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.

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 before you call CGI->new.

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

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 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({});

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.

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).

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 first value.

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.

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',
    );

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.

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'. 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.

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.

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.

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 upload_hook, and as file uploads via "-putdata_upload" option.

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 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 -name argument.

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 :cgi-lib set of function calls (also see the section on CGI-LIB compatibility).

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 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 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 save_parameters().

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 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 param(). Be ready for this!

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 param() and 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:

:cgi

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

: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 not be included in the :all import, you will have to import :cgi-lib to get those)

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

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:

-no_undef_params

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

-utf8

This makes CGI.pm treat all parameters as text strings rather than binary strings (see 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 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 );
    

-putdata_upload / -postdata_upload / -patchdata_upload

Makes "$cgi->param('PUTDATA');", "$cgi->param('PATCHDATA');", and "$cgi->param('POSTDATA');" act like file uploads named PUTDATA, PATCHDATA, and POSTDATA. See "Handling non-urlencoded arguments" and "Processing a file upload field" PUTDATA/POSTDATA/PATCHDATA are also available via upload_hook.

-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.

-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.

-oldstyle_urls

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

-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 /;
    

-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.

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 -type, -status, -expires, and -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 -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 -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 -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 -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. 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 -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 -type to "application/octet-stream".

The -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.

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 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 -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 -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.

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 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()

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 );

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:

-absolute

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

    /path/to/script.cgi
    

-relative

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

    script.cgi
    

-full

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

-path (-path_info)

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

-query (-query_string)

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

-base

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

-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.

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 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 url_param() method. Use it in the same way as 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, "url_param" will not pick up any parameters given on the command line.

Processing a file upload field

Basics

When the form is processed, you can retrieve an 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 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.

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.

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 - 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 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.

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 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;
    }

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 Apache::Request, with the exception that the first argument to the callback is an 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 $data field is optional; it lets you pass configuration information (e.g. a database handle) to your hook callback.

The $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.

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.

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 "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 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.

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()