Php copy file and rename

(PHP 4, PHP 5, PHP 7, PHP 8)

copy — Copies file

Description

copy(string $from, string $to, ?resource $context = null): bool

If you wish to move a file, use the rename() function.

Parameters

Path to the source file.

The destination path. If to is a URL, the copy operation may fail if the wrapper does not support overwriting of existing files.

Warning

If the destination file already exists, it will be overwritten.

A valid context resource created with stream_context_create().

Return Values

Returns true on success or false on failure.

Examples

Example #1 copy() example

$file = 'example.txt';
$newfile = 'example.txt.bak';

if (!

See Also

• move_uploaded_file() - Moves an uploaded file to a new location
• rename() - Renames a file or directory
• The section of the manual about handling file uploads

simonr_at_orangutan_dot_co_dot_uk ¶

18 years ago

Having spent hours tacking down a copy() error: Permission denied , (and duly worrying about chmod on winXP) , its worth pointing out that the 'destination' needs to contain the actual file name ! --- NOT just the path to the folder you wish to copy into.......
DOH !
hope this saves somebody hours of fruitless debugging

cooper at asu dot ntu-kpi dot kiev dot ua ¶

16 years ago

It take me a long time to find out what the problem is when i've got an error on copy(). It DOESN'T create any directories. It only copies to existing path. So create directories before. Hope i'll help,

steve a h ¶

14 years ago

Don't forget; you can use copy on remote files, rather than doing messy fopen stuff.  e.g.

if(!@copy('http://someserver.com/somefile.zip','./somefile.zip'))
{
    $errors= error_get_last();
    echo "COPY ERROR: ".$errors['type'];
    echo "
\n".$errors['message'];
} else {
    echo "File copied from remote!";
}
?>

promaty at gmail dot com ¶

11 years ago

Here is a simple script that I use for removing and copying non-empty directories. Very useful when you are not sure what is the type of a file.

I am using these for managing folders and zip archives for my website plugins.

// removes files and non-empty directories
function rrmdir($dir) {
  if (is_dir($dir)) {
    $files = scandir($dir);
    foreach ($files as $file)
    if ($file != "." && $file != "..") rrmdir("$dir/$file");
    rmdir($dir);
  }
  else if (file_exists($dir)) unlink($dir);
} // copies files and non-empty directories
function rcopy($src, $dst) {
  if (file_exists($dst)) rrmdir($dst);
  if (is_dir($src)) {
    mkdir($dst);
    $files = scandir($src);
    foreach ($files as $file)
    if ($file != "." && $file != "..") rcopy("$src/$file", "$dst/$file");
  }
  else if (file_exists($src)) copy($src, $dst);
}
?>

Cheers!

absorbentshoulderman at gmail dot com ¶

9 years ago

A nice simple trick if you need to make sure the folder exists first:

$srcfile

ASchmidt at Anamera dot net ¶

8 years ago

Below a code snippet for downloading a file from a web server to a local file.

It demonstrates useful customizations of the request (such as setting a User-Agent and Referrer, often required by web sites), and how to download only files if the copy on the web site is newer than the local copy.

It further demonstrates the processing of response headers (if set by server) to determine the timestamp and file name. The file type is checked because some servers return a 200 OK return code with a textual "not found" page, instead of a proper 404 return code.

