=head1 NAME

CGI - Handle Common Gateway Interface requests and responses

=for html
<a href='https://travis-ci.org/leejo/CGI.pm?branch=master'><img src='https://travis-ci.org/leejo/CGI.pm.svg?branch=master' alt='Build Status' /></a>
<a href='https://coveralls.io/r/leejo/CGI.pm'><img src='https://coveralls.io/repos/leejo/CGI.pm/badge.png?branch=master' alt='Coverage Status' /></a>

=head1 SYNOPSIS

    use strict;
    use warnings;

    use CGI;

	# create a CGI object (query) for use
    my $q = CGI->new;

    # Process an HTTP request
    my @values  = $q->multi_param('form_field');
    my $value   = $q->param('param_name');

    my $fh      = $q->upload('file_field');

    my $riddle  = $q->cookie('riddle_name');
    my %answers = $q->cookie('answers');

    # Prepare various HTTP responses
    print $q->header();
    print $q->header('application/json');

    my $cookie1 = $q->cookie(
        -name  => 'riddle_name',
        -value => "The Sphynx's Question"
    );

    my $cookie2 = $q->cookie(
        -name  => 'answers',
        -value => \%answers
    );

    print $q->header(
        -type    => 'image/gif',
        -expires => '+3d',
        -cookie  => [ $cookie1,$cookie2 ]
    );

    print $q->redirect('http://somewhere.else/in/movie/land');

=head1 DESCRIPTION

CGI.pm is a stable, complete and mature solution for processing and preparing
HTTP requests and responses. Major features including processing form
submissions, file uploads, reading and writing cookies, query string generation
and manipulation, and processing and preparing HTTP headers.

CGI.pm performs very well in a vanilla CGI.pm environment and also comes
with built-in support for mod_perl and mod_perl2 as well as FastCGI.

It has the benefit of having developed and refined over 20 years with input
from dozens of contributors and being deployed on thousands of websites.
CGI.pm was included in the perl distribution from perl v5.4 to v5.20, however
is has now been removed from the perl core...

=head1 CGI.pm HAS BEEN REMOVED FROM THE PERL CORE

L<http://perl5.git.perl.org/perl.git/commitdiff/e9fa5a80>

If you upgrade to a new version of perl or if you rely on a
system or vendor perl and get an updated version of perl through a system
update, then you will have to install CGI.pm yourself with cpan/cpanm/a vendor
package/manually. To make this a little easier the L<CGI::Fast> module has been
split into its own distribution, meaning you do not need access to a compiler
to install CGI.pm

The rationale for this decision is that CGI.pm is no longer considered good
practice for developing web applications, B<including> quick prototyping and
small web scripts. There are far better, cleaner, quicker, easier, safer,
more scalable, more extensible, more modern alternatives available at this point
in time. These will be documented with L<CGI::Alternatives>.

For more discussion on the removal of CGI.pm from core please see:

L<http://www.nntp.perl.org/group/perl.perl5.porters/2013/05/msg202130.html>

Note that the v4 releases of CGI.pm will retain back compatibility B<as much>
B<as possible>, however you may need to make some minor changes to your code
if you are using deprecated methods or some of the more obscure features of the
module. If you plan to upgrade to v4.00 and beyond you should read the Changes
file for more information and B<test your code> against CGI.pm before deploying
it.

=head1 HTML Generation functions should no longer be used

B<All> HTML generation functions within CGI.pm are no longer being
maintained. Any issues, bugs, or patches will be rejected unless
they relate to fundamentally broken page rendering.

The rationale for this is that the HTML generation functions of CGI.pm
are an obfuscation at best and a maintenance nightmare at worst. You
should be using a template engine for better separation of concerns.
See L<CGI::Alternatives> for an example of using CGI.pm with the
L<Template::Toolkit> module.

These functions, and perldoc for them, are considered deprecated, they
are no longer being maintained and no fixes or features for them will be
accepted. They will, however, continue to exist in CGI.pm without any
deprecation warnings ("soft" deprecation) so you can continue to use
them if you really want to. All documentation for these functions has
been moved to L<CGI::HTML::Functions>.

=head1 Programming style

There are two styles of programming with CGI.pm, an object-oriented (OO)
style and a function-oriented style. You are recommended to use the OO
style as CGI.pm will create an internal default object when the functions
are called procedurally and you will not have to worry about method names
clashing with perl builtins.

In the object-oriented style you create one or more CGI objects and then
use object methods to create the various elements of the page. Each CGI
object starts out with the list of named parameters that were passed to
your CGI script by the server. You can modify the objects, save them to a
file or database and recreate them. Because each object corresponds to the
"state" of the CGI script, and because each object's parameter list is
independent of the others, this allows you to save the state of the
script and restore it later.

For example, using the object oriented style:

    #!/usr/bin/env perl

    use strict;
    use warnings;

    use CGI;                             # load CGI routines

    my $q = CGI->new;                    # create new CGI object
    print $q->header;                    # create the HTTP header

In the function-oriented style, there is one default CGI object that
you rarely deal with directly. Instead you just call functions to
retrieve CGI parameters, manage cookies, and so on. The following example
is identical to above, in terms of output, but uses the function-oriented
interface. The main differences are that we now need to import a set of
functions into our name space (usually the "standard" functions), and we don't
need to create the CGI object.

    #!/usr/bin/env perl

    use strict;
    use warnings;

    use CGI qw/:standard/;           # load standard CGI routines
    print header();                  # create the HTTP header

The examples in this document mainly use the object-oriented style. See HOW
TO IMPORT FUNCTIONS for important information on function-oriented programming
in CGI.pm

=head2 Calling CGI.pm routines

Most CGI.pm routines accept several arguments, sometimes as many as 20
optional ones! To simplify this interface, all routines use a named
argument calling style that looks like this:

    print $q->header(
        -type    => 'image/gif',
        -expires => '+3d',
    );

Each argument name is preceded by a dash. Neither case nor order matters in
the argument list: -type, -Type, and -TYPE are all acceptable. In fact, only
the first argument needs to begin with a dash. If a dash is present in the
first argument CGI.pm assumes dashes for the subsequent ones.

Several routines are commonly called with just one argument. In the case
of these routines you can provide the single argument without an argument
name. header() happens to be one of these routines. In this case, the single
argument is the document type.

    print $q->header('text/html');

Other such routines are documented below.

Sometimes named arguments expect a scalar, sometimes a reference to an array,
and sometimes a reference to a hash. Often, you can pass any type of argument
and the routine will do whatever is most appropriate. For example, the param()
routine is used to set a CGI parameter to a single or a multi-valued value.
The two cases are shown below:

    $q->param(
        -name  => 'veggie',
        -value => 'tomato',
    );

    $q->param(
        -name  => 'veggie',
        -value => [ qw/tomato tomahto potato potahto/ ],
    );


Many routines will do something useful with a named argument that it doesn't
recognize. For example, you can produce non-standard HTTP header fields by
providing them as named arguments:

    print $q->header(
        -type            => 'text/html',
        -cost            => 'Three smackers',
        -annoyance_level => 'high',
        -complaints_to   => 'bit bucket',
    );

This will produce the following nonstandard HTTP header:

    HTTP/1.0 200 OK
    Cost: Three smackers
    Annoyance-level: high
    Complaints-to: bit bucket
    Content-type: text/html

Notice the way that underscores are translated automatically into hyphens.

