利用Yahoo! Search API开发自已的搜索引擎-php版
美国东部时间3月1日，雅虎公司联合创始人之一的杨致远将宣布公司的搜索网络将进入Web服务。雅虎公司在www.developer.yahoo.com网站建立了Yahoo Search Developer Network，公司计划在此纽约举行的搜索引擎战略大会(Search Engine Strategies Conference)上推出这一计划。该网络将允许开发者在雅虎搜索之上建立新的应用程序，其中包括图像、视频、新闻以及地区搜索等内容。想要使用这项服务的会员必须先去http://api.search.yahoo.com/webservices/register_application 申请一个自已的ID号，注：每个ID号每天只能搜索5000次。
下面我们看一下，如何用PHP脚本调用Yahoo! Search API实现搜索的效果，全部脚本如下：
<?php// Yahoo Web Services PHP Example Code// Rasmus Lerdorf// www.mypchelp.cn
$appid = 'YahooDemo';// 在这输入你申请的ID号
$service = array('image'=>'http://api.search.yahoo.com/ImageSearchService/V1/imageSearch', 'local'=>'http://api.local.yahoo.com/LocalSearchService/V1/localSearch', 'news'=>'http://api.search.yahoo.com/NewsSearchService/V1/newsSearch', 'video'=>'http://api.search.yahoo.com/VideoSearchService/V1/videoSearch', 'web'=>'http://api.search.yahoo.com/WebSearchService/V1/webSearch');?><html><head><title>PHP Yahoo Web Service Example Code</title></head><body><form action='YahooSearchExample.php' method='GET'>Search Term: <input type='text' name='query' /><br />Zip Code: <input type='text' name='zip' /> (for local search)<br /><input type='submit' value=' Go! ' /><select name='type'><?php foreach($service as $name => $val) { if(!empty($_REQUEST['type']) && $name == $_REQUEST['type']) echo '<option SELECTED>$name</option>n'; else echo '<option>$name</option>n';} ?></select></form><?phpfunction done() {?></body></html><?phpexit;}
if(empty($_REQUEST['query']) || !in_array($_REQUEST['type'],array_keys($service))) done();
// Ok, here we go, we have the query and the type of search is valid// First build the query$q = '?query='.rawurlencode($_REQUEST['query']);if(!empty($_REQUEST['zip'])) $q.='&zip='.$_REQUEST['zip'];if(!empty($_REQUEST['start'])) $q.='&start='.$_REQUEST['start'];$q .= '&appid=$appid';
// Then send it to the appropriate service$xml = file_get_contents($service[$_REQUEST['type']].$q);
// Parse the XML and check it for errorsif (!$dom = domxml_open_mem($xml,DOMXML_LOAD_PARSING,$error)) { echo 'XML parse errorn'; foreach ($error as $errorline) { /* For production use this should obviously be logged to a file instead */ echo $errorline['errormessage'].'<br />n'; echo ' Node; : ' . $errorline['nodename'] . '<br />n'; echo ' Line; : ' . $errorline['line'] . '<br />n'; echo ' Column : ' . $errorline['col'] . '<br />n'; } done();}
// Now traverse the DOM with this function// It is basically a generic parser that turns limited XML into a PHP array// with only a couple of hardcoded tags which are common across all the// result xml from the web servicesfunction xml_to_result($dom) { $root = $dom->document_element(); $res['totalResultsAvailable'] = $root->get_attribute('totalResultsAvailable'); $res['totalResultsReturned'] = $root->get_attribute('totalResultsReturned'); $res['firstResultPosition'] = $root->get_attribute('firstResultPosition');
$node = $root->first_child(); $i = 0; while($node) { switch($node->tagname) { case 'Result': $subnode = $node->first_child(); while($subnode) { $subnodes = $subnode->child_nodes(); if(!empty($subnodes)) foreach($subnodes as $k=>$n) { if(empty($n->tagname)) $res[$i][$subnode->tagname] = trim($n->get_content()); else $res[$i][$subnode->tagname][$n->tagname]=trim($n->get_content()); } $subnode = $subnode->next_sibling(); } break; default: $res[$node->tagname] = trim($node->get_content()); $i--; break; } $i++; $node = $node->next_sibling(); }; return $res;}
$res = xml_to_result($dom);
// Ok, now that we have the results in an easy to use format,// display them.; It's quite ugly because I am using a single// display loop to display every type and I don't really understand HTML$first = $res['firstResultPosition'];$last = $first + $res['totalResultsReturned']-1;echo '<p>Matched ${res[totalResultsAvailable]}, showing $first to $last</p>n';if(!empty($res['ResultSetMapUrl'])) { echo '<p>Result Set Map: <a href='https://www.haobala.com/bcjs/${res[ResultSetMapUrl]}'>${res[ResultSetMapUrl]}</a></p>n';}for($i=0; $i<$res['totalResultsReturned']; $i++) { foreach($res[$i] as $key=>$value) { switch($key) { case 'Thumbnail': echo '<img src='https://www.haobala.com/bcjs/${value[Url]}' />n'; break; case 'Cache': echo 'Cache: <a href='https://www.haobala.com/bcjs/${value[Url]}'>${value[Url]}</a> [${value[Size]}]<br />n'; break; case 'PublishDate': echo '<b>$key:</b> '.strftime('%X %x',$value); break; default: if(stristr($key,'url')) echo '<a href='https://www.haobala.com/bcjs/$value'>$value</a><br />n'; else echo '<b>$key:</b> $value<br />'; break; } } echo '<hr />n';}
// Create Previous/Next Page linksif($start > 1) echo '<a href='https://www.haobala.com/YahooSearchExample.php'. '?query='.rawurlencode($_REQUEST['query']). '&zip='.rawurlencode($_REQUEST['zip']). '&type='.rawurlencode($_REQUEST['type']). '&start='.($start-10).''><-Previous Page</a> ';if($last < $res['totalResultsAvailable']) echo '<a href='https://www.haobala.com/YahooSearchExample.php'. '?query='.rawurlencode($_REQUEST['query']). '&zip='.rawurlencode($_REQUEST['zip']). '&type='.rawurlencode($_REQUEST['type']). '&start='.($last+1).''>Next Page-></a>';done();?>
有兴趣的朋友还可以看一下由[动态网站制作指南]所制作的ASP版本：http://www.mypchelp.cn/yahoo/
网公网安备