使用Web浏览器下载文件的PHP脚本

Question

I am working on a small project to download files from a ftp server using a web based query. I have create a HTML form as a front end, which takes from user the ftp server name and the file names, and then interacts with the php script, to connect with ftp server and download the files specified. I have kept both the html and php files on my university webserver. I am able to download files on the machine running the webserver when I run the PHP script directly form the command line from the server. But I am not able to download files on my local computer using a web browser.

| Web-browser running on local machine | <--not able to download file--> | My PHP script running on web-server machine | <--Able to download file-->| FTP Server |

<?php
$targetFile = 'file.txt';
$curl = curl_init();
$fh = fopen(dirname(__FILE__) . '/'.$targetFile,'w+b');
if ($fh == FALSE){ 
      print "File not opened<br>"; 
           exit; 
}

echo "<br>configuring curl...<br>";

curl_setopt($curl, CURLOPT_URL, "ftp://full_path_name_of_file");
curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($curl, CURLOPT_HEADER, 0);
curl_setopt($curl, CURLOPT_VERBOSE, 1);
curl_setopt($curl, CURLOPT_FILE, $fh);
curl_exec($curl);
echo curl_error($curl);
curl_close($curl);
fclose($fh);
?>

File is downloaded successfully when I run this php script from the command line from the server machine. But when I invoke the script from the web browser using my personal machine, I get an error "File not opened".

Can you please tell is there any way I can download the file via my web browser.

Thanks!

Answer 1

1. this might be a file ownership issue

   Please check the permission and the ownership of the file.

   In order to debug this a bit better, you might use parts of the script provided here:

   https://stackoverflow.com/a/10377285/1163786

2. check the php configuration

   there is a difference between the php configuration of the CLI and the one for the webserver. the later might have some restrictions, when compared to the CLI one. please compare or diff the files (to see the configuration difference).

3. the download itself is not initiated

   The script downloads a file via curl from a ftp server and stores it to a folder on your webserver, but your are not pointing the browser (client) to the downloaded file nor initiating a downstream to the browser from the script.

   I would add a check, that the script is not called from the CLI and then do a header forward to the downloaded file.

   if(PHP_SAPI !== 'cli') {
       header("Location: WWW_URL_SERVER . $path_to . $targetFile);
   }
   

   You can test the download standalone by using the "direct link" in your browser to initiate a download. This is again a permission thing, this time the webserver itself serves the static file and needs permission to do so.

   Referencing: Redirect page after process complete in PHP