实时搜索引擎逻辑背后的查询


Live search engine logic behind query

我在使用AJAX和PHP创建多字段实时搜索引擎时遇到了问题。

到目前为止,还不需要搜索多个字段,所以我有一个简单的查询,只适用于一个字段。

由于使用onkeyup函数进行实时搜索,我发现了一些问题。我需要解释的第一个问题是:

表的结构很简单:zipcode | city

例如,输入12345;当然这是邮政编码。但是如果有人输入12345 hometown,那么第一个关键字是邮政编码,第二个是城市呢?

关键字将通过使用preg_split('/['s]+/', $search_term)进行分割,因此我将收到一个包含将被搜索的单个关键字的数组。在上面的例子中,它将是:key[0] => 12345key[1] => hometown

我使用的查询是:

$where = "";
$search_term = preg_split('/['s]+/', $search_term); //splits the whole string into single keywords
$total_search_terms = count($search_term); //counts the array-keys from $search_term
foreach ($search_term as $key=>$single_term) {
        $where .= "`zipcode` LIKE '$single_term%' OR `city` LIKE '$single_term%' ";
        if ($key != ($total_search_terms - 1)){ //adds AND to the query in case in case of a non empty array-key 
            $where .= " AND ";
        }
}
$query = $db->query("SELECT COUNT(*) FROM table WHERE $where");
...

好的,到目前为止一切顺利。现在的问题是关键字可以一次又一次地匹配每个字段。

再举一个例子:

如果从上面输入12345 hometown,表示key[0] => 12345可以匹配zipcodecity字段。这个条件等于key[1] => hometown,甚至可以匹配字段zipcodecity。因此,即使从另一个方向输入,hometown 12345的含义也是相同的。

这是我遇到的第一个问题。

我正在寻找一个逻辑来构建查询。所以在输入12345 hometown的情况下,我想要这样写:

key[0] => 12345匹配zipcode字段时,不要检查key[1] => hometown,如果zipcodecity匹配,因为key[0]已经在zipcode中匹配,所以key[1]需要是city是相当合乎逻辑的。

好的,告诉我的第二个问题,我想让你看看david strachan的答案他提到,当城市包含多个字符串时,就会产生问题。假设搜索字符串是这样的:

12345 New York

键是:

key[0] => 12345, key[1] => New, key[2] => York

好的,现在的问题是,我可以检查其中一个键是否包含整数,如果字符串长度恰好为5,我知道它将是邮政编码。

key[0] => 12345 //if ( stringlen(key[0|) === 5) === true  && is_int(key[0]) === true) {zipcode}

到目前为止一切顺利,但真正的问题是城市字符串背后的逻辑。我的第一个想法是,我可以说所有不包含整数的键都必须是城市,这样我就可以将它们转换成一个键。

key[1] => New, key[2] => York //if ( !is_int(key[1|) === true && !is_int(key[2|) === true)  {$create_one_key = array_fill_keys(key[1], key[1]+key[2]);}

好吧,但是万一将来我想添加街道名称该怎么办呢?我不知道如何区分和测试街道名称,甚至是城市名称。

另一种方法是使用JQuery Autocomplete来搜索MySQL数据库。下面的代码使用了PDO。

<link rel="stylesheet" href="http://code.jquery.com/ui/1.10.2/themes/smoothness/jquery-ui.css" />
<script src="http://code.jquery.com/jquery-1.9.1.js"></script>
<script src="http://code.jquery.com/ui/1.10.2/jquery-ui.js"></script>
<link rel="stylesheet" href="/resources/demos/style.css" />
<script type="text/javascript">
jQuery(document).ready(function(){
    $('#zipsearch').autocomplete({source:'suggest.php', minLength:2});
});
</script>
<style type="text/css">
    li.ui-menu-item { font-size:10px}
</style>
</head>
<body>
<h2>jQuery UI Autocomplete - With MySQL  </h2>
<form onsubmit="return false;">
    Search:
    <input id="zipsearch" type="text" size ="60"/>
</form>

suggest.php

require("dbinfo.php");//db connection strings
// if the 'term' variable is not sent with the request, exit
if ( !isset($_REQUEST['term']) )
    exit;
$term = $_REQUEST['term'];
 if (is_numeric($term)){
    $query = "SELECT * from ziptest WHERE zip LIKE ? OR address LIKE ? LIMIT 0,10";
  }else{
    $query = "SELECT * from ziptest WHERE city LIKE ? OR state LIKE ? LIMIT 0,10";  
  }
$term.="%";  
// connect to the database  
try {  
    $dbh = new PDO("mysql:host=$host;dbname=$database", $username, $password);  
    $dbh->setAttribute( PDO::ATTR_ERRMODE, PDO::ERRMODE_EXCEPTION );  
  // Prepare statement
    $stmt = $dbh->prepare($query);
    // Assign parameters
    $stmt->bindParam(1,$term);
    $stmt->bindParam(2,$term);
    // setting the fetch mode  
    $stmt->setFetchMode(PDO::FETCH_ASSOC); 
    $stmt->execute();  
    $data = array();
        while($row = $stmt->fetch()) {  
            $data[] = array(
                'label' => $row['address'] .', '. $row['city'] .', '. $row['state'] .', '.$row['zip'] ,
                'value' => "ID ".$row['id'] .': '.$row['address'] .', '. $row['city'] .', '. $row['state'] .', '.$row['zip']
            );
        }
    echo json_encode($data);
}  
catch(PDOException $e) {  
    echo "I'm sorry I'm afraid you can't do that.". $e->getMessage() ;// Remove or modify after testing 
    file_put_contents('PDOErrors.txt',date('[Y-m-d H:i:s]'). $e->getMessage()."'r'n", FILE_APPEND);  
}  
// close the connection
$dbh = null;

通过添加0检查$search_term[0]是string(city)还是int(zipcode),然后相应地格式化查询。

$search_term = '1234 paisley';
$search_term = preg_split('/['s]+/', $search_term); //splits the whole string into single keywords
$total_search_terms = count($search_term); //counts the array-keys from $search_term
$test = $search_term[0]+0;//If string returns 0
if($total_search_terms == 1){
    if ($test == 0 ){
        $where = "WHERE `city` LIKE `%$search_term[0]%`";
         }else{
        $where = "WHERE `zipcode` LIKE `%$search_term[0]%` ";
        }   
    }else{
    if ($test == 0 ){
        $where = "WHERE `zipcode` LIKE `%$search_term[1]%` AND `city` LIKE `%$search_term[0]%`";
        }else{
        $where = "WHERE `zipcode` LIKE `%$search_term[0]%` AND `city` LIKE `%$search_term[1]%`";
        }   
}   
 $query = "SELECT COUNT(*) FROM table $where";
echo $query;
一个问题是你如何对待纽约。我将留给你如何解决这个问题。

编辑

New York $total_search_terms > 2