Stephan Szabo <[EMAIL PROTECTED]> writes:
> On Mon, 15 Aug 2005, Dario Bahena Tapia wrote:
>> The final result seems to be the same, I just was curious about the
>> standard behavior. Does the SQl says something about this execution
>> order?
> I believe SQL defines the order to pay attention to parens, so A join (B
> join C) style clauses result in a "table" being derived from B join C and
> another from A joined with that table.
SQL only constrains the results, though. It does not forbid the
implementation from doing the work in whatever way seems best to it,
so long as the results are the same (and "same" does not consider
row ordering).
For example, SQL92 3.3.4.4 says
A conforming implementation is not required to perform the exact
sequence of actions defined in the General Rules, but shall achieve
the same effect on SQL-data and schemas as that sequence.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster
