Unless you are supporting multiple databases, the extra overhead of
ADODB is generally not worth it. Its serializing the value and using
an md5 of the query as the cache key name. Unnecessary overhead and
it will make deleting cache keys a pain in the ass for a large
application.
If your application doesn't need to delete cache keys, then it
wouldn't be too much of a problem.
Since I deal a lot in MVC environments, I created a function called
simple_model(). Since much of the queries it runs are very simple
check memcache, if not run this query and fetch the results. It
really helps cut down on redundant code and simplifies caching.
Yes, I know it doesn't deal with negative cache results :)
function simple_model($query,$memcache_key,$ttl = NULL,$use_memcache =
False)
{
if ($use_memcache)
{
$result = cache::get($memcache_key);
}
if (! $result)
{
$dbr = mysql_query($query);
mysql_check_error($dbr);
while ($row = mysql_fetch_assoc($dbr))
{
$result[] = $row;
}
if ($use_memcache)
{
cache::set($memcache_key,$result,$ttl);
}
}
return $result;
}
On Jul 13, 2009, at 10:34 AM, Patrick May wrote:
Sounds like you are looking for ADODB:
http://phplens.com/lens/adodb/docs-adodb.htm#memcache
On Jun 24, 3:44 am, Markus <[email protected]> wrote:
My apologies if this mail may appear slightly off-topic.
Is there a php database abstraction layer which uses memcached for
caching?
tnx!
