Will there always be a record in the preferences table for every box and every ball?

Yes, There will be atleast one Preference for every box and ball...

As this challenge is Multi-statement solutions, can we create a physical/#table to derive output results.

Hey I came up with a result that fits the preference table that is different from your 'expected result' but you said there is only supposed to be one correct solution.   Did I miss a rule?

Box Ball

Box 1 Ball 1

Box 3 Ball 2

Box 2 Ball 3

Box 5 Ball 4

Box 6 Ball 5

Box 4 Ball 6

Here is an expanded view in the alternate match (with PreferenceIDs).

BoxID Box BallID Ball PreferenceID

1 Box 1 1 Ball 1 1

3 Box 3 2 Ball 2 4

2 Box 2 3 Ball 3 3

5 Box 5 4 Ball 4 12

6 Box 6 5 Ball 5 14

4 Box 4 6 Ball 6 11

Oh duh!  The sort order didn't match.  The expected results are sorted by box but I sorted by Ball ID and got my boxes and balls mixed up.  Well that means it works!  I'm going to get it to work with a cartesian product preferences table now.

can we use the while loop to get the solution.

As there is only 1 correct ball for a box and thus only 1 correct solution, can we assume that there will be no logic testing for "funny data"..

Ie if you have a working solution on the given sample set, it will thus work on all sets as the only difference can be in the size of the set?

I've got the solution, but now how do I submit it?  Maybe it is obvious to others, but I am new to this site (joined today); finding the solution was much easier than finding out how to submit my solution :)

In question 8 above you state that you cannot create your own tally table.

1. Why not?

2. Is this something new starting with this challeng or has this always been the rule?

Jacob,

So you are saying a sub-query in your solution with a series of cross-joins to generate a tally table 'on the fly' is NOT permitted?

Or are you saying that you cannot generate a physical table (temporary or not) that acts as a tally table?

I got an error when running the stored procedure creating script:

Msg 102, Level 15, State 1, Procedure TC_Eval, Line 7

Incorrect syntax near ';'.

If I remove the ';' after GO, it works.

Did I miss something. Thanks.

That should be changed.

GO is not an actual T-SQL command... it's just a batch-end indicator/separator.  It should never have a semi-colon after it.

Jacob or Ramireddy, can you change that and get rid of the semicolon?

Jacob,

The 'Read More' links for 22 links to 21.

When will the results be posted?

tsqlc_Tally?

Where is it?

Solutions cannot be seen (Page Requests fail).

It does.

Hi Jacob,

I am very much confused by the winners list for this challenge....

I tried to execute the couple of solutions from winner list with Load testing data provided and here is the list of what each solutions take time to executive.

Solution -----------> Execution Time

Top winner ( carlos benito ) --------> 1.44 minutes

Runner Up ( Jesse Roberge ) -----> 1.23 minutes

...

...

2nd Last winner ( jmc bkk  ) -------------> 0.01 Minutes

Last winner ( Michael Lewis  ) -------------> 0.00 Minutes

Can you please help me out what criteria applies to identify the winners.

-Pragnesh

Jacob,

I understand with read, write, CPU, duration rank.

But you can see the execution time difference... its a very huge.

In real world scenario any query takes 1.44 minutes to execution is always not acceptable by anyone.

I am still not clear with the weightage given to read, write, CPU, duration rank.