writingscripts.html

Shall高级编程

></DT
><DD
><P
>What is the most efficient way to make change for $1.68,
	      using only coins in common circulations (up to 25c)? It's
	      6 quarters, 1 dime, a nickel, and three cents.</P
><P
>Given any arbitrary command line input in dollars and
	      cents ($*.??), calculate the change, using the minimum
	      number of coins. If your home country is not the United
	      States, you may use your local currency units instead. The
	      script will need to parse the command line input, then
	      change it to multiples of the smallest monetary unit (cents
	      or whatever). Hint: look at <A
HREF="functions.html#EX61"
>Example 23-8</A
>.</P
></DD
><DT
><B
CLASS="COMMAND"
>Quadratic Equations</B
></DT
><DD
><P
>Solve a <SPAN
CLASS="QUOTE"
>"quadratic"</SPAN
> equation of the form
	      <TT
CLASS="PARAMETER"
><I
>Ax^2 + Bx + C = 0</I
></TT
>. Have a script take
	      as arguments the coefficients, <TT
CLASS="USERINPUT"
><B
>A</B
></TT
>,
	      <TT
CLASS="USERINPUT"
><B
>B</B
></TT
>, and <TT
CLASS="USERINPUT"
><B
>C</B
></TT
>,
	      and return the solutions to four decimal places.</P
><P
>Hint: pipe the coefficients to <A
HREF="mathc.html#BCREF"
>bc</A
>, using the well-known formula,
	      <TT
CLASS="PARAMETER"
><I
>x = ( -B +/- sqrt( B^2 - 4AC ) ) / 2A</I
></TT
>.</P
></DD
><DT
><B
CLASS="COMMAND"
>Sum of Matching Numbers</B
></DT
><DD
><P
>Find the sum of all five-digit numbers (in the range
	      10000 - 99999) containing <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>exactly two</I
></SPAN
>
	      out of the following set of digits: { 4, 5, 6 }. These may
	      repeat within the same number, and if so, they count once
	      for each occurrence.</P
><P
>Some examples of matching numbers are
	      42057, 74638, and 89515.</P
></DD
><DT
><B
CLASS="COMMAND"
>Lucky Numbers</B
></DT
><DD
><P
>A "lucky number" is one whose individual digits add
	      up to 7, in successive additions. For example, 62431 is a
	      "lucky number" (6 + 2 + 4 + 3 + 1 = 16, 1 + 6 = 7). Find
	      all the "lucky numbers" between 1000 and 10000.</P
></DD
><DT
><B
CLASS="COMMAND"
>Alphabetizing a String</B
></DT
><DD
><P
>Alphabetize (in ASCII order) an arbitrary string
	      read from the command line.</P
></DD
><DT
><B
CLASS="COMMAND"
>Parsing</B
></DT
><DD
><P
>Parse <TT
CLASS="FILENAME"
>/etc/passwd</TT
>, and output
	      its contents in nice, easy-to-read tabular form.</P
></DD
><DT
><B
CLASS="COMMAND"
>Logging Logins</B
></DT
><DD
><P
>Parse <TT
CLASS="FILENAME"
>/var/log/messages</TT
> to
	      produce a nicely formatted file of user logins and login
	      times. The script may need to run as
	      <I
CLASS="FIRSTTERM"
>root</I
>. (Hint: Search for the string
	      <SPAN
CLASS="QUOTE"
>"LOGIN."</SPAN
>)</P
></DD
><DT
><B
CLASS="COMMAND"
>Pretty-Printing a Data File</B
></DT
><DD
><P
>Certain database and spreadsheet packages use save-files
	      with <I
CLASS="FIRSTTERM"
>comma-separated values</I
>
	      (CSVs). Other applications often need to parse these
	      files.</P
><P
>Given a data file with comma-separated fields,
	      of the form:
	        <TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;Jones,Bill,235 S. Williams St.,Denver,CO,80221,[phone]
   2&nbsp;Smith,Tom,404 Polk Ave.,Los Angeles,CA,90003,[phone]
   3&nbsp;...</PRE
></TD
></TR
></TABLE
>
	      Reformat the data and print it out to
	      <TT
CLASS="FILENAME"
>stdout</TT
> in labeled, evenly-spaced columns.</P
></DD
><DT
><B
CLASS="COMMAND"
>Justification</B
></DT
><DD
><P
>Given ASCII text input either from
	      <TT
CLASS="FILENAME"
>stdin</TT
> or a file, adjust
	      the word spacing to right-justify each line to a
	      user-specified line-width, then send the output to
	      <TT
CLASS="FILENAME"
>stdout</TT
>.</P
></DD
><DT
><B
CLASS="COMMAND"
>Mailing List</B
></DT
><DD
><P
>Using the <A
HREF="communications.html#COMMMAIL1"
>mail</A
> command,
	      write a script that manages a simple mailing list. The
	      script automatically e-mails the monthly company newsletter,
	      read from a specified text file, and sends it to all the
	      addresses on the mailing list, which the script reads from