=head2 Creating a new query object (object-oriented style)

    my $q = CGI->new;

This will parse the input (from POST, GET and DELETE methods) and store
it into a perl5 object called $q. Note that because the input parsing
happens at object instantiation you have to set any CGI package variables
that control parsing B<before> you call CGI->new.

Any filehandles from file uploads will have their position reset to the
beginning of the file.

=head2 Creating a new query object from an input file

    my $q = CGI->new( $input_filehandle );

If you provide a file handle to the new() method, it will read parameters
from the file (or STDIN, or whatever). The file can be in any of the forms
describing below under debugging (i.e. a series of newline delimited
TAG=VALUE pairs will work). Conveniently, this type of file is created by
the save() method (see below). Multiple records can be saved and restored.

Perl purists will be pleased to know that this syntax accepts references to
file handles, or even references to filehandle globs, which is the "official"
way to pass a filehandle. You can also initialize the CGI object with a
FileHandle or IO::File object.

If you are using the function-oriented interface and want to initialize CGI
state from a file handle, the way to do this is with B<restore_parameters()>.
This will (re)initialize the default CGI object from the indicated file handle.

    open( my $in_fh,'<',"test.in") || die "Couldn't open test.in for read: $!";
    restore_parameters( $in_fh );
    close( $in_fh );

You can also initialize the query object from a hash reference:

    my $q = CGI->new( {
        'dinosaur' => 'barney',
        'song'     => 'I love you',
        'friends'  => [ qw/ Jessica George Nancy / ]
    } );

or from a properly formatted, URL-escaped query string:

    my $q = CGI->new('dinosaur=barney&color=purple');

or from a previously existing CGI object (currently this clones the parameter
list, but none of the other object-specific fields, such as autoescaping):

    my $old_query = CGI->new;
    my $new_query = CGI->new($old_query);

To create an empty query, initialize it from an empty string or hash:

    my $empty_query = CGI->new("");

       -or-

    my $empty_query = CGI->new({});

=head2 Fetching a list of keywords from the query

    my @keywords = $q->keywords

If the script was invoked as the result of an ISINDEX search, the parsed
keywords can be obtained as an array using the keywords() method.

=head2 Fetching the names of all the parameters passed to your script

    my @names = $q->multi_param

    my @names = $q->param

If the script was invoked with a parameter list
(e.g. "name1=value1&name2=value2&name3=value3"), the param() / multi_param()
methods will return the parameter names as a list. If the script was invoked
as an ISINDEX script and contains a string without ampersands
(e.g. "value1+value2+value3"), there will be a single parameter named
"keywords" containing the "+"-delimited keywords.

The array of parameter names returned will be in the same order as they were
submitted by the browser. Usually this order is the same as the order in which
the parameters are defined in the form (however, this isn't part of the spec,
and so isn't guaranteed).

=head2 Fetching the value or values of a single named parameter

    my @values = $q->multi_param('foo');

        -or-

    my $value = $q->param('foo');

        -or-

    my @values = $q->param('foo'); # list context, discouraged and will raise
                                   # a warning (use ->multi_param instead)

Pass the param() / multi_param() method a single argument to fetch the value
of the named parameter. When calling param() If the parameter is multivalued
(e.g. from multiple selections in a scrolling list), you can ask to receive
an array. Otherwise the method will return the B<first> value.

B<Warning> - calling param() in list context can lead to vulnerabilities if
you do not sanitise user input as it is possible to inject other param
keys and values into your code. This is why the multi_param() method exists,
to make it clear that a list is being returned, note that param() can still
be called in list context and will return a list for back compatibility.

The following code is an example of a vulnerability as the call to param will
be evaluated in list context and thus possibly inject extra keys and values
into the hash:

    my %user_info = (
        id   => 1,
        name => $q->param('name'),
    );

The fix for the above is to force scalar context on the call to ->param by
prefixing it with "scalar"

    name => scalar $q->param('name'),

If you call param() in list context with an argument a warning will be raised
by CGI.pm, you can disable this warning by setting $CGI::LIST_CONTEXT_WARN to 0
or by using the multi_param() method instead

If a value is not given in the query string, as in the queries "name1=&name2=",
it will be returned as an empty string.

If the parameter does not exist at all, then param() will return undef in scalar
context, and the empty list in a list context.

=head2 Setting the value(s) of a named parameter

    $q->param('foo','an','array','of','values');

This sets the value for the named parameter 'foo' to an array of values. This
is one way to change the value of a field AFTER the script has been invoked
once before.

param() also recognizes a named parameter style of calling described in more
detail later:

    $q->param(
        -name   => 'foo',
        -values => ['an','array','of','values'],
    );

                -or-

    $q->param(
        -name  => 'foo',
        -value => 'the value',
    );

=head2 Appending additional values to a named parameter

    $q->append(
        -name   =>'foo',
        -values =>['yet','more','values'],
    );

This adds a value or list of values to the named parameter. The values are
appended to the end of the parameter if it already exists. Otherwise the
parameter is created. Note that this method only recognizes the named argument
calling syntax.

=head2 Importing all parameters into a namespace

    $q->import_names('R');

This creates a series of variables in the 'R' namespace. For example, $R::foo,
@R:foo. For keyword lists, a variable @R::keywords will appear. If no namespace
is given, this method will assume 'Q'. B<WARNING>: don't import anything into
'main'; this is a major security risk!

NOTE 1: Variable names are transformed as necessary into legal perl variable
names. All non-legal characters are transformed into underscores. If you need
to keep the original names, you should use the param() method instead to access
CGI variables by name.

In fact, you should probably not use this method at all given the above caveats
and security risks.

=head2 Deleting a parameter completely

    $q->delete('foo','bar','baz');

This completely clears a list of parameters. It sometimes useful for resetting
parameters that you don't want passed down between script invocations.

If you are using the function call interface, use "Delete()" instead to avoid
conflicts with perl's built-in delete operator.

=head2 Deleting all parameters

    $q->delete_all();

This clears the CGI object completely. It might be useful to ensure that all
the defaults are taken when you create a fill-out form.

Use Delete_all() instead if you are using the function call interface.

=head2 Handling non-urlencoded arguments

If POSTed data is not of type application/x-www-form-urlencoded or
multipart/form-data, then the POSTed data will not be processed, but instead
be returned as-is in a parameter named POSTDATA. To retrieve it, use code like
this:

    my $data = $q->param('POSTDATA');

Likewise if PUTed and PATCHed data can be retrieved with code like this:

    my $data = $q->param('PUTDATA');

    my $data = $q->param('PATCHDATA');

(If you don't know what the preceding means, worry not. It only affects people
trying to use CGI for XML processing and other specialized tasks)

PUTDATA/POSTDATA/PATCHDATA are also available via
L<upload_hook|/Progress bars for file uploads and avoiding temp files>,
and as L<file uploads|/Processing a file upload field> via L</-putdata_upload>
option.

=head2 Direct access to the parameter list

    $q->param_fetch('address')->[1] = '1313 Mockingbird Lane';
    unshift @{$q->param_fetch(-name=>'address')},'George Munster';

If you need access to the parameter list in a way that isn't covered by the
methods given in the previous sections, you can obtain a direct reference to
it by calling the B<param_fetch()> method with the name of the parameter. This
will return an array reference to the named parameter, which you then can
manipulate in any way you like.

You can also use a named argument style using the B<-name> argument.

