filearchiv.html

一本完整的描述Unix Shell 编程的工具书的所有范例

CLASS="COMMAND"
>compress</B
>.</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
>The <B
CLASS="COMMAND"
>znew</B
> command transforms
	      <I
CLASS="EMPHASIS"
>compressed</I
> files into
	      <I
CLASS="EMPHASIS"
>gzipped</I
> ones.</P
></TD
></TR
></TABLE
></DIV
></DD
><DT
><B
CLASS="COMMAND"
>sq</B
></DT
><DD
><P
>Yet another compression utility, a filter that works
	      only on sorted ASCII word lists. It uses the standard
	      invocation syntax for a filter, <B
CLASS="COMMAND"
>sq &#60; input-file &#62;
	      output-file</B
>.  Fast, but not nearly as efficient
	      as <A
HREF="filearchiv.html#GZIPREF"
>gzip</A
>.  The corresponding
	      uncompression filter is <B
CLASS="COMMAND"
>unsq</B
>, invoked
	      like <B
CLASS="COMMAND"
>sq</B
>.</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
>The output of <B
CLASS="COMMAND"
>sq</B
> may be
	        piped to <B
CLASS="COMMAND"
>gzip</B
> for further
		compression.</P
></TD
></TR
></TABLE
></DIV
></DD
><DT
><B
CLASS="COMMAND"
>zip</B
>, <B
CLASS="COMMAND"
>unzip</B
></DT
><DD
><P
>Cross-platform file archiving and compression utility
	      compatible with DOS <I
CLASS="EMPHASIS"
>pkzip.exe</I
>.
	      <SPAN
CLASS="QUOTE"
>"Zipped"</SPAN
> archives seem to be a more
	      acceptable medium of exchange on the Internet than
	      <SPAN
CLASS="QUOTE"
>"tarballs"</SPAN
>.</P
></DD
><DT
><B
CLASS="COMMAND"
>unarc</B
>, <B
CLASS="COMMAND"
>unarj</B
>, <B
CLASS="COMMAND"
>unrar</B
></DT
><DD
><P
>These Linux utilities permit unpacking archives
	      compressed with the DOS <I
CLASS="EMPHASIS"
>arc.exe</I
>,
	      <I
CLASS="EMPHASIS"
>arj.exe</I
>, and
	      <I
CLASS="EMPHASIS"
>rar.exe</I
> programs.</P
></DD
></DL
></DIV
><DIV
CLASS="VARIABLELIST"
><P
><B
><A
NAME="FAINFORMATION1"
></A
>File Information</B
></P
><DL
><DT
><A
NAME="FILEREF"
></A
><B
CLASS="COMMAND"
>file</B
></DT
><DD
><P
>A utility for identifying file types. The command
	      <TT
CLASS="USERINPUT"
><B
>file file-name</B
></TT
> will return a
	      file specification for <TT
CLASS="FILENAME"
>file-name</TT
>,
	      such as <TT
CLASS="COMPUTEROUTPUT"
>ascii text</TT
> or
	      <TT
CLASS="COMPUTEROUTPUT"
>data</TT
>. It references
	      the <A
HREF="sha-bang.html#MAGNUMREF"
>magic numbers</A
>
	      found in <TT
CLASS="FILENAME"
>/usr/share/magic</TT
>,
	      <TT
CLASS="FILENAME"
>/etc/magic</TT
>, or
	      <TT
CLASS="FILENAME"
>/usr/lib/magic</TT
>, depending on the
	      Linux/UNIX distribution.</P
><P
>The <TT
CLASS="OPTION"
>-f</TT
> option causes
	      <B
CLASS="COMMAND"
>file</B
> to run in batch mode, to read from
	      a designated file a list of filenames to analyze. The
	      <TT
CLASS="OPTION"
>-z</TT
> option, when used on a compressed
	      target file, forces an attempt to analyze the uncompressed
	      file type.</P
><P
>	      <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>file test.tar.gz</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>test.tar.gz: gzip compressed data, deflated, last modified: Sun Sep 16 13:34:51 2001, os: Unix</TT
