Even the logon screen for BO server did not come up. Solution to this is:

Delete all files and folders in location: C:\Users\..\Analysis-workspace\.metadata\.plugins

(some 2 com.sap and 6 org.eclipse folders)

You have lost your user preferences in DS (eg. so start up mode, which windows to see), but DS starts up fine again.

log location: C:\Users\..\Analysis-workspace\.metadata

]]>Hence:

set on Category, condition [Category]=[Parameter Category]

set on Customer, condition MAX( IIF([Category]=[Parameter Category],1, 0) ) = 1

[VAR] * ACOS

(

SIN( RADIANS([Origin-Latitude]) ) * SIN( RADIANS([Destination-Latitude]) ) +

COS( RADIANS([Origin-Latitude]) ) * COS( RADIANS([Destination-Latitude]) ) *

COS( RADIANS([Destination-Longitude]) – RADIANS([Origin-Longitude]) )

)

[VAR]= 3959 for miles

[VAR]= 6371 for kilometers

DATEPART(‘dayofyear’, [Order Date]) <= DATEPART(‘dayofyear’, Today() )

Calculated Field for “Keep up to current month of year”:

DATEPART(‘dayofyear’, [Order Date]) <= DATEPART(‘dayofyear’, Today() ) AND month([Order Date]) = month(Today())

then filter on True for both fields.

]]>In the example I colored the bars by years for example purposes, normally you would color by difference in sales or something to give it a more meaningful analysis.

**Relative Date**: DATEADD(‘year’,DATEDIFF(‘year’,[Order date], today()), [Order date])

This formula will add 4 years to orders made in 2010 (if current year is 2014), 3 years to orders from 2011, 2 years for 2012, and 1 year for 2013. Therefore the year is 2014 for all rows, and the relative date filter is applied across all years of orders.

Have the new calculated field **Relative Date** as column and use existing **Order Date** as color distinction.

- create a sheet for each chart

- create 1 string list parameter, which you place on each sheet, display parameter

- create a calculated field referencing the parameter

- add the calculated field as a filter on each sheet (and verify that only the value selected in the parameter on the sheet is selected)

- create dashboard and place all sheets floating and aligned over each other, display the parameter window and select a value, you’ll notice that the charts will toggle.

Order of Execution in Tableau:

Computed in the database:

1. Context filters

2. Top N and conditional filters

3. Regular quick filters

Computed in Tableau:

4. Table calculations are computed

5. Table calculation filters are applied

6. Reference lines

This means that for a Top N chart that needs to display the Top 10 Sales per State, Average Sales of Top 10 States and the Average Total Sales all States, you have to make sure that the average all states needs to be computed before the filter is applied, you have to use the Window_AVG and add it to dual axis as line. The Top N condition needs to be applied after the Window_Avg calculation. The Windows_AVG is calculated in step 4, so the TOP N needs to be in step 5 or 6. The ‘Normal’ way of calculating TOP N is on database and cannot be used that leaves a table calculation filter. Add the Sales measure to the filter, add a quick table calculation Rank, and then modify this to run along State and keep the sorting in Descending order.

In case you need a filled area chart on Percent of Sales with possible filter option, this execution path means you need to make sure you create a table calculation on your filter dimension else the % of Total will be 100% instead of the correct share of total. You could use e.g. a calculated field like **Region select**: lookup(Max([Region]),0) and then filter on this new field.

Notice that Top N filters are applied before regular quick filters, so this just shows of the top N states, which are in the east region. Notice as well that the Region pill is colored in dimensional blue.

Solution: Click the drop down on the region pill on the filter shelf and select ‘add to context’. Context filters are applied before Top N filters, so this then shows the top N states within the context of the selected region. Notice that the Region pill is now colored in grey.