File: moreadv.html

package info (click to toggle)
abs-guide 10-4
links: PTS, VCS
area: main
in suites: bookworm, bullseye, forky, sid, trixie
size: 6,952 kB
sloc: sh: 14,129; makefile: 81
file content (1738 lines) | stat: -rw-r--r-- 33,954 bytes
parent folder | download | duplicates (3)
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<HTML
><HEAD
><TITLE
>Complex Commands</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.76b+
"><LINK
REL="HOME"
TITLE="Advanced Bash-Scripting Guide"
HREF="index.html"><LINK
REL="UP"
TITLE="External Filters, Programs and Commands"
HREF="external.html"><LINK
REL="PREVIOUS"
TITLE="External Filters, Programs and Commands"
HREF="external.html"><LINK
REL="NEXT"
TITLE="Time / Date Commands"
HREF="timedate.html"><META
HTTP-EQUIV="Content-Style-Type"
CONTENT="text/css"><LINK
REL="stylesheet"
HREF="common/kde-common.css"
TYPE="text/css"><META
HTTP-EQUIV="Content-Type"
CONTENT="text/html; charset=iso-8859-1"><META
HTTP-EQUIV="Content-Language"
CONTENT="en"><LINK
REL="stylesheet"
HREF="common/kde-localised.css"
TYPE="text/css"
TITLE="KDE-English"><LINK
REL="stylesheet"
HREF="common/kde-default.css"
TYPE="text/css"
TITLE="KDE-Default"></HEAD
><BODY
CLASS="SECT1"
BGCOLOR="#FFFFFF"
TEXT="#000000"
LINK="#AA0000"
VLINK="#AA0055"
ALINK="#AA0000"
STYLE="font-family: sans-serif;"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="3"
ALIGN="center"
>Advanced Bash-Scripting Guide: An in-depth exploration of the art of shell scripting</TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="bottom"
><A
HREF="external.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="80%"
ALIGN="center"
VALIGN="bottom"
>Chapter 16. External Filters, Programs and Commands</TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="bottom"
><A
HREF="timedate.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="MOREADV"
></A
>16.2. Complex Commands</H1
><DIV
CLASS="VARIABLELIST"
><P
><B
><A
NAME="CCLISTING1"
></A
>Commands for more advanced users</B
></P
><DL
><DT
><A
NAME="FINDREF"
></A
><B
CLASS="COMMAND"
>find</B
></DT
><DD
><P
><A
NAME="FINDREF0"
></A
></P
><P
>-exec <TT
CLASS="REPLACEABLE"
><I
>COMMAND</I
></TT
> \;</P
><P
>Carries out <TT
CLASS="REPLACEABLE"
><I
>COMMAND</I
></TT
> on
	      each file that <B
CLASS="COMMAND"
>find</B
> matches.  The
	      command sequence terminates with <SPAN
CLASS="TOKEN"
>;</SPAN
> (the
	      <SPAN
CLASS="QUOTE"
>";"</SPAN
> is <A
HREF="escapingsection.html#ESCP"
>escaped</A
> to
	      make certain the shell passes it to <B
CLASS="COMMAND"
>find</B
>
	      literally, without interpreting it as a special character).</P
><P
>	      <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>find ~/ -name '*.txt'</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>/home/bozo/.kde/share/apps/karm/karmdata.txt
 /home/bozo/misc/irmeyc.txt
 /home/bozo/test-scripts/1.txt</TT
>
 	      </PRE
></TD
></TR
></TABLE
>
	  </P
><P
><A
NAME="CURLYBRACKETSREF"
></A
></P
><P
>If <TT
CLASS="REPLACEABLE"
><I
>COMMAND</I
></TT
> contains
	      <SPAN
CLASS="TOKEN"
>{}</SPAN
>, then <B
CLASS="COMMAND"
>find</B
>
	      substitutes the full path name of the selected file for
	      <SPAN
CLASS="QUOTE"
>"{}"</SPAN
>.</P
><P
>          <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;find ~/ -name 'core*' -exec rm {} \;
   2&nbsp;# Removes all core dump files from user's home directory.</PRE
></TD
></TR
></TABLE
>
	  </P
><P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;find /home/bozo/projects -mtime -1
   2&nbsp;#                               ^   Note minus sign!
   3&nbsp;#  Lists all files in /home/bozo/projects directory tree
   4&nbsp;#+ that were modified within the last day (current_day - 1).
   5&nbsp;#
   6&nbsp;find /home/bozo/projects -mtime 1
   7&nbsp;#  Same as above, but modified *exactly* one day ago.
   8&nbsp;#
   9&nbsp;#  mtime = last modification time of the target file
  10&nbsp;#  ctime = last status change time (via 'chmod' or otherwise)
  11&nbsp;#  atime = last access time
  12&nbsp;
  13&nbsp;DIR=/home/bozo/junk_files
  14&nbsp;find "$DIR" -type f -atime +5 -exec rm {} \;
  15&nbsp;#                          ^           ^^
  16&nbsp;#  Curly brackets are placeholder for the path name output by "find."
  17&nbsp;#
  18&nbsp;#  Deletes all files in "/home/bozo/junk_files"
  19&nbsp;#+ that have not been accessed in *at least* 5 days (plus sign ... +5).
