- many DB connections - the module can connect to many databases on different servers using different DB driver modules (mysql, postgres, ...) at the same time.
- many DB results - the module can store many results of different SQL queries in separate structures at the same time. Thus is possible to work in parallel with several DB results.
- access via pseudo-variables - the content of SQL query result is accessible via pseudo-variables.
- array indexes - fast access to result values via array possition: [row,column].
- persistence in process space - a result can be used many times in same worker process. Query once, use many times.
Here is an example of fetching and printing the content of a table:
...
modparam("sqlops","sqlcon","ca=>mysql://openser:abc@localhost/openser")
...
sql_query("ca", "select * from domain", "ra");
xlog("rows: $dbr(ra=>rows) cols: $dbr(ra=>cols)\n");
if($dbr(ra=>rows)>0)
{
$var(i) = 0;
while($var(i)<$dbr(ra=>cols))
{
xlog("--- SCRIPT: column[$var(i)] = $dbr(ra=>colname[$var(i)])\n");
$var(i) = $var(i) + 1;
}
$var(i) = 0;
while($var(i)<$dbr(ra=>rows))
{
$var(j) = 0;
while($var(j)<$dbr(ra=>cols))
{
xlog("[$var(i),$var(j)] = $dbr(ra=>[$var(i),$var(j)])\n");
$var(j) = $var(j) + 1;
}
$var(i) = $var(i) + 1;
}
}
sql_result_free("ra");
...
Other advantages than the ones listed above against old avp_db_query() function:
- does not mess up fields position if the value in db is null
- use private memory, no locking
- does not iterate through linked (avp) list to get each field, direct
reference to each data structure in result array, thus much faster.
More info at:
No comments:
Post a Comment