From the earlier System.vdb changes, does Teiid still support
"Groups"
and "elements"? If not we already have conflict in
supporting the old
VDBs right?
If we would need to support both "sys" and "system", we need two VDBs
one with both and one with "sys"? and "sys" being the default?
Thanks
Ramesh..
On Mon, 2010-03-01 at 13:32 -0500, Steven Hawkins wrote:
So it sounds like there's agreement with adding additional
reserved word. Including most if not all of SQL2003, SQL/XML, and SQL/MED.
To resolve internal conflicts, I would propose that the system schema be renamed sys
(this will also allow for an easier drop-in of a legacy system schema should a customer
want it) and to deprecate the use of input as the input group name with the replacement of
inputs (which is more consistent with the other plural group name variables).
----- Original Message -----
From: "Steven Hawkins" <shawkins(a)redhat.com>
To: "teiid-dev" <teiid-dev(a)lists.jboss.org>, "teiid-users"
<teiid-users(a)lists.jboss.org>
Sent: Sunday, February 28, 2010 10:19:34 PM GMT -06:00 US/Canada Central
Subject: additional reserved words for Teiid 7.0
Hello all,
I would like to propose adding additional reserved words. Some of which would be used
immediately others would be place holders for future features and would help prevent users
from using identifier names that will conflict. There is the potential for breaking
existing vdbs though that use any of these words as an unqualified identifier.
This includes:
bigint - type alias for long
decimal - type alias for bigdecimal
real - type alias for real
smallint - type alias for short
tinyint - type alias for byte
varchar - type alias for string
call - to be used as a standard alternative to exec/execute a stored procedure
extract function (to be added) targets:
day
hour
minute
month
second
timezone_hour
timezone_minute
year
Other possible reserved words could include all of
http://savage.net.au/SQL/sql-2003-2.bnf.html#reserved%20word. One minor issue is that the
reserved word input conflicts with our "input" variable group used in update
procedures. Since it wouldn't be used as a table reference, it would break existing
vdbs, but the generated output strings for parsed commands would include input in double
quotes.
Any thoughts?
_______________________________________________
teiid-dev mailing list
teiid-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/teiid-dev