使用预匹配推送到 foreach 中的多维数组

Question

我在使用 preg_match 构建多维数组时遇到了一些困难。

我试图将一个段落分解成句子。 然后对于段落的每个部分/句子，我想将每个单词和标点符号分解为数组的另一个级别。

@Toto 昨天帮助我使用预匹配来分解字符串，同时保留标点符号作为元素。

但是，我一直在努力构建我想要的数组。

考虑这样一个段落：

First section. This section, and this. How about this section? And a section; split in two.

期望输出

作为回报，结果如下所示：

Array ( [0] => 
     Array ( [0] => First [1] => section [2] => . )
Array ( [1] =>
     Array ( [0] => This [1] => section [2] => , [3] => and [4] => this [2] => . ) 
Array ( [2] => 
     Array ( [0] => How [1] => about [2] => this [3] => section [4] => ? ) 
Array ( [3] =>
     Array ( [0] => And [1] => a [2] => section [3] => ; [4] => split 
     [5] => in [6] => two [7] => . )
)))

到目前为止我的代码/我尝试过的

这是行不通的。 我不太确定一旦我构建了第二维，我将如何删除 $s 的内容，但现在我对复制每个部分并将它们添加到数组 [0] 的数组感到更加困惑？

$m = '    First section. This section, and this. How about this section? And a section; split in two.'

$s = preg_split('/\s*[!?.]\s*/u', $m, -1, PREG_SPLIT_NO_EMPTY);

foreach ($s as $x => $var) {
    preg_match_all('/(\w+|[.;?!,:]+)/', $var, $a);
    array_push($s, $a);
}

print_r($s);

Answer 1

你快接近了，我刚刚添加了PREG_SPLIT_DELIM_CAPTURE并更改了preg_split的正则表达式。 所以你可以这样使用：

$str = 'First section. This section, and this. How about this section? And a section; split in two.';

$matchDelim = preg_split("/([^.?!]+[.?!]+)/", $str, -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY);

$finalArr = [];

foreach ($matchDelim as $match) {
    preg_match_all('/(\w+|[.;?!,:])/', $match, $matches);   
    $finalArr[] = $matches[0];
}

print_r($finalArr);

结果：

Array
(
    [0] => Array
        (
            [0] => First
            [1] => section
            [2] => .
        )

    [1] => Array
        (
            [0] => This
            [1] => section
            [2] => ,
            [3] => and
            [4] => this
            [5] => .
        )

    [2] => Array
        (
            [0] => How
            [1] => about
            [2] => this
            [3] => section
            [4] => ?
        )

    [3] => Array
        (
            [0] => And
            [1] => a
            [2] => section
            [3] => ;
            [4] => split
            [5] => in
            [6] => two
            [7] => .
        )

)