From: ZyX <kp-pav@yandex.ru>
To: Eric Cook <llua@gmx.com>, "zsh-users@zsh.org" <zsh-users@zsh.org>
Subject: Re: Equivalent of set -- *(DN) in sh
Date: Mon, 19 Jan 2015 19:16:49 +0300 [thread overview]
Message-ID: <1178831421684209@web6h.yandex.ru> (raw)
In-Reply-To: <1136081421683325@web6h.yandex.ru>
19.01.2015, 19:02, "ZyX" <kp-pav@yandex.ru>:
> 19.01.2015, 18:55, "ZyX" <kp-pav@yandex.ru>:
>> 19.01.2015, 18:51, "ZyX" <kp-pav@yandex.ru>:
>>> 18.01.2015, 23:53, "Eric Cook" <llua@gmx.com>:
>>>> On 01/18/2015 01:28 PM, Nikolai Weibull wrote:
>>>>> Hi!
>>>>>
>>>>> Is there any way to get the equivalent of Zsh’s
>>>>>
>>>>> set -- *(DN)
>>>>>
>>>>> in sh? Most important here would be NULL_GLOB, as, by default, sh
>>>>> simply leaves the * if there are no files to match.
>>>>>
>>>>> Thanks!
>>>> match() {
>>>> test "$#" -gt 2 && return
>>>> test -e "$1" && return
>>>> return 1
>>>> }
>>>>
>>>> set --
>>>> for pat in '.[^.]*' '*'; do # *(DN) ignores . and ..
>>> `..foo` is a valid name, but it is being excluded. You need to add `'.??*'` to the list of patterns.
>> No, this may make duplicates. Then `'..?*'`.
>
> And you must replace `[^.]` with `[!.]`. mksh does not support `[^]` and treats this as `[\^.]`, but other shells I have (dash, ksh, zsh (in sh emulation mode), bash, busybox ash) are fine with both `[!.]` and `[^.]`.
>>>> if match $pat; then
>>>> set -- "$@" $pat
>>>> fi
>>>> done
>>>> unset pat
>>>>
>>>> test "$#" -gt 0 && printf '%s\n' "$@"
Also note that if you need *full* equivalent of *(DN) you need to do something with ordering. I have checked all the shells with
for sh in dash ksh mksh zsh bash bb ; do $sh set--sDN.sh > set--sDN.$sh.log ; done
with
emulate -L sh &>/dev/null
match() {
test "$#" -gt 2 && return
test -e "$1" && return
return 1
}
set --
for pat in '.[!.]*' '..?*' '*'; do # *(DN) ignores . and ..
if match $pat; then
set -- "$@" $pat
fi
done
unset pat
test "$#" -gt 0 && printf '>>%s<<\n' "$@"
in set--sDN.sh file and have the following md5sums:
% md5sum set--sDN.*.log | sort
4601e6d9c845fe88e5166db3be04cc75 set--sDN.bash.log
4601e6d9c845fe88e5166db3be04cc75 set--sDN.ksh.log
4601e6d9c845fe88e5166db3be04cc75 set--sDN.zsh.log
ac2c245cdac664121a348c6b05ea2f44 set--sDN.bb.log
ac2c245cdac664121a348c6b05ea2f44 set--sDN.dash.log
ac2c245cdac664121a348c6b05ea2f44 set--sDN.mksh.log
: two variants of ordering:
bash/ksh/zsh:
>>.abc<<
>>.b<<
>>.^foo<<
>>..foo<<
>>
<<
>>1<<
>>17<<
>>2<<
>>append-remove.bash<<
>>getnumbers.bash<<
>>grepdir<<
>>nroff<<
>>parinput.bash<<
>>qwerty<<
>>rl.sh<<
>>script_path.bash<<
>>set--sDN.bash.log<<
>>set--sDN.bb.log<<
>>set--sDN.dash.log<<
>>set--sDN.ksh.log<<
