NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/comp/unix/shell/5467

Text File | 1993-01-23 | 1.7 KB | 54 lines

Newsgroups: comp.unix.shell
Path: sparky!uunet!zaphod.mps.ohio-state.edu!pacific.mps.ohio-state.edu!linac!att!mcdchg!laidbak!katzung
From: [e-mail] (Brian Katzung)
Subject: Re: Is there a "foreach" loop control in ksh?
Message-ID: <[e-mail]>
Sender: [e-mail] (Usenet News)
Nntp-Posting-Host: laitnite.i88.isc.com
Organization: Interactive Systems Corporation, Naperville, IL
References: <[e-mail]>
Date: Fri, 22 Jan 1993 17:37:54 GMT
Lines: 41

In article <[e-mail]> [e-mail] (Chaim Manaster) writes:
|I want to do a series of comands on each file in the current
|directory, with some sort of "foreach" construct that will loop
|over every file in some list (something like in perl), or in some
|cases every file in the current directory. How would one do this in
|ksh?

For every file except, regardless of type, try:

    for file in `find . -print -prune`
    do
        ...
    done

For regular files, use:

    for file in `find ..?* .[!.]* * -prune -type f -print 2> /dev/null`
    do
        ...
    done

You can also use the one above without the "-type f" if you want
all types of files, but don't want ./ in front of each one.  Just
omit the "." patterns if they are excessive for your needs.

|In particular, I want to run compress on each file in the directory
|that does not already end in a .Z

    find ..?* .[!.]* * -prune ! -type f -name '*.Z' -print 2> /dev/null |
    xargs compress

|(dito for uncompress in
|reverse).

    find ..?* .[!.]* * -prune -type f -name '*.Z' -print 2> /dev/null |
    xargs uncompress

There is nothing ksh-specific here BTW, so these should work in sh just
as well.

  -- Brian Katzung  [e-mail]