Anyway to make the code more efficient...

ajdegans

$return="SELECT * FROM bl_counter WHERE hits>'0' AND hits<='5'";
$return2=mysql_query($return) or die("Could not select returning hits 1");
$return3=mysql_num_rows($return2);
print "<tr class='mainrow'><td colspan='3'>$return3 returned 1 to 5 times";

$return_2="SELECT * FROM bl_counter WHERE hits>='6' AND hits<='10'";
$return_22=mysql_query($return_2) or die("Could not select returning hits 2");
$return_23=mysql_num_rows($return_22);
print "<br>$return_23 returned 5 to 10 times";

$return_3="SELECT * FROM bl_counter WHERE hits>'10'";
$return_32=mysql_query($return_3) or die("Could not select returning hits 1");
$return_33=mysql_num_rows($return_32);
print "<br>$return_33 returned over 10 times";

Hey there, is there a way to make this somewhat more efficient?
As you might have seen its used to show returning visitors from a database (mysql), but a query for each number is not very efficient i think. Does anyone know a way to make it better? I cant think of any...

Thanks in advance 😃

AstroTeg

You could use SQL's COUNT() function in the select clause, which would also require a group by clause. Then you can eliminate your mysql_num_rows() function call. That could speed things up a little.

MySQL doesn't handle subselects very well, which is what you'd want to use here. Your alternative is to group all your data together and count it all. Then store it in a temp table and then run your 3 queries against the temp table. This would work if you're working with a ton of data.

I'd say hold onto your current chunk of code and experiment and keep track of the amount of time various approaches take. Then pick the fastest one...

ajdegans

ah ok... one thing though

with your suggested temp table wouldnt i have 4 queries then?

my point was to lessen that...

ANyway, ill look into the count thing... THanks for the ideas 🙂

AstroTeg

With MySQL, speed isn't always created from reducing the number of queries (in other RDBMSs this usually isn't the case).

The idea with creating the temp table is to do a few things: have MySQL only do the grouping and counting ONCE. Then you have that data tossed in a temp table (which really means you're tossing it in RAM which is usually uber-fast). Then when you run your 4 queries, the brunt of the work has already been done and the query time should be pretty fast.

Although without testing, this can be easily argued. It depends on how your data is structured and how much of it you're dealing with.