Good Day Experts!
I have an interesting problem that I'm not sure is really solve-able, at least not at my skill level. The tables have multiple identifier keys because they describe myriad units that are very similar to each other, yet contain little details that make them different. Here is what the three major tables look like: (and how they relate)
The idea is to figure out what all units (for a particular MODEL_CODE) have in common, even if the results do not describe a full, completed unit. I've got lots of mathematical ideas, but I'm stumped trying to relate the data in the given tables properly. It's probably possible, I just don't know how to accomplish this.
And yes, my hope is to understand the SQL statement well enough to (afterward) determine which elements didn't make it into the previous set, and (somehow) separate them off as options. And I'm sure there will be a few (ancient) units in this database that will not belong to any of the common units - but wouldn't that break the "all common elements" set?
The [DETAILS] table contains pieces/parts that make up a unit described in [PRODUCT], and the units described in [PRODUCT] are grouped into various MODEL_CODEs at the [MODEL] level.
I'm trying to say that - for all of the units in [MODEL] group "A" they have these [DETAIL] elements in common. Elements would be made up of PART_NUM.PART_REV combinations. You see - I can describe it, I just can't SELECT it - I keep foundering at INNER JOINS between the tables, trying to make the DETAILS elements "stack up" correctly, then relate back to the PRODUCTs, so that I can group them (or filter them) by MODEL_CODE.
Thanks (in advance) for your timely assistance!
- The Lurking LongFist