R - date sequence with condition - r

Question

I have this dataframe
test <-
data.frame(
id = c(4, 6, 9, 12),
open = c(as.Date("2011-01-01"), as.Date("2011-01-01"), as.Date("2011-01-01"), as.Date("2011-01-01")),
closed = c(as.Date("2011-12-01"), as.Date("2011-12-31"), as.Date("2012-01-01"), as.Date("2015-12-31"))
)
My goal is to get each date that overlapped or reached last day in year. Since id 4 was started at 2011 and ended before last day, there should be NA for it. id 6 reached last day in 2011 but not in 2012, same for id 9.
Result should be
summary <-
data.frame(
id = c(4, 6, 9, 12),
open = c(as.Date("2011-01-01"), as.Date("2011-01-01"), as.Date("2011-01-01"), as.Date("2011-01-01")),
closed = c(as.Date("2011-12-01"), as.Date("2011-12-31"), as.Date("2012-01-01"), as.Date("2015-12-31")),
open_summary = c(NA, 2011, 2011, 2011),
closed_summary = c(NA, 2011, 2011, 2015)
)
Then I'd like to create a sequence from these dates so result should be
result <-
data.frame(
y = c(2011, 2011, 2011, 2012, 2013, 2014, 2015),
id = c(6, 9, 12, 12, 12, 12, 12)
)

Answer 1

Here is a tidyverse solution, also using lubridate (for the year function)...
library(tidyverse)
library(lubridate)
summary <- test %>%
mutate(open_summary = year(open) * (year(open) > year(open - 1)),
closed_summary = (year(closed + 1) - 1) * (year(closed + 1) > year(open)))
output <- summary %>%
filter(open_summary * closed_summary > 1) %>%
mutate(open_year = map2(open_summary, closed_summary, seq)) %>%
select(id, open_year) %>%
unnest(c(open_year))
summary
id open closed open_summary closed_summary
1 4 2011-01-01 2011-12-01 2011 0
2 6 2011-01-01 2011-12-31 2011 2011
3 9 2011-01-01 2012-01-01 2011 2011
4 12 2011-01-01 2015-12-31 2011 2015
output
id open_year
1 6 2011
2 9 2011
3 12 2011
4 12 2012
5 12 2013
6 12 2014
7 12 2015
If either open_summary or closed_summary is zero, that is equivalent to your NA row.

Answer 2

Here is a approach using data.table:
library(data.table)
#create a lookup table of year end dates
yrend <- data.table(YR_END=seq(as.Date("2010-12-31"), as.Date("2015-12-31"), by="1 year"))[,
YR := year(YR_END)]
setDT(test)
#create open_summary column since its just the end of the year of the open column
test[, open_summary := year(open)]
#lookup the year for the closed date
test[, closed_summary := yrend[test, on=.(YR_END>=open, YR_END<=closed), mult="last", YR]]
#create the sequence in part 2 of the qn
test[!is.na(open_summary) & !is.na(closed_summary),
.(y=open_summary:closed_summary), id]
test output:
id open closed open_summary closed_summary
1: 4 2011-01-01 2011-12-01 2011 NA
2: 6 2011-01-01 2011-12-31 2011 2011
3: 9 2011-01-01 2012-01-01 2011 2011
4: 12 2011-01-01 2015-12-31 2011 2015
the other output:
id y
1: 6 2011
2: 9 2011
3: 12 2011
4: 12 2012
5: 12 2013
6: 12 2014
7: 12 2015