[ic] Performance problem found - Help needed solving

[Christian Brink]

Ok from my SQL profiling I found:
|         COSTS        |IDX|     Query              |
Total  |  Avg  | High  |   |                        |
28101.1 5620.22 5620.22 SEQ  select * from products

I pretty sure this comes from the component 'random'
[loop
  random="[control matches 2]"
  search="
    fi=products
    st=db
    ra=yes
    ml=[control pool 1000]
  "
]

I'm guessing 'ml' doesn't use PostgreSQL's 'limit' since it non-portable

My questions are:
1. Can I use PostgreSQL's limit in a loop/search or do have to go to a query?
2. Can I limit this based on an area or cat in the loop/search?
3. One of the reasons I'm asking the previous questions is I can not find
the syntax for 'search' I've picked up 'ra' mean return all 'fi' is file,
and 'ml' is limit from reading the docs but other than that I'm at a loss.
Where can I find a guide on the 'search' syntax?

Thanks
Christian