Spotfire: Convert string date to datetime - calculated-columns

Question

I have a feeling this is easy but I just can't crack it and am spending too much time on it. I am trying to convert w2037.4 09:00 to a date time.
I ultimately would like to have the above be 09/10/2020 09:00.
I've tried ParseDate(RXReplace([value],"w"," ","i"),"yyww.d HH:mm") but this is definitely not it.
Any help is appreciated, thanks.

Answer 1

I hope this helps, even if it does not provide a complete answer.
I think the date you are looking for is 10th September (and not 9th October, as I had initially thought - please remember to specify date formats as they vary across countries).
From my understanding, your original column is made of
a w character
last two digits of year 2020
week 37
day of week 4
then the time portion
I cannot find in Spotfire a function that gives you the date from week and day of week. Can you use a TERR expression?
This one worked for me for the specific example you gave, but it is not bullet proof - weeks and days of week are tricky as they depend on your local/regional settings. In my case, I subtracted one day to make it work but you probably don't want it. Also, open source R and TERR give different results with week formats.
So the TERR Expression function I used is:
mydatetime=sapply(input1,function(x) sub('w','',x)) #remove the w
turnToDate = function(x) {
x.vector=strsplit(x,' ')[[1]] #separate date and time parts
x.date=x.vector[1] #store date portion
#is this correct? Remove if not!
x.date=as.character(as.numeric(x.date)+.1) #add a day
x.time=x.vector[2] #store time portion
y.date=as.Date(x.date,format = "%y%W.%w",tz='GMT') #convert week to date
y.datetime=paste(y.date,x.time) #add time as string
return (as.POSIXct(y.datetime,origin="1970-01-01 00:00:00"))
}
#do not use sapply as dates turn to numbers
output=Reduce(c,lapply(mydatetime,turnToDate))
I created it (from the Data>Data Function properties>Expression Functions menu) with the name TERR_convert, as a column function returning a DateTime. Then created a calculated column as :
TERR_convert([value])