gnumed-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnumed-devel] re: tricky


From: Karsten Hilbert
Subject: Re: [Gnumed-devel] re: tricky
Date: Wed, 1 Jun 2005 17:10:43 +0200
User-agent: Mutt/1.3.22.1i

On Wed, Jun 01, 2005 at 10:27:53PM +0800, Syan Tan wrote:

> here's a first implementation of a med record sql exporter.
:-)) This is what I had in mind !

> 1. prevent dumping "domain info" data e.g  type tables , vaccination knowledge
> schemas.
> Also prevents backtracking from static data onto other xlnk_identity or 
> identity 
> It's hard to get a program to distinguish domain data from "instance" data , 
> without
> specifically labelling things.

> 2. cater for xlnk_identity using xfk_identity as both an alternate primary key
> and a foreign key

> 3. order the inserts in dependency order.
>  - the set constraints all deferred; didn't seem to do anything , so had to 
> order
> inserts.

I knew you could do it ;-)

> It uses a temp table of (table, old_id, new_id) at the start of the script,
> to remap the ids, and then "select new_id from id_remap where table=x and 
> old_id
> = id"  for all the key fields ).
Ha, I think that's an excellent idea !

> The script also tries to insert test_type_unified entries to ensure they
> are there before inserting the identity data, but doesn't fail if they are 
> there,
> because they are not part of the transaction.
Agree.

> To test it worked, a modified run of cygwin/install_schema.sh was used which
> had no test data in it.
I just ran the *export* against the public DB at salaam and it
worked flawlessly. Didn't run the import yet, though.

Feel free to check this in !

Thanks,
Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346




reply via email to

[Prev in Thread] Current Thread [Next in Thread]