[jayz]

hope someone can help with this.

I am trying to create a a flatfile database which stores a line number as a unique id

Example:

1 | $Name | $Email | $Date |
2 | $Name | $Email | $Date |
3 | $Name | $Email | $Date |
4 | $Name | $Email | $Date |

and so on...

im have trouble trying to get the next save location, when an entry has been removed.

Example:

if entry 2 has been removed & the next entry is saved it is saved as

5 instead of 2.

I've tried a number of things but all fail.

i have tried doin this with a function

example:

PHP Code:

$x =1; 





function FindNextID($z)

{

$ffile = "dbfile.dat";

$fdata = file($ffile);

for ($q = 0; $q < sizeof($fdata); $q++)

{

list ($id, $name, $email, $date) = explode("|", trim($fdata[$q]));

if ($id >= $z)

{

$z = $id+1;

}

}

DoSave($z);   // Take the next ID and save

}



FindNextID($x); 

This code will find the next id location but does not take into account if id 2 is missing

so i tried somethin like this

PHP Code:

$x =1;



function FindNextID($z)

{

$dosave = 1;

$ffile = "dbfile.dat";

$fdata = file($ffile);

for ($q = 0; $q < sizeof($fdata); $q++)

{

list ($id, $name, $email, $date) = explode("|", trim($fdata[$q]));



if ($id == $z)

{

$dosave = 0;

$z = $z+1;

FindNext($z);

}

}

if ($dosave)

{

DoSave($z);

}

}



FindNextID($x);  // can be called from another function 

this is as close as i can get .... it does save the next location but!!

after about 2 saves it does mental lol! and saves the same data over again with a new id

I know there is somethin missing but not sure what

any help would be great

thanks!!

[Keith]

I hope you're not planning on having a large amount of records. Is there any reason you'd be opposed to a MySQL database? It'd save you loads of headaches.

...
And shouldn't the next entry be 5 anyhow? Why would you replace the deleted #2? That kind of kills the point of unique IDs.

[jayz]

hi thanks for the reply

sorry but sql is not an option if it had i wouldnt be writtin this lol

'And shouldn't the next entry be 5 anyhow?

yes thats what i said!

but i want the code to check if an entry has been removed like in the example.

if 2 is missing then call the next save id 2 - is it really that difficult to understand?

the whole point is to stop the ids from spiraling out of control

if there are 200 entries and 100 have been removed previously the id will be above 300! instead of 200.

makes sence to me.

the size of the file at this stage is not important!

thanks

[Keith]

I understood you the first time. The thing I fail to agree on is how you see single-incrementing values as sprawling out of control on what is evidently not going to become such a large file anyhow.



What you'll need to do is a bit crazy:

First, you'll need to explode the contents of the file into an array by line.

Then you'll have to loop through the array and get the ID of each line as you go. But save the ID to a temporary variable. You'll need it.

When you get to the next line and have the ID, you'll have to add 1 to the temp variable and compare it with the current variable. If they match, save that ID to the temp variable and move on. If they don't match, you've found a deleted value, and you now have your variable.

Then what do you do? You'll have to rewrite your file to that point and append the value to the end of the file - or write it there and then continue rewriting the entire file. If my guess is correct, you'll be rewriting your entire database file every time you insert, update, and delete. That is insane.

And I still don't get how the size of the file is not important, but you're concerned about the IDs sprawling out of control? This makes no sense.

But there you have it, an easy enough solution if you have an extremely low-traffic database and few extra hours on your hands.

[jayz]

ok i get what your saying about rewrite the entire file..

would it be more stable to save the line data to its own file rather then all data to a single file... example

file_1.dat < inside has the data .. $name | $email | $date .....

what would you say ?

and not mysql lol

[Keith]

Well, rather than save it pipe-delimited, I'd either var_export() or serialize() the array. That way you don't have to worry about replacing characters or checking for pipes in the user input.

To save the data:

PHP Code:

<?php



// dummy up some user input... could easily be _POST data

$data = array

(

    'name'  => 'Joe Smith',

    'email' => 'joe@smith.com',

    'date'  => '12/25/2007'

);



// get your current ID

$id = ( int ) $_REQUEST['id'];



// prepare your data array

$data = var_export( $data, 1 ); // $data = serialize( $data );



// set the current working file

$file = 'file_' . $id . '.dat';



// write the data array to the file

$fp = @fopen( $file, 'w' );

if ( $fp )

{

    fwrite( $fp, $data );

    fclose( $fp );

}

else

    echo 'Could not save.';

To read the data:

PHP Code:

<?php



// get your current ID

$id = ( int ) $_REQUEST['id'];



// set the current working file

$file = 'file_' . $id . '.dat';



// read your data array

if ( file_exists( $file ) )

    eval( '$data = ' . file_get_contents( $file ) . ';' ); // $data = unserialize( file_get_contents( $file ) );

else

    $data = array();



// $data now has the original saved array



?>

[jayz]

ok...

so save each submitted date to its own file is the best way yeah....

i will have to look into var_export & serialize as i dont know anythin about them

$id... in your example what is the $id where are you getting this from a form or somthin?

[Keith]

Yeah, using $_REQUEST, it will pick from $_GET, $_POST, and $_COOKIE, so you could use:

Code:

file.php?id=3

or in a form:

Code:

<input type="hidden" name="id" value="3" />

To make the two work together:

PHP Code:

<?php



$id = ( int ) $_REQUEST['id'];



?>



<form action="<?= basename( $_SERVER['PHP_SELF'] ); ?>" method="post">

    <input type="hidden" name="id" value="<?= $id; ?>" />

    <!-- REST OF YOUR FORM -->

</form>

[jayz]

wow thats interesting about 'REQUEST'

sorry for sounding stupid but what is the ( int )

by doin it this way will solve my problem to begin with too

example:

PHP Code:

$x=1;



// Check if file already exists - Start

function checkfile($id)

{



$fpath = "file-$id.dat";    // Save File



if (file_exists($fpath))    // Does the ID already exist?

{

$id++;                      // Yes! so increase it

$x = $id;

checkfile($x);              // Check the new id

}

else

{

dosave ($fpath);            // Found new ID save it!

}

}

// Check if file already exists - End





checkfile($x); 

thanks for your help - i will look into var_export it sounds just what i need to not have delimiters

[coolmark18]

I suggest not using comments such as

- is it really that difficult to understand?

when people are voluntarily giving up their time to try and help you.