  20&nbsp;#
  21&nbsp;#  "-type filetype", where
  22&nbsp;#  f = regular file
  23&nbsp;#  d = directory
  24&nbsp;#  l = symbolic link, etc.
  25&nbsp;#
  26&nbsp;#  (The 'find' manpage and info page have complete option listings.)</PRE
></TD
></TR
></TABLE
>
          </P
><P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;find /etc -exec grep '[0-9][0-9]*[.][0-9][0-9]*[.][0-9][0-9]*[.][0-9][0-9]*' {} \;
   2&nbsp;
   3&nbsp;# Finds all IP addresses (xxx.xxx.xxx.xxx) in /etc directory files.
   4&nbsp;# There a few extraneous hits. Can they be filtered out?
   5&nbsp;
   6&nbsp;# Possibly by:
   7&nbsp;
   8&nbsp;find /etc -type f -exec cat '{}' \; | tr -c '.[:digit:]' '\n' \
   9&nbsp;| grep '^[^.][^.]*\.[^.][^.]*\.[^.][^.]*\.[^.][^.]*$'
  10&nbsp;#
  11&nbsp;#  [:digit:] is one of the character classes
  12&nbsp;#+ introduced with the POSIX 1003.2 standard. 
  13&nbsp;
  14&nbsp;# Thanks, Stphane Chazelas. </PRE
></TD
></TR
></TABLE
></P
><DIV
CLASS="NOTE"
><TABLE
CLASS="NOTE"
WIDTH="90%"
BORDER="0"
><TR
><TD
WIDTH="25"
ALIGN="CENTER"
VALIGN="TOP"
><IMG
SRC="common/note.png"
HSPACE="5"
ALT="Note"></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The <TT
CLASS="OPTION"
>-exec</TT
> option to
	      <B
CLASS="COMMAND"
>find</B
> should not be confused with the <A
HREF="internal.html#EXECREF"
>exec</A
> shell builtin.</P
></TD
></TR
></TABLE
></DIV
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="EX57"
></A
><P
><B
>Example 16-3. <I
CLASS="FIRSTTERM"
>Badname</I
>, eliminate file names
		in current directory containing bad characters and <A
HREF="special-chars.html#WHITESPACEREF"
>whitespace</A
>.</B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;# badname.sh
   3&nbsp;# Delete filenames in current directory containing bad characters.
   4&nbsp;
   5&nbsp;for filename in *
   6&nbsp;do
   7&nbsp;  badname=`echo "$filename" | sed -n /[\+\{\;\"\\\=\?~\(\)\&#60;\&#62;\&#38;\*\|\$]/p`
   8&nbsp;# badname=`echo "$filename" | sed -n '/[+{;"\=?~()&#60;&#62;&#38;*|$]/p'`  also works.
   9&nbsp;# Deletes files containing these nasties:     + { ; " \ = ? ~ ( ) &#60; &#62; &#38; * | $
  10&nbsp;#
  11&nbsp;  rm $badname 2&#62;/dev/null
  12&nbsp;#             ^^^^^^^^^^^ Error messages deep-sixed.
  13&nbsp;done
  14&nbsp;
  15&nbsp;# Now, take care of files containing all manner of whitespace.
  16&nbsp;find . -name "* *" -exec rm -f {} \;
  17&nbsp;# The path name of the file that _find_ finds replaces the "{}".
  18&nbsp;# The '\' ensures that the ';' is interpreted literally, as end of command.
  19&nbsp;
  20&nbsp;exit 0
  21&nbsp;
  22&nbsp;#---------------------------------------------------------------------
  23&nbsp;# Commands below this line will not execute because of _exit_ command.
  24&nbsp;
  25&nbsp;# An alternative to the above script:
  26&nbsp;find . -name '*[+{;"\\=?~()&#60;&#62;&#38;*|$ ]*' -maxdepth 0 \
  27&nbsp;-exec rm -f '{}' \;
  28&nbsp;#  The "-maxdepth 0" option ensures that _find_ will not search
  29&nbsp;#+ subdirectories below $PWD.
  30&nbsp;
  31&nbsp;# (Thanks, S.C.)</PRE
></TD
></TR
></TABLE
><HR></DIV
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="IDELETE"
></A
><P
><B
>Example 16-4. Deleting a file by its <I
CLASS="FIRSTTERM"
>inode</I
>
	        number</B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;# idelete.sh: Deleting a file by its inode number.
   3&nbsp;
   4&nbsp;#  This is useful when a filename starts with an illegal character,
   5&nbsp;#+ such as ? or -.
   6&nbsp;
   7&nbsp;ARGCOUNT=1                      # Filename arg must be passed to script.
   8&nbsp;E_WRONGARGS=70
   9&nbsp;E_FILE_NOT_EXIST=71
  10&nbsp;E_CHANGED_MIND=72
  11&nbsp;
  12&nbsp;if [ $# -ne "$ARGCOUNT" ]
  13&nbsp;then
  14&nbsp;  echo "Usage: `basename $0` filename"
  15&nbsp;  exit $E_WRONGARGS
  16&nbsp;fi  
  17&nbsp;
  18&nbsp;if [ ! -e "$1" ]
  19&nbsp;then
  20&nbsp;  echo "File \""$1"\" does not exist."
