The London Perl and Raku Workshop takes place on 26th Oct 2024. If your company depends on Perl, please consider sponsoring and/or attending.


HTTP::Lite - Lightweight HTTP implementation


    use HTTP::Lite;
    $http = HTTP::Lite->new;
    $req = $http->request("") 
        or die "Unable to get document: $!";
    print $http->body();


Note: you should look at HTTP::Tiny or LWP before using this module.

HTTP::Lite is a stand-alone lightweight HTTP/1.1 implementation for perl. It is not intended as a replacement for the fully-featured LWP module. Instead, it is intended for use in situations where it is desirable to install the minimal number of modules to achieve HTTP support, or where LWP is not a good candidate due to CPU overhead, such as slower processors. HTTP::Lite is also significantly faster than LWP.

HTTP::Lite is ideal for CGI (or mod_perl) programs or for bundling for redistribution with larger packages where only HTTP GET and POST functionality are necessary.

HTTP::Lite supports basic POST and GET operations only. As of 0.2.1, HTTP::Lite supports HTTP/1.1 and is compliant with the Host header, necessary for name based virtual hosting. Additionally, HTTP::Lite now supports Proxies.

As of 2.0.0 HTTP::Lite now supports a callback to allow processing of request data as it arrives. This is useful for handling very large files without consuming memory.

If you require more functionality, such as FTP or HTTPS, please see libwwwperl (LWP). LWP is a significantly better and more comprehensive package than HTTP::Lite, and should be used instead of HTTP::Lite whenever possible.



This is the constructor for HTTP::Lite. It presently takes no arguments. A future version of HTTP::Lite might accept parameters.


request ( $url, $data_callback, $cbargs )

Initiates a request to the specified URL.

Returns undef if an I/O error is encountered, otherwise the HTTP status code will be returned. 200 series status codes represent success, 300 represent temporary errors, 400 represent permanent errors, and 500 represent server errors.

See for detailed information about HTTP status codes.

The $data_callback parameter, if used, is a way to filter the data as it is received or to handle large transfers. It must be a function reference, and will be passed: a reference to the instance of the http request making the callback, a reference to the current block of data about to be added to the body, and the $cbargs parameter (which may be anything). It must return either a reference to the data to add to the body of the document, or undef.

If set_callback is used, $data_callback and $cbargs are not used. $cbargs may be either a scalar or a reference.

The data_callback is called as: &$data_callback( $self, $dataref, $cbargs )

An example use to save a document to file is:

  # Write the data to the filehandle $cbargs
  sub savetofile {
    my ($self,$phase,$dataref,$cbargs) = @_;
    print $cbargs $$dataref;
    return undef;

  $url = "$testpath/bigbinary.dat";
  open OUT, '>','bigbinary.dat';
  $res = $http->request($url, \&savetofile, OUT);
  close OUT;
set_callback ( $functionref, $dataref )

At various stages of the request, callbacks may be used to modify the behaviour or to monitor the status of the request. These work like the $data_callback parameter to request(), but are more versatile. Using set_callback disables $data_callback in request()

The callbacks are called as: callback ( $self, $phase, $dataref, $cbargs )

The current phases are:

  connect - connection has been established and headers are being

  content-length - return value is used as the content-length.  If undef,
            and prepare_post() was used, the content length is

  done-headers - all headers have been sent

  content - return value is used as content and is sent to client.  Return
            undef to use the internal content defined by prepare_post().

  content-done - content has been successfuly transmitted.

  data - A block of data has been received.  The data is referenced by
            $dataref.  The return value is dereferenced and replaces the
            content passed in.  Return undef to avoid using memory for large

  done - Request is done.
prepare_post ( $hashref )

Takes a reference to a hashed array of post form variables to upload. Create the HTTP body and sets the method to POST.

http11_mode ( 0 | 1 )

Turns on or off HTTP/1.1 support. This is off by default due to broken HTTP/1.1 servers. Use 1 to enable HTTP/1.1 support.

add_req_header ( $header, $value )
get_req_header ( $header )
delete_req_header ( $header )

Add, Delete, or get a HTTP header(s) for the request. These functions allow you to override any header. Presently, Host, User-Agent, Content-Type, Accept, and Connection are pre-defined by the HTTP::Lite module. You may not override Host, Connection, or Accept.

If you call add_req_header() with $value set to undef, then the header won't be added.

To provide (proxy) authentication or authorization, you would use:

    use HTTP::Lite;
    use MIME::Base64;
    $http = HTTP::Lite->new;
    $encoded = encode_base64('username:password');
    $http->add_req_header("Authorization", $encoded);

NOTE: The present implementation limits you to one instance of each header.


Returns the body of the document returned by the remote server.


Returns an array of the HTTP headers returned by the remote server.


Returns a string representation of the HTTP headers returned by the remote server.

get_header ( $header )

Returns an array of values for the requested header.

NOTE: HTTP requests are not limited to a single instance of each header. As a result, there may be more than one entry for every header.


Returns the HTTP protocol identifier, as reported by the remote server. This will generally be either HTTP/1.0 or HTTP/1.1.

proxy ( $proxy_server )

The URL or hostname of the proxy to use for the next request.


Returns the HTTP status code returned by the server. This is also reported as the return value of request().


Returns the textual description of the status code as returned by the server. The status string is not required to adhere to any particular format, although most HTTP servers use a standard set of descriptions.


You must call this prior to re-using an HTTP::Lite handle, otherwise the results are undefined.

local_addr ( $ip )

Explicity select the local IP address. (default) lets the system choose.

local_port ( $port )

Explicity select the local port. 0 (default and recommended) lets the system choose.

method ( $method )

Explicity set the method. Using prepare_post or reset overrides this setting. Usual choices are GET, POST, PUT, HEAD


    # Get and print out the headers and body of the CPAN homepage
    use HTTP::Lite;
    $http = HTTP::Lite->new;
    $req = $http->request("")
        or die "Unable to get document: $!";
    die "Request failed ($req): ".$http->status_message()
      if $req ne "200";
    @headers = $http->headers_array();
    $body = $http->body();
    foreach $header (@headers)
      print "$header$CRLF";
    print "$CRLF";
    print "$body$CRLF";

    # POST a query to the dejanews USENET search engine
    use HTTP::Lite;
    $http = HTTP::Lite->new;
    %vars = (
             "QRY" => "perl",
             "ST" => "MS",
             "svcclass" => "dncurrent",
             "DBS" => "2"
    $req = $http->request("")
      or die "Unable to get document: $!";
    print "req: $req\n";
    print $http->body();


    - FTP 
    - HTTPS (SSL)
    - Authenitcation/Authorizaton/Proxy-Authorization
      are not directly supported, and require MIME::Base64.
    - Redirects (Location) are not automatically followed
    - multipart/form-data POSTs are not directly supported (necessary
      for File uploads).


Some broken HTTP/1.1 servers send incorrect chunk sizes when transferring files. HTTP/1.1 mode is now disabled by default.


Roy Hooper <>

Now co-maintained by Neil Bowers <>.


This module (HTTP::Lite) is almost certainly not the best module for your needs.

For most uses HTTP::Tiny is a good choice. If you need more features, then look at LWP.

You could also read this review of CPAN modules for making HTTP requests.


Copyright (c) 2000-2002 Roy Hooper. All rights reserved.

Some parts copyright 2009 - 2010 Adam Kennedy.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.