Php array_multisort对2个相同的多维数组进行不同排序
我试图通过将两个csv文件导入多维数组并使用array_diff函数找出差异来比较php中的两个csv文件 我使用的方法是 1） 获取预期csv的每条记录并转储到arr1中 2） 获取实际csv的每条记录并转储到arr2中 3） 使用数组_multisort对数组1进行排序 4） 使用array\u multisort对array2进行排序 5） 使用数组_diff函数比较每条记录（例如arr1[0][1]与arr2[0][1]） 我的目标是使用php脚本在尽可能短的时间内比较文件。我发现上面的方法是最短的（最初尝试将th csv内容转储到MySQL中，并使用db查询进行比较，但由于一些未知的原因，这些查询的运行速度非常慢，以至于超时后会导致Apache服务器崩溃） 我的csv文件大小高达300mb，但通常是70k记录，20列，10mb大小 我正在粘贴我所做工作的代码（w.r.t上述步骤） 更新：两个csv文件可能包含不同的日期格式，并且每个文件可能以不同的格式表示数字，如1.csv可以将2013年1月12日和0.01作为第一行……2.csv可以将2013年12月1日和.01日作为第一行Php array_multisort对2个相同的多维数组进行不同排序,php,Php,我试图通过将两个csv文件导入多维数组并使用array_diff函数找出差异来比较php中的两个csv文件 我使用的方法是 1） 获取预期csv的每条记录并转储到arr1中 2） 获取实际csv的每条记录并转储到arr2中 3） 使用数组_multisort对数组1进行排序 4） 使用array\u multisort对array2进行排序 5） 使用数组_diff函数比较每条记录（例如arr1[0][1]与arr2[0][1]） 我的目标是使用php脚本在尽可能短的时间内比较文件。我发现上面的
因此，我认为哈希不起作用确定这两个文件是不同的吗？首先，我会使用md5_文件并比较这两个文件的md5哈希值，以检查它们是否有任何不同 如果它们不同，我会做如下事情：
$csv_1_path = 'file_1.csv';
$csv_2_path = 'file_2.csv';
$fh_csv_1 = fopen($csv_1_path, 'r');
$fh_csv_2 = fopen($csv_2_path, 'r');
$md5_1 = array();
$md5_2 = array();
while( !feof($fh_csv_1) ) {
$md5_1[] = md5(fgets($fh_csv_1));
}
while( !feof($fh_csv_1) ) {
$md5_2[] = md5(fgets($fh_csv_2));
}
$common_records = array_intersect($md5_1, $md5_2);
$records_diff_count = 0;
foreach($md5_1 as $row_index => $md5_rec_1) {
if ( !in_array($md5_rec_1, $common_records) ) {
print "Record in file $csv_1_path, row $row_index has no match.\n";
$records_diff_count++;
}
}
foreach($md5_2 as $row_index => $md5_rec_2) {
if ( !in_array($md5_rec_2, $common_records) ) {
print "Record in file $csv_2_path, row $row_index has no match.\n";
$records_diff_count++;
}
}
找到每个文件的行索引后，可以对文件之间的差异进行更深入的分析。快速观察
$csvA = "a.log";
$csvB = "b.log";
echo PHP_EOL;
$hashA = readCSVFile($csvA);
$hashB = readCSVFile($csvB);
// Lines in A not in B
$hash = array_diff($hashA, $hashB);
if (($fp = fopen($csvA, "r")) !== FALSE) {
foreach ( $hash as $p => $v ) {
fseek($fp, $p);
echo implode("|",array_map("trim", fgetcsv($fp, 2024, "|"))), PHP_EOL;
}
fclose($fp);
}
236|INPQR|31-AUG-12|200|INR|6653|AAAAAA,PPPPP|0|0|0|0|0
225|INPZQ|31-AUG-12|200|USD|6655|AAAAAA,PPPPP|0|0|0|0|0
225|INPZQ|31-AUG-12|200|INR|6654|AAAAAA,PPPPP|0|0|0|0|0
- 您不需要排序来获得差异
- 我不确定如何将
有效加载到300MB
，但显然您没有内存问题，因为我建议您改用PHP应用程序中
或SQLMap Reduce - 您的
有点滑稽，很难断定CSV
或|
是分隔符。空格- 因为你有
wow。。。对diff使用哈希将比整个20列更好。。然后，您可以确定检索内容的位置20列
- 因为你有
$csvA = "a.log";
$csvB = "b.log";
echo PHP_EOL;
$hashA = readCSVFile($csvA);
$hashB = readCSVFile($csvB);
// Lines in A not in B
$hash = array_diff($hashA, $hashB);
if (($fp = fopen($csvA, "r")) !== FALSE) {
