PHP：根据与键值相关的术语对多维数组排序

我有以下数组，我想根据$term=“geo”对其进行排序，仅与

[result\u title]

相关

Array
(
    [0] => Array
        (
            [result_title] => Agathoklis Georgiou
            [result_subtext] => Active Employee
        )

    [1] => Array
        (
            [result_title] => Frixos Georgiou
            [result_subtext] => Active Employee
        )

    [2] => Array
        (
            [result_title] => George Ellinas
            [result_subtext] => Active Employee
        )

    [3] => Array
        (
            [result_title] => Georgi Georgiev
            [result_subtext] => Active Employee
        )

    [4] => Array
        (
            [result_title] => Charalambos Georgiou
            [result_subtext] => Former Employee
        )

    [5] => Array
        (
            [result_title] => Georgia Kantouna
            [result_subtext] => Former Employee
        )
)

预期结果应为：

Array
(
    [0] => Array
        (
            [result_title] => George Ellinas
            [result_subtext] => Active Employee

        )

    [1] => Array
        (
            [result_title] => Georgi Georgiev
            [result_subtext] => Active Employee


        )

    [2] => Array
        (
            [result_title] => Georgia Kantouna
            [result_subtext] => Former Employee
        )

    [3] => Array
        (
            [result_title] => Agathoklis Georgiou
            [result_subtext] => Active Employee
        )

    [4] => Array
        (
            [result_title] => Charalambos Georgiou
            [result_subtext] => Former Employee
        )

    [5] => Array
        (
            [result_title] => Frixos Georgiou
            [result_subtext] => Active Employee
        )
)

我尝试过各种方法，例如：

usort($data, function($a, $b) use ($term) {
    $x = strpos($a["result_title"], $term) === false;
    $y = strpos($b["result_title"], $term) === false;
    if ($x && !$y) return 1;
    if ($y && !$x) return -1;

    // use this if you want to sort alphabetically after the keyword sort:
    return strcmp($a["result_title"], $b["result_title"]);

    // or if you only want to sort by whether or not the keyword was found:
    return 0;
});

及

及

及

还有更多的没有任何适当的结果。或者也许我不明白实现我的结果的方法

---

我也检查过了：但是答案给出了我想要的结果。

这里有一个函数，我认为它可以满足您的需要。我假定您希望在单词开头找到

$term

的值。此代码提取标题中包含

$term

的任何关键字，然后根据是否找到该关键字进行排序，然后根据关键字的顺序进行排序，或者如果两者在标题上相同，则进行排序

$term = 'geo';
usort($data, function ($a, $b) use ($term) {
    // find the term in first entry
    $t1 = preg_match("/^.*?\b($term\w*)\b.*\$/i", $a['result_title'], $matches) ? $matches[1] : '';
    // find the term in second entry
    $t2 = preg_match("/^.*?\b($term\w*)\b.*\$/i", $b['result_title'], $matches) ? $matches[1] : '';
    // check if the terms were found
    if ($t1 == '' && $t2 != '') return 1;
    if ($t1 != '' && $t2 == '') return -1;
    // found in both - if not the same, just sort on the keyword
    if ($t1 != $t2) return strcmp($t1, $t2);
    // found the same keyword, sort on the whole title
    return strcmp($a['result_title'], $b['result_title']);
});

---

由于输出很长（这是您要求的），所以我省略了它，但我做了一个修改。

您可以将源数组作为PHP代码提供吗？@MarcinOrlowski您的意思是数组最初是如何形成的？它来自何处并不重要。把它放在有效的PHP数组中，这样我就可以自己尝试不浪费时间将这些数据转换成代码：）数组（[0]=>array（[result\u title]=>Agathoklis Georgiou[result\u subtext]=>Active Employee）[1]=>array（[result\u title]=>Frixos Georgiou[result\u subtext]=>Active Employee）[2]=>array（[result\u title]=>George Ellinas[result\u subtext]=>在职员工[3]=>数组（[result\u title]=>Georgi Georgiev[result\u subtext]=>Active Employee）[4]=>数组（[result\u title]=>Charalambos Georgiou[result\u subtext]=>前任员工]）@MarcinOrlowski我希望这就是你的意思：我不知道如何感谢你。在过去的48小时里我一直在努力。非常感谢。如果我将术语字符串设置为较低，会有不同吗？这其中有一些困惑example@jQuerybeast我不确定“将术语字符串设置为较低”是什么意思？@jQuerybeast我已对答案进行了编辑，以允许术语是整个单词。@jQuerybeast代码的工作方式它不会按这种方式排序，因为

'LIA'<'LIA'

。如果希望它不区分大小写，请将

strcmp

更改为

strcasecmp

usort($data, function ($a, $b) use ($term) {
    $levA = levenshtein($term, $a["result_title"]);
    $levB = levenshtein($term, $b["result_title"]);

    return $levA === $levB ? 0 : ($levA > $levB ? 1 : -1);
});

usort($data, function($a, $b){ return $a["result_title"] - $b["result_title"]; }); 

$term = 'geo';
usort($data, function ($a, $b) use ($term) {
    // find the term in first entry
    $t1 = preg_match("/^.*?\b($term\w*)\b.*\$/i", $a['result_title'], $matches) ? $matches[1] : '';
    // find the term in second entry
    $t2 = preg_match("/^.*?\b($term\w*)\b.*\$/i", $b['result_title'], $matches) ? $matches[1] : '';
    // check if the terms were found
    if ($t1 == '' && $t2 != '') return 1;
    if ($t1 != '' && $t2 == '') return -1;
    // found in both - if not the same, just sort on the keyword
    if ($t1 != $t2) return strcmp($t1, $t2);
    // found the same keyword, sort on the whole title
    return strcmp($a['result_title'], $b['result_title']);
});