=head2 Fetching the parameter list as a hash

    my $params = $q->Vars;
    print $params->{'address'};
    my @foo = split("\0",$params->{'foo'});
    my %params = $q->Vars;

    use CGI ':cgi-lib';
    my $params = Vars();

Many people want to fetch the entire parameter list as a hash in which the keys
are the names of the CGI parameters, and the values are the parameters' values.
The Vars() method does this. Called in a scalar context, it returns the
parameter list as a tied hash reference. Changing a key changes the value of
the parameter in the underlying CGI parameter list. Called in a list context,
it returns the parameter list as an ordinary hash. This allows you to read the
contents of the parameter list, but not to change it.

When using this, the thing you must watch out for are multivalued CGI
parameters. Because a hash cannot distinguish between scalar and list context,
multivalued parameters will be returned as a packed string, separated by the
"\0" (null) character. You must split this packed string in order to get at the
individual values. This is the convention introduced long ago by Steve Brenner
in his cgi-lib.pl module for perl version 4, and may be replaced in future
versions with array references.

If you wish to use Vars() as a function, import the I<:cgi-lib> set of function
calls (also see the section on CGI-LIB compatibility).

=head2 Saving the state of the script to a file

    $q->save(\*FILEHANDLE)

This will write the current state of the form to the provided filehandle. You
can read it back in by providing a filehandle to the new() method. Note that
the filehandle can be a file, a pipe, or whatever.

The format of the saved file is:

    NAME1=VALUE1
    NAME1=VALUE1'
    NAME2=VALUE2
    NAME3=VALUE3
    =

Both name and value are URL escaped. Multi-valued CGI parameters are represented
as repeated names. A session record is delimited by a single = symbol. You can
write out multiple records and read them back in with several calls to B<new>.
You can do this across several sessions by opening the file in append mode,
allowing you to create primitive guest books, or to keep a history of users'
queries. Here's a short example of creating multiple session records:

    use strict;
    use warnings;
    use CGI;

    open (my $out_fh,'>>','test.out') || die "Can't open test.out: $!";
    my $records = 5;
    for ( 0 .. $records ) {
        my $q = CGI->new;
        $q->param( -name => 'counter',-value => $_ );
        $q->save( $out_fh );
    }
    close( $out_fh );

    # reopen for reading
    open (my $in_fh,'<','test.out') || die "Can't open test.out: $!";
    while (!eof($in_fh)) {
        my $q = CGI->new($in_fh);
        print $q->param('counter'),"\n";
    }

The file format used for save/restore is identical to that used by the Whitehead
Genome Center's data exchange format "Boulderio", and can be manipulated and
even databased using Boulderio utilities. See L<Boulder> for further details.

If you wish to use this method from the function-oriented (non-OO) interface,
the exported name for this method is B<save_parameters()>.

=head2 Retrieving cgi errors

Errors can occur while processing user input, particularly when processing
uploaded files. When these errors occur, CGI will stop processing and return
an empty parameter list. You can test for the existence and nature of errors
using the I<cgi_error()> function. The error messages are formatted as HTTP
status codes. You can either incorporate the error text into a page, or use
it as the value of the HTTP status:

    if ( my $error = $q->cgi_error ) {
        print $q->header( -status => $error );
        print "Error: $error";
        exit 0;
    }

When using the function-oriented interface (see the next section), errors may
only occur the first time you call I<param()>. Be ready for this!

=head2 Using the function-oriented interface

To use the function-oriented interface, you must specify which CGI.pm
routines or sets of routines to import into your script's namespace.
There is a small overhead associated with this importation, but it
isn't much.

    use strict;
    use warnings;

    use CGI qw/ list of methods /;

The listed methods will be imported into the current package; you can
call them directly without creating a CGI object first. This example
shows how to import the B<param()> and B<header()>
methods, and then use them directly:

    use strict;
    use warnings;

    use CGI qw/ param header /;
    print header('text/plain');
    my $zipcode = param('zipcode');

More frequently, you'll import common sets of functions by referring
to the groups by name. All function sets are preceded with a ":"
character as in ":cgi" (for CGI protocol handling methods).

Here is a list of the function sets you can import:

=over 4

=item B<:cgi>

Import all CGI-handling methods, such as B<param()>, B<path_info()>
and the like.

=item B<:all>

Import all the available methods. For the full list, see the CGI.pm
code, where the variable %EXPORT_TAGS is defined. (N.B. the :cgi-lib
imports will B<not> be included in the :all import, you will have to
import :cgi-lib to get those)

=back

Note that in the interests of execution speed CGI.pm does B<not> use
the standard L<Exporter> syntax for specifying load symbols. This may
change in the future.

=head2 Pragmas

In addition to the function sets, there are a number of pragmas that you can
import. Pragmas, which are always preceded by a hyphen, change the way that
CGI.pm functions in various ways. Pragmas, function sets, and individual
functions can all be imported in the same use() line. For example, the
following use statement imports the cgi set of functions and enables
debugging mode (pragma -debug):

    use strict;
    use warninigs;
    use CGI qw/ :cgi -debug /;

The current list of pragmas is as follows:

=over 4

=item -no_undef_params

This keeps CGI.pm from including undef params in the parameter list.

=item -utf8

This makes CGI.pm treat all parameters as text strings rather than binary
strings (see L<perlunitut> for the distinction), assuming UTF-8 for the
encoding.

CGI.pm does the decoding from the UTF-8 encoded input data, restricting this
decoding to input text as distinct from binary upload data which are left
untouched. Therefore, a ':utf8' layer must B<not> be used on STDIN.

If you do not use this option you can manually select which fields are
expected to return utf-8 strings and convert them using code like this:

    use strict;
    use warnings;

    use CGI;
    use Encode qw/ decode /;

    my $cgi   = CGI->new;
    my $param = $cgi->param('foo');
    $param    = decode( 'UTF-8',$param );

=item -putdata_upload / -postdata_upload / -patchdata_upload

Makes C<<< $cgi->param('PUTDATA'); >>>, C<<< $cgi->param('PATCHDATA'); >>>,
and C<<< $cgi->param('POSTDATA'); >>> act like file uploads named PUTDATA,
PATCHDATA, and POSTDATA. See L</Handling non-urlencoded arguments> and
L</Processing a file upload field> PUTDATA/POSTDATA/PATCHDATA are also available
via L<upload_hook|/Progress bars for file uploads and avoiding temp files>.

=item -nph

This makes CGI.pm produce a header appropriate for an NPH (no parsed header)
script. You may need to do other things as well to tell the server that the
script is NPH. See the discussion of NPH scripts below.

=item -newstyle_urls

Separate the name=value pairs in CGI parameter query strings with semicolons
rather than ampersands. For example:

    ?name=fred;age=24;favorite_color=3

Semicolon-delimited query strings are always accepted, and will be emitted by
self_url() and query_string(). newstyle_urls became the default in version
2.64.

=item -oldstyle_urls

Separate the name=value pairs in CGI parameter query strings with ampersands
rather than semicolons. This is no longer the default.

=item -no_debug

This turns off the command-line processing features. If you want to run a CGI.pm
script from the command line, and you don't want it to read CGI parameters from
the command line or STDIN, then use this pragma:

   use CGI qw/ -no_debug :standard /;

=item -debug

This turns on full debugging. In addition to reading CGI arguments from the
command-line processing, CGI.pm will pause and try to read arguments from STDIN,
producing the message "(offline mode: enter name=value pairs on standard input)"
features.

See the section on debugging for more details.