// $fURI:         URL to a file located on a web server
// $target_file:    Path to a local file    if ( file_exists( $target_file ) ) {
    $ifmodhdr = 'If-Modified-Since: '.date( "r", filemtime( $target_file ) )."\r\n";
}
else {
    $ifmodhdr = '';
}// set request header for GET with referrer for modified files, that follows redirects           
$arrRequestHeaders = array(
    'http'=>array(
        'method'        =>'GET',
        'protocol_version'    =>1.1,
        'follow_location'    =>1,
        'header'=>    "User-Agent: Anamera-Feed/1.0\r\n" .
                     "Referer: $source\r\n" .
            $ifmodhdr
            )
        );
$rc = copy( $fURI, $target_file, stream_context_create($arrRequestHeaders) );// HTTP request completed, preserve system error, if any
if( $rc ) {
    if ( fclose( $rc ) ) {
        unset( $err );
    }
    else {
        $err = error_get_last();
    }
}
else {
    $err = error_get_last();
}// Parse HTTP Response Headers for  HTTP Status, as well filename, type, date information
// Need to start from rear, to get last set of headers after possible sets of redirection headers
if ( $http_response_header ) {
    for ( $i = sizeof($http_response_header) - 1; $i >= 0; $i-- ) {
        if ( preg_match('@^http/\S+ (\S{3,}) (.+)$@i', $http_response_header[$i], $http_status) > 0 ) {
            // HTTP Status header means we have reached beginning of response headers for last request
            break;
        }
        elseif ( preg_match('@^(\S+):\s*(.+)\s*$@', $http_response_header[$i], $arrHeader) > 0 ) {
            switch ( $arrHeader[1] ) {
                case 'Last-Modified':
                    if ( !isset($http_content_modtime) ) {
                        $http_content_modtime = strtotime( $arrHeader[2] );
                    }                       
                    break;
                case 'Content-Type':
                    if ( !isset($http_content_image_type) ) {
                        if ( preg_match('@^image/(\w+)@ims', $arrHeader[2], $arrTokens) > 0 ) {
                            if ( in_array(strtolower($arrTokens[1]), $arrValidTypes)) {
                                $http_content_image_type = $arrTokens[1];
                                break;
                            }
                        }
                        throw new Exception( "Error accessing file $fURI; invalid content type: $arrHeader[2]", 2);
                    }
                    break;
                case 'Content-Disposition':
                    if ( !isset($http_content_filename) && preg_match('@filename\\s*=\\s*(?|"([^"]+)"|([\\S]+));?@ims', $arrHeader[2], $arrTokens) > 0 ) {
                        $http_content_filename = basename($arrTokens[1]);
                    }                       
                    break;
            }
        }
    }
}

if (

tom at r dot je ¶

14 years ago

It's worth noting that copy() sets the destination file's last modified time/date.

gimmicklessgpt at gmail dot com ¶

13 years ago

Here's a simple recursive function to copy entire directories

Note to do your own check to make sure the directory exists that you first call it on.

function recurse_copy($src,$dst) {
    $dir = opendir($src);
    @mkdir($dst);
    while(false !== ( $file = readdir($dir)) ) {
        if (( $file != '.' ) && ( $file != '..' )) {
            if ( is_dir($src . '/' . $file) ) {
                recurse_copy($src . '/' . $file,$dst . '/' . $file);
            }
            else {
                copy($src . '/' . $file,$dst . '/' . $file);
            }
        }
    }
    closedir($dir);
}
?>

eng-ayman at aymax dot net ¶

13 years ago

some hosts disable copy() function and say its for security

and for some copy is important so this is and simple function that do same as copy function effect

how smart php can help us l like php

    function copyemz($file1,$file2){
          $contentx =@file_get_contents($file1);
                   $openedfile = fopen($file2, "w");
                   fwrite($openedfile, $contentx);
                   fclose($openedfile);
                    if ($contentx === FALSE) {
                    $status=false;
                    }else $status=true;

                                        return

jim dot mcgowen at cox dot net ¶

14 years ago

On Windows (not sure about Linux) copy will overwrite an existing file but will not change the case of the existing filename.

In other words if I have a file named "Myfile.txt" and I overwrite it using copy with a file named "MyFile.txt" it will overwrite it but the filename will remain "Myfile.txt".

If this is a problem (as it was for me) use unlink to delete the existing file first.

hugo_2000 at gmx dot at ¶

7 years ago

If you try to copy a file to itself - e.g. if the target directory is just a symlink to the source directory - copy will return false. just like on the command line.

2mobile at gmail dot com ¶

5 years ago

If you have problems with "failed to open stream: Permission denied"

Check if the user that PHP runs under have "X" permission on every directory of the file path.
It will need it to access the file

If your file is: /path/to/test-in.txt
You should have X permission on:

/path
/path/to
and read permission on /path/to/test-in.txt

Sina Salek ¶

13 years ago

Thanks for all the comments, i wrote this function for fully supporting file and directory copy.
As you may have noticed there are feature that didn't have time to implement , but if you have time to implement them or even add more cool features, please notify me as well :). you can find me here http://sina.salek.ws/en/contact

PS : It was very useful for me, hope you find it useful as well.

