[Babase] UPLOAD pregs error
Karl O. Pinc
kop at meme.com
Fri Jan 4 14:11:09 EST 2008
On 01/04/2008 12:56:18 PM, kfenn wrote:
> Hi again Karl,
>
> So can we rename the database Hydrabase 2.0? Every time I try to fix
> a problem I have two more grow in it's place.
:-)
>>
>> What goes in the pregs table are Cpids (CYCPOINTS keys), not
>> Cids (CYCLES keys). I suspect you put in a Cpid of a CYCPOINTS
>> that belonged to LYE.
> Yes, I was using the cid, although I really though I had copied the
> dcpid from mtd_cycles. Thanks for the fix.
>
> Here's my latest issue:
> So let's review what I've been doing:
> I tried to update the census table.
> But I can't load new individuals until they appear in Biograph.
> So I tried to load new baboons into Biograph.
> But I couldn't load them in until I had loaded the pregs from which
> they came.
> So I tried to load the pregs. But I can't load the pregs until I have
> the cycles loaded.
> So I tried to load the cycles.
> But I can't load the cycles until I insert the maturedates.
> So I tried to load the maturedates.
> But I can't load the maturedates becaues they all fall after the last
> census days and seen ti violate some date rule.
> So I tried to update the census table.....ooops I seem to be at the
> top of my list again.
>
> I think I'm going to have to do the update in two stages....one for
> old individuals and one for new ones because of the loop. I don't
> have a grand plan at the moment...so I'll work on this and you can
> weigh in with anything that might be helpful.
I swear at one point during the conversion we had a grand plan to
cover this.
Might it have been:
Insert new individuals without a pid?
Do census
Do maturedates
Do cycles
Do pregs. Insert fake gaps if a preg has a resume but
is not yet referenced in biograph. (This should be
a rare occurence.)
Put pids into new individuals in biograph.
Remove fake gaps, if any.
?
Fake gaps arn't needed if the time periods are short enough.
In other words, if you run through the whole list with
only a week or a month of data rather than 6 months or
a year.
I think 6 months is short enough, but you could also
cycle through a month at a time and then I don't know
that the fake gap issue would ever come up. Time will
tell.
With short time intervals you also don't need
to go back and put in the pids later.
That was something like what I came up with for the
conversion, which dealt with exactly the same issues.
At the time I thought I wrote stuff regarding this.
You may be able to come up with a better approach.
The "cycle" breaks if the time intervals entered are
small enough. When data goes in in a strictly chronological
order, across all tables, then there's no issues at all.
Obviously, we don't enter data that way. Mostly, I think
it's the cycles that need 'big chunks' of data entry,
because they need review over time so particular dates
can be picked.
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