=back

=head1 GENERATING DYNAMIC DOCUMENTS

Most of CGI.pm's functions deal with creating documents on the fly. Generally
you will produce the HTTP header first, followed by the document itself. CGI.pm
provides functions for generating HTTP headers of various types.

Each of these functions produces a fragment of HTTP which you can print out
directly so that it is processed by the browser, appended to a string, or saved
to a file for later use.

=head2 Creating a standard http header

Normally the first thing you will do in any CGI script is print out an HTTP
header. This tells the browser what type of document to expect, and gives other
optional information, such as the language, expiration date, and whether to
cache the document. The header can also be manipulated for special purposes,
such as server push and pay per view pages.

    use strict;
    use warnings;

    use CGI;

    my $cgi = CGI->new;

    print $cgi->header;

        -or-

    print $cgi->header('image/gif');

        -or-

    print $cgi->header('text/html','204 No response');

        -or-

    print $cgi->header(
        -type       => 'image/gif',
        -nph        => 1,
        -status     => '402 Payment required',
        -expires    => '+3d',
        -cookie     => $cookie,
        -charset    => 'utf-8',
        -attachment => 'foo.gif',
        -Cost       => '$2.00'
    );

header() returns the Content-type: header. You can provide your own MIME type
if you choose, otherwise it defaults to text/html. An optional second parameter
specifies the status code and a human-readable message. For example, you can
specify 204, "No response" to create a script that tells the browser to do
nothing at all. Note that RFC 2616 expects the human-readable phase to be there
as well as the numeric status code.

The last example shows the named argument style for passing arguments to the CGI
methods using named parameters. Recognized parameters are B<-type>, B<-status>,
B<-expires>, and B<-cookie>. Any other named parameters will be stripped of
their initial hyphens and turned into header fields, allowing you to specify
any HTTP header you desire. Internal underscores will be turned into hyphens:

    print $cgi->header( -Content_length => 3002 );

Most browsers will not cache the output from CGI scripts. Every time the browser
reloads the page, the script is invoked anew. You can change this behavior with
the B<-expires> parameter. When you specify an absolute or relative expiration
interval with this parameter, some browsers and proxy servers will cache the
script's output until the indicated expiration date. The following forms are all
valid for the -expires field:

    +30s                                  30 seconds from now
    +10m                                  ten minutes from now
    +1h                                   one hour from now
    -1d                                   yesterday (i.e. "ASAP!")
    now                                   immediately
    +3M                                   in three months
    +10y                                  in ten years time
    Thursday, 25-Apr-2018 00:40:33 GMT    at the indicated time & date

The B<-cookie> parameter generates a header that tells the browser to provide
a "magic cookie" during all subsequent transactions with your script. Some
cookies have a special format that includes interesting attributes such as
expiration time. Use the cookie() method to create and retrieve session cookies.

The B<-nph> parameter, if set to a true value, will issue the correct headers
to work with a NPH (no-parse-header) script. This is important to use with
certain servers that expect all their scripts to be NPH.

The B<-charset> parameter can be used to control the character set sent to the
browser. If not provided, defaults to ISO-8859-1. As a side effect, this sets
the charset() method as well. B<Note> that the default being ISO-8859-1 may not
make sense for all content types, e.g.:

    Content-Type: image/gif; charset=ISO-8859-1

In the above case you need to pass -charset => '' to prevent the default being
used.

The B<-attachment> parameter can be used to turn the page into an attachment.
Instead of displaying the page, some browsers will prompt the user to save it
to disk. The value of the argument is the suggested name for the saved file. In
order for this to work, you may have to set the B<-type> to
"application/octet-stream".

The B<-p3p> parameter will add a P3P tag to the outgoing header. The parameter
can be an arrayref or a space-delimited string of P3P tags. For example:

    print $cgi->header( -p3p => [ qw/ CAO DSP LAW CURa / ] );
    print $cgi->header( -p3p => 'CAO DSP LAW CURa' );

In either case, the outgoing header will be formatted as:

    P3P: policyref="/w3c/p3p.xml" cp="CAO DSP LAW CURa"

CGI.pm will accept valid multi-line headers when each line is separated with a
CRLF value ("\r\n" on most platforms) followed by at least one space. For
example:

    print $cgi->header( -ingredients => "ham\r\n\seggs\r\n\sbacon" );

Invalid multi-line header input will trigger in an exception. When multi-line
headers are received, CGI.pm will always output them back as a single line,
according to the folding rules of RFC 2616: the newlines will be removed, while
the white space remains.

=head2 Generating a redirection header

    print $q->redirect( 'http://somewhere.else/in/movie/land' );

Sometimes you don't want to produce a document yourself, but simply redirect
the browser elsewhere, perhaps choosing a URL based on the time of day or the
identity of the user.

The redirect() method redirects the browser to a different URL. If you use
redirection like this, you should B<not> print out a header as well.

You are advised to use full URLs (absolute with respect to current URL or even
including the http: or ftp: part) in redirection requests as relative URLs
are resolved by the user agent of the client so may not do what you want or
expect them to do.

You can also use named arguments:

    print $q->redirect(
        -uri    => 'http://somewhere.else/in/movie/land',
        -nph    => 1,
        -status => '301 Moved Permanently'
    );

All names arguments recognized by header() are also recognized by redirect().
However, most HTTP headers, including those generated by -cookie and -target,
are ignored by the browser.

The B<-nph> parameter, if set to a true value, will issue the correct headers
to work with a NPH (no-parse-header) script. This is important to use with
certain servers, such as Microsoft IIS, which expect all their scripts to be
NPH.

The B<-status> parameter will set the status of the redirect. HTTP defines
several different possible redirection status codes, and the default if not
specified is 302, which means "moved temporarily." You may change the status
to another status code if you wish.

Note that the human-readable phrase is also expected to be present to conform
with RFC 2616, section 6.1.

=head2 Creating a self-referencing url that preserves state information

    my $myself = $q->self_url;
    print qq(<a href="$myself">I'm talking to myself.</a>);

self_url() will return a URL, that, when selected, will re-invoke this script
with all its state information intact. This is most useful when you want to
jump around within the document using internal anchors but you don't want to
disrupt the current contents of the form(s). Something like this will do the
trick:

     my $myself = $q->self_url;
     print "<a href=\"$myself#table1\">See table 1</a>";
     print "<a href=\"$myself#table2\">See table 2</a>";
     print "<a href=\"$myself#yourself\">See for yourself</a>";

If you want more control over what's returned, using the B<url()> method
instead.

You can also retrieve a query string representation of the current object
state with query_string():

    my $the_string = $q->query_string();

The behavior of calling query_string is currently undefined when the HTTP method
is something other than GET.

If you want to retrieved the query string as set in the webserver, namely the
environment variable, you can call env_query_string()

=head2 Obtaining the script's url

    my $full_url      = url();
    my $full_url      = url( -full =>1 );  # alternative syntax
    my $relative_url  = url( -relative => 1 );
    my $absolute_url  = url( -absolute =>1 );
    my $url_with_path = url( -path_info => 1 );
    my $url_path_qry  = url( -path_info => 1, -query =>1 );
    my $netloc        = url( -base => 1 );

B<url()> returns the script's URL in a variety of formats. Called without any
arguments, it returns the full form of the URL, including host name and port
number

    http://your.host.com/path/to/script.cgi

You can modify this format with the following named arguments:

=over 4

=item B<-absolute>

If true, produce an absolute URL, e.g.

