[Babase] new sexskin error_death while ppa

Karl O. Pinc kop at meme.com
Wed May 14 14:54:27 EDT 2008


On 05/14/2008 01:24:02 PM, kfenn wrote:
> Hi again Karl,
> 
> The sexskin sizes are no longer failing to upload due to the Mdate  
> error..  I don't know if the mdates are actually going in correctly  
> or not.  That is still on my list of things to check, but I'm not get  
> stuck on that old error message.

Ok, good.

> My newest error is rather befuddling:
> 
> ERROR: Sxid 12946: Cid 20263: 2000-02-14: There is no sexual cycle  
> (CYCLES) to which this row can be assigned
> 
> I checked a few things:
> SOU  last zdate is 1999-08-22.
> She gives birth on 2000-02-14
> All her observed sizes after the problematic date (2000-02-14)  are  
> size 0 (obviously she's ppa).
> She dies in 2000-11-27.
> 
> I have run into a similar error with two other females (SOU and FED)  
> and there are a  few recurring themes here:  1)  the program starts  
> complaining about "no sexual cycle" after the birth 2) The female  
> dies relatively soon (within a few months or a year) or the date in  
> the error message without having a cycle.
> 
> So the female is ppa and there shouldn't be a row for her in cycles,  
> yet there are observations that she is ppa and we don't want to  
> simply lost those data.  How are other ppa females treated and why do  
> you suppose death during ppa is causing problems with the CYCLES  
> validations?

Death during ppa means that there is no sexual cycle after the
conception cycle.  Our new rule says that any sexkins
measurements after birth belong to the next cycle.  When there is
no next cycle then there's just no cycle, so the sexskins
measurements can't go in.

How about this?  We change the rule about assignment to a sexual
cycle for sexskins so that any sexskins measurements > 0 on or
after birth belong to the next cycle, but if it's a 0 measurement
then it belongs to the conception cycle.

This might do odd things if for some reason your looking at
lengths of cycles by looking at sexskins dates.  Better
check with Jeanne (at least) to see if this is a good idea.
(I don't have a better one at the moment.  Maybe somebody
else does?)  It does not sound like a bad idea, but...

I suppose we could not record the observations.
I guess that depends on the field protocols and when they
write down a 0 and when they don't bother.

I also wonder what was done during the conversion.  Did we
throw out data in a data-cleanup fit?  (pause)  No.  We
didn't have the pregnancy test in the sexual cycle assignment
code so all ppa sexskin measurements were assigned to the
conception cycle.  Right?

If we make this change you'll have to do your "update cid=cid"
thingie again to get all the sexskins  re-assigned to
cycles in a consistent fashion.

Note that because our new rule is sexskin value specific
I'll have to fiddle with the documentation a bit.  It
describes generic rules based strictly on date, rules
that don't take into account the measurement value.

Karl <kop at meme.com>
Free Software:  "You don't pay back, you pay forward."
                  -- Robert A. Heinlein



More information about the Babase mailing list