root/trunk/owa_httpRequest.php

<?php

//
// Open Web Analytics - An Open Source Web Analytics Framework
//
// Copyright 2006 Peter Adams. All rights reserved.
//
// Licensed under GPL v2.0 http://www.gnu.org/copyleft/gpl.html
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// $Id$
//

if(!class_exists('snoopy')):
        require_once(OWA_INCLUDE_DIR.'/Snoopy.class.php');
endif;

/**
 * Wrapper for Snoopy http request class
 * 
 * @author      Peter Adams <peter@openwebanalytics.com>
 * @copyright   Copyright &copy; 2006 Peter Adams <peter@openwebanalytics.com>
 * @license     http://www.gnu.org/copyleft/gpl.html GPL v2.0
 * @category    owa
 * @package     owa
 * @version             $Revision$            
 * @since               owa 1.0.0
 */

class owa_http extends Snoopy {
        
        /**
         * Configuration
         *
         * @var array
         */
        var $config;
        
        /**
         * Error handler
         *
         * @var object
         */
        var $e;
        
        /**
         * The length of text contained in the snippet
         *
         * @var string
         */
        var $snip_len = 100;
        
        /**
         * The string that is added to the beginning and
         * end of snippet text.
         *
         * @var string
         */
        var $snip_str = '...';
        
        /**
         * Anchor information for a particular link
         *
         * @var array
         */
        var $anchor_info;
        
        function owa_http() {
                
                $c = &owa_coreAPI::configSingleton();
                $this->config = $c->fetch('base');
                $this->e = &owa_coreAPI::errorSingleton();
                $this->agent = $this->config['owa_user_agent'];
                
                return;
        }
        
        /**
         * Searches a fetched html document for the anchor of a specific url
         *
         * @param string $link
         */
        function extract_anchor($link) {
                
                $matches = '';
                $regex = '/<a[^>]*href=\"%s\"[^>]*>(.*?)<\/a>/i';
                
                //$escaped_link = str_replace(array("/", "?"), array("\/", "\?"), $link);

                $pattern = trim(sprintf($regex, preg_quote($link, '/')));
                $search = preg_match($pattern, $this->results, $matches);
                //$this->e->debug('pattern: '.$pattern);
                //$this->e->debug('link: '.$link);
                
                
                if (empty($matches)):
                        if (substr($link, -1) === '/'):
                                $link = substr($link, 0, -1);
                                $pattern = trim(sprintf($regex, preg_quote($link, '/')));
                                $search = preg_match($pattern, $this->results, $matches);
                                //$this->e->debug('pattern: '.$pattern);
                                //$this->e->debug('link: '.$link);
                        endif;
                endif;
                
                $this->e->debug('ref search: '.$search);
                //$this->e->debug('ref matches: '.print_r($this->results, true));
                //$this->e->debug('ref matches: '.print_r($matches, true));
                                
                $this->anchor_info =  array('anchor_tag' => $matches[0], 'anchor_text' => owa_lib::inputFilter($matches[0]));
                
                $this->e->debug('Anchor info: '.print_r($this->anchor_info, true));
                
                return;
        }
        
        /**
         * Creates a text snippet of the portion of page where the 
         * specific link is found.
         * 
         * Takes fully qualified URL for the link to search for.
         *
         * @param string $link
         * @return string
         */
        function extract_anchor_snippet($link){
                
                // Search the page for a specific anchor
                $this->extract_anchor($link);
        
                if(!empty($this->anchor_info['anchor_tag'])):
                        
                        // drop certain HTML entitities and their content
                        $this->results = $this->strip_selected_tags($this->results, array('title', 'head', 'script', 'object', 'style', 'meta', 'link', 'rdf:'), true);
                        
                        //$this->e->debug('Refering page content after certain html entities were dropped: '.$this->results);
                
                        // strip html from doc
                        $nohtml = $this->results;
                        
                        // calc len of the anchor text
                        $atext_len = strlen($this->anchor_info['anchor_tag']);
                        
                        // find position within document of the anchor text
                        $start = strpos($nohtml, $this->anchor_info['anchor_tag']);
                        
                        if ($start < $this->snip_len):
                                $part1_start_pos = 0;
                                $part1_snip_len = $start;
                        else:
                                $part1_start_pos = $start;
                                $part1_snip_len = $this->snip_len;
                        endif;
                        
                        
                        // Create first segment of snippet
                        $first_part = substr($nohtml, 0, $part1_start_pos);
                        $first_part = str_replace(array('\r\n', '\n\n', '\t', '\r', '\n'), '', $first_part); 
                        $first_part = strip_tags(owa_lib::inputFilter($first_part));
                        //$part1 = trim(substr($nohtml, $part1_start_pos, $part1_snip_len));
                        $part1 = substr($first_part,-$part1_snip_len, $part1_snip_len);
                        
                        //$part1 = str_replace(array('\r\n', '\n\n', '\t', '\r', '\n'), '', $part1);
                        //$part1 = owa_lib::inputFilter($part1);
                        // Create second segment of snippet
                        $part2 = trim(substr($nohtml, $start + $atext_len, $this->snip_len+300));
                        $part2 = str_replace(array('\r\n', '\n\n', '\t', '\r', '\n'), '', $part2);
                        $part2 = substr(strip_tags(owa_lib::inputFilter($part2)),0, $this->snip_len);

                        // Put humpty dumpy back together again and create actual snippet
                        $snippet =  $this->snip_str.$part1.' <span class="snippet_anchor">'.owa_lib::inputFilter($this->anchor_info['anchor_tag']).'</span> '.$part2.$this->snip_str;
                
                else:
                
                        $snippet = '';
                        
                endif;
                
                return $snippet;
                
        }
        
        function extract_title() {
                
                preg_match('~(</head>|<body>|(<title>\s*(.*?)\s*</title>))~i', $this->results, $m);
                
                $this->e->debug("referer title extract: ". print_r($m, true));
                
        return $m[3];
        }
        
         function strip_selected_tags($str, $tags = array(), $stripContent = false) {

       foreach ($tags as $k => $tag){
       
           if ($stripContent == true) {
                        $pattern = sprintf('#(<%s.*?>)(.*?)(<\/%s.*?>)#is', preg_quote($tag), preg_quote($tag));
               $str = preg_replace($pattern,"",$str);
           }
           $str = preg_replace($pattern, ${2},$str);
       }
       
       return $str;
   }
        
}


?>