    /path/to/script.cgi

=item B<-relative>

Produce a relative URL. This is useful if you want to re-invoke your
script with different parameters. For example:

    script.cgi

=item B<-full>

Produce the full URL, exactly as if called without any arguments. This overrides
the -relative and -absolute arguments.

=item B<-path> (B<-path_info>)

Append the additional path information to the URL. This can be combined with
B<-full>, B<-absolute> or B<-relative>. B<-path_info> is provided as a synonym.

=item B<-query> (B<-query_string>)

Append the query string to the URL. This can be combined with B<-full>,
B<-absolute> or B<-relative>. B<-query_string> is provided as a synonym.

=item B<-base>

Generate just the protocol and net location, as in http://www.foo.com:8000

=item B<-rewrite>

If Apache's mod_rewrite is turned on, then the script name and path info
probably won't match the request that the user sent. Set -rewrite => 1 (default)
to return URLs that match what the user sent (the original request URI). Set
-rewrite => 0 to return URLs that match the URL after the mod_rewrite rules have
run. 

=back

=head2 Mixing post and url parameters

    my $color = url_param('color');

It is possible for a script to receive CGI parameters in the URL as well as in
the fill-out form by creating a form that POSTs to a URL containing a query
string (a "?" mark followed by arguments). The B<param()> method will always
return the contents of the POSTed fill-out form, ignoring the URL's query
string. To retrieve URL parameters, call the B<url_param()> method. Use it in
the same way as B<param()>. The main difference is that it allows you to read
the parameters, but not set them.

Under no circumstances will the contents of the URL query string interfere with
similarly-named CGI parameters in POSTed forms. If you try to mix a URL query
string with a form submitted with the GET method, the results will not be what
you expect.

If running from the command line, C<url_param> will not pick up any
parameters given on the command line.

=head2 Processing a file upload field

=head3 Basics

When the form is processed, you can retrieve an L<IO::File> compatible handle
for a file upload field like this:

    use autodie;

    # undef may be returned if it's not a valid file handle
    if ( my $io_handle = $q->upload('field_name') ) {
        open ( my $out_file,'>>','/usr/local/web/users/feedback' );
        while ( my $bytesread = $io_handle->read($buffer,1024) ) {
            print $out_file $buffer;
        }
    }

In a list context, upload() will return an array of filehandles. This makes it
possible to process forms that use the same name for multiple upload fields.

If you want the entered file name for the file, you can just call param():

    my $filename = $q->param('field_name');

Different browsers will return slightly different things for the name. Some
browsers return the filename only. Others return the full path to the file,
using the path conventions of the user's machine. Regardless, the name returned
is always the name of the file on the I<user's> machine, and is unrelated to
the name of the temporary file that CGI.pm creates during upload spooling
(see below).

When a file is uploaded the browser usually sends along some information along
with it in the format of headers. The information usually includes the MIME
content type. To retrieve this information, call uploadInfo(). It returns a
reference to a hash containing all the document headers.

    my $filehandle = $q->upload( 'uploaded_file' );
    my $type       = $q->uploadInfo( $filehandle )->{'Content-Type'};
    if ( $type ne 'text/html' ) {
        die "HTML FILES ONLY!";
    }

Note that you must use ->upload or ->param to get the file-handle to pass into
uploadInfo as internally this is represented as a File::Temp object (which is
what will be returned by ->upload or ->param). When using ->Vars you will get
the literal filename rather than the File::Temp object, which will not return
anything when passed to uploadInfo. So don't use ->Vars.

If you are using a machine that recognizes "text" and "binary" data modes, be
sure to understand when and how to use them (see the Camel book). Otherwise
you may find that binary files are corrupted during file uploads.

=head3 Accessing the temp files directly

When processing an uploaded file, CGI.pm creates a temporary file on your hard
disk and passes you a file handle to that file. After you are finished with the
file handle, CGI.pm unlinks (deletes) the temporary file. If you need to you
can access the temporary file directly. You can access the temp file for a file
upload by passing the file name to the tmpFileName() method:

    my $filehandle  = $q->upload( 'uploaded_file' );
    my $tmpfilename = $q->tmpFileName( $filehandle );

As with ->uploadInfo, using the reference returned by ->upload or ->param is
preferred, although unlike ->uploadInfo, plain filenames also work if possible
for backwards compatibility.

The temporary file will be deleted automatically when your program exits unless
you manually rename it or set $CGI::UNLINK_TMP_FILES to 0. On some operating
systems (such as Windows NT), you will need to close the temporary file's
filehandle before your program exits. Otherwise the attempt to delete the
temporary file will fail.

=head3 Changes in temporary file handling (v4.05+)

CGI.pm had its temporary file handling significantly refactored, this logic is
now all deferred to File::Temp (which is wrapped in a compatibility object,
CGI::File::Temp - B<DO NOT USE THIS PACKAGE DIRECTLY>). As a consequence the
PRIVATE_TEMPFILES variable has been removed along with deprecation of the
private_tempfiles routine and B<complete> removal of the CGITempFile package.
The $CGITempFile::TMPDIRECTORY is no longer used to set the temp directory,
refer to the perldoc for File::Temp if you want to override the default
settings in that package (the TMPDIR env variable is still available on some
platforms). For Windows platforms the temporary directory order remains
as before: TEMP > TMP > WINDIR ( > TMPDIR ) so if you have any of these in
use in existing scripts they should still work.

The Fh package still exists but does nothing, the CGI::File::Temp class is
a subclass of both File::Temp and the empty Fh package, so if you have any
code that checks that the filehandle isa Fh this should still work.

When you get the internal file handle you will receive a File::Temp object,
this should be transparent as File::Temp isa IO::Handle and isa IO::Seekable
meaning it behaves as previously. If you are doing anything out of the ordinary
with regards to temp files you should test your code before deploying this
update and refer to the File::Temp documentation for more information.

=head3 Handling interrupted file uploads

There are occasionally problems involving parsing the uploaded file. This
usually happens when the user presses "Stop" before the upload is finished. In
this case, CGI.pm will return undef for the name of the uploaded file and set
I<cgi_error()> to the string "400 Bad request (malformed multipart POST)". This
error message is designed so that you can incorporate it into a status code to
be sent to the browser. Example:

    my $file = $q->upload( 'uploaded_file' );
    if ( !$file && $q->cgi_error ) {
        print $q->header( -status => $q->cgi_error );
        exit 0;
    }

=head3 Progress bars for file uploads and avoiding temp files

CGI.pm gives you low-level access to file upload management through a file
upload hook. You can use this feature to completely turn off the temp file
storage of file uploads, or potentially write your own file upload progress
meter.

This is much like the UPLOAD_HOOK facility available in L<Apache::Request>,
with the exception that the first argument to the callback is an
L<Apache::Upload> object, here it's the remote filename.

    my $q = CGI->new( \&hook [,$data [,$use_tempfile]] );

    sub hook {
        my ( $filename, $buffer, $bytes_read, $data ) = @_;
        print "Read $bytes_read bytes of $filename\n";
    }

The C<< $data >> field is optional; it lets you pass configuration information
(e.g. a database handle) to your hook callback.

The C<< $use_tempfile >> field is a flag that lets you turn on and off CGI.pm's
use of a temporary disk-based file during file upload. If you set this to a
FALSE value (default true) then $q->param('uploaded_file') will no longer work,
and the only way to get at the uploaded data is via the hook you provide.