    /**
     * Copy file or folder from source to destination, it can do
     * recursive copy as well and is very smart
     * It recursively creates the dest file or directory path if there weren't exists
     * Situtaions :
     * - Src:/home/test/file.txt ,Dst:/home/test/b ,Result:/home/test/b -> If source was file copy file.txt name with b as name to destination
     * - Src:/home/test/file.txt ,Dst:/home/test/b/ ,Result:/home/test/b/file.txt -> If source was file Creates b directory if does not exsits and copy file.txt into it
     * - Src:/home/test ,Dst:/home/ ,Result:/home/test/** -> If source was directory copy test directory and all of its content into dest     
     * - Src:/home/test/ ,Dst:/home/ ,Result:/home/**-> if source was direcotry copy its content to dest
     * - Src:/home/test ,Dst:/home/test2 ,Result:/home/test2/** -> if source was directoy copy it and its content to dest with test2 as name
     * - Src:/home/test/ ,Dst:/home/test2 ,Result:->/home/test2/** if source was directoy copy it and its content to dest with test2 as name
     * @todo
     *     - Should have rollback technique so it can undo the copy when it wasn't successful
     *  - Auto destination technique should be possible to turn off
     *  - Supporting callback function
     *  - May prevent some issues on shared enviroments : http://us3.php.net/umask
     * @param $source //file or folder
     * @param $dest ///file or folder
     * @param $options //folderPermission,filePermission
     * @return boolean
     */
    function smartCopy($source, $dest, $options=array('folderPermission'=>0755,'filePermission'=>0755))
    {
        $result=false;

                if (

kadnan at yahoo dot com ¶

18 years ago

you can also try xcopy command by using Shell to move/copy files/folders from one place to another
here is the code:

exec('xcopy c:\\myfolder d:\\myfolder /e/i', $a, $a1);
?>

by executing this command, it will move folder along with all contents to destination.

-adnan

allasso residing at signalmesa dot com ¶

13 years ago

As far as recursive copy, something like this seems to work fine for me:

$output = shell_exec( " cp -r -a dir_source/* dir_dest 2>&1 " )
echo $output
?>

Of course you need to get all your permissions clear.  You can do the necessary stuff to use variables.

You could also do this to create the destination directory:

shell_exec( " cp -r -a dir_source dir_dest 2>&1 " )
?>

This will create a new directory called "dir_dest" if it does not already exist. This is a bit risky though if your situation is ambiguous, and you want to continue to make backups etc, 'cause if you do it twice you end up with:

dir_destination/dir_source

to avoid that one could do something like:

shell_exec( " mkdir dir_dest;  cp -r -a dir_source/* dir_dest 2>&1 " )
?>

Maybe someone can tell me when or why it would be better to use all that PHP code I see here.

Vinicio Coletti ¶

6 years ago

Copying large files under Windows 8.1, from one NTFS filesystem to another NTFS filesystem, results in only the first 4 GiB copied and the rest of the file is ignored.

So, if you think to have files larger than 4 GiB, instead of doing:
   copy($source,$destination);
it is much better to do something like:
   exec("xcopy $source $destination");

I will check to see if this issue is valid also under Linux.
It depends on PHP not being compiled in 64 bit mode?

someone at terrasim dot com ¶

11 months ago

On Windows, php-7.4.19-Win32-vc15-x64 - copy() corrupted a 6GB zip file. Our only recourse was to write:

function file_win_copy( $src, $dst ) {
shell_exec( 'COPY "'.$src.'" "'.$dst.'"');
return file_exists($dest);
}

jtaylor -at- ashevillenc -dot- com ¶

15 years ago

It seems as though you can only use move_uploaded_file() once on a temporary file that has been uploaded through a form. Most likely the action this function takes destroys the temporary file after it has been moved, assuming permanent placement on a file system.

Attempting to use the function again in the same PHP script will return false and not move the file.

I ran into this when an image that was uploaded did not need resizing (smaller than a size threshold) and after moving the temporary upload to the "originals" directory, an attempt was made to again move the temporary file to another folder.

This behavior is understandable, but be careful - in this instance, I simply used copy() on the file that was already uploaded.

nensa at zeec dot biz ¶

13 years ago

When I recently had to copy a few millions of small files (< 1kb) from one NAS to another and for some reasons had to do that file by file I compared the following function with copy($src, $dest) and shell_exec("cp -r $src $dest").

