当我在 R 代码中有 NA 数据时生成变量的问题

Question

I have a question regarding the code below. Note that my input data was day = 30/06 , Category = FDE and DTT = Hol , and I can get SPV ( First Code ). However, when I do day = 30/06 , Category = ABC and DTT = NA , I can't get SPV ( Second Code ). I would need to show the line corresponding to that date/category/dtt. How to adjust this?

Executable code below:

For 30/06, FDE, Hol

library(dplyr)

df1 <- structure(
  list(date1= c("2021-06-28","2021-06-28","2021-06-28"),
       date2 = c("2021-06-30","2021-06-30","2021-07-02"),
       DTT= c(NA,"Hol","Hol"),
       Week= c("Wednesday","Wednesday","Friday"),
       Category = c("ABC","FDE","ABC"),
       DR1 = c(4,1,1),
       DR01 = c(4,1,2), DR02= c(4,2,0),DR03= c(9,5,0),
       DR04 = c(5,4,3),DR05 = c(5,4,0)),
  class = "data.frame", row.names = c(NA, -3L))


dmda<-"2021-06-30"
CategoryChosse<-"FDE"
DTest<-"Hol"

  x<-df1 %>% select(starts_with("DR0"))
  
  x<-cbind(df1, setNames(df1$DR1 - x, paste0(names(x), "_PV")))
  PV<-select(x, date2,Week, Category, DTT, DR1, ends_with("PV"))
  
  med<-PV %>%
    group_by(Category,Week,DTT) %>%
    summarize(across(ends_with("PV"), median))
  
  SPV<-df1%>%
    inner_join(med, by = c('Category', 'Week','DTT')) %>%
    mutate(across(matches("^DR0\\d+$"), ~.x + 
                    get(paste0(cur_column(), '_PV')),
                  .names = '{col}_{col}_PV')) %>%
    select(date1:Category, DR01_DR01_PV:last_col())
  
  SPV<-data.frame(SPV)
  
  mat1 <- df1 %>%
    filter(date2 == dmda, Category == CategoryChosse, DTT==DTest) %>%
    select(starts_with("DR0")) %>%
    pivot_longer(cols = everything()) %>%
    arrange(desc(row_number())) %>%
    mutate(cs = cumsum(value)) %>%
    filter(cs == 0) %>%
    pull(name)
  
  (dropnames <- paste0(mat1,"_",mat1, "_PV"))
  
  SPV <- SPV %>%
    filter(date2 == dmda, Category == CategoryChosse, DTT==DTest) %>%
    select(-any_of(dropnames))
  if(length(grep("DR0", names(SPV))) == 0) {
    SPV[mat1] <- NA_real_
  }

> SPV
       date1      date2 DTT      Week Category DR01_DR01_PV DR02_DR02_PV DR03_DR03_PV DR04_DR04_PV DR05_DR05_PV
1 2021-06-28 2021-06-30 Hol Wednesday      FDE            1            1            1            1    

For 30/06, ABC, NA

dmda<-"2021-06-30"
CategoryChosse<-"ABC"
DTest<-NA

x<-df1 %>% select(starts_with("DR0"))

x<-cbind(df1, setNames(df1$DR1 - x, paste0(names(x), "_PV")))
PV<-select(x, date2,Week, Category, DTT, DR1, ends_with("PV"))

med<-PV %>%
    group_by(Category,Week,DTT) %>%
    summarize(across(ends_with("PV"), median))

SPV<-df1%>%
    inner_join(med, by = c('Category', 'Week','DTT')) %>%
    mutate(across(matches("^DR0\\d+$"), ~.x + 
                      get(paste0(cur_column(), '_PV')),
                  .names = '{col}_{col}_PV')) %>%
    select(date1:Category, DR01_DR01_PV:last_col())

SPV<-data.frame(SPV)

mat1 <- df1 %>%
    filter(date2 == dmda, Category == CategoryChosse, DTT==DTest) %>%
    select(starts_with("DR0")) %>%
    pivot_longer(cols = everything()) %>%
    arrange(desc(row_number())) %>%
    mutate(cs = cumsum(value)) %>%
    filter(cs == 0) %>%
    pull(name)

(dropnames <- paste0(mat1,"_",mat1, "_PV"))

SPV <- SPV %>%
    filter(date2 == dmda, Category == CategoryChosse, DTT==DTest) %>%
    select(-any_of(dropnames))
if(length(grep("DR0", names(SPV))) == 0) {
    SPV[mat1] <- NA_real_
}

> SPV
[1] date1        date2        DTT          Week         Category     DR01_DR01_PV DR02_DR02_PV DR03_DR03_PV
[9] DR04_DR04_PV DR05_DR05_PV
<0 lines>

Answer 1

You can write your own function which is like == unless one element is NA , and in that case only returns TRUE when both are NA . Then use that function in filter instead of == .

In the future please attempt to create a minimal reproducible example like the one below, so that you are asking a specific question rather than asking people to fix your code for you. Almost none of the code you have posted has to do with the question you are asking.

See How to make a great R reproducible example

library(dplyr, warn.conflicts = FALSE)

same <- function(x, y){
  case_when(
    is.na(x) != is.na(y) ~ FALSE,
    is.na(x) ~ TRUE,
    TRUE ~ x == y)
}

df <- data.frame(x = c('hol', NA))

x_want <- 'hol'

df %>% 
  filter(same(x, x_want))
#>     x
#> 1 hol

x_want <- NA

df %>% 
  filter(same(x, x_want))
#>      x
#> 1 <NA>

^{Created on 2021-12-20 by the reprex package (v2.0.1)}