Here is a simple function to "peek" at the position of the internal pointer in a query result:
function mysql_pointer_position($result_set) {
$num_rows = mysql_num_rows($result_set);
$i = 0;
while($result = mysql_fetch_array($result_set)) {
$i++;
}
$pointer_position = $num_rows - $i;
//Return pointer to original position
if($pointer_position <= $num_rows - 1) {
mysql_data_seek($result_set, $pointer_position);
}
return $pointer_position;
}
mysql_data_seek
(PHP 4, PHP 5, PECL mysql:1.0)
mysql_data_seek — Déplace le pointeur interne de résultat MySQL
Description
mysql_data_seek() déplace le pointeur interne de résultat, dans le résultat associé à l'identifiant de résultat result_identifier . Il le fait pointer à la ligne row_number . Le prochain appel à une fonction MySQL de récupération de données, comme la fonction mysql_fetch_assoc() retournera cette ligne.
row_number commence à 0. row_number doit être une valeur qui va de 0 à mysql_num_rows() - 1. Cependant, si le résultat est vide, un row_number de 0 échouera avec une erreur E_WARNING et mysql_data_seek() retournera FALSE.
Liste de paramètres
- result
-
La ressource de résultat qui vient d'être évaluée. Ce résultat vient de l'appel à la fonction mysql_query().
- row_number
-
La position de la ligne désirée pour le nouveau pointeur de résultats.
Valeurs de retour
Cette fonction retourne TRUE en cas de succès, FALSE en cas d'échec.
Exemples
Exemple #1 Exemple avec mysql_data_seek()
<?php
$link = mysql_connect('localhost', 'mysql_user', 'mysql_password');
if (!$link) {
die('Connexion impossible : ' . mysql_error());
}
$db_selected = mysql_select_db('sample_db');
if (!$db_selected) {
die('Sélection de base de données impossible : ' . mysql_error());
}
$query = 'SELECT nom_famille, prenom FROM amis';
$result = mysql_query($query);
if (!$result) {
die('Requête échec : ' . mysql_error());
}
/* Récupération des lignes en ordre inverse */
for ($i = mysql_num_rows($result) - 1; $i >= 0; $i--) {
if (!mysql_data_seek($result, $i)) {
echo "Ne peut pointer vers la ligne $i : " . mysql_error() . "\n";
continue;
}
if (!($row = mysql_fetch_assoc($result))) {
continue;
}
echo $row['nom_famille'] . ' ' . $row['prenom'] . "<br />\n";
}
mysql_free_result($result);
?>
Notes
Note: La fonction mysql_data_seek() peut être utilisée en conjonction avec mysql_query(), mais pas avec la fonction mysql_unbuffered_query().
mysql_data_seek
28-Aug-2008 11:59
27-Jun-2007 10:26
I needed to "peek" at the next record in order to see if fetching it would go too far. So I want to do a fetch, followed by seek(-1).
I could find no function to move the internal row pointer relative to it's current position, or to retrieve it as a row number as required by mysql_data_seek(). This limits the function's usefulness to resetting the row to 0, unless you track the row number yourself.
If you use a While loop to step through the results, you can increment a tracking index at the bottom of the loop. But be sure never to use Continue; which would bypass your index. And document this restriction for the person who needs to maintain your code. It's probably better to use a For loop, which makes the index explicit.
In either case be sure to range check the index when you manipulate it. E.G. When I "peek" at the next record I must check for index>=count (end of data). Or if I decrement the index, make sure it does not go negative. Again, document why you are coding it this way, so the next programmer doesn't "correct" the inelegant code.
A helpful note about the 'resource' data type.
Since the 'resource' variable is pointing to a row in a result set at any given time, you can think of it as being passed to this function by reference every time you pass it or assign it to a variable.
<?
$sql = "SELECT * from <table>";
$result = mysql_query($sql);
$temp_result = $result;
while ($row = mysql_fetch_assoc($temp_result)) {
// do stuff with $row
}
while ($row = mysql_fetch_assoc($result)) {
// This code will never run because the 'resource' variable is pointing past the end of the result set,
// even though it was *not* assigned by reference to $result2.
}
?>
Therefore, the following snipits are functionally identical:
<?
// Start snipit 1
$sql = "SELECT * from <table>";
$result = mysql_query($sql);
while ($row = mysql_fetch_assoc($result)) {
// do stuff with $row
}
mysql_data_seek($result, 0);
while ($row = mysql_fetch_assoc($result)) {
// do other stuff with $row
}
// Start snipit 2
$sql = "SELECT * from <table>";
$result = mysql_query($sql);
$temp_result = $result;
while ($row = mysql_fetch_assoc($temp_result)) {
// do stuff with $row
}
mysql_data_seek($result, 0);
while ($row = mysql_fetch_assoc($temp_result)) {
// do other stuff with $row
}
?>
27-Jun-2005 02:40
/*
helpfull for real time databases query
- Query one time
- Retreive data twice from the same query
- mysql_data_seek *
*/
include("p_MySql_Connection.php");
$v_Query = "SELECT f1 from t1";
$v_Result = mysql_query($v_Query, $v_RS);
/*
First loop for one single query
*/
while ($row = mysql_fetch_array($v_Result,MYSQL_NUM)) {
$v_total = $v_total + $row[1];
}
echo $v_total;
/*
Retreive data
*/
$v_Re = mysql_data_seek($v_Result,0);
if (!$v_Re){
echo 'MySql data seek Error' . mysql_error();
}
/*
Second loop for one single query
*/
while ($row = mysql_fetch_array($v_Result,MYSQL_NUM)) {
echo $row[0];
}
21-Apr-2005 05:53
hello, this script would be easy to understand for those that are novice in php whose want to understand about this function:
the table "user" have 2 columns "id" and "name".
"user" content:
position 0: "id"=195342481 "name"='Arthur'
position 1: "id"=179154675 "name"='John'
>>position 2<<: "id"=157761949 "name"='April' >>third row<<
position 3: "id"=124492684 "name"='Tammy'
position 4: "id"=191346457 "name"='Mike'
<?php
mysql_connect("localhost", "root")
mysql_select_db("test");
$sql = mysql_query("select * from user");
mysql_data_seek($sql, 2);
echo "<table border=1>";
while ($row = mysql_fetch_row($sql)){
echo "<tr><td>$row[0]</td><td>$row[1]</td></tr>";
}
echo "</tabla>";
?>
explanation:
mysql_data_seek move internal result pointer to the third row of table user. Thus mysql_fetch_row will begin by april´s row.
09-Dec-2004 08:09
to kennethnash1134 at yahoo dot com
your loop can be done like this as well and i guess this is faster:
$r=mysql_query("select user,id,ip from accounts limit 10");
unset($users); // Just to be sure
while($users[] = mysql_fetch_row);
array_pop($users); // Drop the last entry which is FALSE
26-Mar-2004 09:12
/*here is a nice function for converting a mysql result row set into a 2d array, a time saver if need small data from several rows, saves you from having to do Alot of queries... would be nice to have this built into PHP future versions :) */
// simple example query
$r=mysql_query("select user,id,ip from accounts limit 10");
//starts the for loop, using mysql_num_rows() to count total
//amount of rows returned by $r
for($i=0; $i<mysql_num_rows($r); $i++){
//advances the row in the mysql resource $r
mysql_data_seek($r,$i);
//assigns the array keys, $users[row][field]
$users[$i]=mysql_fetch_row($r);
}
//simple, hope someone can use it :)
// -Kenneth Nash
