在SQL Server中上传CSV文件而不清空数据库中的表

Question

I want to upload csv files in sql server database. I want to upload data one after the other without emptying the table before inserting....How can i modify that code to achieve that...

Thanks.

$deleterecords = "TRUNCATE TABLE csvTable"; //empty the table of its current records
$params = array();
$options =  array( "Scrollable" => SQLSRV_CURSOR_KEYSET );
$stmt = sqlsrv_query($conn, $deleterecords, $params, $options ); 
//$stmt=odbc_exec($conn,$deleterecords);




//Upload File
if (isset($_POST['submit'])) {
    if (is_uploaded_file($_FILES['filename']['tmp_name'])) {
        echo "<h1>" . "File ". $_FILES['filename']['name'] ." uploaded successfully." . "</h1>";
        echo "<h2>Displaying contents:</h2>";
        readfile($_FILES['filename']['tmp_name']);
    }

    //Import uploaded file to Database
    $handle = fopen($_FILES['filename']['tmp_name'], "r");

    while (($data = fgetcsv($handle, 1000, ",")) !== FALSE) {
        $import="INSERT into csvTable(Name,city) values('$data[0]','$data[1]')";
        $params = array();
        $options =  array( "Scrollable" => SQLSRV_CURSOR_KEYSET );
        $stmt = sqlsrv_query($conn, $import, $params, $options ); 
        //$stmt=odbc_exec($conn,$import);
    }

    fclose($handle);

    print "Import done";

Answer 1

I think, that this is the actual problem

now i want the data to remain in the table and the next insert to go just below the data already inserted in the table

A SQL-Server table does not have any implicit order!

There is no just below the data ...

What happens, if you just do not delete the old data? Am I right, that your actual problem is: How can I know which records are newly inserted?

You might

• add a column ImportID INT IDENTITY which will ensure a running number. If you use ORDER BY ImportID in your queries, you'll get new data just below old data
• add a column to your target table and pass in a mark (sessionID, datetime, whatever) to allow groupings later on
• add a column with a DEFAULT CONSTRAINT to set the current time to allow groupings on this (only if there are no parallel inserts possible!)
• use a staging table to first store your values in an emptied table (as before) and move your data into the real data with a stored procedure (with any additional logic you like)