All my texts talk about on delete and update properties for foreign keys. Do they work with primary keys, too?

You should probably read up on foreign keys, what they represent, and the details on update/delete issues. Primary keys are a completely different thing in this respect.

rbfree, could you do us a favour? please use a descriptive title for your threads

Re. the reading advice, I've given no shortage of time to the reading, but as always with learning I have some kinks to work out. In this case, I worked it out after rethinking it. Nevertheless, I always appreciate a fresh perspective.

And, certainly yes on the descriptive titles.

Actually, no foreign keys without primary keys (or at least, "alternate" keys)!
That is, a foreign key on table C refers to a (sort of) primary key in table P.
Otherwise said, a foreign key is a property of a pair of tables, (C,P); it "points" from table C to table P.
(To complicate things even further, tables C and P might even be the same table ;-)

Delete and update properties of a foreign key are thus properties of this "pointer" between the two tables C and P. They will limit some of the allowed actions on one of the tables, or "trigger" additional actions on one of the two tables.
Which actions and what restrictions? That's exactly what the property tells us. E.g., an "update cascade" will have different restrictions and side-effects than a "delete restrict".

regarding foreign key actions, perhaps this might help -- Relational Integrity -- r937.com

Rudy,

Nice work on the article on relational integrity. Believe it or not, I did manage to get a primitive but working concept of relational integrity and even normalization via reading a few texts and pestering the kind people on this list.

But then came the next learning phase which is to learn to use sql (ddl) to create these structures, these normalized (and then denormalized), interrelated tables. That created a few confusions that I expressed when posting this thread. But after a few mental revisions, I managed somehow to get a primitive but working understanding of the process (relating a formal conceptual design to a physical db using sql). And, with help by you and others, many texts, and some samples from the web, I've managed to create most of the ddl code I'll need.

Now, I need to test it (and parts of it) which requires getting mysql established (installation and connection) on machine... which turns out to be more complex and contingent than I would have imagined.

Sooo... when I get over that hump I'll probably be back with more questions. Meanwhile, thanks for all help.