  21&nbsp;  exit $E_FILE_NOT_EXIST
  22&nbsp;fi  
  23&nbsp;
  24&nbsp;inum=`ls -i | grep "$1" | awk '{print $1}'`
  25&nbsp;# inum = inode (index node) number of file
  26&nbsp;# -----------------------------------------------------------------------
  27&nbsp;# Every file has an inode, a record that holds its physical address info.
  28&nbsp;# -----------------------------------------------------------------------
  29&nbsp;
  30&nbsp;echo; echo -n "Are you absolutely sure you want to delete \"$1\" (y/n)? "
  31&nbsp;# The '-v' option to 'rm' also asks this.
  32&nbsp;read answer
  33&nbsp;case "$answer" in
  34&nbsp;[nN]) echo "Changed your mind, huh?"
  35&nbsp;      exit $E_CHANGED_MIND
  36&nbsp;      ;;
  37&nbsp;*)    echo "Deleting file \"$1\".";;
  38&nbsp;esac
  39&nbsp;
  40&nbsp;find . -inum $inum -exec rm {} \;
  41&nbsp;#                           ^^
  42&nbsp;#        Curly brackets are placeholder
  43&nbsp;#+       for text output by "find."
  44&nbsp;echo "File "\"$1"\" deleted!"
  45&nbsp;
  46&nbsp;exit 0</PRE
></TD
></TR
></TABLE
><HR></DIV
><P
>The <B
CLASS="COMMAND"
>find</B
> command also works
	      without the <TT
CLASS="OPTION"
>-exec</TT
> option.</P
><P
>	    <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;#  Find suid root files.
   3&nbsp;#  A strange suid file might indicate a security hole,
   4&nbsp;#+ or even a system intrusion.
   5&nbsp;
   6&nbsp;directory="/usr/sbin"
   7&nbsp;# Might also try /sbin, /bin, /usr/bin, /usr/local/bin, etc.
   8&nbsp;permissions="+4000"  # suid root (dangerous!)
   9&nbsp;
  10&nbsp;
  11&nbsp;for file in $( find "$directory" -perm "$permissions" )
  12&nbsp;do
  13&nbsp;  ls -ltF --author "$file"
  14&nbsp;done</PRE
></TD
></TR
></TABLE
>
	    </P
><P
>See <A
HREF="filearchiv.html#EX48"
>Example 16-30</A
>, <A
HREF="special-chars.html#EX58"
>Example 3-4</A
>,
	      and <A
HREF="loops.html#FINDSTRING"
>Example 11-10</A
> for scripts using
	      <B
CLASS="COMMAND"
>find</B
>. Its <A
HREF="external.html#MANREF"
>manpage</A
> provides more detail
	      on this complex and powerful command.</P
></DD
><DT
><A
NAME="XARGSREF"
></A
><B
CLASS="COMMAND"
>xargs</B
></DT
><DD
><P
>A filter for feeding arguments to a command, and also
	      a tool for assembling the commands themselves. It breaks
	      a data stream into small enough chunks for filters and
	      commands to process.  Consider it as a powerful replacement
	      for <A
HREF="commandsub.html#BACKQUOTESREF"
>backquotes</A
>.
	      In situations where <A
HREF="commandsub.html#COMMANDSUBREF"
>command
	      substitution</A
> fails with a <SPAN
CLASS="ERRORNAME"
>too
	      many arguments</SPAN
> error,
	      substituting <B
CLASS="COMMAND"
>xargs</B
> often
	      works.
	        <A
NAME="AEN10465"
HREF="#FTN.AEN10465"
>[1]</A
>
	      Normally, <B
CLASS="COMMAND"
>xargs</B
> reads from
	      <TT
CLASS="FILENAME"
>stdin</TT
> or from a pipe, but it can also
	      be given the output of a file.</P
><P
>The default command for <B
CLASS="COMMAND"
>xargs</B
> is
	      <A
HREF="internal.html#ECHOREF"
>echo</A
>. This means that input
	      piped to <B
CLASS="COMMAND"
>xargs</B
> may have linefeeds and
	      other whitespace characters stripped out.</P
><P
>	      <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>ls -l</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>total 0
 -rw-rw-r--    1 bozo  bozo         0 Jan 29 23:58 file1
 -rw-rw-r--    1 bozo  bozo         0 Jan 29 23:58 file2</TT
>
 
 
 
 <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>ls -l | xargs</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>total 0 -rw-rw-r-- 1 bozo bozo 0 Jan 29 23:58 file1 -rw-rw-r-- 1 bozo bozo 0 Jan...</TT
>
 
 
 
 <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>find ~/mail -type f | xargs grep "Linux"</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>./misc:User-Agent: slrn/0.9.8.1 (Linux)
 ./sent-mail-jul-2005: hosted by the Linux Documentation Project.
