Ralph Shumaker wrote: > All this talk of perl tutorials (on the main list) moves me to ask a > question about tutorials for Base (Open Office). > > I'm not sure if I want to buy a manual at this point because I don't > know yet if I want to really get into it beyond my current need. > > So far, the tutorials I have found online are not that great. The one I > decided to try first was written by someone who seems to be only about > one or two levels above me. That would be all right except that he has > aimed his tutorials at people who are many levels below, people who are > *completely* green. _AND_ he diverges often. > > Can anyone point me to good tutorials that assume at least a basic > understanding of databases? > > One thing I'd like to do may not be doable, at least not the way I > imagine it. One small part of what I'd like to do is this: I will have > a table of keys for cars. (Since key means somethings else in database > context, I will refer to the brass kind as "kee".) Some of these kees > (k1, k2, k3) have substitutes (k1a, k1b, k2a, k3a, k3b, k3c) that can be > used. Some of the substitutes (k1b, k3a, k3c) have secondary > substitutes. Some of those secondary substitutes are the original kee > (k1, k3). If a car requires k1, I want to show that it could also take > k1a or k1b. If another car calls for k3a, I want to show that it could > also take k3, and by extension, k3b and k3c. I would prefer to *not* > have two separate tables with data entered twice (increasing the risk of > typos, and preventing corrections in one list from propagating to the > other). But the only kind of referencing I can think of ends up being > circular since I don't know of a way to get one record to point to > another record as an acceptable substitute. It's complicated. It's not > a one to one relationship either, which complicates it further. It > would be nice to have a second table which has each record linked > > table-1 > --------- > rec kee > num info > --------- > 0 k1 > 1 k2 > 2 k3 > 3 k4 > 4 k1a > 5 k1b > 6 k2a > 7 k3a > 8 k3b > 9 k3c > > table-2 > ------------- > rec pri sec > num kee kee > ------------- > 0 0 4 > 1 0 5 > 2 1 6 > 3 2 7 > 4 2 8 > 5 2 9 > 6 5 0 > 7 7 2 > 8 9 2 > > Line 3 of table-1 isn't even referenced in table-2 because it has no > substitutes. > > That's only a small part of what I want to do. > > I have some knowledge of database work, but not graphical. I worked in > the guts of FoxBasePlus (not Pro). I took a program for work and > tripled its size (or more) by taking away the ugly stuff, adding > features, and making it easier to work with. I had *zero* knowledge of > database programming before that. But good knowledge of BASIC gave me a > *huge* leg up on it because of similarity in terms and function. I only > had to learn the stuff that was *not* similar, and most of that was > self-explanatory from the context in the old program. And the rest I > learned from other sources. >
I haven't the slightest acquaintance with OpenOffice database capabilities. ..BUT.. Database design questions are pretty fun! In your example, it's not quite clear what the meaning or significance of original and substitute really is. What I see is kees and cars (kars?) where any given car has zero-or-more associated kees. kee car --- --- k1 bluecar k1a bluecar k1b bluecar k2 greencar k2a greencar k2b greencar <NULL> batmobile ... etc You presumably want to describe something more. Can you elaborate? Perhaps it would help to pose particular questions you want the database to be able to answer. What kees mate with car=bluecar? Regards, ..jim -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
