Details
-
Task
-
Status: Stalled (View Workflow)
-
Minor
-
Resolution: Unresolved
-
None
-
10.2.4-4
Description
With a few exceptions, most native aggregate functions are supported as window functions.
https://mariadb.com/kb/en/library/aggregate-functions-as-window-functions/
In MDEV-7773, support for creating of custom aggregate functions was added.
This task proposes to extend that feature and allow custom aggregate functions to be used as window functions
An example of a creating a custom aggregate function is given below:
create aggregate function agg_sum(x INT) returns double |
begin
|
declare z double default 0; |
declare continue handler for not found return z; |
loop
|
fetch group next row; |
set z = z + x; |
end loop; |
end| |
This functions can be used in the following query:
create table balances (id int, amount int); |
insert into balances values (1, 10), (2, 20), (3, 30); |
|
|
select agg_sum(amount) from balances; |
After this task is complete the following must also work:
select agg_sum(amount) over (order by id); |
Attachments
Issue Links
- relates to
-
-
- Closed
-
Activity
Hi Arhant
You can find the classes using your IDE's search feature or a regular expression search tool like grep, ack, etc.
I suggested in my previous post to investigate the add() method and try to figure out how it is used inside the server. Did you manage to set up a breakpoint and have the debugger stop at it when a query to the server is issued? What did you try to do so far? Providing more details as to what actually got you stuck will help in getting a more coherent reply 
Hi Varun and Vincent, was any progress made on this issue during the last iteration ? I have built the project and will start looking at the window and aggregate functions touch points in the code. I will also pick up a recent beginner friendly task to get familiar with the codebase.
tavneet No progress has been done since last GSoC. Feel free to have a look and ask any questions you might have. I recommend sending an email to maria-developers mailing list. You stand a much higher chance of getting a reply there, rather than though JIRA only.
Hi @Vicențiu Ciorbaru @Varun,
I have set up the debugger and have some idea about the code flow for the regular aggregate function (sum) and the custom aggregate function posted by Varun.
I had a couple of queries which I have asked on the email and I am also posting here.
- In do_add from decimal.c, there are three parts with comments - /* part 1 - MY_MAX(frac) ... min (frac) /, / part 2 - MY_MIN(frac) ... MY_MIN(intg) */. Can someone please elaborate on what do the comments mean ?
- In Item_sum_sum::add_helper, there is an unlikely branch for variable direct_added. Can someone please give an idea about when will direct added be true ? In fact in all the uses for direct_added, it is always in an unlikely branch in Item_sum.cc.
- What is the design choice behind the struct decimal_t in decimal.h ? Why are we representing the numbers in this format (Is it helpful for handling floating point inaccuracies and overflows )?
Thanks Vicentiu
Please clarify me.
As you said " start looking into the Item_sum class and particularly subclasses of it like Item_sum_sum".
Can you please detail about it and where should I look for these subclasses.