Rank Function in load script

[#1]

Hi guys

I am looking to Rank Rejection reasons as per the frequency they occur.

I have the Rejection reasons loading from an excel spreadsheet , where the rejection reasons are all listed in one cell - so I have split out each reason on load using the below.

if(trim(upper(subfield([Rejection Reasons], ',')))='',null(),trim(upper(subfield([Rejection Reasons], ',')))) as ReasonSub

I now would like to rank it in the script ,as I do not want the rankings to change when certain filters are selected in the chart .

Can anyone advise if this can be done and if so any pointers greatly appreciated

Thanks

A

[#2]

Using the script as I pasted it in , this is the chart

ReasonSubs	Number of Rejections	Rank	Correct  Rank
Incorrect Docket	363	2	1
Fire Damage	150	5	2
Under use	142	4	3
Bad Rate	123	3	4
Trade Green	118	7	5
Knife Sharp	112	6	6

So Rank is the result and Correct Rank is the desired result

To be honest Im not too concerned about ranking at customer level , I just want to see the most popular rejection reasons

In the chart I used expression

rank(sum({$<[Approved? N/Y] = {"*No*"}>}Numbers),ReasonSubs,1)

And this worked fine - but I really would like to show the ranking whilst not effected by filters

Thanks

[nagaiank]

Rank function is available for use in chart expressions only.It is not available in load scripts. In load script, you may use Order By.

[swuehl · Accepted solution]

I am slow from time to time, but I am not seeing how just repeating your message helps me here.

In your expression

rank(sum({$<[Approved? N/Y] = {"*No*"}>}Numbers),ReasonSubs,1)

how do you derive Numbers? Can't see it in your script.

What does the ReasonSubsCount show when compared to your Number of Rejections?

If I understand your request correctly, I still think that calculating the rank in the script should be possible using a count (or whatever appropriate) in a group by load and the assign a rank number in a order by load.

If it's all about creating a rank expression in the front end that is not sensitive to selections, you can try using set analysis (set identifier 1).

See attached.

[#5]

OMG Swuehl

Sorry I didnt mean to repost ( that must look rude - sorry !!!) Sometimes my I get errors when posting a message onto the forum and then it posts twice ,

Let me have a look at making a dash that explains my issue

Thanks

A

[#6]

Hi guys

Thanks for your time on this -- the issue is with my data model

I have rewrite most of the script to accomodate what I require now

Thanks for your posts they will deffo assist when I go to redesign the dash

A

[#7]

Hi guys

Thanks for your time on this -- the issue is with my data model

I have rewrite most of the script to accomodate what I require now

Thanks for your posts they will deffo assist when I go to redesign the dash

A