>
 
 <TT
CLASS="PROMPT"
>bash </TT
><TT
CLASS="USERINPUT"
><B
>file -z test.tar.gz</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>test.tar.gz: GNU tar archive (gzip compressed data, deflated, last modified: Sun Sep 16 13:34:51 2001, os: Unix)</TT
>
 	      </PRE
></TD
></TR
></TABLE
>
	    </P
><P
>	      <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;# Find sh and Bash scripts in a given directory:
   2&nbsp;
   3&nbsp;DIRECTORY=/usr/local/bin
   4&nbsp;KEYWORD=Bourne
   5&nbsp;# Bourne and Bourne-Again shell scripts
   6&nbsp;
   7&nbsp;file $DIRECTORY/* | fgrep $KEYWORD
   8&nbsp;
   9&nbsp;# Output:
  10&nbsp;
  11&nbsp;# /usr/local/bin/burn-cd:          Bourne-Again shell script text executable
  12&nbsp;# /usr/local/bin/burnit:           Bourne-Again shell script text executable
  13&nbsp;# /usr/local/bin/cassette.sh:      Bourne shell script text executable
  14&nbsp;# /usr/local/bin/copy-cd:          Bourne-Again shell script text executable
  15&nbsp;# . . .</PRE
></TD
></TR
></TABLE
>
	    </P
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="STRIPC"
></A
><P
><B
>Example 12-29. Stripping comments from C program files</B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;# strip-comment.sh: Strips out the comments (/* COMMENT */) in a C program.
   3&nbsp;
   4&nbsp;E_NOARGS=0
   5&nbsp;E_ARGERROR=66
   6&nbsp;E_WRONG_FILE_TYPE=67
   7&nbsp;
   8&nbsp;if [ $# -eq "$E_NOARGS" ]
   9&nbsp;then
  10&nbsp;  echo "Usage: `basename $0` C-program-file" &#62;&#38;2 # Error message to stderr.
  11&nbsp;  exit $E_ARGERROR
  12&nbsp;fi  
  13&nbsp;
  14&nbsp;# Test for correct file type.
  15&nbsp;type=`file $1 | awk '{ print $2, $3, $4, $5 }'`
  16&nbsp;# "file $1" echoes file type . . .
  17&nbsp;# Then awk removes the first field of this, the filename . . .
  18&nbsp;# Then the result is fed into the variable "type".
  19&nbsp;correct_type="ASCII C program text"
  20&nbsp;
  21&nbsp;if [ "$type" != "$correct_type" ]
  22&nbsp;then
  23&nbsp;  echo
  24&nbsp;  echo "This script works on C program files only."
  25&nbsp;  echo
  26&nbsp;  exit $E_WRONG_FILE_TYPE
  27&nbsp;fi  
  28&nbsp;
  29&nbsp;
  30&nbsp;# Rather cryptic sed script:
  31&nbsp;#--------
  32&nbsp;sed '
  33&nbsp;/^\/\*/d
  34&nbsp;/.*\*\//d
  35&nbsp;' $1
  36&nbsp;#--------
  37&nbsp;# Easy to understand if you take several hours to learn sed fundamentals.
  38&nbsp;
  39&nbsp;
  40&nbsp;#  Need to add one more line to the sed script to deal with
  41&nbsp;#+ case where line of code has a comment following it on same line.
  42&nbsp;#  This is left as a non-trivial exercise.
  43&nbsp;
  44&nbsp;#  Also, the above code deletes non-comment lines with a "*/" --
  45&nbsp;#+ not a desirable result.
  46&nbsp;
  47&nbsp;exit 0
  48&nbsp;
  49&nbsp;
  50&nbsp;# ----------------------------------------------------------------
  51&nbsp;# Code below this line will not execute because of 'exit 0' above.