	      another specified file.</P
></DD
><DT
><B
CLASS="COMMAND"
>Generating Passwords</B
></DT
><DD
><P
>Generate pseudorandom 8-character passwords, using
	      characters in the ranges [0-9], [A-Z], [a-z]. Each password
	      must contain at least two digits.</P
></DD
><DT
><B
CLASS="COMMAND"
>Monitoring a User</B
></DT
><DD
><P
>You suspect that one particular user on the network
	      has been abusing his privileges and possibly attempting to
	      hack the system. Write a script to automatically monitor
	      and log his activities when he's signed on. The log file
	      will save entries for the previous week, and delete those
	      entries more than seven days old.</P
><P
>You may use <A
HREF="system.html#LASTREF"
>last</A
>,
             <A
HREF="system.html#LASTLOGREF"
>lastlog</A
>, and <A
HREF="system.html#LASTCOMMREF"
>lastcomm</A
> to aid your
             surveillance of the suspected malefactor.</P
></DD
><DT
><B
CLASS="COMMAND"
>Checking for Broken Links</B
></DT
><DD
><P
>Using <A
HREF="communications.html#LYNXREF"
>lynx</A
> with the
	      <TT
CLASS="OPTION"
>-traversal</TT
> option, write a script that
	      checks a Web site for broken links.</P
></DD
></DL
></DIV
><DIV
CLASS="VARIABLELIST"
><P
><B
><A
NAME="EXDIFFICULT1"
></A
>DIFFICULT</B
></P
><DL
><DT
><B
CLASS="COMMAND"
>Testing Passwords</B
></DT
><DD
><P
>Write a script to check and validate passwords. The object
	      is to flag <SPAN
CLASS="QUOTE"
>"weak"</SPAN
> or easily guessed password
	      candidates.</P
><P
>A trial password will be input to the script as a
	       command line parameter. To be considered acceptable,
	       a password must meet the following minimum qualifications:

            <UL
><LI
><P
>Minimum length of 8 characters</P
></LI
><LI
><P
>Must contain at least one numeric character</P
></LI
><LI
><P
>Must contain at least one of the following
		  non-alphabetic characters: <SPAN
CLASS="TOKEN"
>@</SPAN
>,
		  <SPAN
CLASS="TOKEN"
>#</SPAN
>, <SPAN
CLASS="TOKEN"
>$</SPAN
>, <SPAN
CLASS="TOKEN"
>%</SPAN
>,
		  <SPAN
CLASS="TOKEN"
>&#38;</SPAN
>, <SPAN
CLASS="TOKEN"
>*</SPAN
>, <SPAN
CLASS="TOKEN"
>+</SPAN
>,
		  <SPAN
CLASS="TOKEN"
>-</SPAN
>, <SPAN
CLASS="TOKEN"
>=</SPAN
></P
></LI
></UL
></P
><P
>Optional:

            <UL
><LI
><P
>Do a dictionary check on every sequence of at least
		  four consecutive alphabetic characters in the password under
		  test. This will eliminate passwords containing embedded
		  <SPAN
CLASS="QUOTE"
>"words"</SPAN
> found in a standard dictionary.</P
></LI
><LI
><P
>Enable the script to check all the passwords on your
		  system. These may or may not reside in
		  <TT
CLASS="FILENAME"
>/etc/passwd</TT
>.</P
></LI
></UL
></P
><P
>This exercise tests mastery of <A
HREF="regexp.html#REGEXREF"
>Regular Expressions</A
>.</P
></DD
><DT
><B
CLASS="COMMAND"
>Cross Reference</B
></DT
><DD
><P
>Write a script that generates a
	      <I
CLASS="FIRSTTERM"
>cross-reference</I
>
	      (<I
CLASS="FIRSTTERM"
>concordance</I
>) on a target file.
	      The output will be a listing of all word occurrences in
	      the target file, along with the line numbers in which
	      each word occurs. Traditionally, <I
CLASS="FIRSTTERM"
>linked
	      list</I
> constructs would be used in such
	      applications. Therefore, you should investigate <A
HREF="arrays.html#ARRAYREF"
>arrays</A
> in the course of
	      this exercise. <A
HREF="textproc.html#WF"
>Example 15-12</A
> is probably
	      <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>not</I
></SPAN
> a good place to start.</P
></DD
><DT
><A
NAME="NEWTONSQRT"
></A
><B
CLASS="COMMAND"
>Square Root</B
></DT
><DD
><P
>Write a script to calculate square roots of numbers
              using <I
CLASS="FIRSTTERM"
>Newton's Method</I
>.</P
><P
>The algorithm for this, expressed as a snippet of Bash
              <A
HREF="assortedtips.html#PSEUDOCODEREF"
>pseudo-code</A
> is:</P
><P
><TABLE
BORDER="0"
BGCOLOR="#E0E0E0"
WIDTH="90%"
><TR
><TD
><PRE
CLASS="PROGRAMLISTING"
>   1&nbsp;#  (Isaac) Newton's Method for speedy extraction
   2&nbsp;#+ of square roots.
   3&nbsp;
   4&nbsp;guess = $argument
   5&nbsp;#  $argument is the number to find the square root of.