foreach ( $hash as $p => $v ) {
fseek($fp, $p);
echo implode("|",array_map("trim", fgetcsv($fp, 2024, "|"))), PHP_EOL;
}
fclose($fp);
}
236|INPQR|31-AUG-12|200|INR|6653|AAAAAA,PPPPP|0|0|0|0|0
225|INPZQ|31-AUG-12|200|USD|6655|AAAAAA,PPPPP|0|0|0|0|0
225|INPZQ|31-AUG-12|200|INR|6654|AAAAAA,PPPPP|0|0|0|0|0
$csvA = "a.log";
$csvB = "b.log";
echo PHP_EOL;
$hashA = readCSVFile($csvA);
$hashB = readCSVFile($csvB);
// Lines in A not in B
$hash = array_diff($hashA, $hashB);
if (($fp = fopen($csvA, "r")) !== FALSE) {
foreach ( $hash as $p => $v ) {
fseek($fp, $p);
echo implode("|",array_map("trim", fgetcsv($fp, 2024, "|"))), PHP_EOL;
}
fclose($fp);
}
236|INPQR|31-AUG-12|200|INR|6653|AAAAAA,PPPPP|0|0|0|0|0
225|INPZQ|31-AUG-12|200|USD|6655|AAAAAA,PPPPP|0|0|0|0|0
225|INPZQ|31-AUG-12|200|INR|6654|AAAAAA,PPPPP|0|0|0|0|0
function readCSVFile($file, $size = 2024, $delimiter = "|") {
$hash = array();
if (($fp = fopen($file, "r")) !== FALSE) {
$t = ftell($fp);
while ( ($data = fgetcsv($fp, $size, $delimiter)) !== FALSE ) {
$hash[$t] = sha1(implode(array_map("trim", $data)));
$t = ftell($fp);
}
fclose($fp);
}
return $hash;
}
您可以使用下面的行对数据进行正确排序
<?php
function readCSV($fileName, $delimiter, $exclude_cols = array()) {
$data = array();
$fh = fopen($fileName, 'r');
while ($line = fgetcsv($fh, 0, $delimiter)) {
if (count($line) == 1 && $line[0] == '') {
continue;
}
for ($i = 0; $i < count($line); $i++) {
$line[$i] = in_array($i, $exclude_cols, true) ? 'NTBT' : trim($line[$i]);
}
$data[] = $line;
}
fclose($fh);
return $data;
}
function sort2dArray($data) {
$tmp = array();
$lineCount = count($data);
foreach ($data as $lineNum => $lineData) {
foreach ($lineData as $column => $value) {
$tmp[$column][$lineNum] = $value;
}
}
$multiSortArgs = array();
foreach ($tmp as $column => &$columnData) {
array_push($multiSortArgs, &$columnData, SORT_ASC);
}
$multiSortArgs[] = &$data;
call_user_func_array('array_multisort', $multiSortArgs);
return $data;
}
// ========= Reading and sorting
// The expected data
$data_Exp = readCSV($fileExp, $_POST['dl1']);
$rarr1 = sort2dArray($data_Exp);
// The actual data
$data_Act = readCSV($fileAct, $_POST['dl2']);
$rarr2 = sort2dArray($data_Act);
有许多不同的方法来比较两个CSV文件。我使用了一种方法来检查两个文件中的不同行。我考虑到您希望从行中删除某些列 我没有使用排序，因为我检查一行是否在另一个文件中，而不是它是否在同一位置。原因很简单：如果有一行不匹配，并在文件开头排序，则该行之后的所有行都将不同 例如：
file1: file2:
1|a 1|a
2|b 2|b
3|c 3|c
4|d 4|d
5|e 1|e
After sorting
file1: file2:
1|a 1|a
2|b 1|e
3|c 2|b
4|d 3|c
5|e 4|d
Now the rows 2, 3, 4, and 5 are all marked as different, because they do not match if you check per line. But in fact only 1 row is different.