Surprisingly stream_copy seamed to be slightly faster (at least in this specific context).

    function stream_copy($src, $dest)
    {
        $fsrc = fopen($src,'r');
        $fdest = fopen($dest,'w+');
        $len = stream_copy_to_stream($fsrc,$fdest);
        fclose($fsrc);
        fclose($fdest);
        return $len;
    }
?>

Bas Vijfwinkel ¶

7 years ago

Paths and filenames with Japanese characters are not handled correctly if you are running Apache/PHP on a Windows machine.
With the following code you can convert e.g. the network path to the appropriate encoding so that Windows understands where to look:

function convertPath($path)
{
    //split networkpath into parts
    $parts = explode('\\',$path);
    // convert each part to SJIS
    foreach($parts as $index => $part)
    {
         $parts[$index] = iconv('UTF-8','SJIS//IGNORE',$part);
    }   
    // put the network path back together
    return implode('\\',$parts);
}$oldname = convertPath('c:/Temp/ほげほげ.pdf');
$newname  = convertPath('\\\\PFSV0100\\DATA\\06：個人別\\333328_ほげほげ\\test_あいうえお.pdf');copy($oldname, $newname);
?>

There are however a number of characters that cannot be converted correctly like '②' and '﨑' because there are no SJIS equivalents.

cory at veck dot ca ¶

11 years ago

My own 'cp -R' written in PHP.

Hopefully it will work for your situation. I'm using it in a web based file manager for my CMS.

    define

mishawagon at gmail dot com ¶

3 years ago

Copy failed for me until I added dirname(__FILE__) to the file paths.
copy(dirname(__FILE__).$tempdata,dirname(__FILE__).$filepath)

mspreij ¶

15 years ago

This function creates a new filename to use for a copy of the given filename, its behaviour was mostly sto^Wborrowed from how the OS X Finder (*1) does it.
Note it *doesn't* actually copy the file, it just returns the new name. I needed it to work regardless of data source (filesystem, ftp, etc).

It also tries to match the current name as neatly as possible:
foo.txt -> foo copy.txt -> foo copy 1.txt -> foo copy 2.txt [etc]
foo.bar.baz.jpg -> foo.bar.baz copy.jpg
foobar -> foobar copy -> foobar copy 1 [etc]
".txt" -> .txt copy, and "txt." -> txt. copy
file.longextension -> file.longextension copy

It keeps trying until it finds a name that is not yet taken in $list, or until it looped 500 times (change as needed).

If the renamed file becomes longer than max filename length, it starts chopping away at the end of the part before where it adds " copy": reallylong...filename.txt -> reallylong...filena copy.txt

// $orig = current name, of course
// $list = array of filenames in the target directory (if none given, it will still return a new name)
// $max = max length of filename
function duplicate_name($orig, $list = array(), $max = 64) {
  $ext = '';
  $counter = 0;
  $list = (array) $list;
  $max = (int) $max;
  $newname = $orig;
  do {
    $name = $newname; # name in, newname out
    if (preg_match('/ copy$| copy \d+$/', $name, $matches)) {
      // don't even check for extension, name ends with " copy[ digits]"
    // preg hereunder matches anything with at least one period in the middle and an extension of 1-5 characters
    }elseif (preg_match('/(.+)\.([^.]{1,5})$/', $name, $parts)) {
      // split to name & extension
      list($name, $ext) = array($parts[1], $parts[2]);
    }
    if (preg_match('/ copy (\d+)$/', $name, $digits)) {
      $newname = substr($name, 0, - strlen($digits[1])) . ($digits[1] + 1);
      # $cutlen is only used for the bit at the end where it checks on max filename length
      $cutlen = 6 + strlen($digits[1]+1); // ' copy ' + digits
    }elseif(preg_match('/ copy$/', $name, $digits)) {
      $newname = $name . ' 1';
      $cutlen = 7; // ' copy' + ' 1'
    }else{
      $newname = $name . ' copy';
      $cutlen = 5; // ' copy'
    }
    if ($ext) {
      $newname .= '.' . $ext;
      $cutlen += strlen($ext) + 1;
    }
    if ($max > 0) {
      if (strlen($newname) > $max) {
        $newname = substr($newname, 0, max($max - $cutlen, 0)) . substr($newname, -$cutlen);
        if (strlen($newname) > $max) {echo "duplicate_name() error: Can't keep the new name under given max length.\n"; return false;}
      }
    }
    if ($counter++ > 500) {echo "duplicate_name() error: Too many similarly named files or infinite while loop.\n"; return false;}
  } while (in_array($newname, $list));
  return $newname;
} ?>

*1) The Finder seems to check the extension vs a list of known extensions, this function considers it valid if it's 5 or fewer characters long.

ps. sorry for taking up so much space! :-)