Copying data from one version over to another in same cube
Posted: Tue Aug 31, 2010 7:24 pm
I am trying to get handle on using TM1. So what I am trying to do, is emulate in TM1 an EP model I built in Analyst and improvise it (where possible). This my first post on this forum, hopefully it comes out right!
I have a cube (C1) that collects information about items and has 3 versions - Input, Overwrite and Final. In Input version, we ask users to enter details related to the product like its Name, Vendor, Type, Purchase Month, 'Purchase Price' etc. There is another cube 'C2', which calculates the cash flow amount based on a 12 month cycle & purchase month. It then sends the calced amount to 'C1'. Sometimes there are extra charges in a month that a Cost Center has to bear for the product. That's where overwrite comes into picture. In Overwrite version, users have the ability to enter their own amount. The Final version will contain either Overwrite amount (if present) or calculated amount.
Between these 3 versions, only the cash flow amount is different; the attibutes related to an item like - name, vendor etc are fixed. Currently in this cube 'C1', I have written rule like this -
['Overwrite', 'Vendor Name'] = S: DB('C1', 'Input', !PurchaseType, 'Vendor Name');
['Overwrite', 'Product Name'] = S: DB('C1', 'Input', !PurchaseType, 'Product Name');
...
...
and so on for all the 11 attributes. Then repeat this for 'Final'. Thus, 22 lines to tranfer this information. I could create a subset for 'Final' and 'Overwrite' and may be able to reduce it to 11.
I have 2 questions -
1) Using rules, is there a better way to transfer these attributes of an item from 'Input' version to other 2 versions - 'Overwrite' & 'Final'?
2) If you notice in the rule, I am using a DB reference to itself. Is there a way it avoid hardcoding the cube name in DB calculation? If we decide to rename the cube (this was pretty easy in Analyst), I am afraid I have to modify the rule with this approach.
I'd appreciate any advice from you. Thank you.
TM1 9.5, Win XP, 32 bit, Excel 2003
I have a cube (C1) that collects information about items and has 3 versions - Input, Overwrite and Final. In Input version, we ask users to enter details related to the product like its Name, Vendor, Type, Purchase Month, 'Purchase Price' etc. There is another cube 'C2', which calculates the cash flow amount based on a 12 month cycle & purchase month. It then sends the calced amount to 'C1'. Sometimes there are extra charges in a month that a Cost Center has to bear for the product. That's where overwrite comes into picture. In Overwrite version, users have the ability to enter their own amount. The Final version will contain either Overwrite amount (if present) or calculated amount.
Between these 3 versions, only the cash flow amount is different; the attibutes related to an item like - name, vendor etc are fixed. Currently in this cube 'C1', I have written rule like this -
['Overwrite', 'Vendor Name'] = S: DB('C1', 'Input', !PurchaseType, 'Vendor Name');
['Overwrite', 'Product Name'] = S: DB('C1', 'Input', !PurchaseType, 'Product Name');
...
...
and so on for all the 11 attributes. Then repeat this for 'Final'. Thus, 22 lines to tranfer this information. I could create a subset for 'Final' and 'Overwrite' and may be able to reduce it to 11.
I have 2 questions -
1) Using rules, is there a better way to transfer these attributes of an item from 'Input' version to other 2 versions - 'Overwrite' & 'Final'?
2) If you notice in the rule, I am using a DB reference to itself. Is there a way it avoid hardcoding the cube name in DB calculation? If we decide to rename the cube (this was pretty easy in Analyst), I am afraid I have to modify the rule with this approach.
I'd appreciate any advice from you. Thank you.
TM1 9.5, Win XP, 32 bit, Excel 2003