Subject | Re: [Firebird-Architect] Re: Java Stored Procedures |
---|---|
Author | Ann W. Harrison |
Post date | 2005-10-20T18:59:34Z |
paulruizendaal wrote:
is that the system flag is currently 0 for user tables, 1 for
core system tables, and 2 for QLI tables. The qli table and field
names are of the format QLI$xxx and are added on first reference.
That might not be a bad model for JAVA tables - a JAVA$ prefix
and system flag value of 3.
The second thought - probably way to late for Firebird2 - is to
make all the flag fields in the system tables NOT NULL. As far
as I know, the only way to create a null value in those fields
is a direct system table update, but there's a fair bit of code
that checks for nulls....
Cheers,
Ann
>That brings to mind a couple of system table issues. The first
> For me it is *not* a priority, but adding a RDB$JAVA table, with
> name, source and bytecode would be an option....
>
> The RDB$JAVA table could be created on the fly when first referenced
> by the SQL compiler and/or the meta data system.
>
is that the system flag is currently 0 for user tables, 1 for
core system tables, and 2 for QLI tables. The qli table and field
names are of the format QLI$xxx and are added on first reference.
That might not be a bad model for JAVA tables - a JAVA$ prefix
and system flag value of 3.
The second thought - probably way to late for Firebird2 - is to
make all the flag fields in the system tables NOT NULL. As far
as I know, the only way to create a null value in those fields
is a direct system table update, but there's a fair bit of code
that checks for nulls....
Cheers,
Ann