   6&nbsp;#  $guess is each successive calculated "guess" -- or trial solution --
   7&nbsp;#+ of the square root.
   8&nbsp;#  Our first "guess" at a square root is the argument itself.
   9&nbsp;
  10&nbsp;oldguess = 0
  11&nbsp;# $oldguess is the previous $guess.
  12&nbsp;
  13&nbsp;tolerance = .000001
  14&nbsp;# To how close a tolerance we wish to calculate.
  15&nbsp;
  16&nbsp;loopcnt = 0
  17&nbsp;# Let's keep track of how many times through the loop.
  18&nbsp;# Some arguments will require more loop iterations than others.
  19&nbsp;
  20&nbsp;
  21&nbsp;while [ ABS( $guess $oldguess ) -gt $tolerance ]
  22&nbsp;#       ^^^^^^^^^^^^^^^^^^^^^^^ Fix up syntax, of course.
  23&nbsp;
  24&nbsp;#      "ABS" is a (floating point) function to find the absolute value
  25&nbsp;#+      of the difference between the two terms.
  26&nbsp;#             So, as long as difference between current and previous
  27&nbsp;#+            trial solution (guess) exceeds the tolerance, keep looping.
  28&nbsp;
  29&nbsp;do
  30&nbsp;   oldguess = $guess  # Update $oldguess to previous $guess.
  31&nbsp;
  32&nbsp;#  =======================================================
  33&nbsp;   guess = ( $oldguess + ( $argument / $oldguess ) ) / 2.0
  34&nbsp;#        = 1/2 ( ($oldguess **2 + $argument) / $oldguess )
  35&nbsp;#  equivalent to:
  36&nbsp;#        = 1/2 ( $oldguess + $argument / $oldguess )
  37&nbsp;#  that is, "averaging out" the trial solution and
  38&nbsp;#+ the proportion of argument deviation
  39&nbsp;#+ (in effect, splitting the error in half).
  40&nbsp;#  This converges on an accurate solution
  41&nbsp;#+ with surprisingly few loop iterations . . .
  42&nbsp;#+ for arguments &#62; $tolerance, of course.
  43&nbsp;#  =======================================================
  44&nbsp;
  45&nbsp;   (( loopcnt++ ))     # Update loop counter.
  46&nbsp;done</PRE
></TD
></TR
></TABLE
></P
><P
>It's a simple enough recipe, and
              <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>seems</I
></SPAN
> at first glance easy enough to
              convert into a working Bash script. The problem, though,
              is that Bash has <A
HREF="operations.html#NOFLOATINGPOINT"
>no native
              support for floating point numbers</A
>. So, the script
              writer needs to use <A
HREF="mathc.html#BCREF"
>bc</A
> or
              possibly <A
HREF="awk.html#AWKREF"
>awk</A
> to convert the
              numbers and do the calculations. It may get rather messy
              . . .</P
></DD
><DT
><B
CLASS="COMMAND"
>Logging File Accesses</B
></DT
><DD
><P
>Log all accesses to the files in <TT
CLASS="FILENAME"
>/etc</TT
> during the course of
	      a single day. This information should include the filename,
	      user name, and access time. If any alterations to the
	      files take place, that should be flagged. Write this data
	      as neatly formatted records in a logfile.</P
></DD
><DT
><B
CLASS="COMMAND"
>Monitoring Processes</B
></DT
><DD
><P
>Write a script to continually monitor all running
	      processes and to keep track of how many child processes each
	      parent spawns. If a process spawns more than five children,
	      then the script sends an e-mail to the system administrator
	      (or <I
CLASS="FIRSTTERM"
>root</I
>) with all relevant
	      information, including the time, PID of the parent, PIDs
	      of the children, etc. The script appends a report to a log
	      file every ten minutes.  </P
></DD
><DT
><B
CLASS="COMMAND"
>Strip Comments</B
></DT
><DD
><P
>Strip all comments from a shell script whose name
	      is specified on the command line. Note that the <SPAN
CLASS="QUOTE"
>"#!
	      line"</SPAN
> must not be stripped out.</P
></DD
><DT
><B
CLASS="COMMAND"
>Strip HTML Tags</B
></DT
><DD
><P
>Strip all HTML tags from a specified HTML file, then
	      reformat it into lines between 60 and 75 characters
	      in length. Reset paragraph and block spacing, as
	      appropriate, and convert HTML tables to their approximate
	      text equivalent.</P
></DD
><DT
><B
CLASS="COMMAND"
>XML Conversion</B
></DT
><DD
><P
>Convert an XML file to both HTML and text format.</P
></DD
><DT
><A
NAME="CSPAMMERS"
></A
><B
CLASS="COMMAND"
>Chasing Spammers</B
></DT
><DD
><P
> Write a script that analyzes a spam e-mail by doing
	      DNS lookups on the IP addresses in the headers to identify
	      the relay hosts as well as the originating ISP. The
	      script will forward the unaltered spam message to the
	      responsible ISPs. Of course, it will be necessary to
	      filter out <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>your own ISP's IP address</I