zsh/bash 上不区分大小写的 Glob
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/156916/
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
Case-insensitive Glob on zsh/bash
提问by Agnel Kurian
I need to list all files whose names start with 'SomeLongString'. But the case of 'SomeLongString' can vary. How?
我需要列出名称以“SomeLongString”开头的所有文件。但是“SomeLongString”的情况可能会有所不同。如何？
I am using zsh, but a bash solution is also welcome.
我正在使用 zsh，但也欢迎使用 bash 解决方案。
采纳答案by jkramer
ZSH:
ZSH：
$ unsetopt CASE_GLOB
Or, if you don't want to enable case-insensitive globbing in general, you can activate it for only the varying part:
或者，如果您不想启用不区分大小写的通配符，您可以只为不同的部分激活它：
$ print -l (#i)(somelongstring)*
This will match any file that starts with "somelongstring" (in any combination of lower/upper case). The case-insensitive flag applies for everything between the parentheses and can be used multiple times. Read the manual zshexpn(1)for more information.
这将匹配以“somelongstring”开头的任何文件（小写/大写的任意组合）。不区分大小写的标志适用于括号之间的所有内容，并且可以多次使用。阅读手册zshexpn(1)以获取更多信息。
UPDATEAlmost forgot, you have to enable extendend globbing for this to work:
更新几乎忘记了，您必须启用扩展通配符才能使其正常工作：
setopt extendedglob
回答by Jacek Szymański
bash:
重击：
shopt -s nocaseglob
回答by Horst Gutmann
Depending on how deep you want to have this listing, findoffers quite a lot
in this regard:
根据您希望此列表的深度，find在这方面提供了很多：
find . -iname 'SomeLongString*' -maxdepth 1
This will only give you the files in the current directory. Important here is
the -inameparameter instead of -name.
这只会为您提供当前目录中的文件。这里重要的是-iname参数而不是-name。
回答by Modern Hacker
$ function i () {
> shopt -s nocaseglob; $*; shopt -u nocaseglob
> }
$ ls *jtweet*
ls: cannot access *jtweet*: No such file or directory
$ i ls *jtweet*
JTweet.pm JTweet.pm~ JTweet2.pm JTweet2.pm~
回答by michael
For completeness (and frankly surprised it's not mentioned yet, even though all the other answers are better and/or "more correct"), obviously one can also use (especially for grepaficionados):
为了完整性（坦率地说，它尚未提及，即使所有其他答案都更好和/或“更正确”），显然也可以使用（特别是对于grep爱好者）：
$ ls | egrep -i '^SomeLongString'
One might also stick in a redundant ls -1(that's option "one", not "ell"), but when passed to a pipe, each entry is already going to be one per line, anyway. I'd typically use something like this (vs set) in shell scripts, eg in a for/whileloop: for i in $(ls | grep -i ...). However, the other answer using findwould be preferable & more flexible in that circumstance, because you can, for example, omit directories (or set other restrictions): for i in $(find . -type f -iname 'SomeString*' -print -maxdepth 1)...or even forgo the loop altogether and just use the power of findall by itself, eg: find ... -exec do_stuff {} \; ..., but I do digress (again, for completeness.)
一个人也可能坚持一个多余的ls -1（这是选项“一”，而不是“ell”），但是当传递到管道时，无论如何，每个条目已经是每行一个。我通常会使用这样的事情（VS setShell脚本），例如在for/while循环：for i in $(ls | grep -i ...)。但是，find在这种情况下，使用其他答案会更可取且更灵活，因为例如，您可以省略目录（或设置其他限制）：for i in $(find . -type f -iname 'SomeString*' -print -maxdepth 1)...或者甚至完全放弃循环并仅使用findall 本身的功能，例如：find ... -exec do_stuff {} \; ...，但我确实离题了（再次，为了完整性。）
