Push versus Pull

[wiskers]

I'd like to know if there's alot of difference (performance wise) between:

- executing a query, looping the result rows & adding them to an array, then assigning them to the template, then looping through them once more to display.

- executing a query, assigning result to template, then looping over result set (mysql_fetch_assoc()) to display each row.

Also, second method: Will there be an accumulation of memory if I assign like 5 different result resources before I begin looping over them? ($result = mysql_query(..); assign to template)

Thanks

[Gods Lonely Man]

- executing a query, assigning result to template, then looping over result set (mysql_fetch_assoc()) to display each row.

I think this is faster because of the extra loop. Other than that they are the same?

Every variable in a script comsumses memory, until you "free" them. So yeah, there will be an accumulation of memory if you have 5 different result resources.

[kyberfabrikken]

- executing a query, looping the result rows & adding them to an array, then assigning them to the template, then looping through them once more to display.

- executing a query, assigning result to template, then looping over result set (mysql_fetch_assoc()) to display each row.

#1 will load the entire result into php's memory, whereas #2 will load on demand. For large resultsets, this can have quite some impact on memory-use. The processing overhead of the extra loop is quite negligible in comparison.

From a design perspective, I wouldn't want to use mysql_* functions from within a template - You could get around this by using a wrapper for the resultsets. If you're using PHP5, you could let the wrapper implement Iterator, which will let you treat it like it was an array (use foreach)