Re: [Gnumed-devel] stress test for gnumed

[Karsten Hilbert]

On Sun, Oct 30, 2005 at 10:02:16PM +0800, Syan Tan wrote:

> could someone give a synopsis of how to do the client calls
> in order to load gnumed with a large amount of test data ?
The steps are easy enough but do require work:

1) format a copy of your original data such that you can
generate SQL commands inserting certain pieces of data into
the appropriate tables, for this to work you need to look at
the tables and decide which bit goes where, for an example
look at server/sql/test_data/

2) actually run the inserts

> The test data has the following :-
> 
> past history
table health_issue

> allergies
table allergy

> , rx, scripts,
no tables available, you may want to store textual
representations in clin_narrative to at least keep the
information

> progress notes,
clin_narrative

> letters,
doc_med/doc_obj

> pathol, 
test_result, perhaps lab_request if applicable

> document meta and images of documents,
doc_med/doc_obj

> immunizations,
table vaccination etc

> diagnoses,
depends, may make sense to use as name for clin_episode but
there's also clin_diag

> pap smears,
either test_result or doc_med/doc_obj depending on original data

> measurements.
test_result

> I would like a description on how to call the api

a) instantiate gmPerson.cPerson with the proper patient
b) emr = pat.get_emr()

for vaccinations:

c) vacc = emr.add_vaccination(...)              # instantiate new vaccination 
row
d) vacc[...] = ...                                              # fill with 
optional data if any available
e) vacc.save_payload()                                  # and store it

quite similar for other items

client/business/ should be helpful, eg see gmMedDoc.py for blobs handling

> rx and scripts ? - seems to be an avoided schema topic
yes, no one has taken up the complex task of doing that
properly, however, some tables for "current medication"
exist and are fairly done

> Of course, it would be nice to add the ability to browse other kinds of clin
> items from the emr_browser tree , such as
> pathology and letters and document meta and images
There already is a browser for documents, it's just not in
0.1 (but it does work).

> :- would this clash with the intentions of the gnumed designers ?
No. Not to my knowledge. It much sounds like the first step
on the road to an importer for the data source you have.

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