Re: [Gnumed-devel] Data pack for ICD-9 :-)

[Karsten Hilbert]

On Wed, Sep 28, 2011 at 09:51:00AM -0700, Jim Busser wrote:

> >> Following is the text of the various tooltips, which we
> >> might better regularize to cover the dual-case of the
> >> tooltip having to both
> > 
> > Done.
> > 
> > Karsten
> 
> I hope you were not too fast. : - )
> 
> I am just realizing that it is possible to know a code such as ICD-9
> 
>       250 (Diabetes mellitus)
>       401 (Hypertension)
> 
> and therefore to quickly enter in some cases, from simple familiarity
> 
>       250;401
> 
> into the code box. It is both possible to do this faster than the phrasewheel 
> will show, and it is also possible that in spite of the phrasewheel having 
> opened on the first match, to (inappropriately) ignore the phrasewheel.
> 
> While focus remains in the code box, despite that a phrasewheel is open with 
> a match highlighted, pressing
> 
>       ;
> 
> after 250 will cause the phrasewheel to disappear. Inputting additional 
> characters
> 
>       250;401
> 
> will reinvoke the phrasewheel on a match to 401. However focus remains in the 
> code box, and it is possible to exit the code box *without* pressing 'return' 
> (enter), and therefore *without* 'confirming' any highlighted phrasewheel 
> items, yet keeping in the code box the values
> 
>       250;401

That's possible, yes. As an improvement I have added the
following behaviour:

Within multi-phrase phrasewheels if there does not exists a
data item for each of the strings it will turn yellow.

> and so what I am wondering is whether (once a pending bug is squashed) the 
> backend will store whatever might have been the first match in the 
> phrasewheel even if the user did not specifically 'select' it (if they even 
> saw it).
> 
> IOW is it necessary to wait for the phrasewheel and to 'confirm' a value from 
> within it? It seems to me that dates of birth work in a similar fashion i.e. 
> that there is either a certainty or at least a risk that unless one confirms 
> from these phrasewheels, that value won't be properly stored and recognized

Unless code *outside* of the generic phrasewheel code
explicitely implements "snapping" to the first value (or
any, that is), no, this will not happen with multi-phrase
phrasewheels.

Karsten
-- 
GPG key ID E4071346 @ gpg-keyserver.de
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346