Hello all . . .

I'm creating a database schema and i need some help.

I have two tables

The first table looks like

Table1
tb1_ID
tb1_data1
tbl1_data2
tbl1_data3

The second table looks like

table2
tb2_ID
tb2_TB1_ID
tb2_data1
tb2_data2
tb2_data...
tb2_data25

The tb2_TB1_ID is foreign key to table1

Now the problem that i have is the following

each of the tb2_data(column_number) must have one to three sub records in an other table.

How can i design this stracture ? ? ?

Thanks a lot . . . .

start by splitting off the repeating data columns from table2

create table table2data
( tb2_ID integer
, tb2_TB1_ID integer
, tb2_number tinyint
, primary key ( tb2_ID, tb2_number )
, tb2_data varchar(99)
)

so instead of 25 columns, you have 25 rows

as for your "sub records" you will need to explain this more clearly

Coo - so mike_bike_kite finally caught your imagination then eh? EAVtastic

Seriously though - are you sure this is a first normal form problem or could it be that meranios has chosen an unfortunate way to label his\ her attributes for the purposes of the question?

pootle flump ur codings are working excelent.

huh? caught my imagination? what are you on about?

Good Lord. He could start by dropping the table prefixes from his column names.

about tbX_ is not prefix i gave that names because my original names are more deficault to write. are very long names and too many columns.

The third table must have the following stracture

table3
tb3_id
tb3_data1
tb3_data2
tb3_data3

in table3 i will inserting records for each column for each record of the table2.

The ammount of records in table3 will be from 1 record to 3 records for each column/record in table2

merianos - please could you put your genuine field names down? Rudy and I are interpreting your names differently and he is inferring that you have a not met first normal form because of it. data1, data2, data3 etc implies repeating groups.

pootle flump ur codings are working excelent.

must???

okay, then, good luck to you

Is realy very very deficault to read all together . . .

My main problem is on table2 and table3

tha table2 must storing information for 25 deferent thinks. all of that 25 columns in table2 are storing only boolean values.

My client told me "For each boolean information in table2 i have to store one or two or three other informations."

If i make a table like

table2
tb2_ID
tb2_TB1_ID
tb2_data1
tb2_data1_extra_information1
tb2_data1_extra_information2
tb2_data1_extra_information3
tb2_data2
tb2_data2_extra_information1
tb2_data2_extra_information2
tb2_data2_extra_information3
tb2_data...
tb2_data..._extra_information1
tb2_data..._extra_information2
tb2_data..._extra_information3
tb2_data25
tb2_data25_extra_information1
tb2_data25_extra_information2
tb2_data25_extra_information3

i think is too wrong.

so the problem is the table2 and table3.

How can i do that ? ? ?

To be fair - merianos's first language is clearly not English.

Sorry merianos - without the requested information I can't help you. I don't know what you are modelling since you are using generic names & numbers.

pootle flump ur codings are working excelent.

Whether that design is wrong depends upon the nature of the data you are storing. Presumably you want your data normalized, but from the generic table and column names you have given we can't tell whether your schema is normalized or not.
What kind of data are your storing? What is the purpose of this database?

The database is for a doctor.

The table2 storing history data.

For example :

patient_lactation
patient_sneeze
patient_itch

the above columns are a sort example of my table2 columns. In total are 25 columns similar to this columns and all storing boolean values.

Now for each column, if the value is true, then i want to store, at the most trhee diferent types of drugs (medicens) for each case seperatly.

ie.
patient_lactation = true

drugs for this column
drug_name_1
drug_name_2
drug_name_3

This is what i want to make.

And this is the realy table but in sort view.

Thanks a lot

For a normalized schema, you will need a structure similar to this:
Note: I used surrogate keys in this example. Others (Rudy...) may advise you to use natural keys. The important thing is to understand how these table relationships give your database application maximum flexibility.

the only table of all those you've listed, blindman, which potentially has a useful natural key is Medications

a quick google turns up a couple of US patents on the code, but no information on the code itself

use a surrogate

This is a very small part of my database project.

That i have to do is add one to three sub records for each column in table SYSTEMATIC_RACH_EPISODES.

For example if

SYRE_SKIN_PALM_ITCH = true

then the doctor will deside if he will give drugs to the patient for that symptom. And if he deside to give drugs he can give to him the most three drugs. In SYSTEMATIC_RACH_EPISODES will adding records for each visit of the patient. And in any visit will giving other drugs to the patient for each column of the table, or he will stop to giving drugs to the patient for the some columns.

Any idea on how to do that ? ? ?

http://www.datacenterhellas.eu/070820071135.jpg