 ./sent-mail-jul-2005: (Linux Documentation Project Site, rtf version)
 ./sent-mail-jul-2005: Subject: Criticism of Bozo's Windows/Linux article
 ./sent-mail-jul-2005: while mentioning that the Linux ext2/ext3 filesystem
 . . .</TT
>
 	      </PRE
></TD
></TR
></TABLE
>
	      </P
><P
><TT
CLASS="USERINPUT"
><B
>ls | xargs -p -l gzip</B
></TT
> <A
HREF="filearchiv.html#GZIPREF"
>gzips</A
> every file in current
	      directory, one at a time, prompting before each
	      operation.</P
><P
><A
NAME="XARGSONEATATIME"
></A
></P
><DIV
CLASS="NOTE"
><TABLE
CLASS="NOTE"
WIDTH="90%"
BORDER="0"
><TR
><TD
WIDTH="25"
ALIGN="CENTER"
VALIGN="TOP"
><IMG
SRC="common/note.png"
HSPACE="5"
ALT="Note"></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Note that <I
CLASS="FIRSTTERM"
>xargs</I
> processes the
	      arguments passed to it sequentially, <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>one at
	      a time</I
></SPAN
>.</P
><P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>find /usr/bin | xargs file</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>/usr/bin:          directory
 /usr/bin/foomatic-ppd-options:          perl script text executable
 . . .</TT
>
 	      </PRE
></TD
></TR
></TABLE
>
	      </P
></TD
></TR
></TABLE
></DIV
><P
><A
NAME="XARGSLIMARGS"
></A
></P
><DIV
CLASS="TIP"
><TABLE
CLASS="TIP"
WIDTH="90%"
BORDER="0"
><TR
><TD
WIDTH="25"
ALIGN="CENTER"
VALIGN="TOP"
><IMG
SRC="common/tip.png"
HSPACE="5"
ALT="Tip"></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>An interesting <I
CLASS="FIRSTTERM"
>xargs</I
>
	      option is <TT
CLASS="OPTION"
>-n <TT
CLASS="REPLACEABLE"
><I
>NN</I
></TT
></TT
>,
	      which limits to <TT
CLASS="REPLACEABLE"
><I
>NN</I
></TT
> the number
	      of arguments passed.</P
><P
><TT
CLASS="USERINPUT"
><B
>ls | xargs -n 8 echo</B
></TT
> lists the files in the
	      current directory in <TT
CLASS="LITERAL"
>8</TT
> columns.</P
></TD
></TR
></TABLE
></DIV
><P
><A
NAME="XARGSWS"
></A
></P
><DIV
CLASS="TIP"
><TABLE
CLASS="TIP"
WIDTH="90%"
BORDER="0"
><TR
><TD
WIDTH="25"
ALIGN="CENTER"
VALIGN="TOP"
><IMG
SRC="common/tip.png"
HSPACE="5"
ALT="Tip"></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>Another useful option is
	      <TT
CLASS="OPTION"
>-0</TT
>, in combination with <TT
CLASS="USERINPUT"
><B
>find
	      -print0</B
></TT
> or <TT
CLASS="USERINPUT"
><B
>grep -lZ</B
></TT
>. This
	      allows handling arguments containing whitespace or
	      quotes.</P
><P
>	    <TT
CLASS="USERINPUT"
><B
>find / -type f -print0 | xargs -0 grep -liwZ GUI | xargs -0 rm -f</B
></TT
>
	    </P
><P
>	    <TT
CLASS="USERINPUT"
><B
>grep -rliwZ GUI / | xargs -0 rm -f</B
></TT
>
	    </P
><P
>Either of the above will remove any file containing <SPAN
CLASS="QUOTE"
>"GUI"</SPAN
>.
	      <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>(Thanks, S.C.)</I
></SPAN
></P
><P
>Or:
	      <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;cat /proc/"$pid"/"$OPTION" | xargs -0 echo
   2&nbsp;#  Formats output:         ^^^^^^^^^^^^^^^
   3&nbsp;#  From Han Holl's fixup of "get-commandline.sh"
   4&nbsp;#+ script in "/dev and /proc" chapter.</PRE
></TD
></TR
></TABLE
></P
></TD
></TR
></TABLE
></DIV
><DIV
CLASS="TIP"
><TABLE
CLASS="TIP"
WIDTH="90%"
BORDER="0"
><TR
><TD
WIDTH="25"
ALIGN="CENTER"
VALIGN="TOP"
><IMG
SRC="common/tip.png"
HSPACE="5"
ALT="Tip"></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
><A
NAME="XARGSMULTIPROCESS"
></A
></P
><P
>The <TT
CLASS="OPTION"
>-P</TT
> option to
		      <I
CLASS="FIRSTTERM"
>xargs</I
> permits running
		      processes in parallel. This speeds up execution
		      in a machine with a multicore CPU.</P
><P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;
   3&nbsp;ls *gif | xargs -t -n1 -P2 gif2png
   4&nbsp;# Converts all the gif images in current directory to png.
   5&nbsp;
   6&nbsp;# Options:
   7&nbsp;# =======
   8&nbsp;# -t    Print command to stderr.
   9&nbsp;# -n1   At most 1 argument per command line.
  10&nbsp;# -P2   Run up to 2 processes simultaneously.
  11&nbsp;
  12&nbsp;# Thank you, Roberto Polli, for the inspiration.</PRE
></TD
></TR
></TABLE
></P
></TD
></TR
></TABLE
></DIV
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="EX41"
></A
><P
><B
>Example 16-5. Logfile: Using <I
CLASS="FIRSTTERM"
>xargs</I
> to monitor system log</B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;
   3&nbsp;# Generates a log file in current directory
   4&nbsp;# from the tail end of /var/log/messages.