If using the function-oriented interface, call the CGI::upload_hook() method
before calling param() or any other CGI functions:

    CGI::upload_hook( \&hook [,$data [,$use_tempfile]] );

This method is not exported by default. You will have to import it explicitly
if you wish to use it without the CGI:: prefix.

=head3 Troubleshooting file uploads on Windows

If you are using CGI.pm on a Windows platform and find that binary files get
slightly larger when uploaded but that text files remain the same, then you
have forgotten to activate binary mode on the output filehandle. Be sure to call
binmode() on any handle that you create to write the uploaded file to disk.

=head3 Older ways to process file uploads

This section is here for completeness. if you are building a new application
with CGI.pm, you can skip it.

The original way to process file uploads with CGI.pm was to use param(). The
value it returns has a dual nature as both a file name and a lightweight
filehandle. This dual nature is problematic if you following the recommended
practice of having C<use strict> in your code. perl will complain when you try
to use a string as a filehandle. More seriously, it is possible for the remote
user to type garbage into the upload field, in which case what you get from
param() is not a filehandle at all, but a string.

To solve this problem the upload() method was added, which always returns a
lightweight filehandle. This generally works well, but will have trouble
interoperating with some other modules because the file handle is not derived
from L<IO::File>. So that brings us to current recommendation given above,
which is to call the handle() method on the file handle returned by upload().
That upgrades the handle to an IO::File. It's a big win for compatibility for
a small penalty of loading IO::File the first time you call it.

=head1 HTTP COOKIES

CGI.pm has several methods that support cookies.

A cookie is a name=value pair much like the named parameters in a CGI query
string. CGI scripts create one or more cookies and send them to the browser
in the HTTP header. The browser maintains a list of cookies that belong to a
particular Web server, and returns them to the CGI script during subsequent
interactions.

In addition to the required name=value pair, each cookie has several optional
attributes:

=over 4

=item 1. an expiration time

This is a time/date string (in a special GMT format) that indicates when a
cookie expires. The cookie will be saved and returned to your script until this
expiration date is reached if the user exits the browser and restarts it. If an
expiration date isn't specified, the cookie will remain active until the user
quits the browser.

=item 2. a domain

This is a partial or complete domain name for which the cookie is valid. The
browser will return the cookie to any host that matches the partial domain name.
For example, if you specify a domain name of ".capricorn.com", then the browser
will return the cookie to Web servers running on any of the machines
"www.capricorn.com", "www2.capricorn.com", "feckless.capricorn.com", etc. Domain
names must contain at least two periods to prevent attempts to match on top
level domains like ".edu". If no domain is specified, then the browser will
only return the cookie to servers on the host the cookie originated from.

=item 3. a path

If you provide a cookie path attribute, the browser will check it against your
script's URL before returning the cookie. For example, if you specify the path
"/cgi-bin", then the cookie will be returned to each of the scripts
"/cgi-bin/tally.pl", "/cgi-bin/order.pl", and
"/cgi-bin/customer_service/complain.pl", but not to the script
"/cgi-private/site_admin.pl". By default, path is set to "/", which causes the
cookie to be sent to any CGI script on your site.

=item 4. a "secure" flag

If the "secure" attribute is set, the cookie will only be sent to your script
if the CGI request is occurring on a secure channel, such as SSL.

=back

The interface to HTTP cookies is the B<cookie()> method:

    my $cookie = $q->cookie(
        -name    => 'sessionID',
        -value   => 'xyzzy',
        -expires => '+1h',
        -path    => '/cgi-bin/database',
        -domain  => '.capricorn.org',
        -secure  => 1
    );

    print $q->header( -cookie => $cookie );

B<cookie()> creates a new cookie. Its parameters include:

=over 4

=item B<-name>

The name of the cookie (required). This can be any string at all. Although
browsers limit their cookie names to non-whitespace alphanumeric characters,
CGI.pm removes this restriction by escaping and unescaping cookies behind the
scenes.

=item B<-value>

The value of the cookie. This can be any scalar value, array reference, or even
hash reference. For example, you can store an entire hash into a cookie this
way:

    my $cookie = $q->cookie(
        -name  => 'family information',
        -value => \%childrens_ages
    );

=item B<-path>

The optional partial path for which this cookie will be valid, as described
above.

=item B<-domain>

The optional partial domain for which this cookie will be valid, as described
above.

=item B<-expires>

The optional expiration date for this cookie. The format is as described in the
section on the B<header()> method:

    "+1h"  one hour from now

=item B<-secure>

If set to true, this cookie will only be used within a secure SSL session.

=back

The cookie created by cookie() must be incorporated into the HTTP header within
the string returned by the header() method:

    use strict;
    use warnings;

    use CGI;

    my $q      = CGI->new;
    my $cookie = ...
    print $q->header( -cookie => $cookie );

To create multiple cookies, give header() an array reference:

    my $cookie1 = $q->cookie(
        -name  => 'riddle_name',
        -value => "The Sphynx's Question"
    );

    my $cookie2 = $q->cookie(
        -name  => 'answers',
        -value => \%answers
    );

    print $q->header( -cookie => [ $cookie1,$cookie2 ] );

To retrieve a cookie, request it by name by calling cookie() method without the
B<-value> parameter. This example uses the object-oriented form:

    my $riddle  = $q->cookie('riddle_name');
    my %answers = $q->cookie('answers');

Cookies created with a single scalar value, such as the "riddle_name" cookie,
will be returned in that form. Cookies with array and hash values can also be
retrieved.

The cookie and CGI namespaces are separate. If you have a parameter named
'answers' and a cookie named 'answers', the values retrieved by param() and
cookie() are independent of each other. However, it's simple to turn a CGI
parameter into a cookie, and vice-versa:

    # turn a CGI parameter into a cookie
    my $c = cookie( -name => 'answers',-value => [$q->param('answers')] );
    # vice-versa
    $q->param( -name => 'answers',-value => [ $q->cookie('answers')] );

If you call cookie() without any parameters, it will return a list of
the names of all cookies passed to your script:

    my @cookies = $q->cookie();

See the B<cookie.cgi> example script for some ideas on how to use cookies
effectively.

=over 4

=item B<$CGI::COOKIE_CACHE>

If set to a non-negative integer, this variable will cause CGI.pm to use the
cached cookie details from the previous call to cookie(). By default this
cache is off to retain backwards compatibility.

=back

=head1 DEBUGGING

If you are running the script from the command line or in the perl debugger,
you can pass the script a list of keywords or parameter=value pairs on the
command line or from standard input (you don't have to worry about tricking
your script into reading from environment variables). You can pass keywords
like this:

    your_script.pl keyword1 keyword2 keyword3

or this:

   your_script.pl keyword1+keyword2+keyword3

or this:

    your_script.pl name1=value1 name2=value2

or this:

    your_script.pl name1=value1&name2=value2

To turn off this feature, use the -no_debug pragma.

To test the POST method, you may enable full debugging with the -debug pragma.
This will allow you to feed newline-delimited name=value pairs to the script on
standard input.

When debugging, you can use quotes and backslashes to escape characters in the
familiar shell manner, letting you place spaces and other funny characters in
your parameter=value pairs:

    your_script.pl "name1='I am a long value'" "name2=two\ words"

Finally, you can set the path info for the script by prefixing the first
name/value parameter with the path followed by a question mark (?):

    your_script.pl /your/path/here?name1=value1&name2=value2

=head1 FETCHING ENVIRONMENT VARIABLES

Some of the more useful environment variables can be fetched through this
interface. The methods are as follows:

=over 4

=item B<Accept()>

Return a list of MIME types that the remote browser accepts. If you give this
method a single argument corresponding to a MIME type, as in
Accept('text/html'), it will return a floating point value corresponding to the
browser's preference for this type from 0.0 (don't want) to 1.0. Glob types
(e.g. text/*) in the browser's accept list are handled correctly.

