Let's face it; Perl is hard. So, this book provides an easy, straightforward introduction to Perl with scads and scads of examples. This book goes far beyond the Perl basics, tackling the tough issues that you'll need to solve to get your work done. I tell you where Perl really is evil and how you can get around it. I also provide a lot of handy Perl routines you can reuse repeatedly, not only as examples to learn from but in your actual work. And, I try to shed some light on the confusion that is Perl. Perl doesn't have to be difficult if you avoid the confusing constructs.

If you've ever been intimidated by Perl, this is the book for you.

Cross-Platform Perl was the first book to cover Perl on both UNIX and Windows as well as using Perl/Tk for creating graphical user interfaces. The newly-expanded second edition includes new chapters, more coverage of Windows and Linux, and a fully-updated text. You'll see a lot more on networking, Windows-specific modules, Linux, and XML.

• Using Perl for Cross-Platform Development
• When to use Perl
• Perl Versions
• Acquiring Perl
• Installing Perl from the CD-ROMs
• Learning Perl
• Contacting the Author

• A First Perl Script
• Creating Perl Scripts
• Editing On Windows
• Editing on UNIX and Linux
• The Perl Interpreter
• Verifying Perl is Installed
• Running Perl Programs
• The First Program in Depth
• The print Command
• Getting Input In Perl Scripts
• Perl Variables
• Standard Input
• Chomping the Data
• Using If to Control the Flow of Your Scripts
• If Expressions
• How Perl Runs Programs
• Problems with Perl Scripts
• Finding Out More About Perl's Commands
• Making Your Scripts Into Commands
• Maning an Executable Script on UNIX and Linux
• Launching Perl Scripts on Windows

• The Perl You Need to Know
• Controlling Your Perl Scripts
• The while command
• The until statement
• The unless statement
• The for loop
• The foreach statement
• The continue block
• Extra Flow Control
• Ending Modifiers
• Math Operations
• Floating Point
• Hexadecimal and octal numbers
• Math Functions
• Operations On Text
• The Difference Between Single
  and Double Quotes
• Problems With Quotes
• String Functions
• String Operators
• Variables and Arrays
• Scalar Variables
• Arrays
• Associative Arrays Or Hashes
• Built-In Variables
• Command-Line Arguments
• Subroutines
• The sub Statement
• Returning Data From Subroutines
• Accessing Variables in Subroutines
• Local Variables in Subroutines
• Passing Parameters to Subroutines
• Changing the Value of
  Subroutine Parameters
• Returning Arrays From Subroutines

• Files
• Opening Files and File Handles
• Or Else, Die!
• Writing Files
• Reading Files
• Using Directories
• Reading the Contents of a Directory
• Creating, Removing, and Changing Directories
• Getting Information on a File
• Testing Files
• Manipulating Files in a Directory
• Symbolic Links
• Packages for Working With Files and Directories
• Copying Files
• Extracting File Names From Paths
• Detecting the Current Directory
• Setting File Attributes on Windows
• Getting Information on Modules

• Searching for Data in Text
• Regular Expressions
• Searching From the Beginning and the End
• Checking For Multiple Occurrences
• Greedy Searches
• Changing the Delimiter
• Pattern-Matching Options
• Using Patterns with Substitutions
• Controlling the Delimiter
• Extracting Data with Substitutions
• Transforming Data with tr
• Using tr to Convert File Names

• Understanding Formats
• Using Formats
• Defining Formats
• Page Headers
• Controlling Pagination
• Selecting File Handles
• Other Means to Control Data Output
• Using sprintf
• Putting the Date and Time Into Your Reports
• Databases and Perl
• DBM Databases
• Using AnyDBM_File
• Connecting Other Databases to Perl

• Packages
• Inside a Package
• Modules
• Extensions
• Common Packages, Modules,
  and Extensions
• Using Modules
• References and Objects
• Object-Oriented Scripting
• The English Module
• Downloading and Installing Modules
• Using the CPAN Module to
  Download and Install Modules
• Downloading and Installing Windows Modules
• Detecting Whether Your Script is
  Running on Windows or UNIX
• Checking $^O
• Using the Config Module
• Checking Environment Variables

• Launching Processes
• Calling exec
• Calling system
• Using Backticks to Run Programs
• Controlling the Command Line with eval

• Using start
• Using Win32::Spawn
• Creating Processes with Win32::Process
• Launching Perl Scripts as Batch Files
• Creating Shortcuts

• Getting Information on Windows
• Network Administration
• Domain Controllers
• Auto-Discovery
• Network Shares
• Working With Users
• Login Names
• Listing All Users
• The Windows Registry
• Reading the Registry
• Reading the Registry Remotely
• Modifying the Registry
• Controlling and Monitoring
  Windows Services
• The Win32::Service Module
• Checking the Status Of Services
• Checking the Status Of Services
  on Remote Systems
• Stopping and Starting Services
• Checking the Windows Event Log
• Other Administration Tasks

• Accessing the Windows Clipboard
• OLE Automation With Perl
• The Win32::OLE Module
• Scripting Microsoft Excel
• Loading In Files From Disk
• Scripting Microsoft Word

