Re: Feedback Re: [Gnumed-devel] GNUmed Release 0.6.rc1

[Karsten Hilbert]

On Mon, Nov 30, 2009 at 01:11:34PM -0800, Jim Busser wrote:

> > We can do that but only at the cost of having two virtual
> > messages (one for each category) when there are both normal
> > and abnormal unreviewed messages.
> 
> In absence of two virtual messages, I gather you saying that we have only a 
> single virtual message which cannot in a single query distinguish normal and 
> abnormal. Can it however decide that *any* unreviewed abnormals will cause 
> this patient's virtual message to carry this info such as
> 
>       "unreviewed AND abnormal results for patient [...
> 
> (I am guessing no, by virtue of how the query is done?)

To set this straight:

We used to have one virtual message per patient if there
were any unreviewed results for that patient.

We now have one virtual message for each group of normal and
abnormal unreviewed results per patient.

My level of SQL prowess does not allow me to merge those
into one with the message conditional on whether there is
any one abnormal result or not.

> > I raised the priority of abnormal results, though :)
> 
> Does the above refer to a sort-order within the messages
> that are displayed in the inbox and if so can the basis
> (presence) of what accounts for the prioritization be made
> visible?

sort order is .importance, then .received_when (most recent
on top)

Importance is denoted by presence or absence of "!" in the
first column of the inbox

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