Date: prev next · Thread: first prev next last
2013 Archives by date, by thread · List index

On 02/11/2013 05:31 PM, Uwe Brauer wrote:
"Dan" == Dan Lewis <> writes:
    > `
    > On 02/11/2013 09:18 AM, Uwe Brauer wrote:
    >> Uwe Brauer
    >      How do you define an "intrinsic database"? I do not understand
    >      the term.

Well in my understanding there are two possibilities,

     -  either Base can connect to an external database such as mysql via
        a number of drivers or

     -  it uses a database it ships (I think it is based in
        hsql). Because of a lack of a word I denoted this as intrinsic.

    >      Primary-foreign key pairs are created when a table is
    > normalized. (There several levels of normalized tables.) These pairs
    > are based upon the relationships between the fields of the original
    > table.
    >      So, you probably should put all the fields of the two tables into
    > a single table with the exception of any fields that belong to both
    > tables. For example the single table contains only one primary key
    > ("ID") and one field named "Notes". Then look for any fields (other
    > than "ID") that determine the values of other fields. (In other words,
    > begin to normalize the table.) Without knowing all of the fields that
    > you are using, it would be difficult to know what fields these might
    > be.
    >      FYI, very seldom will two tables have a 1:1 relationship (linking
    > the primary keys of the two tables). Very likely, yours doesn't
    > either. Most often, these relationships involve a primary key of one
    > table and a foreign key of the other.

I am not entirely sure I understand: I set up a new base, no the primary
keys are not any existing field of the table (autovalue is On)

My tables are the following (they are just toy models in order to
understand the relation ship concept)

Name Type
Id   (generated by OO as a primary key)
Task Id Integer
Description Memo Longchar
Notes Memo Longchar

Name Type
Id (generated by OO as a primary key)
Author Id Integer
FirstName Memo Longchar
Nationality Notes Memo Longchar

Now I try to make a relation between the two tables, and in the GUI I
select the field Task Id and Author id in the other table, since both
fields have the same type.

I receive  the following error:

SQL Status: 23000
Error code: -177

Integrity constraint violation - no parent 1, table: Authors in
statement [ALTER TABLE "Authors" ADD  FOREIGN KEY ("AuthorID")
REFERENCES "Tasks" ("ID")]

Now the gui allow me to add more fields so I add both primary keys, gain
an error:
SQL Status: S0011
Error code: -170

Primary or unique constraint required on main table: "Tasks" in
statement [ALTER TABLE "Authors" ADD FOREIGN KEY ("AuthorID","ID")
REFERENCES "Tasks" ("TaskID","ID")]

The only one which works, using the GUI is between the primaries keys!!!

Did you try it out your self and you are able to set more relations
between the tables?

As I said, in another mail,  when I connect the data base to mysql via
ODBC then I can set more relations, but they are not saved..

      Some basics about creating a database:
1) You begin with data that you want to use for some purpose or purposes.
2) Then you design the database to organize this data so that it will be useful.

For example: an address database:
In the beginning, the database consists of information such as names, addresses, email address, phone numbers, etc. The first thing you would want to do is to organize this data so that a name is linked to its address, its email address, its phone numbers, etc. Now you have rows of data, each one of them containing information about a single person (a relationship exists between the data for each row). Each of these pieces of information is a field. If you look at these rows, you will see that they also contain some things in common. Several rows can contain data about your friends, others are about your relatives, and others are about companies you do business with. You probably know of other possibilities. So, these rows can be organized into groups based upon what they have in common. (This too is a relationship.) The next thing you should do is to further organize these rows in each one of these groups to make them more useful. You could form a table using these as the column headings: name, address, email address, and phone number. ( Now you have a table for each group of rows. So, if you want data about a particular contact, you can go to the table that contains it. If you want to look at the data for a relative, you go to the table containing all of your relatives. This is basically how a flat database is created. Relational databases begin in the same way. At this point, each of these tables are checked to see whether they are "normal" or not. For example, a contact is likely to have multiple phone numbers or other possible multiple entries. Each of these have something in common (another relationship). We can remove the field containing multiple entries forming a new table. To keep the relationship between this new table with the original table, we create a primary-foreign key pair (the primary key is for the new table, and the foreign key is added to the original table in place of the fields we removed.) This is what is done to make a table "first normal form". There are several levels of a table being normal: "first normal form" to "fourth normal form" and beyond. "Fourth normal form" is considered to be the standard against which a table should be judged. The point being that we do not create tables with their fields and then define the relationship between them. We begin the the fields that we know we will need, combine the fields based upon relationships between data, and combine the fields into tables based upon relationships between the fields. The the tables are normalized up to "fourth normal form" creating new tables and modifying the old ones. Primary-foreign key pairs are used to define the relationship of the new tables and the modified old table from which the new table came.
     Hopefully this will help some.


For unsubscribe instructions e-mail to:
Posting guidelines + more:
List archive:
All messages sent to this list will be publicly archived and cannot be deleted


Privacy Policy | Impressum (Legal Info) | Copyright information: Unless otherwise specified, all text and images on this website are licensed under the Creative Commons Attribution-Share Alike 3.0 License. This does not include the source code of LibreOffice, which is licensed under the Mozilla Public License (MPLv2). "LibreOffice" and "The Document Foundation" are registered trademarks of their corresponding registered owners or are in actual use as trademarks in one or more countries. Their respective logos and icons are also subject to international copyright laws. Use thereof is explained in our trademark policy.