• THe UNIX Process Model
• Launching Programs on UNIX and Linux
• Waiting for Child Processes To exit
• Controlling Processes
• Killing Processes
• Piping Data Between Processes
• Alarm Clocks and Sleeping
• Handling Command-Line Parameters
• Controlling the Variables
  Used by the Getopt::Std Module
• Supporting More Than Single-
  Letter Command-Line Switches

• Working With Users
• Login Names
• UNIX and Linux Password File Data
• Modules to Help You Access User Information
• Parsing Log Files
• Monitoring Your System
• Checking the List of Processes
• Using SNMP
• Other Administration Tasks

• Controlling Command-Line Parameters
• Scripting Command-Line Based Applications
• Using the Net::Telnet Module

• How Web pages Work
• The Common Gateway Interface
• The CGI URL
• How CGI Scripts Work
• HTML, The HyperText Markup Language
• Using Perl for CGI Scripts
• Creating a Web Page from a Script
• Launching the Perl Script
  from Your Web Browser
• Security and CGI Scripts
• Using the CGI Module
• Testing the Script
• Data-Entry Forms in Web Pages
• Web Forms
• A Perl Script to Process a Form
• More Data-Entry Options
• Scrolled Lists and Option Menus
• Entering More Text
• Adding a Reset Button
• Querying All the Parameters
• XML: The Future of the Web?
• RDF: XML and Data Channels
• Making MyPerl? Writing Code
  to Handle XML Channel Data
• Converting Channel Data to
  HTML For Display
• The Self-Maintaining Web Page
• Building a Web Page from Major Sections
• Generating a Table of
  Contents from the Topics
• Allowing Users to Add Topics
• Responding to the User's Input
• Creating New Topics
• Generating a Thank-You List
• Putting It All Together: The
  Self-Maintaining Web Page
• Regenerating the Web Page

• Automating the Software Development Process
• Building Your Software With Help From Perl
• Sending E-Mail When A Task Completes
• Automatically Rebuilding
  When the Code Changes
• Making Makefiles
• Using Perl in Your Software Environment
• Automated Testing
• Scanning for Include Files
  in C and C++ Programs
• Generating Documentation with Perl
• Generating Code with Perl
• Perl Modules to Access Source
  Code Control Systems

• Networking with Perl
• Internet Protocols
• Socket-Based Communication
• Using the Socket Module
• Building a Server
• Internet Addresses
• Checking for An Open Port
• Building a Client
• Testing the Client and Server Scripts
• Perl Modules for Networking with Sockets
• Using the IO::Socket Module
• Getting the Hostname
• Perl Modules for Network Protocols
• File Transfers with the Net::FTP Module

• Adding Graphical Interfaces
  to Your Perl Scripts
• Event-Driven Scripts
• Using Callbacks to Execute Perl Code
• Tk Widgets
• acquiring and Installing Perl/Tk
• Scripting with Perl/Tk
• The Main Window
• Creating Widgets
• Changing Widget Attributes
• Choosing Colors
• Making Widgets Appear and
  Controlling Widget Placement
• Using pack to Control Widget Placement
• Using grid to Control Widget Placement
• Adding More Widgets
• Menus and the Menu Bar
• MAnaging The Status Area
• Creating a Scrolled Text Widget
• Opening Files with a File Dialog
• Single-Line Text Entry
• For More Information on Perl/Tk

Index
License Agreement
GNU General Public License
CD-ROM Installation Instructions

Thanks to Glenn Carr, Tom Grydeland, Bruce Murray, Patrick Riley, and Kathy Smart for pointing them out. If you have found any problems, please send me a message. Thanks.

Page 9 Contacting the Author

I've moved my Web pages to:

http://foster-johnson.com

Page 39-40 Code examples for Ending Modifiers

The code examples shouldn't have the semi colons right after the statement. These code examples from pages 39 (the end) and 40 (the top) should read:

            
                statement if condition;
            
                statement unless condition;
                statement while condition;
                statement until condition;
                

Page 56 correction for typo at the end of the page

The last text paragraph should read:

To access individual values in an associative array, you use a different syntax. Any value in an associative array can be accessed through its key name, using the following syntax:

                $assoc{keyname}
                

Page 57 in keys.pl

Kathy Smart reports that the keys.pl script won't work with Windows NT 4.0 with IIS 3.0 unless you rename it to something like keys2.pl. if anyone has any more information on why keys.pl would be a reserved script name, please send me a message. Thanks.

Page 106 Table 4.2

/[^aA]/ returns true if any character is found that is not a or A.

Page 107 Table 4.4

/a\b/ Match a at end of word.

/a\B/ Match a not at end of word.

Page 107 Warning Icon

/[^a]/ will return true if there is any character not a anywhere in the word.

Pages 109, 111, 116 in pattern.pl and path.pl

While you can use | as a delimiter for both m and s, it is a special character for regular expressions, so you probably don't want to do this. A better approach would be to use open and close delimiters like the following:

            m<^/usr/local>
                

or:

            
            $path =~ s</usr/bin> {/usr/local/bin};
                

Page 117 in tr code example.

