Hi,

Ok, I am creating a small application that handles a staff rota and have some data that needs normalising, I am familiar with the process however I've never normalised grid data before, let me explain..

here are the fields:

userID (this is included so each staff member can log in independantly, coinciding with other parts of the application)
Location (this is typically the same)
Week Beginning
Week
Dept (typically the same)
staffName
Monday
Tuesday
Wednesday
Friday
Saturday
Sunday
Date
Hours Owed (calculated, so not to be included)


The raw data is a simple excel spreadsheet, with the days of the week across the top, the staff names down the side and shifts populating the rest i.e 8-8, 12-4 etc.

What I need to figure out is the repeating data, the rest will follow

I spoke with my tutor who reckoned I go with the days of the week as the repeating data, I dont agree. Here is what I reckon to be the repeating data, indented (as the data is repeated over and over):

userID
Location
Week Beginning
Week
Dept
staffName
Monday
Tuesday
Wednesday
Friday
Saturday
Sunday
Date


although even that I'm struggling to make sense of, as the repeating group also needs taking to 1NF.

Any thoughts?

lol, i did indent it however obviously not correct here are the fields i think to be repeating:


userID
Location
Week Beginning
Week
Dept
staffName

if you look at the rota data independently of how you want it displayed, then it normalizes quite easily --

Location
userID
assigneddate
shift

in other words, you are assigning a shift value (i.e. 8-8, 12-4 etc.) to a particular person at a particular location on a particular date

that's the repeating data taken care of

then you'll have other normalizations to do, such as not stuffing the user's name into this table

displaying the results nicely in "rota" style, one page per location per week, days across the top, staff down the left -- that's the job of the application code

thanks for taking a look at it, makes sense now!!

Your right once you get the fields down its just like any other data, i can get cracking with this now, no problem!

Yeah that was poor foresight on my part about the 'staffName'!

thanks so much again!!