Re: [Gnumed-devel] (no subject)re: reverting

[Karsten Hilbert]

> As I said before, I'm in the process of reconstructing the old directories 
> by downloading each previous versioned
> file in a "revert" branch, which can be overlayed over the client main 
> branch;  so if the consensus is to restore old files, just let me know, and 
I think we are rather trying to work with what happened
to get in there now.

> Unfortunately, the cvs doesn't distinguish its own commit comments from the 
> rest of the program, so any changes in these areas will cause a commit and 
> comment change: much of the commit was new files, and gmClinicalRecord, 
> gmDemographicRecord, gmEditArea, gmGP_PastHistory, gmGP_Allergies,
> gmDemographics. The others had no or little code changes.
But they still got committed without a meaningful commit
message.

> The point of the exercise was to produce a client that called the modules in 
> the business directory ( and existing code - Karsten's or Ian's - if the 
The intention is understood and acceptable.

> I'm not so sure it helps to understand the code by reading the comments; so 
No, but it helps to decide whether I need to look at the
changes in order to find out whether my code is affected. The
commit messages are supposed to tell me what has been changed.

> There are parts of the gui which I don't understand, like the StikoBrowser, 
Simple a browser for an HTML file from our National
Vaccination Committee.

> and EMRTextDump
A text dump of the content of all descendant tables of
clin_root_item. Ordered by age of commit to database.

> The client behaviour that has been committed in is:
> login as test-doc.  Choose first menu item and patient select.
IMHO "new patient" should be a menu item, too.

> and  produces a blank patient with dummy names ( yes , I did change the sql 
> in gmTmpPatient,
gmTmpPatient ceased to exist several weeks ago

> to just insert into v_basic_person , as this worked easily)
create_dummy_identity() is named as such for a reason. It
creates a dummy identity, not a basic_person. This was done
after some thought and quite a bit of discussion (with Ian, on
ICQ). Not because it worked easily. We are not in the business
of reinventing Windows. I reverted to the old behaviour and
thusly intentionally broke code that may have relied on the
new behaviour thus making it easy to find.

> Edit the 
> patient name, sex, date of birth. Select an address type ( home, work ) and 
> edit the number, street , suburb, postcode. Street , suburb and postcode
> fields have PhraseWheel functionality ( if the suburb table is loaded).
What is the suburb table ?

> Press add address to add the address. Can add another address of a different
> type. Double click an existing address in the list of patient addresses to 
> edit the address. click delete address
> to remove address. 
> Click save to save the person. 
Sounds good all in all. Sorely needed functionality.

> Currently , updates to allergies will appear in the editarea list, but will 
> only appear in the top panel if the patient is re-selected
That's a bug in the top panel then. It's supposed to listen to
allergy-update intra-frontend signals. Not sure whether I
implemented that yet.

> ( by pressing enter in the focussed patient 
> select field in the top panel).
I removed reload from gmCurrentPatient.__init__() as that's a
workaround for broken logic and laziness to fix.

> Really, not much has been done, because the gui (as a proprosed functional 
> client) is ahead of the business layer.
True.

> If the current minor functionality added is acceptable, then someone should 
> work on a basic prescriptions business
> object , and possibly a vaccinations object. Someone could re-do the 
> allergies and past history business objects
They aren't really business objects. They are widgets with the
GUI parts moved into other files.

> I wrote, but I think trying to get a small part working perfectly rather 
Not perfectly but rather cleanly.

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