   5&nbsp;
   6&nbsp;# Note: /var/log/messages must be world readable
   7&nbsp;# if this script invoked by an ordinary user.
   8&nbsp;#         #root chmod 644 /var/log/messages
   9&nbsp;
  10&nbsp;LINES=5
  11&nbsp;
  12&nbsp;( date; uname -a ) &#62;&#62;logfile
  13&nbsp;# Time and machine name
  14&nbsp;echo ---------------------------------------------------------- &#62;&#62;logfile
  15&nbsp;tail -n $LINES /var/log/messages | xargs | fmt -s &#62;&#62;logfile
  16&nbsp;echo &#62;&#62;logfile
  17&nbsp;echo &#62;&#62;logfile
  18&nbsp;
  19&nbsp;exit 0
  20&nbsp;
  21&nbsp;#  Note:
  22&nbsp;#  ----
  23&nbsp;#  As Frank Wang points out,
  24&nbsp;#+ unmatched quotes (either single or double quotes) in the source file
  25&nbsp;#+ may give xargs indigestion.
  26&nbsp;#
  27&nbsp;#  He suggests the following substitution for line 15:
  28&nbsp;#  tail -n $LINES /var/log/messages | tr -d "\"'" | xargs | fmt -s &#62;&#62;logfile
  29&nbsp;
  30&nbsp;
  31&nbsp;
  32&nbsp;#  Exercise:
  33&nbsp;#  --------
  34&nbsp;#  Modify this script to track changes in /var/log/messages at intervals
  35&nbsp;#+ of 20 minutes.
  36&nbsp;#  Hint: Use the "watch" command. </PRE
></TD
></TR
></TABLE
><HR></DIV
><P
><A
NAME="XARGSCURLYREF"
></A
></P
><P
><A
HREF="moreadv.html#CURLYBRACKETSREF"
>As in
	      <B
CLASS="COMMAND"
>find</B
></A
>, a curly bracket
	      pair serves as a placeholder for replacement text.</P
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="EX42"
></A
><P
><B
>Example 16-6. Copying files in current directory to another</B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;# copydir.sh
   3&nbsp;
   4&nbsp;#  Copy (verbose) all files in current directory ($PWD)
   5&nbsp;#+ to directory specified on command-line.
   6&nbsp;
   7&nbsp;E_NOARGS=85
   8&nbsp;
   9&nbsp;if [ -z "$1" ]   # Exit if no argument given.
  10&nbsp;then
  11&nbsp;  echo "Usage: `basename $0` directory-to-copy-to"
  12&nbsp;  exit $E_NOARGS
  13&nbsp;fi  
  14&nbsp;
  15&nbsp;ls . | xargs -i -t cp ./{} $1
  16&nbsp;#            ^^ ^^      ^^
  17&nbsp;#  -t is "verbose" (output command-line to stderr) option.
  18&nbsp;#  -i is "replace strings" option.
  19&nbsp;#  {} is a placeholder for output text.
  20&nbsp;#  This is similar to the use of a curly-bracket pair in "find."
  21&nbsp;#
  22&nbsp;#  List the files in current directory (ls .),
  23&nbsp;#+ pass the output of "ls" as arguments to "xargs" (-i -t options),
  24&nbsp;#+ then copy (cp) these arguments ({}) to new directory ($1).  
  25&nbsp;#
  26&nbsp;#  The net result is the exact equivalent of
  27&nbsp;#+   cp * $1
  28&nbsp;#+ unless any of the filenames has embedded "whitespace" characters.
  29&nbsp;
  30&nbsp;exit 0</PRE
></TD
></TR
></TABLE
><HR></DIV
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="KILLBYNAME"
></A
><P
><B
>Example 16-7. Killing processes by name</B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;# kill-byname.sh: Killing processes by name.
   3&nbsp;# Compare this script with kill-process.sh.
   4&nbsp;
   5&nbsp;#  For instance,
   6&nbsp;#+ try "./kill-byname.sh xterm" --
   7&nbsp;#+ and watch all the xterms on your desktop disappear.
   8&nbsp;
   9&nbsp;#  Warning:
  10&nbsp;#  -------
  11&nbsp;#  This is a fairly dangerous script.
  12&nbsp;#  Running it carelessly (especially as root)
  13&nbsp;#+ can cause data loss and other undesirable effects.
  14&nbsp;
  15&nbsp;E_BADARGS=66
  16&nbsp;
  17&nbsp;if test -z "$1"  # No command-line arg supplied?
  18&nbsp;then
  19&nbsp;  echo "Usage: `basename $0` Process(es)_to_kill"
  20&nbsp;  exit $E_BADARGS
  21&nbsp;fi
  22&nbsp;
  23&nbsp;
  24&nbsp;PROCESS_NAME="$1"
  25&nbsp;ps ax | grep "$PROCESS_NAME" | awk '{print $1}' | xargs -i kill {} 2&#38;&#62;/dev/null
  26&nbsp;#                                                       ^^      ^^
  27&nbsp;
  28&nbsp;# ---------------------------------------------------------------
  29&nbsp;# Notes:
  30&nbsp;# -i is the "replace strings" option to xargs.
  31&nbsp;# The curly brackets are the placeholder for the replacement.
