Error in `[[<-`(`*tmp*`, id_index, value = HD_before) : no such index at level 1; How can I solve this problem?

When I enter the code in R as below

library(dplyr)
library(readxl)

fn<-function(data, id_variable_name)
{ 
list<-list()
ids<-unique(data[,id_variable_name]) 
for(id_index in ids){
 data_by_group<-data %>% filter(id==id_index)
 data_plus<-data_by_group %>% filter(timedif>0)
 HD_before<-arrange(data_plus) %>% head(1)
 data_minus<-data_by_group %>% filter(timedif<0)
 HD_after<-arrange(data_minus) %>% head(1)
 HD_before_after<-rbind(HD_before, HD_after)
 list[[id_index]]<-HD_before_after
 }
output<-do.call(rbind.data.frame, list)
return(output)
}

temp<-fn(data_temp, "id")

an error code

Error in `[[<-`(`*tmp*`, id_index, value = HD_before_after) : 
  no such index at level 1

appears.

What is wrong with my code and how can I fix it?

Details of data_temp is

 id timedif
1 -5850 
1 -4975
1 -7855
1 -8817
1 -5969
2 1500
2 1708
2 -1442
2 1749
3 -4110
3 -4050
3 -4140
3 -4094
4 -8651
4 -8576
4 -8561
4 -8555
4 -8546
4 -5621
4 -4901
4 -8728
4 -8548
4 -5693
4 -5513

Both id and timedif are numeric.

Perhaps a simpler approach:

arrange(data_temp) %>%
  group_by(id, g = timedif > 0) %>%
  slice_head(n = 1) %>%
  ungroup() %>%
  select(-g)
# # A tibble: 5 × 2
#      id timedif
#   <int>   <int>
# 1     1   -5850
# 2     2   -1442
# 3     2    1500
# 4     3   -4110
# 5     4   -8651

Functionized as:

library(rlang) # enquo
fn <- function(data, id_variable_name, time) {
  arrange(data) %>%
    group_by(!!enquo(id_variable_name), g = !!enquo(time) > 0) %>%
    slice_head(n = 1) %>%
    ungroup() %>%
    select(-g)
}
fn(data_temp, id, timedif)
# # A tibble: 5 × 2
#      id timedif
#   <int>   <int>
# 1     1   -5850
# 2     2   -1442
# 3     2    1500
# 4     3   -4110
# 5     4   -8651

Notice that this is using non-standard evaluation (NSE) for identifying the column names, and that I added time=.