  52&nbsp;
  53&nbsp;# Stephane Chazelas suggests the following alternative:
  54&nbsp;
  55&nbsp;usage() {
  56&nbsp;  echo "Usage: `basename $0` C-program-file" &#62;&#38;2
  57&nbsp;  exit 1
  58&nbsp;}
  59&nbsp;
  60&nbsp;WEIRD=`echo -n -e '\377'`   # or WEIRD=$'\377'
  61&nbsp;[[ $# -eq 1 ]] || usage
  62&nbsp;case `file "$1"` in
  63&nbsp;  *"C program text"*) sed -e "s%/\*%${WEIRD}%g;s%\*/%${WEIRD}%g" "$1" \
  64&nbsp;     | tr '\377\n' '\n\377' \
  65&nbsp;     | sed -ne 'p;n' \
  66&nbsp;     | tr -d '\n' | tr '\377' '\n';;
  67&nbsp;  *) usage;;
  68&nbsp;esac
  69&nbsp;
  70&nbsp;#  This is still fooled by things like:
  71&nbsp;#  printf("/*");
  72&nbsp;#  or
  73&nbsp;#  /*  /* buggy embedded comment */
  74&nbsp;#
  75&nbsp;#  To handle all special cases (comments in strings, comments in string
  76&nbsp;#+ where there is a \", \\" ...) the only way is to write a C parser
  77&nbsp;#+ (using lex or yacc perhaps?).
  78&nbsp;
  79&nbsp;exit 0</PRE
></TD
></TR
></TABLE
><HR></DIV
></DD
><DT
><A
NAME="WHICHREF"
></A
><B
CLASS="COMMAND"
>which</B
></DT
><DD
><P
><B
CLASS="COMMAND"
>which command-xxx</B
> gives the full path
	      to <SPAN
CLASS="QUOTE"
>"command-xxx"</SPAN
>. This is useful for finding
	      out whether a particular command or utility is installed
	      on the system.</P
><P
><TT
CLASS="USERINPUT"
><B
>$bash which rm</B
></TT
>
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="COMPUTEROUTPUT"
>/usr/bin/rm</TT
></PRE
></TD
></TR
></TABLE
>
	  </P
></DD
><DT
><B
CLASS="COMMAND"
>whereis</B
></DT
><DD
><P
>Similar to <B
CLASS="COMMAND"
>which</B
>, above,
	      <B
CLASS="COMMAND"
>whereis command-xxx</B
> gives the
	      full path to <SPAN
CLASS="QUOTE"
>"command-xxx"</SPAN
>, but also to its
	      <I
CLASS="EMPHASIS"
>manpage</I
>.</P
><P
><TT
CLASS="USERINPUT"
><B
>$bash whereis rm</B
></TT
>
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="COMPUTEROUTPUT"
>rm: /bin/rm /usr/share/man/man1/rm.1.bz2</TT
></PRE
></TD
></TR
></TABLE
>
	  </P
></DD
><DT
><A
NAME="WHATISREF"
></A
><B
CLASS="COMMAND"
>whatis</B
></DT
><DD
><P
><B
CLASS="COMMAND"
>whatis filexxx</B
> looks up
	      <SPAN
CLASS="QUOTE"
>"filexxx"</SPAN
> in the
	      <TT
CLASS="REPLACEABLE"
><I
>whatis</I
></TT
> database. This is useful
	      for identifying system commands and important configuration
	      files. Consider it a simplified <B
CLASS="COMMAND"
>man</B
>
	      command.</P
><P
><TT
CLASS="USERINPUT"
><B
>$bash whatis whatis</B
></TT
>
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="COMPUTEROUTPUT"
>whatis               (1)  - search the whatis database for complete words</TT
></PRE
></TD
></TR
></TABLE
>
	  </P
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="WHAT"
></A
><P
><B
>Example 12-30. <B
CLASS="COMMAND"
>Exploring <TT
CLASS="FILENAME"
>/usr/X11R6/bin</TT
></B
></B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;
   3&nbsp;# What are all those mysterious binaries in /usr/X11R6/bin?
   4&nbsp;
   5&nbsp;DIRECTORY="/usr/X11R6/bin"
   6&nbsp;# Try also "/bin", "/usr/bin", "/usr/local/bin", etc.
   7&nbsp;
   8&nbsp;for file in $DIRECTORY/*
   9&nbsp;do
  10&nbsp;  whatis `basename $file`   # Echoes info about the binary.
