php - 在两个 csv 文件之间用 PHP 实现左连接

Question

由于此解决方案改编自投票得很好的答案 elsewhere ，我没想到会遇到问题。

问题:我想LEFT JOIN file0.csv与 file1.csv .

文件0.csv

+-----------------+-------------+--------------+
| Manufacturer ID |    image    | description  |
+-----------------+-------------+--------------+
| SKU231          | image1.jpg  | A box.       |
| SKUAG1          | image22.jpg | Another box. |
| SKU21D          | image7a.png | A third box. |
+-----------------+-------------+--------------+

文件 1.csv:

+--------+--------+--------+-------+-------+
|  mpn   | length | height | width | title |
+--------+--------+--------+-------+-------+
| SKU231 |     22 |     14 |    10 | Box 1 |
| SKUAG1 |     12 |      6 |     6 | Box 2 |
| SKU21D |      5 |      8 |     5 | Box 3 |
+--------+--------+--------+-------+-------+

期望的结果(file2.csv):

+-----------------+-------------+--------------+--------+--------+--------+-------+-------+
| Manufacturer ID |    image    | description  |  mpn   | length | height | width | title |
+-----------------+-------------+--------------+--------+--------+--------+-------+-------+
| SKU231          | image1.jpg  | A box.       | SKU231 |     22 |     14 |    10 | Box 1 |
| SKUAG1          | image22.jpg | Another box. | SKUAG1 |     12 |      6 |     6 | Box 2 |
| SKU21D          | image7a.png | A third box. | SKU21D |      5 |      8 |     5 | Box 3 |
+-----------------+-------------+--------------+--------+--------+--------+-------+-------+

LEFT JOIN 的 PHP 函数 file0.csv 与 Manufacturer ID 和 上的 file1.csv mpn 分别为:

function my_csv_join(array $csv_input_file_array, $csv_output_file, $html_preview, $left_join_on, $right_join_on = NULL) {
    if (count($csv_input_file_array) > 2) {
        echo 'This function probably only works for 2 csv files being joined at a time.  Reapply iteratively if needed.  Exiting now.';
        exit;
    } else {
        for ($x = 0; $x <= 1; $x++) {
            //get csv file to array 
            ${'file' . $x} = fopen($csv_input_file_array[$x], "r"); //Dynamic variables using braces: https://stackoverflow.com/a/9257536/9095603
    
            while (!feof(${'file' . $x})) {
                ${'csv_array' . $x}[] = fgetcsv(${'file' . $x});
            }
            fclose(${'file' . $x});
    
            ${'csv_array' . $x} = array_filter(${'csv_array' . $x}); // gets rid of last empty array in case present
            //CREATE HEADERED ARRAY
            ${'header' . $x} = array_shift(${'csv_array' . $x});
            foreach (${'csv_array' . $x} as ${'product_data_array' . $x} {
                ${'headered_array' . $x}[] = array_combine(${'header' . $x}, ${'product_data_array' . $x});
            }        
        }
        // How to simulate the SQL LEFT JOIN operation using PHP arrays?  https://stackoverflow.com/a/25837426

        //function to simulate the left join
        $left = $headered_array0;
        $right = $headered_array1;
        $final = array();

        if (empty($right_join_on)) // if $right_join_on omitted implies $right_join_on has the same value as $left_join_on
            $right_join_on = $left_join_on;
        foreach ($left AS $k => $v) {
            $final[$k] = $v; //basically keeping everything; $left just becomes $final
            foreach ($right AS $kk => $vv) {
                if ($v[$left_join_on] == $vv[$right_join_on]) { 
                    foreach ($vv AS $key => $val)
                        $final[$k][$key] = $val; 
                } else {
                    foreach ($vv AS $key => $val)
                   $final[$k][$key] = NULL; 
                }
            }
        }
        if ($html_preview == 'y') {
            echo '<pre>';
                var_dump($final);
            echo '</pre>';
        }
    
        // REINSTATE HEADERS
        // var_dump($final[0]);
        $indented_header = array(
            0 => array_keys($final[0])
        );
        $re_headered_array = array_merge($indented_header, $final);
    
        // write array to csv file
        $file2 = fopen($csv_output_file, "w");
        foreach ($re_headered_array as $line) {
            fputcsv($file2, $line);
        }
    
        fclose($file2);
    }
}

调用函数 my_csv_join():

my_csv_join(array('C:\xampp\htdocs\kalugi\file0.csv','C:\xampp\htdocs\kalugi\file1.csv'), 'C:\xampp\htdocs\kalugi\file2.csv','y','Manufacturer ID','mpn');

