[Babase] The order in which sexual cycle tables load

Karl O. Pinc kop at meme.com
Tue Sep 26 16:27:37 EDT 2006


On 09/22/2006 10:23:54 AM, kfenn wrote:
> Hi Karl,
> 
> I had line command nightmares all last night....I hope you're happy.

Welcome to the inner circle. :-)

> 
> Anyway, I'm recovering.  Thanks for the conversion sequence.  I've  
> found it very helpful.  I have some follow-up questions:
> 
> What about other female reproductive tables....where do they fit in?   
> The cycles table, a primary source table, didn't appear in you list.   
> When does that get loaded?

The cycles table in foxpro turns into the cycpoints and cycles tables
in postgres.  In some sense it does not really get loaded at all
because it's pretty much constructed from the cycpoints table by the
database itself.  The exception is the sname.  Anyway, you can think
of it as being loaded at the same time as the cycpoints table,
concurrently.

> Also what about the analyzed tables (repstats, cycstats, cycgapdays,  
> mm and md intervals)?  Are they auto-generated as part of this  
> conversion process or does that happen after we get the bugs with the  
> primary source tables worked out?

Cycgapdays gets auto-generated.  Repstats, cycstats, mm, and md
must be generated manually by running a program.  When I was down
at Duke I showed Leah how to do repstats and cycstats.  The last
item on Susan's list from our last phone meeting is to get those
into the documentation.  the Mmintervals and  Mdintervals tables have
been overlooked until now.  I'm not sure whether there's a
stand-alone build program for them.  I'd like to do those next,
along with the automatic generation of those tables and of repstats
and cycstats.

> 
> Finally, I'm curious about the "other stuff happens here" part of  
> your email.  I don't need to know much about the specifics, but very  
> generally what types of "other stuff" are happening?

After biograph comes the census and demog tables and members is
automatically generated.  Then somewhere around maturedates,
rankdates and dispersedates also goes in.  Also, the
statdate and status on biograph get set, which keeps
interpolation from taking too long because otherwise it's always
working with the individuals full lifespans as the
census tables load.

Finally after everything listed below the interactions and
point sample data gets loaded.

> 
> 
> Tabby
> 
> Karl O. Pinc wrote:
>> Tabby asked me for a document showing the order
>> in which sexual cycle related tables and data
>> loads.  This is that:
>> 
>> biograph, except for Pid
>> (other stuff happens here)
>> maturedates
>> (other stuff happens here.)
>> cycgaps, (from real_gaps, the real gaps we want)
>>          (The latest conversion errors
>>           are from starting the coversion with the
>>           loading of cycgaps.  Errors from earlier
>>           do not show up (and are not checked.))
>> cycpoints
>> sexskins
>> cycgaps, (fake gaps (fake_gaps), so we can put in pregs)
>> pregs, except for Resume
>> biograph, the Pids are put in and biograph is now complete
>> cycgaps, (undo_fake_gaps) the fake gaps are removed.
>>          When the fake gaps are removed the pregs.Resume
>>          is automatically set by the triggers.
>>          This leaves pregs complete.
>> 

Sorry it took so long to respond.

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