[Maypole] Auto Increment Primary Keys (and Documentation)

[Gordon Haverland]

The way I thought I had to set things up, {\it vis a vis} database 
tables started requiring quite involved SQL queries for what 
should be fairly standard operations.  So, I had to backtrack and 
change things.

I have a couple of tables which are related, one of which has an 
auto_increment primary key.  I can create a tuple (row/object) of 
data without ever mentioning this primary key as I understand 
things.  Fine.  But, is there some easy way to find out just what 
value was assigned to that variable?  For all the power of all 
the modules behind Maypole, do I really need to search for the 
maximum value in the primary key column after I create the tuple?  
It seems to me, that if there is a high access rate on the 
application, that this isn't going to get me the primary key 
value I want anyway, some other process may have created another 
tuple in that same table between when my process created the 
tuple and when it searched to find the maximum value of the 
primary key.

I would have thought this would be a common requirement, since the 
value of the primary key column in a tuple is often used as a 
foreign key in other tables.  But I just don't see it anywhere.  
Are the creates for the associated tables supposed to get this 
value automagically?

On the wishlist side of things......

Documentation or examples in general of all Maypole stuff are 
still needed.  Beer (and Maypole) looks very much like: we want 
almost everybody to be able to look at every table.  Flox looks a 
bit more like my (first) application, I really only want an 
authenticated user to be able to access a single row of a view of 
some part of the database tables, most of the time.  Since SQLite 
only has read-only views, this means doing everything explicitly 
as near as I can tell (and not trying to build a SQL view).  But 
searching the documentation usually produces little or nothing at 
most turns, so I am guessing a lot.  My guessing is getting 
better, but I still wish the docs were better.  At this point, 
the rewrites required I think are producing a better product.  
But it would be nice to be able to get this to the point where 
(l)users (of my project) could start trying it out.  :-)

Gord