实际结果显示并非所有在 file0."Manufacturer ID"= file1.mpn 上匹配的记录都被匹配。因此，一些满足连接条件的预期行未被连接。我们用 NULL 代替它们的值:

var_dump 结果:

array(3) {
  [0]=>
  array(8) {
    ["Manufacturer ID"]=>
    string(6) "SKU231"
    ["image"]=>
    string(10) "image1.jpg"
    ["description"]=>
    string(6) "A box."
    ["mpn"]=>
    NULL
    ["length"]=>
    NULL
    ["height"]=>
    NULL
    ["width"]=>
    NULL
    ["title"]=>
    NULL
  }
  [1]=>
  array(8) {
    ["Manufacturer ID"]=>
    string(6) "SKUAG1"
    ["image"]=>
    string(11) "image22.jpg"
    ["description"]=>
    string(12) "Another box."
    ["mpn"]=>
    NULL
    ["length"]=>
    NULL
    ["height"]=>
    NULL
    ["width"]=>
    NULL
    ["title"]=>
    NULL
  }
  [2]=>
  array(8) {
    ["Manufacturer ID"]=>
    string(6) "SKU21D"
    ["image"]=>
    string(11) "image7a.png"
    ["description"]=>
    string(12) "A third box."
    ["mpn"]=>
    string(6) "SKU21D"
    ["length"]=>
    string(1) "5"
    ["height"]=>
    string(1) "8"
    ["width"]=>
    string(1) "5"
    ["title"]=>
    string(5) "Box 3"
  }
}

写到 $file2 (file2.csv) 的结果:

    +-----------------+-------------+--------------+--------+--------+--------+-------+-------+
    | Manufacturer ID |    image    | description  |  mpn   | length | height | width | title |
    +-----------------+-------------+--------------+--------+--------+--------+-------+-------+
    | SKU231          | image1.jpg  | A box.       |        |        |        |       |       |
    | SKUAG1          | image22.jpg | Another box. |        |        |        |       |       |
    | SKU21D          | image7a.png | A third box. | SKU21D |      5 |      8 |     5 | Box 3 |
    +-----------------+-------------+--------------+--------+--------+--------+-------+-------+

尽管 file0."Manufacturer ID"= file1.mpn 已满足，但为什么有 2 行没有连接？

Answer 1

我将把它简化为 SSCCE所以它很容易理解和重现。

我认为您在这里使解决方案过于复杂，这可能是导致您出现如此多奇怪错误的原因。解决方法其实很简单。您需要做的就是获取每个 csv 文件，将其解析为行和列，然后将这些行中的每一行合并为一行以将其写回另一个 csv。

这是最简单的方法:

$file0 = <<<'FILE0'
Manufacturer ID,image,description
SKU231,image1.jpg,A box.
SKUAG1,image22.jpg,Another box.
SKU21D,image7a.png,A third box.
FILE0;

$file1 = <<<'FILE1'
mpn,length,height,width,title
SKU231,22,14,10,Box 1
SKUAG1,12,6,6,Box 2
SKU21D,5,8,5,Box 3
FILE1;

$csv1 = array_map('str_getcsv', explode("\n", $file0));
$csv2 = array_map('str_getcsv', explode("\n", $file1));

$fd = fopen("/tmp/test.csv", 'w');


foreach ($csv1 as $row => $columns) {

    $newRow = array_merge($columns, $csv2[$row]);
    fputcsv($fd, $newRow);

}

$fd = fopen("/tmp/test.csv", "r");

while (($line = fgets($fd)) !== false) {
    echo $line;
}

结果:

"Manufacturer ID",image,description,mpn,length,height,width,title
SKU231,image1.jpg,"A box.",SKU231,22,14,10,"Box 1"
SKUAG1,image22.jpg,"Another box.",SKUAG1,12,6,6,"Box 2"
SKU21D,image7a.png,"A third box.",SKU21D,5,8,5,"Box 3"

---

|Manufacturer ID | image       | description  | mpn    | length | height | width | title |
|----------------|-------------|--------------|--------|--------|--------|-------|-------|
| SKU231         | image1.jpg  | A box.       | SKU231 | 22     | 14     | 10    | Box 1 |
| SKUAG1         | image22.jpg | Another box. | SKUAG1 | 12     | 6      | 6     | Box 2 |
| SKU21D         | image7a.png | A third box. | SKU21D | 5      | 8      | 5     | Box 3 |