result set, result pointer

[rs51]

i've no programming background whatsoever, but i like php+mysql, though i'm very newbie at both.
(i must confess i have a cms but was mostly copy&paste)
i learned how to connect to mysql from php and then wondered how to list the databases i've at mysql.
in the manual i found this:

[tt]<?php
error_reporting(E_ALL);
$db_list = mysql_list_dbs();
$i = 0;
$cnt = mysql_num_rows($db_list);
while ($i < $cnt) {
echo mysql_db_name($db_list, $i) . &quot;\n&quot;;
$i++;
}
?>[/tt]

of course this worked, but i want to understand what i'm doing, so:
there i read this:
&quot;mysql_list_dbs() will return a result pointer containing the databases available from the current mysql daemon&quot;
(i guess daemon is my mysql)
As i didnt understand what a pointer is, i googled and found this:
&quot;A pointer is a variable that holds a memory address&quot;
hmm... so now i know that that function returns a kind of a group of variables...
hmm...is this a kind of 'the' RECORDSET used in vb?
well... i made:

[tt]$cnt = mysql_num_rows($db_list);[/tt]

and got how many databases i had.
Next i read this (in the manual):
&quot;The row parameter is an index into the result set&quot;
Now i'm getting more and more confused!
What's the diference between result set and result pointer?
i really feel very confused!
dont get that diference and dont get this last piece of sentence:
&quot;...is an index into the result set&quot;
Can someone help me please?
thanks in advance

 

[KarveR]

look at it a different way, sometimes odd angles help

// get an array of database names from the local mysql service
$db_list = mysql_list_dbs();

// while you have an array, assign each item or sub array to a pointer ($myrow)       
while ($myrow = mysql_fetch_array($db_list)) {

// print out the value of the first element of $myrow ([0])          
 echo &quot;$myrow[0]<br>&quot;;

}





______________________________________________________________________
There's no present like the time, they say. - Henry's Cat.

 

[sleipnir214]

PHP's manual is a little unclear on nomenclature sometimes. This is particularly so in the database sections.

In general, a pointer points to something. In the case of PHP, when you use the fopen() function to open a file, what is returned is a &quot;handle&quot; to the file.

&quot;Handle&quot; is a good metaphor. It's like the handle on a suitcase. The clothes in the suitcase is the contents of a file. The suitcase itself is the file on your file system. The handle is the thing you use to manipulate the suitcase -- or the file.

The same is true of database access in PHP. mysql_query() returns a handle to a result set. It's not the actual data, just a way to manipulate the data. It requires special functions, like mysql_fetch_row() to get at the data.



The result set itself contains a data pointer. That pointer simply points to the &quot;current&quot; row of the data returned, so that if you read a row from the result, you know where to read from each time you access the data in the result set.


So, when you query a database to fetch data, the database server returns a result set. PHP holds on to that result set for you and maintains a result pointer, which keeps track of the last place you read from in the data. PHP then gives to your code a result handle, which you use to manipulate the data.

Some simple code with comments:

<?php
$DB_conn_handle = mysql_connect('localhost', 'test', 'test');
//$DB_conn_handle now contains a database connection
//handle.  Connection handles are different from result
//handles.

mysql_select_db ('test', $DB_conn_handle);

$query = &quot;SELECT foo FROM bar&quot;;

$result_handle = mysql_query($query, $DB_conn_handle);
//mysql_query() has passed my query to the server, and
//fetched the result set.  It returns a result handle,
//which I am storing in $result_handle

while ($the_row_of_data = mysql_fetch_assoc($result_handle))
{
   print $the_row_of_data['foo'];
   print '<br>';
}
//in each iteration of the loop, mysql_fetch_assoc() gets the
//current row of data from the result set.  It uses the result pointer,
//which is initialized to point to the first row of the result set.
//Each time mysql_fetch_assoc() fetches a row from the result set, the
//result set pointer is moved to the next row.  When there is
//nothing more to fetch, mysql_fetch_assoc() returns FALSE.
?>

Want the best answers? Ask the best questions:

http://www.catb.org/~esr/faqs/smart-questions.html

TANSTAAFL!!

 

[rs51]

for sure this way looks nice and clear!
let me see if i got it right (i'll transpose your code to a case where i want to obtain values from a select statment):

$sql = 'SELECT * FROM pai';

$result = mysql_query($sql, $cnx);
// i got an array that contains so to speak
// the 'recordset'/result set of my query

// now i'm gonna use a pointer (cursor?) called $row
// i guess each 'item'/piece of the array is gonna
// be put into my pointer

while ($row = mysql_fetch_array($result)) {
echo ('<p>'. $row['nome'].' '.$row['apelido'].'</p>');
}
// the loop prints every element of the pointer array

sorry about my ignorance, but i dont get what you mean when you say:
&quot;// print out the value of the first element of $myrow ([0])&quot;

i guess the loop prints ALL, not just the $row([0]) but all $row(), beeing i=0 till i='eof', or = number of array members.
i say this because i tested exactly your code and prints all databases - and that's what we want

 

[rs51]

sleipnir214:
what i wrote above was made before i had a chance to see your post
Thanks for your explanation.
I read it very carefully and fully apreciated it. Thanks a lot!
i just have a doubt:
at some point you say:
&quot;...The result set itself contains a data pointer...&quot;
after, you say:
&quot;...PHP holds on to that result set for you and maintains a result pointer,...&quot;
you mean data pointer the same as result pointer?
i guess yes
thanks a lot

 

[sleipnir214]

Yes.

Sorry about that. I was trying to keep my nomenclature standardized. I tend to use &quot;pointer&quot; as my default term -- my &quot;organic&quot; programming language is c-language, where you do lots of things using explicit pointers.

But without trying to muddy the waters, everything we've been discussing up to now are all pointers: the result handles, the result pointers, the connection handles. It's just metaphorically useful to discuss them as very differeent things.

What you can do with a pointer is decided by what that pointer points to. PHP keeps pretty good track of what a pointer points to, and will keep straight what can be done with each type. Other languages, like c-language, will let you do all kinds of squirrely things with pointers.

But to keep all the metaphors straight:
[ul][li]mysql_connect() returns a connection handle.[/li][li]mysql_query() can (with a SELECT query) return a result handle.[/li][li]mysql_fetch_[row|array|assoc|object]() will return a row of the result set as a numeric array, a combination numeric and associative array, an associative array, or an object, respectively.[/li][li]PHP (or more correctly the MySQL communications libraries) maintains a result pointer that keeps track of the &quot;current&quot; row of a result set.[li][/ul]


Want the best answers? Ask the best questions:

http://www.catb.org/~esr/faqs/smart-questions.html

TANSTAAFL!!

 

[rs51]

that's perfect, thank you!