Could you tell me the way to realize the following in q?
As given the following table,
t:( sym:(); register_val:(); remain_val:())
[Note]
Sym is either A or B.
Type of register_val and remain_val is float.
Every time inserting a record into the table t, if values of the sym in existing record is the same as new record’s sym and remain_val is greater than 0, then subtract new record’s register_val from oldest record’s remain_val and update the table t with the result of subtraction and register the new record as it is.
For example,
- Initial condition
sym register_val remain_val
- Inset first record
sym register_val remain_val
A 1 1
- Insert second record
sym register_val remain_val
A 1 1
B 3 3
- Insert third record
sym register_val remain_val
A 1 0
B 3 3
A 1 1
- Insert fourth record
sym register_val remain_val
A 1 0
B 3 1
A 1 1
B 2 2
- Insert fitth record
sym register_val remain_val
A 1 0
B 3 0
A 1 1
B 2 0
B 3 3
Could you advise the way to realize above.
Regards
Hi,
You can write a function which updates the table in the way you want then inserts your new data, for example
f:{[x;y;z] update (remain_val:0) from (update (remain_val):remain_val-z from t where sym = x) where remain_val\<0; t insert (x;y;z)}
This function subtracts the new_remain val from the existing values in the table with the same sym, then converts any negative values to 0 and finally inserts your data into the table, e.g. if your table is
sym register_val remain_val
A 1 0
B 3 0
A 1 1
B 2 0
B 3 3
Run the function as
f[`B;4;4]
And the resulting table with look like-
sym register_val remain_val
A 1 0
B 3 0
A 1 1
B 2 0
B 3 0
B 4 4
Regards,
Lee
Hi Lee,
Thank you for your helpful advice.
In this case I have to sequentially subtract the register_val of the record to be inserted from remain_vals of existing records according to ascending order of time.
For example,
- Inset first record
sym register_val remain_val
A 3 3
- Insert second record
sym register_val remain_val
A 3 1
A 2 2
- Insert third record
sym register_val remain_val
A 3 0
A 2 2
A 1 1
I guess if I apply your function to this example Ill get the following, which I wont wish.
sym register_val remain_val
A 3 0
A 2 1
A 1 1
In addition if the register_val of the record to be inserted is greater than the full amount of existing remain_vals, all of existing remain_vals have to be update as 0.
For example,
- Inset first record
sym register_val remain_val
A 5 5
- Insert second record
sym register_val remain_val
A 5 4
A 1 1
- Insert third record
sym register_val remain_val
A 5 0
A 1 0
A 10 10
Regards