NIST Test Suite
Christopher Kings-Lynne <chriskl@familyhealth.com.au> writes:
Are we able to run more NIST tests now?
http://www.itl.nist.gov/div897/ctg/sql_form.htm
I thought we'd extracted all the interesting juice from the NIST tests
a couple years ago. Specifically I recall this fix came out of NIST
testing done by Red Hat:
2003-06-06 11:04 tgl
Implement outer-level
aggregates to conform to the SQL spec, with extensions to support
our historical behavior. An aggregate belongs to the closest query
level of any of the variables in its argument, or the current query
level if there are no variables (e.g., COUNT(*)). The
implementation involves adding an agglevelsup field to Aggref, and
treating outer aggregates like outer variables at planning time.
regards, tom lane
I thought we'd extracted all the interesting juice from the NIST tests
a couple years ago. Specifically I recall this fix came out of NIST
testing done by Red Hat:2003-06-06 11:04 tgl
Implement outer-level
aggregates to conform to the SQL spec, with extensions to support
our historical behavior. An aggregate belongs to the closest query
level of any of the variables in its argument, or the current query
level if there are no variables (e.g., COUNT(*)). The
implementation involves adding an agglevelsup field to Aggref, and
treating outer aggregates like outer variables at planning time.
I was just chatting with Neil C on IRC and he mentioned that back when
they were using it at RedHat, PostgreSQL didn't have schemas so most
stuff failed. Was just wondering if that's true and we need to re-run
them or something.
Chris
Christopher Kings-Lynne <chriskl@familyhealth.com.au> writes:
I thought we'd extracted all the interesting juice from the NIST tests
a couple years ago.
I was just chatting with Neil C on IRC and he mentioned that back when
they were using it at RedHat, PostgreSQL didn't have schemas so most
stuff failed.
I think he's remembering an earlier iteration. The mid-2003 patch I
cited is well past 7.3's release date, so it was surely inspired by
testing with a schema-aware PG.
My own recollection is that the other stuff we found in that test cycle
had to do with locale/collation features (eg per-column collation
options), and so there's no point in retesting until some progress is
made on that front ...
regards, tom lane