<?php
set_time_limit(0);
//create a function to create the CSV arrays.
//If you create the code twice like you did, you are bound to make a mistake or change something in one place and not the other.
//Obviously that could lead to sorting two equal files differently.
function CsvToArray($file)
{
$exclude_cols = array(2); //you didnt provide it, so for testig i remove the date col because its always the same
//load file contents into variable and trim it
$data = trim(implode('', file($file)));
//strip \r new line to make sure only \n is used
$data = str_replace("\r", "", $data);
//strip all spaces from |
$data = preg_replace('/\s\s+\|/', '|', $data);
$data = preg_replace('/\|\s\s+/', '|', $data);
//strip all spaces from each line
$data = preg_replace('/\s\s+\n/', "\n", $data);
$data = preg_replace('/\n\s\s+/', "\n", $data);
//each line to seperate row
$data = explode("\n", $data);
//each col to seperate record
//This is only needed for comparisment if you want to remove certain cols
//if thats not needed, you can skip this part
foreach($data as $k=>$v)
$data[$k] = explode('|', $v);
//get the header. Its always the first row
//array_shift will return the first element and remove it from the dataset
$header = array_shift($data);
//turn the array around, by making the row the key and count howmany times it shows
$ar = array();
foreach ($data as $row) {
//remove unwanted cols
//if you dont want to remove certain cols, skip this and the implode part and use $ar[$row]++
foreach($exclude_cols as $c)
$row[$c] = '';
//implode the remaining
$key = implode('', $row);
//you can use str_to_lower($key) for case insensive matching
$ar[$key]++;
}
return $ar;
}
function CompareTwoCsv($file1, $file2)
{
$start = microtime(true);
$ar1 = CsvToArray($file1);
$ar2 = CsvToArray($file2);
//check for differences.
$diff = 0;
foreach($ar1 as $k=>$v) {
//the second array doesnt contain the key (is row) so there is a difference
if (!array_key_exists($k, $ar2)) {
$diff+=$v; //all rows that are in the first array are different
continue;
}
$c2 = $ar2[$k];
if ($v == $c2) //row is in both file an equal number of times
continue;
$diff += max($v, $c2) - min($v, $c2); //add the number of different rows
}
$ar1_count = count($ar1);
$ar2_count = count($ar2);
//if ar2 has more records. Every row that is more, is different.
if ($ar2_count>$ar1_count)
$diff += $ar2_count - $ar1_count;
$end = microtime(true);
$difftime = $end - $start;
//debug output
echo "We found ".$diff." differences in the files. it took ".$difftime." seconds<hr>";
}
//test and test2 are two files with ~100.000 rows based on the data you supplied.
//They have many equal rows in the files, so the array returned from CsvToArray is small
CompareTwoCsv("test.txt", "test.txt");
//We found 0 differences in the files. it took 5.6848769187927 seconds
CompareTwoCsv("test.txt", "test2.txt");
//We found 17855 differences in the files. it took 6.6002569198608 seconds
CompareTwoCsv("test2.txt", "test.txt");
//We found 17855 differences in the files. it took 7.5223989486694 seconds
//randomly generated files with 100.000 rows. Very little duplicate data;
CompareTwoCsv("largescv1.txt", "largescv2.txt");
//We found 98250 differences in the files. it took 5.4302139282227 seconds
?>
结果：你就不能比较一下
md5