  32&nbsp;# 2&#38;&#62;/dev/null suppresses unwanted error messages.
  33&nbsp;#
  34&nbsp;# Can  grep "$PROCESS_NAME" be replaced by pidof "$PROCESS_NAME"?
  35&nbsp;# ---------------------------------------------------------------
  36&nbsp;
  37&nbsp;exit $?
  38&nbsp;
  39&nbsp;#  The "killall" command has the same effect as this script,
  40&nbsp;#+ but using it is not quite as educational.</PRE
></TD
></TR
></TABLE
><HR></DIV
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="WF2"
></A
><P
><B
>Example 16-8. Word frequency analysis using
	      <I
CLASS="FIRSTTERM"
>xargs</I
></B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;# wf2.sh: Crude word frequency analysis on a text file.
   3&nbsp;
   4&nbsp;# Uses 'xargs' to decompose lines of text into single words.
   5&nbsp;# Compare this example to the "wf.sh" script later on.
   6&nbsp;
   7&nbsp;
   8&nbsp;# Check for input file on command-line.
   9&nbsp;ARGS=1
  10&nbsp;E_BADARGS=85
  11&nbsp;E_NOFILE=86
  12&nbsp;
  13&nbsp;if [ $# -ne "$ARGS" ]
  14&nbsp;# Correct number of arguments passed to script?
  15&nbsp;then
  16&nbsp;  echo "Usage: `basename $0` filename"
  17&nbsp;  exit $E_BADARGS
  18&nbsp;fi
  19&nbsp;
  20&nbsp;if [ ! -f "$1" ]       # Does file exist?
  21&nbsp;then
  22&nbsp;  echo "File \"$1\" does not exist."
  23&nbsp;  exit $E_NOFILE
  24&nbsp;fi
  25&nbsp;
  26&nbsp;
  27&nbsp;
  28&nbsp;#####################################################
  29&nbsp;cat "$1" | xargs -n1 | \
  30&nbsp;#  List the file, one word per line. 
  31&nbsp;tr A-Z a-z | \
  32&nbsp;#  Shift characters to lowercase.
  33&nbsp;sed -e 's/\.//g'  -e 's/\,//g' -e 's/ /\
  34&nbsp;/g' | \
  35&nbsp;#  Filter out periods and commas, and
  36&nbsp;#+ change space between words to linefeed,
  37&nbsp;sort | uniq -c | sort -nr
  38&nbsp;#  Finally remove duplicates, prefix occurrence count
  39&nbsp;#+ and sort numerically.
  40&nbsp;#####################################################
  41&nbsp;
  42&nbsp;#  This does the same job as the "wf.sh" example,
  43&nbsp;#+ but a bit more ponderously, and it runs more slowly (why?).
  44&nbsp;
  45&nbsp;exit $?</PRE
></TD
></TR
></TABLE
><HR></DIV
></DD
><DT
><A
NAME="EXPRREF"
></A
><TT
CLASS="USERINPUT"
><B
>expr</B
></TT
></DT
><DD
><P
>All-purpose expression evaluator:
	      Concatenates and evaluates the arguments according
	      to the operation given (arguments must be separated
	      by spaces). Operations may be arithmetic, comparison,
	      string, or logical.</P
><DIV
CLASS="VARIABLELIST"
><DL
><DT
><TT
CLASS="USERINPUT"
><B
>expr 3 + 5</B
></TT
></DT
><DD
><P
>returns <TT
CLASS="LITERAL"
>8</TT
></P
></DD
><DT
><TT
CLASS="USERINPUT"
><B
>expr 5 % 3</B
></TT
></DT
><DD
><P
>returns 2</P
></DD
><DT
><TT
CLASS="USERINPUT"
><B
>expr 1 / 0</B
></TT
></DT
><DD
><P
>returns the error message, <SPAN
CLASS="ERRORCODE"
>expr: division by
		    zero</SPAN
></P
><P
>Illegal arithmetic operations not allowed.</P
></DD
><DT
><TT
CLASS="USERINPUT"
><B
>expr 5 \* 3</B
></TT
></DT
><DD
><P
>returns 15</P
><P
>The multiplication operator
		  must be escaped when used in an arithmetic expression
		  with <B
CLASS="COMMAND"
>expr</B
>.</P
></DD
><DT
><TT
CLASS="USERINPUT"
><B
>y=`expr $y + 1`</B
></TT
></DT
><DD
><P
>Increment a variable, with the same effect
		    as <TT
CLASS="USERINPUT"
><B
>let y=y+1</B
></TT
> and
		    <TT
CLASS="USERINPUT"
><B
>y=$(($y+1))</B
></TT
>. This is an
		    example of <A
HREF="arithexp.html#ARITHEXPREF"
>arithmetic
		    expansion</A
>.</P
></DD
><DT
><A
NAME="EXPEXTRSUB"
></A
><TT
CLASS="USERINPUT"
><B
>z=`expr substr
		$string $position $length`</B
></TT
></DT
><DD
><P
>Extract substring of $length characters, starting
		    at $position.</P
></DD
></DL
></DIV
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="EX45"
></A
><P
><B
>Example 16-9. Using <I
CLASS="FIRSTTERM"
>expr</I
></B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;
   3&nbsp;# Demonstrating some of the uses of 'expr'
   4&nbsp;# =======================================
   5&nbsp;
   6&nbsp;echo
   7&nbsp;
   8&nbsp;# Arithmetic Operators
   9&nbsp;# ---------- ---------
  10&nbsp;
  11&nbsp;echo "Arithmetic Operators"
  12&nbsp;echo
  13&nbsp;a=`expr 5 + 3`
  14&nbsp;echo "5 + 3 = $a"
  15&nbsp;
  16&nbsp;a=`expr $a + 1`
  17&nbsp;echo
  18&nbsp;echo "a + 1 = $a"
  19&nbsp;echo "(incrementing a variable)"
  20&nbsp;
  21&nbsp;a=`expr 5 % 3`
  22&nbsp;# modulo
  23&nbsp;echo
  24&nbsp;echo "5 mod 3 = $a"
  25&nbsp;
  26&nbsp;echo
  27&nbsp;echo
  28&nbsp;
  29&nbsp;# Logical Operators
  30&nbsp;# ------- ---------
  31&nbsp;
  32&nbsp;#  Returns 1 if true, 0 if false,
  33&nbsp;#+ opposite of normal Bash convention.
