Am 02.10.2012 22:52, schrieb Rene Kluwen:
The patch is a little bit difficult to read, but I get the concept.
One disadvantage of this patch is that it is backward incompatible. You need
to re-create the sqlbox tables and thereby you lose your old information.
If you are unaware of this incompatibility, and not recreate your tables,
sqlbox will suddenly fail to work.
Have you thought of a solution for this problem?
that's a viable argument, I agree. We need to inform the user that he
needs to ALTER the table schema in order to have the 'foreign_id' available.
I would suggest even to aim a bit higher here:
a) remove the explicit msg struct field usage in the sqlbox code base,
and use our pre-processor magic around the gw/msg-decl.h code to create
all the required string dependencies here.
b) this allows us also to perform "field exists" tests on the table
schema, and if a field doesn't exist we can create/delete it via ALTER
SQL command.
Doing a) is considerable easy, it's only some effort to make it right
here. Doing b) is a bit harder, since this will be DB type specific I guess.
Anyone having ideas on how schema changes can be "tested" against via SQL?
Stipe
--
-------------------------------------------------------------------
Kölner Landstrasse 419
40589 Düsseldorf, NRW, Germany
tolj.org system architecture Kannel Software Foundation (KSF)
http://www.tolj.org/ http://www.kannel.org/
mailto:st_{at}_tolj.org mailto:stolj_{at}_kannel.org
-------------------------------------------------------------------
