[Novalug] Database advice

John Holland jbholland@gmail.com
Tue Mar 16 09:47:52 EDT 2010


On 03/15/2010 09:32 PM, Jon LaBadie wrote:
> On Mon, Mar 15, 2010 at 03:54:29PM -0400, Bonnie Dalzell wrote:
>    
>> On Mon, 15 Mar 2010, John Holland wrote:
>>
>>      
> ...
>    
>>> Litters could be identified by birthday, without requiring an addtional table
>>> (mom + birthday =  unique litter) , or there could be an additional Litter
>>> table with links from the dogs to that as well. (a dog has a Litter ID
>>> attribute)
>>> And probably links from the litter table to the mother&  father dogs.
>>>        
>> Actually now that we do DNA parent testing on litters it is possible to
>> have pups with different fathers born on the same day from the same
>> mother.
>>      
> Might there also be multiple birthdates for the same litter.  Birthing could
> certainly begin on one evening and end the next day.
>
>    
OK, how about: LITTER_ID (autogenerated)   and    LITTER_BIRTH_START 
(DateTime) LITTER_BIRTH_END (DateTime)
Dogs could have their own BIRTH(DateTime) which should be within the 
range, although that would
not be a typical referential integrity constraint.





More information about the Novalug mailing list