  34&nbsp;
  35&nbsp;echo "Logical Operators"
  36&nbsp;echo
  37&nbsp;
  38&nbsp;x=24
  39&nbsp;y=25
  40&nbsp;b=`expr $x = $y`         # Test equality.
  41&nbsp;echo "b = $b"            # 0  ( $x -ne $y )
  42&nbsp;echo
  43&nbsp;
  44&nbsp;a=3
  45&nbsp;b=`expr $a \&#62; 10`
  46&nbsp;echo 'b=`expr $a \&#62; 10`, therefore...'
  47&nbsp;echo "If a &#62; 10, b = 0 (false)"
  48&nbsp;echo "b = $b"            # 0  ( 3 ! -gt 10 )
  49&nbsp;echo
  50&nbsp;
  51&nbsp;b=`expr $a \&#60; 10`
  52&nbsp;echo "If a &#60; 10, b = 1 (true)"
  53&nbsp;echo "b = $b"            # 1  ( 3 -lt 10 )
  54&nbsp;echo
  55&nbsp;# Note escaping of operators.
  56&nbsp;
  57&nbsp;b=`expr $a \&#60;= 3`
  58&nbsp;echo "If a &#60;= 3, b = 1 (true)"
  59&nbsp;echo "b = $b"            # 1  ( 3 -le 3 )
  60&nbsp;# There is also a "\&#62;=" operator (greater than or equal to).
  61&nbsp;
  62&nbsp;
  63&nbsp;echo
  64&nbsp;echo
  65&nbsp;
  66&nbsp;
  67&nbsp;
  68&nbsp;# String Operators
  69&nbsp;# ------ ---------
  70&nbsp;
  71&nbsp;echo "String Operators"
  72&nbsp;echo
  73&nbsp;
  74&nbsp;a=1234zipper43231
  75&nbsp;echo "The string being operated upon is \"$a\"."
  76&nbsp;
  77&nbsp;# length: length of string
  78&nbsp;b=`expr length $a`
  79&nbsp;echo "Length of \"$a\" is $b."
  80&nbsp;
  81&nbsp;# index: position of first character in substring
  82&nbsp;#        that matches a character in string
  83&nbsp;b=`expr index $a 23`
  84&nbsp;echo "Numerical position of first \"2\" in \"$a\" is \"$b\"."
  85&nbsp;
  86&nbsp;# substr: extract substring, starting position &#38; length specified
  87&nbsp;b=`expr substr $a 2 6`
  88&nbsp;echo "Substring of \"$a\", starting at position 2,\
  89&nbsp;and 6 chars long is \"$b\"."
  90&nbsp;
  91&nbsp;
  92&nbsp;#  The default behavior of the 'match' operations is to
  93&nbsp;#+ search for the specified match at the BEGINNING of the string.
  94&nbsp;#
  95&nbsp;#       Using Regular Expressions ...
  96&nbsp;b=`expr match "$a" '[0-9]*'`               #  Numerical count.
  97&nbsp;echo Number of digits at the beginning of \"$a\" is $b.
  98&nbsp;b=`expr match "$a" '\([0-9]*\)'`           #  Note that escaped parentheses
  99&nbsp;#                   ==      ==             #+ trigger substring match.
 100&nbsp;echo "The digits at the beginning of \"$a\" are \"$b\"."
 101&nbsp;
 102&nbsp;echo
 103&nbsp;
 104&nbsp;exit 0</PRE
></TD
></TR
></TABLE
><HR></DIV
><DIV
CLASS="IMPORTANT"
><TABLE
CLASS="IMPORTANT"
WIDTH="90%"
BORDER="0"
><TR
><TD
WIDTH="25"
ALIGN="CENTER"
VALIGN="TOP"
><IMG
SRC="common/important.png"
HSPACE="5"
ALT="Important"></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
><P
>The <A
HREF="special-chars.html#NULLREF"
>:
	      (<I
CLASS="FIRSTTERM"
>null</I
>)</A
> operator
	      can substitute for <B
CLASS="COMMAND"
>match</B
>. For example,
	      <TT
CLASS="USERINPUT"
><B
>b=`expr $a : [0-9]*`</B
></TT
> is the
	      exact equivalent of <TT
CLASS="USERINPUT"
><B
>b=`expr match $a
	      [0-9]*`</B
></TT
> in the above listing.</P
><P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;
   3&nbsp;echo
   4&nbsp;echo "String operations using \"expr \$string : \" construct"
   5&nbsp;echo "==================================================="
   6&nbsp;echo
   7&nbsp;
   8&nbsp;a=1234zipper5FLIPPER43231
   9&nbsp;
  10&nbsp;echo "The string being operated upon is \"`expr "$a" : '\(.*\)'`\"."