>>set--sDN.mksh.log<<
>>set--sDN.sh<<
>>set--sDN.zsh.log<<
>>tes.bash<<
>>test<<
>>test2.bash<<
>>test3.bash<<
>>test.bash<<
>>testdir<<
>>test-rename.sh<<
>>test-so-2914220.bash<<
>>t.tar<<
>>ttttt.tar.bz2<<
>>ttttt.tar.gz<<
bb/dash/mksh:
>>.^foo<<
>>.abc<<
>>.b<<
>>..foo<<
>>
<<
>>1<<
>>17<<
>>2<<
>>append-remove.bash<<
>>getnumbers.bash<<
>>grepdir<<
>>nroff<<
>>parinput.bash<<
>>qwerty<<
>>rl.sh<<
>>script_path.bash<<
>>set--sDN.bash.log<<
>>set--sDN.bb.log<<
>>set--sDN.dash.log<<
>>set--sDN.ksh.log<<
>>set--sDN.mksh.log<<
>>set--sDN.sh<<
>>set--sDN.zsh.log<<
>>t.tar<<
>>tes.bash<<
>>test<<
>>test-rename.sh<<
>>test-so-2914220.bash<<
>>test.bash<<
>>test2.bash<<
>>test3.bash<<
>>testdir<<
>>ttttt.tar.bz2<<
>>ttttt.tar.gz<<
(note position of t.tar).
I have verified that all tests emit essentially the same result by using `for f in set--sDN.*.log ; do cat $f | sort | md5sum ; done | uniq`. But both orderings are different from the zsh one:
% zsh -f -c 'for f in *(DN); do echo ">>$f<<" ; done'
>>
<<
>>1<<
>>17<<
>>2<<
>>.abc<<
>>append-remove.bash<<
>>.b<<
>>.^foo<<
>>..foo<<
>>getnumbers.bash<<
>>grepdir<<
>>nroff<<
>>parinput.bash<<
>>qwerty<<
>>rl.sh<<
>>script_path.bash<<
>>set--sDN.bash.log<<
>>set--sDN.bb.log<<
>>set--sDN.dash.log<<
>>set--sDN.ksh.log<<
>>set--sDN.mksh.log<<
>>set--sDN.sh<<
>>set--sDN.zsh.log<<
>>tes.bash<<
>>test<<
>>test2.bash<<
>>test3.bash<<
>>test.bash<<
>>testdir<<
>>test-rename.sh<<
>>test-so-2914220.bash<<
>>t.tar<<
>>ttttt.tar.bz2<<
>>ttttt.tar.gz<<
(note: I have filename with newline, this is why I added `>><<`.)
next prev parent reply other threads:[~2015-01-19 16:27 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-01-18 18:28 Nikolai Weibull
2015-01-18 19:07 ` Roman Neuhauser
2015-01-18 19:43 ` Nikolai Weibull
2015-01-18 20:32 ` Roman Neuhauser
2015-01-18 19:31 ` ZyX
2015-01-18 19:46 ` Nikolai Weibull
2015-01-18 20:46 ` Eric Cook
2015-01-19 15:51 ` ZyX
2015-01-19 15:55 ` ZyX
2015-01-19 16:02 ` ZyX
2015-01-19 16:16 ` ZyX [this message]
2015-01-19 21:44 ` Eric Cook
[not found] ` <54BD7ABB.5070501__36205.2317861982$1421704010$gmane$org@gmx.com>
2015-01-19 23:05 ` Stephane Chazelas
2015-01-22 7:44 ` Nikolai Weibull
[not found] ` <CADdV=MsvuSAQMJVsr17Y7g2Nfjy95CQ007opQqv-7=RHjgjaKw__35342.9068615243$1421913204$gmane$org@mail.gmail.com>
2015-01-22 15:21 ` Stephane Chazelas
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1178831421684209@web6h.yandex.ru \
--to=kp-pav@yandex.ru \
--cc=llua@gmx.com \
--cc=zsh-users@zsh.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://git.vuxu.org/mirror/zsh/
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).