Repeat row and mutate multiple observations on conditional statement-CodePudding

I'm trying write a conditional statement that duplicates the row where df$food=1. Then changes the value "df$weight" of initial row to the value of df$prior_w and not the duplicate. Also i need to change the df$food value to 0 on the duplicate and prior_w to NA on the duplicate

df <- data.frame(date=("2022-01-01","2022-01-02","2022-01-03","2022-01-04","2022-01-05"),
food=(0,0,0,1,0),
prior_w=(NA,NA,NA,2,NA),
weight=(5,4,3,6,4))

Id like to have a data frame like this

df_2 <- data.frame(date=("2022-01-01","2022-01-02","2022-01-03","2022-01-04","2022-01-04","2022-01-05"),
food=(0,0,0,1,0,0),
prior_w=(NA,NA,NA,NA,2,NA),
weight=(5,4,3,2,6,4))

Im will translate what i need in words (Not actual code, sorry im struggling). I looked at a lot of stack overflow questions and answers but i cant seem to find the perfect mix. I know the rep function repeats, and that i can write conditional statements with case_when or ifelse.

df_1 <- df %>%
  repeat row case_when df$food==1 %>%
   mutate (the_first_row (df$weight=prior_w),
          second_row (df$food=0, df$prior_w = NA))

thank you for your help

CodePudding user response：

We may use uncount to replicate the rows and then change the values based on the number of rows by group

library(dplyr)
library(tidyr)
df %>% 
  uncount(1   (food == 1)) %>% 
  group_by(date) %>%
  mutate(food = if(n() > 1) replace(food, -1, 0) else food,
          weight = if(n() == 2) replace(weight, 1, prior_w[n()]) else weight,
          prior_w = if(n() ==2 ) lag(prior_w) else prior_w) %>%
  ungroup

-output

# A tibble: 6 × 4
  date        food prior_w weight
  <chr>      <dbl>   <dbl>  <dbl>
1 2022-01-01     0      NA      5
2 2022-01-02     0      NA      4
3 2022-01-03     0      NA      3
4 2022-01-04     1      NA      2
5 2022-01-04     0       2      6
6 2022-01-05     0      NA      4