Using dplyr to remove duplicates conditionally - r

Question

I have a dataset in longformat that contains both visit and measure dates for each ID. What I want is to remove the duplicate visit dates for each ID conditionally, namely:
IF visit date - measure date does not equal 0, then I want the to include the first visit date.
IF visit date - measure date is a draw, however, then I want to include the lastest visit date.
I already wrote part of the code using dplyr. However, I cannot seem to figure out how to code the second part of the condition.
Any help would be very much appreciated.
library(dplyr)
df <- data.frame(ID = c(1, 1),
VISIT = c(as.Date("2020-01-01"), as.Date("2020-01-01")),
MEASURE = c(as.Date("2020-01-01"), as.Date("2020-01-01")),
VALUE = c(5, 10))
df2 <- df %>%
mutate(DIFF = abs(VISIT - MEASURE)) %>%
arrange(DIFF) %>%
group_by(ID) %>%
group_by(VISIT) %>%
# If DIFF dates is != 0, I want the first value
# If DIFF dates is a draw, I want the latest value
slice(1) %>%
ungroup()

Answer 1

I am not sure what exactly you try to achieve, but maybe this could help you. I adjusted the example dataframe a bit, maybe you will need to edit yours in your question such that it makes sense. In your example data DIFF dates is never unequal to 0.
library(dplyr)
df <- data.frame(
ID = c(1, 1, 2, 2),
VISIT = c(
as.Date("2020-01-01"),
as.Date("2020-01-01"),
as.Date("2020-01-01"),
as.Date("2020-01-02")
),
MEASURE = c(
as.Date("2020-01-01"),
as.Date("2020-01-01"),
as.Date("2020-01-01"),
as.Date("2020-01-03")
),
VALUE = c(5, 10, 15, 20)
)
df2 <- df %>%
group_by(ID) %>%
mutate(
DIFF = abs(VISIT - MEASURE),
# get days as a digit
DIFF = stringr::str_extract(DIFF, "\\d+") %>% as.numeric(),
# your if conditions
DIFF_filter = case_when(
DIFF != 0 ~ min(VISIT),
DIFF == 0 ~ max(VISIT)
)
)