Re: pgbench more operators & functions

Hello,

> Sorry for the changing the status of the patch against to the current 
> status. While going through the recent mails, I thought that there is 
> some disagreement from committer.

If so, I'm willing to explain again why these operators are useful for 
writing some benchmarks, for instance, this paragraph taken randomly from 
the TPC-B specification, on page 16:

"""  The Account_ID is generated as follows:  • A random number X is generated within [0,1]  • If X<0.85 or branches =
1,a random Account_ID is selected over all <Branch_ID> accounts.  • If X>=0.85 and branches > 1, a random Account_ID is
selectedover all non-<Branch_ID> accounts.
 
"""

This extracy suggests clearly that having some comparison operators and 
some ability to act upon the comparison result is required to implement 
this particular benchmark, which is beyond pgbench current capabilities.

> Moved to next CF with "ready for committer" status.

Ok. We'll see next time what becomes of it...

-- 
Fabien.