊的转义。只需将命令完全按照编写在函数中的方式放入即可：

lastLog() {
ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}'
}

您可以在交互式shell中键入多行函数定义，就像这样。但是，如果你想把它作为一行，它应该是这样的，在结束大括号之前在命令后面附加一个分号：

lastLog() { ls -tl | head -n20 | grep $(basename $(pwd | egrep '/environmentsDir/env/[^/]+' -o)) | head -n1 | awk '{print $(NF)}'; }

Answer 2

您的函数可以使用shell的参数扩展来简化：

lastLog() {
local dirname=${PWD#/environmentDir/env/}     # remove the prefix
dirname=${dirname%%/*}                        # remove the suffix
ls -tl *"$dirname"* | head -n1
}