With the following statement:

            $newname =~ tr/[A-Z]/[a-z]/;
                

You do not need the brackets in tr expressions, so the code should look like:

            $newname =~ tr/A-Z/a-z/;
                    

You should make similar changes to tolower.pl on page and toupper.pl on page 121.

Page 112 in Table 4.6

A better explanation for the m and s options comes from Tom Grydeland:

/pattern/m	Allows ^ to match after every newline and $ to match before every newline in the string being searched. This makes "beginning of line" and "end of line" meaningful, even in multi-line strings.
/pattern/s	Allows . to match newline. This is useful for matching a pattern with . in a multi-line string as if it were a single line.

Page 141 in time1.pl

Comment out the line that increments the variable hour, as shown below (with the changed line in bold):

            
#
# Using time and localtime.
#
$t = time();

($sec, $min, $hour, $dom, $mon,
 $year,$wday, $yday, $isdst) =
  localtime($t);

# Convert data to normal values.
 # Comment out the next line.
 #$hour++;   # from 0..23  

$year += 1900;

# Provide English equivalents.
@months = ("January", "February",
           "March", "April", "May",
           "June", "July", "August",
           "September", "October",
           "November", "December");

@week = ("Sunday", "Monday", "Tuesday",
         "Wednesday", "Thursday",
         "Friday", "Saturday");

# Print data.


printf ("Time is: %2.2d:%2.2d:%2.2d\n",
    $hour, $min, $sec);

printf("Date is: %s, %d-%s-%d\n",
    $week[$wday], $dom, $months[$mon],
    $year);

printf("%d days since 1 January\n",
    $yday);

if ($isdst) {
    print "Is daylight savings time.\n";
} else {
    print "No daylight savings time.\n";
}

# time1.pl
            

Page 157 in backtick.pl

You need to remove the extra newline returned from the pwd command. Change the backtick.pl script to the following:

#
# Using back-ticks to
# capture program output.
#
$dir = `pwd`;
chomp($dir);

print "Directory: $dir\n";

# backtick.pl
            
                    

Pages 158-160, with fork

The following line is wrong and will actually undefine the variable (dumb typo on my part):

} elsif (undef $pid) {
        

Change this (on all the pages mentioned above) to:

} elsif (! defined $pid) {
        

Pages 162-163

I should insert a note that it's better to use the UNIX forward slash for most Windows path names. The double backslash is a bit tedious, but I want to stress that users will expect to enter the backslash directory separator and Perl coders need to deal with this.

Pages 162-163, in winapp.pl

You need to add use Win32::WinError; and call Win32::WinError::GetLastError instead of Win32::GetLastError, as shown below (with the changed lines in bold):

Note: I just tried this with Perl for Win32 build 306 and the original script worked. So, try the script first. If you have problems, try the changes as listed.

#
# Launch a Windows process from Win32 module.
# See online docs for more on this subject.
#
use Win32;
use Win32::Process;
use Win32::WinError;

#Create the process object.
Win32::Process::Create($ProcessObj,
    "C:\\WINNT35.0\\NOTEPAD.EXE",
    "NOTEPAD",
    0, # Don't inherit.
    DETACHED_PROCESS,
    ".") ||  # current dir.
    die &print_error;

#Wait for the process to end. No timeout.
$ProcessObj->Wait(INFINITE) ||
    warn &print_error;

$ProcessObj->GetExitCode($ExitCode) ||
    warn &print_error;

print "Notepad exited with $ExitCode\n";
   print Win32::FormatMessage( Win32::WinError::GetLastError() );

sub print_error {
    print Win32::FormatMessage( Win32::WinError::GetLastError() );
}

# winapp.pl
            

Page 169, in config.pl

This should also check for Solaris and SunOS, two common flavors of UNIX.

Page 189, The file msexcel.pl is missing from the CD-ROM

Sorry about that! Thanks to Bruce Murray for pointing this out. Here is the file:

            
#
# Use OLE to control MS Excel.
#
use OLE;

# Create an instance of Excel.
$app = CreateObject OLE "Excel.Application" ||
    die "Unable to open Excel.";

# Make application visible.
$app->{'Visible'} = 1;

# Create a new workbook.
$app->Workbooks->Add();

# Set values in a "range".
$app->Range("A1")->{'Value'} = "Developer";
$app->Range("B1")->{'Value'} = "Bugs";

$app->Range("B2")->{'Value'} = "Assigned";
$app->Range("C2")->{'Value'} = "Fixed";

$app->Range("A3")->{'Value'} = "Eric F-J.";
$app->Range("B3")->{'Value'} = 10;
$app->Range("C3")->{'Value'} = 10;

# Leave Excel running. Use $app->Quit() to exit.

# msexcel.pl

            

Page 264 Missing part of a sentence.

If you're using an integrated environment, such as Microsoft Visual C++ or SunSoft Workshop, you still need to initiate the environment to rebuild the application.

Page 337, the home page for the Win32 port of Perl

This has moved to: http://www.activestate.com