  11&nbsp;#     Escaped parentheses grouping operator.            ==  ==
  12&nbsp;
  13&nbsp;#       ***************************
  14&nbsp;#+          Escaped parentheses
  15&nbsp;#+           match a substring
  16&nbsp;#       ***************************
  17&nbsp;
  18&nbsp;
  19&nbsp;#  If no escaped parentheses ...
  20&nbsp;#+ then 'expr' converts the string operand to an integer.
  21&nbsp;
  22&nbsp;echo "Length of \"$a\" is `expr "$a" : '.*'`."   # Length of string
  23&nbsp;
  24&nbsp;echo "Number of digits at the beginning of \"$a\" is `expr "$a" : '[0-9]*'`."
  25&nbsp;
  26&nbsp;# ------------------------------------------------------------------------- #
  27&nbsp;
  28&nbsp;echo
  29&nbsp;
  30&nbsp;echo "The digits at the beginning of \"$a\" are `expr "$a" : '\([0-9]*\)'`."
  31&nbsp;#                                                             ==      ==
  32&nbsp;echo "The first 7 characters of \"$a\" are `expr "$a" : '\(.......\)'`."
  33&nbsp;#         =====                                          ==       ==
  34&nbsp;# Again, escaped parentheses force a substring match.
  35&nbsp;#
  36&nbsp;echo "The last 7 characters of \"$a\" are `expr "$a" : '.*\(.......\)'`."
  37&nbsp;#         ====                  end of string operator  ^^
  38&nbsp;#  (In fact, means skip over one or more of any characters until specified
  39&nbsp;#+  substring found.)
  40&nbsp;
  41&nbsp;echo
  42&nbsp;
  43&nbsp;exit 0</PRE
></TD
></TR
></TABLE
></P
></TD
></TR
></TABLE
></DIV
></DD
></DL
></DIV
><P
>The above script illustrates how
	      <B
CLASS="COMMAND"
>expr</B
> uses the <I
CLASS="FIRSTTERM"
>escaped
	      parentheses -- \( ... \) --</I
> grouping operator
	      in tandem with <A
HREF="regexp.html#REGEXREF"
>regular
	      expression</A
> parsing to match a substring.
	      Here is a another example, this time from <SPAN
CLASS="QUOTE"
>"real
	      life."</SPAN
>

	        <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="100%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;# Strip the whitespace from the beginning and end.
   2&nbsp;LRFDATE=`expr "$LRFDATE" : '[[:space:]]*\(.*\)[[:space:]]*$'`
   3&nbsp;
   4&nbsp;#  From Peter Knowles' "booklistgen.sh" script
   5&nbsp;#+ for converting files to Sony Librie/PRS-50X format.
   6&nbsp;#  (http://booklistgensh.peterknowles.com)</PRE
></TD
></TR
></TABLE
>

	      </P
><P
><A
HREF="wrapper.html#PERLREF"
>Perl</A
>,
	      <A
HREF="sedawk.html#SEDREF"
>sed</A
>, and <A
HREF="awk.html#AWKREF"
>awk</A
> have far superior string
	      parsing facilities. A short <B
CLASS="COMMAND"
>sed</B
> or
	      <B
CLASS="COMMAND"
>awk</B
> <SPAN
CLASS="QUOTE"
>"subroutine"</SPAN
> within
	      a script (see <A
HREF="wrapper.html"
>Section 36.2</A
>) is an attractive
	      alternative to <B
CLASS="COMMAND"
>expr</B
>.</P
><P
>See <A
HREF="manipulatingvars.html#STRING-MANIPULATION"
>Section 10.1</A
> for more on
              using <B
CLASS="COMMAND"
>expr</B
> in string operations.</P
></DIV
><H3
CLASS="FOOTNOTES"
>Notes</H3
><TABLE
BORDER="0"
CLASS="FOOTNOTES"
WIDTH="100%"
><TR
><TD
ALIGN="LEFT"
VALIGN="TOP"
WIDTH="5%"
><A
NAME="FTN.AEN10465"
HREF="moreadv.html#AEN10465"
>[1]</A
></TD
><TD
ALIGN="LEFT"
VALIGN="TOP"
WIDTH="95%"
><P
>And even when <I
CLASS="FIRSTTERM"
>xargs</I
> is
		not strictly necessary, it can speed up execution of a command
		involving <A
HREF="timedate.html#BATCHPROCREF"
>batch-processing</A
> of multiple
		files.</P
></TD
></TR
></TABLE
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="external.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="timedate.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>External Filters, Programs and Commands</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="external.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>Time / Date Commands</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>