  11&nbsp;done
  12&nbsp;
  13&nbsp;exit 0
  14&nbsp;
  15&nbsp;# You may wish to redirect output of this script, like so:
  16&nbsp;# ./what.sh &#62;&#62;whatis.db
  17&nbsp;# or view it a page at a time on stdout,
  18&nbsp;# ./what.sh | less</PRE
></TD
></TR
></TABLE
><HR></DIV
><P
>See also <A
HREF="loops.html#FILEINFO"
>Example 10-3</A
>.</P
></DD
><DT
><B
CLASS="COMMAND"
>vdir</B
></DT
><DD
><P
>Show a detailed directory listing. The effect is similar to
	      <A
HREF="external.html#LSREF"
>ls -l</A
>.</P
><P
>This is one of the GNU <I
CLASS="EMPHASIS"
>fileutils</I
>.</P
><P
>	      <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>vdir</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>total 10
 -rw-r--r--    1 bozo  bozo      4034 Jul 18 22:04 data1.xrolo
 -rw-r--r--    1 bozo  bozo      4602 May 25 13:58 data1.xrolo.bak
 -rw-r--r--    1 bozo  bozo       877 Dec 17  2000 employment.xrolo</TT
>
 
 <TT
CLASS="PROMPT"
>bash </TT
><TT
CLASS="USERINPUT"
><B
>ls -l</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>total 10
 -rw-r--r--    1 bozo  bozo      4034 Jul 18 22:04 data1.xrolo
 -rw-r--r--    1 bozo  bozo      4602 May 25 13:58 data1.xrolo.bak
 -rw-r--r--    1 bozo  bozo       877 Dec 17  2000 employment.xrolo</TT
>
 	      </PRE
></TD
></TR
></TABLE
>
	      </P
></DD
><DT
><B
CLASS="COMMAND"
>locate</B
>, <B
CLASS="COMMAND"
>slocate</B
></DT
><DD
><P
>The <B
CLASS="COMMAND"
>locate</B
> command searches for files using a
	      database stored for just that purpose. The
	      <B
CLASS="COMMAND"
>slocate</B
> command is the secure version of
	      <B
CLASS="COMMAND"
>locate</B
> (which may be aliased to
	      <B
CLASS="COMMAND"
>slocate</B
>).</P
><P
><TT
CLASS="USERINPUT"
><B
>$bash locate hickson</B
></TT
>
<TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="COMPUTEROUTPUT"
>/usr/lib/xephem/catalogs/hickson.edb</TT
></PRE
></TD
></TR
></TABLE
></P
></DD
><DT
><B
CLASS="COMMAND"
>readlink</B
></DT
><DD
><P
>Disclose the file that a symbolic link points to.</P
><P
>	      <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="SCREEN"
> <TT
CLASS="PROMPT"
>bash$ </TT
><TT
CLASS="USERINPUT"
><B
>readlink /usr/bin/awk</B
></TT
>
 <TT
CLASS="COMPUTEROUTPUT"
>../../bin/gawk</TT
>
 	      </PRE
></TD
></TR
></TABLE
>
	    </P
></DD
><DT
><B
CLASS="COMMAND"
>strings</B
></DT
><DD
><P
>Use the <B
CLASS="COMMAND"
>strings</B
> command to find
	      printable strings in a binary or data file. It will list
	      sequences of printable characters found in the target
	      file. This might be handy for a quick 'n dirty examination
	      of a core dump or for looking at an unknown graphic image
	      file (<TT
CLASS="USERINPUT"
><B
>strings image-file | more</B
></TT
> might
	      show something like <TT
CLASS="COMPUTEROUTPUT"
>JFIF</TT
>,
	      which would identify the file as a <I
CLASS="EMPHASIS"
>jpeg</I
>
	      graphic). In a script, you would probably
	      parse the output of <B
CLASS="COMMAND"
>strings</B
>
	      with <A
HREF="textproc.html#GREPREF"
>grep</A
> or <A
HREF="sedawk.html#SEDREF"
>sed</A
>. See <A
HREF="loops.html#BINGREP"
>Example 10-7</A
>
	      and <A
HREF="loops.html#FINDSTRING"
>Example 10-9</A
>.</P
><DIV
CLASS="EXAMPLE"
><HR><A
NAME="WSTRINGS"
></A
><P
><B
>Example 12-31. An <SPAN
CLASS="QUOTE"
>"improved"</SPAN
> <I
CLASS="EMPHASIS"
>strings</I
>
	        command</B
></P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#!/bin/bash
   2&nbsp;# wstrings.sh: "word-strings" (enhanced "strings" command)
   3&nbsp;#
   4&nbsp;#  This script filters the output of "strings" by checking it
   5&nbsp;#+ against a standard word list file.