Note that the capitalization changed between version 2.43 and 2.44 in order to
avoid conflict with perl's accept() function.

=item B<raw_cookie()>

Returns the HTTP_COOKIE variable. Cookies have a special format, and this
method call just returns the raw form (?cookie dough). See cookie() for ways
of setting and retrieving cooked cookies.

Called with no parameters, raw_cookie() returns the packed cookie structure.
You can separate it into individual cookies by splitting on the character
sequence "; ". Called with the name of a cookie, retrieves the B<unescaped>
form of the cookie. You can use the regular cookie() method to get the names,
or use the raw_fetch() method from the CGI::Cookie module.

=item B<env_query_string()>

Returns the QUERY_STRING variable, note that this is the original value as set
in the environment by the webserver and (possibly) not the same value as
returned by query_string(), which represents the object state

=item B<user_agent()>

Returns the HTTP_USER_AGENT variable. If you give this method a single
argument, it will attempt to pattern match on it, allowing you to do something
like user_agent(Mozilla);

=item B<path_info()>

Returns additional path information from the script URL. E.G. fetching
/cgi-bin/your_script/additional/stuff will result in path_info() returning
"/additional/stuff".

NOTE: The Microsoft Internet Information Server is broken with respect to
additional path information. If you use the perl DLL library, the IIS server
will attempt to execute the additional path information as a perl script. If
you use the ordinary file associations mapping, the path information will be
present in the environment, but incorrect. The best thing to do is to avoid
using additional path information in CGI scripts destined for use with IIS. A
best attempt has been made to make CGI.pm do the right thing.

=item B<path_translated()>

As per path_info() but returns the additional path information translated into
a physical path, e.g. "/usr/local/etc/httpd/htdocs/additional/stuff".

The Microsoft IIS is broken with respect to the translated path as well.

=item B<remote_host()>

Returns either the remote host name or IP address if the former is unavailable.

=item B<remote_ident()>

Returns the name of the remote user (as returned by identd) or undef if not set

=item B<remote_addr()>

Returns the remote host IP address, or 127.0.0.1 if the address is unavailable.

=item B<request_uri()>

Returns the interpreted pathname of the requested document or CGI (relative to
the document root). Or undef if not set.

=item B<script_name()>

Return the script name as a partial URL, for self-referring scripts.

=item B<referer()>

Return the URL of the page the browser was viewing prior to fetching your
script.

=item B<auth_type()>

Return the authorization/verification method in use for this script, if any.

=item B<server_name()>

Returns the name of the server, usually the machine's host name.

=item B<virtual_host()>

When using virtual hosts, returns the name of the host that the browser
attempted to contact

=item B<server_port()>

Return the port that the server is listening on.

=item B<server_protocol()>

Returns the protocol and revision of the incoming request, or defaults to
HTTP/1.0 if this is not set

=item B<virtual_port()>

Like server_port() except that it takes virtual hosts into account. Use this
when running with virtual hosts.

=item B<server_software()>

Returns the server software and version number.

=item B<remote_user()>

Return the authorization/verification name used for user verification, if this
script is protected.

=item B<user_name()>

Attempt to obtain the remote user's name, using a variety of different
techniques. May not work in all browsers.

=item B<request_method()>

Returns the method used to access your script, usually one of 'POST', 'GET'
or 'HEAD'.  If running from the command line it will be undef.

=item B<content_type()>

Returns the content_type of data submitted in a POST, generally
multipart/form-data or application/x-www-form-urlencoded

=item B<http()>

Called with no arguments returns the list of HTTP environment variables,
including such things as HTTP_USER_AGENT, HTTP_ACCEPT_LANGUAGE, and
HTTP_ACCEPT_CHARSET, corresponding to the like-named HTTP header fields in the
request. Called with the name of an HTTP header field, returns its value.
Capitalization and the use of hyphens versus underscores are not significant.

For example, all three of these examples are equivalent:

    my $requested_language = $q->http('Accept-language');

    my $requested_language = $q->http('Accept_language');

    my $requested_language = $q->http('HTTP_ACCEPT_LANGUAGE');

=item B<https()>

The same as I<http()>, but operates on the HTTPS environment variables present
when the SSL protocol is in effect. Can be used to determine whether SSL is
turned on.

=back

=head1 USING NPH SCRIPTS

NPH, or "no-parsed-header", scripts bypass the server completely by sending the
complete HTTP header directly to the browser. This has slight performance
benefits, but is of most use for taking advantage of HTTP extensions that are
not directly supported by your server, such as server push and PICS headers.

Servers use a variety of conventions for designating CGI scripts as NPH. Many
Unix servers look at the beginning of the script's name for the prefix "nph-".
The Macintosh WebSTAR server and Microsoft's Internet Information Server, in
contrast, try to decide whether a program is an NPH script by examining the
first line of script output.

CGI.pm supports NPH scripts with a special NPH mode. When in this mode, CGI.pm
will output the necessary extra header information when the header() and
redirect() methods are called.

The Microsoft Internet Information Server requires NPH mode. As of version 2.30,
CGI.pm will automatically detect when the script is running under IIS and put
itself into this mode. You do not need to do this manually, although it won't
hurt anything if you do.

=over 4

=item In the B<use> statement 

Simply add the "-nph" pragma to the list of symbols to be imported into
your script:

    use CGI qw(:standard -nph)

=item By calling the B<nph()> method:

Call B<nph()> with a non-zero parameter at any point after using CGI.pm in your
program.

    CGI->nph(1)

=item By using B<-nph> parameters

in the B<header()> and B<redirect()>  statements:

    print header(-nph=>1);

=back

=head1 SERVER PUSH

CGI.pm provides four simple functions for producing multipart documents of the
type needed to implement server push. These functions were graciously provided
by Ed Jordan <ed@fidalgo.net>. To import these into your namespace, you must
import the ":push" set. You are also advised to put the script into NPH mode
and to set $| to 1 to avoid buffering problems.

Here is a simple script that demonstrates server push:

    #!/usr/bin/env perl

    use strict;
    use warnings;

    use CGI qw/:push -nph/;

    $| = 1;
    print multipart_init( -boundary=>'----here we go!' );
    for (0 .. 4) {
        print multipart_start( -type=>'text/plain' ),
            "The current time is ",scalar( localtime ),"\n";
        if ($_ < 4) {
            print multipart_end();
        } else {
            print multipart_final();
        }
        sleep 1;
    }

This script initializes server push by calling B<multipart_init()>. It then
enters a loop in which it begins a new multipart section by calling
B<multipart_start()>, prints the current local time, and ends a multipart
section with B<multipart_end()>. It then sleeps a second, and begins again.
On the final iteration, it ends the multipart section with
B<multipart_final()> rather than with B<multipart_end()>.

=over 4

=item multipart_init()

    multipart_init( -boundary => $boundary, -charset => $charset );

Initialize the multipart system. The -boundary argument specifies what MIME
boundary string to use to separate parts of the document. If not provided,
CGI.pm chooses a reasonable boundary for you.

