I would like to be able to cycle between date (a timestamp) and week number in a graph. I use the function weekname(date) to calculate the weeknumber. However I would like my graph to show dates with missing data (for instance due to a particular selection) - this means I need to use the continuous option for this axis. However for the weeks option, this gives a strange number representation on the axis. Then I need to have the continuous option unchecked, which is not what I want... The solution needs to be able to cope with shift of year, so that week 50 in 2009 is interpreted as 'younger' than let's say week 1 in 2010.
Phrased differently, how can I cylce between date and week and still have an axis that 'makes sense'?
You could make a master calendar using autogenerate table with all the dates and join your current data, in that way you will have all weeks even with no data.