   6&nbsp;#  This effectively eliminates gibberish and noise,
   7&nbsp;#+ and outputs only recognized words.
   8&nbsp;
   9&nbsp;# ===========================================================
  10&nbsp;#                 Standard Check for Script Argument(s)
  11&nbsp;ARGS=1
  12&nbsp;E_BADARGS=65
  13&nbsp;E_NOFILE=66
  14&nbsp;
  15&nbsp;if [ $# -ne $ARGS ]
  16&nbsp;then
  17&nbsp;  echo "Usage: `basename $0` filename"
  18&nbsp;  exit $E_BADARGS
  19&nbsp;fi
  20&nbsp;
  21&nbsp;if [ ! -f "$1" ]                      # Check if file exists.
  22&nbsp;then
  23&nbsp;    echo "File \"$1\" does not exist."
  24&nbsp;    exit $E_NOFILE
  25&nbsp;fi
  26&nbsp;# ===========================================================
  27&nbsp;
  28&nbsp;
  29&nbsp;MINSTRLEN=3                           #  Minimum string length.
  30&nbsp;WORDFILE=/usr/share/dict/linux.words  #  Dictionary file.
  31&nbsp;                                      #  May specify a different
  32&nbsp;                                      #+ word list file
  33&nbsp;                                      #+ of one-word-per-line format.
  34&nbsp;
  35&nbsp;
  36&nbsp;wlist=`strings "$1" | tr A-Z a-z | tr '[:space:]' Z | \
  37&nbsp;tr -cs '[:alpha:]' Z | tr -s '\173-\377' Z | tr Z ' '`
  38&nbsp;
  39&nbsp;# Translate output of 'strings' command with multiple passes of 'tr'.
  40&nbsp;#  "tr A-Z a-z"  converts to lowercase.
  41&nbsp;#  "tr '[:space:]'"  converts whitespace characters to Z's.
  42&nbsp;#  "tr -cs '[:alpha:]' Z"  converts non-alphabetic characters to Z's,
  43&nbsp;#+ and squeezes multiple consecutive Z's.
  44&nbsp;#  "tr -s '\173-\377' Z"  converts all characters past 'z' to Z's
  45&nbsp;#+ and squeezes multiple consecutive Z's,
  46&nbsp;#+ which gets rid of all the weird characters that the previous
  47&nbsp;#+ translation failed to deal with.
  48&nbsp;#  Finally, "tr Z ' '" converts all those Z's to whitespace,
  49&nbsp;#+ which will be seen as word separators in the loop below.
  50&nbsp;
  51&nbsp;#  ****************************************************************
  52&nbsp;#  Note the technique of feeding the output of 'tr' back to itself,
  53&nbsp;#+ but with different arguments and/or options on each pass.
  54&nbsp;#  ****************************************************************
  55&nbsp;
  56&nbsp;
  57&nbsp;for word in $wlist                    # Important:
  58&nbsp;                                      # $wlist must not be quoted here.
  59&nbsp;                                      # "$wlist" does not work.
  60&nbsp;                                      # Why not?
  61&nbsp;do
  62&nbsp;
  63&nbsp;  strlen=${#word}                     # String length.
  64&nbsp;  if [ "$strlen" -lt "$MINSTRLEN" ]   # Skip over short strings.
  65&nbsp;  then
  66&nbsp;    continue
  67&nbsp;  fi
  68&nbsp;
  69&nbsp;  grep -Fw $word "$WORDFILE"          #  Match whole words only.
  70&nbsp;#      ^^^                            #  "Fixed strings" and
  71&nbsp;                                      #+ "whole words" options. 
  72&nbsp;
  73&nbsp;done  
  74&nbsp;
  75&nbsp;
  76&nbsp;exit $?</PRE
></TD
></TR
></TABLE
><HR></DIV
></DD
></DL
></DIV
><DIV
CLASS="VARIABLELIST"
><P
><B
><A
NAME="COMPARISONN1"
></A
>Comparison</B
></P