The -charset provides the character set, if not provided this will default to
ISO-8859-1

=item multipart_start()

    multipart_start( -type => $type, -charset => $charset );

Start a new part of the multipart document using the specified MIME type and
charset. If not specified, text/html ISO-8859-1 is assumed.

=item multipart_end()

    multipart_end()

End a part. You must remember to call multipart_end() once for each
multipart_start(), except at the end of the last part of the multipart document
when multipart_final() should be called instead of multipart_end().

=item multipart_final()

    multipart_final()

End all parts. You should call multipart_final() rather than multipart_end()
at the end of the last part of the multipart document.

=back

Users interested in server push applications should also have a look at the
CGI::Push module.

=head1 AVOIDING DENIAL OF SERVICE ATTACKS

A potential problem with CGI.pm is that, by default, it attempts to process
form POSTings no matter how large they are. A wily hacker could attack your
site by sending a CGI script a huge POST of many gigabytes. CGI.pm will attempt
to read the entire POST into a variable, growing hugely in size until it runs
out of memory. While the script attempts to allocate the memory the system may
slow down dramatically. This is a form of denial of service attack.

Another possible attack is for the remote user to force CGI.pm to accept a huge
file upload. CGI.pm will accept the upload and store it in a temporary directory
even if your script doesn't expect to receive an uploaded file. CGI.pm will
delete the file automatically when it terminates, but in the meantime the remote
user may have filled up the server's disk space, causing problems for other
programs.

The best way to avoid denial of service attacks is to limit the amount of
memory, CPU time and disk space that CGI scripts can use. Some Web servers come
with built-in facilities to accomplish this. In other cases, you can use the
shell I<limit> or I<ulimit> commands to put ceilings on CGI resource usage.

CGI.pm also has some simple built-in protections against denial of service
attacks, but you must activate them before you can use them. These take the
form of two global variables in the CGI name space:

=over 4

=item B<$CGI::POST_MAX>

If set to a non-negative integer, this variable puts a ceiling on the size of
POSTings, in bytes. If CGI.pm detects a POST that is greater than the ceiling,
it will immediately exit with an error message. This value will affect both
ordinary POSTs and multipart POSTs, meaning that it limits the maximum size of
file uploads as well. You should set this to a reasonably high
value, such as 10 megabytes.

=item B<$CGI::DISABLE_UPLOADS>

If set to a non-zero value, this will disable file uploads completely. Other
fill-out form values will work as usual.

=back

To use these variables, set the variable at the top of the script, right after
the "use" statement:

    #!/usr/bin/env perl

    use strict;
    use warnings;

    use CGI;

    $CGI::POST_MAX = 1024 * 1024 * 10;  # max 10MB posts
    $CGI::DISABLE_UPLOADS = 1;          # no uploads

An attempt to send a POST larger than $POST_MAX bytes will cause I<param()> to
return an empty CGI parameter list. You can test for this event by checking
I<cgi_error()>, either after you create the CGI object or, if you are using the
function-oriented interface, call <param()> for the first time. If the POST was
intercepted, then cgi_error() will return the message "413 POST too large".

This error message is actually defined by the HTTP protocol, and is designed to
be returned to the browser as the CGI script's status code. For example:

    my $uploaded_file = $q->param('upload');
    if ( !$uploaded_file && $q->cgi_error() ) {
        print $q->header( -status => $q->cgi_error() );
        exit 0;
   }

However it isn't clear that any browser currently knows what to do with this
status code. It might be better just to create a page that warns the user of
the problem.

=head1 MODULE FLAGS

There are a number of global module flags which affect how CGI.pm operates.

=over 4

=item B<$CGI::APPEND_QUERY_STRING>

If set to a non-zero value, this will add query string parameters to a POST
forms parameters hence allowing I<param()> to return values from the query
string as well as from the decoded POST request instead of having to use
I<url_param> instead. This makes it easier to get the value of a parameter
when you don't know the source.

=back

=head1 COMPATIBILITY WITH CGI-LIB.PL

To make it easier to port existing programs that use cgi-lib.pl the
compatibility routine "ReadParse" is provided. Porting is simple:

OLD VERSION

    require "cgi-lib.pl";
    &ReadParse;
    print "The value of the antique is $in{antique}.\n";

NEW VERSION

    use CGI;
    CGI::ReadParse();
    print "The value of the antique is $in{antique}.\n";

CGI.pm's ReadParse() routine creates a tied variable named %in, which can be
accessed to obtain the query variables. Like ReadParse, you can also provide
your own variable. Infrequently used features of ReadParse, such as the creation
of @in and $in variables, are not supported.

Once you use ReadParse, you can retrieve the query object itself this way:

    my $q = $in{CGI};

This allows you to start using the more interesting features of CGI.pm without
rewriting your old scripts from scratch.

An even simpler way to mix cgi-lib calls with CGI.pm calls is to import both the
C<:cgi-lib> and C<:standard> method:

    use CGI qw(:cgi-lib :standard);
    &ReadParse;
    print "The price of your purchase is $in{price}.\n";
    print textfield(-name=>'price', -default=>'$1.99');

=head2 Cgi-lib functions that are available in CGI.pm

In compatibility mode, the following cgi-lib.pl functions are
available for your use:

    ReadParse()
    PrintHeader()
    SplitParam()
    MethGet()
    MethPost()

=head1 LICENSE

The CGI.pm distribution is copyright 1995-2007, Lincoln D. Stein. It is
distributed under the Artistic License 2.0. It is currently maintained
by Lee Johnson (LEEJO) with help from many contributors.

=head1 CREDITS

Thanks very much to:

=over 4

=item Mark Stosberg (mark@stosberg.com)

=item Matt Heffron (heffron@falstaff.css.beckman.com)

=item James Taylor (james.taylor@srs.gov)

=item Scott Anguish (sanguish@digifix.com)

=item Mike Jewell (mlj3u@virginia.edu)

=item Timothy Shimmin (tes@kbs.citri.edu.au)

=item Joergen Haegg (jh@axis.se)

=item Laurent Delfosse (delfosse@delfosse.com)

=item Richard Resnick (applepi1@aol.com)

=item Craig Bishop (csb@barwonwater.vic.gov.au)

=item Tony Curtis (tc@vcpc.univie.ac.at)

=item Tim Bunce (Tim.Bunce@ig.co.uk)

=item Tom Christiansen (tchrist@convex.com)

=item Andreas Koenig (k@franz.ww.TU-Berlin.DE)

=item Tim MacKenzie (Tim.MacKenzie@fulcrum.com.au)

=item Kevin B. Hendricks (kbhend@dogwood.tyler.wm.edu)

=item Stephen Dahmen (joyfire@inxpress.net)

=item Ed Jordan (ed@fidalgo.net)

=item David Alan Pisoni (david@cnation.com)

=item Doug MacEachern (dougm@opengroup.org)

=item Robin Houston (robin@oneworld.org)

=item ...and many many more...

for suggestions and bug fixes.

=back

=head1 BUGS

Address bug reports and comments to: L<https://github.com/leejo/CGI.pm/issues>

See the L<https://github.com/leejo/CGI.pm/blob/master/CONTRIBUTING.md> file for information
on raising issues and contributing

The original bug tracker can be found at:
L<https://rt.cpan.org/Public/Dist/Display.html?Queue=CGI.pm>

=head1 SEE ALSO

L<CGI::Carp> - provides L<Carp> implementation tailored to the CGI environment.

L<CGI::Fast> - supports running CGI applications under FastCGI

=cut