R：如果有连续的列相等，如何只保留一个并分配一个新的列名

Question

我有一个客户数据集，他们在哪些商店购物、他们在每家商店购买了什么以及在哪一天。

Shop_list <- data.frame (Names  = c('Adam','Eve','Lucy','Ricky','Gomez','Morticia','Adam','Eve','Lucy','Ricky',
                        'Adam','Eve','Ricky','Gomez','Adam','Eve','Lucy','Adam','Eve','Lucy','Adam','Eve','Lucy'),
                  Day = c(1,1,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,5,5,5,6,6,6),
                  Store= c('None','None','None','None','None','None','Lowes',
                          'Home Depot','Lowes','Home Depot','Lowes',
                          'Home Depot','Home Depot','Lowes','None',
                          'Home Depot','None','None','None','Home Depot',
                          'Home Depot','None','None'),
                  Item= c('None','None','None','None','None','None','Wood','Soil','Nails','Pots','Nails',
                        'Pots','Soil','Wood','None','Seeds','None','None','None','Seeds','Seeds','None','None'),
              stringsAsFactors=FALSE
)

我写了一个函数来总结这些数据，

library(dplyr)
library(flextable)

Shop_fcn <- function(data){
  data %>%
    group_by(Day) %>%
    mutate(N_nam = n_distinct(Names)) %>%
    group_by(Names, Day, N_nam, Store, Item) %>%
    summarize(n_item = n()) %>%
    group_by(Day, N_nam, Store, Item) %>%
    summarize(n_nam = n(),
              n_item = sum(n_item))%>%
    mutate(pct = round(n_nam/N_nam*100,digits = 1),
           txt = paste0( n_nam, " (", pct, "%)"),
           Day_n = (paste0("Day ", Day," (N=",  N_nam, ")")))%>%
    ungroup %>% select(Day_n , Store, Item, txt) %>%
    pivot_wider(values_from = txt, names_from = Day_n) %>%
    mutate_at(vars(starts_with(c("Day"))), ~if_else(is.na(.), "", .)) %>%
    arrange(Store, Item) %>% 
    group_by(store2 = Store) %>% 
    mutate(Store = if_else(row_number() != 1, "", Store))%>%
    ungroup() %>% select(-store2)
}

Shop_day <- Shop_list %>%
  bind_rows(Shop_list) %>%
  Shop_fcn ()

flextable(Shop_day)

我得到以下输出。

第 2 天和第 3 天的列是相等的，第 4 天、第 5 天和第 6 天的列也是如此。我正在努力使具有相同信息的列的列标题读取为第 2-3 天(N=4) 和第 4 - 6 天 (N=3)。

到目前为止，我已经尝试删除重复的列

Shop_nodup <- Shop_day[!duplicated(as.list(Shop_day))]
flextable(Shop_nodup)

这给了我

重复的列消失了，但我想不出一种方法来在列标题中指定该列涵盖的天数范围（第 2 - 3 天（N=4）和第 4 - 6 天（ N=3))

Answer 1

如果我们需要更改标题，请进行如下更改

library(stringr)
Shop_fcn <- function(data){
  data %>%
    group_by(Day) %>%
    mutate(N_nam = n_distinct(Names)) %>%
    group_by(Names, Day, N_nam, Store, Item) %>%
    summarize(n_item = n()) %>%
    group_by(Day, N_nam, Store, Item) %>%
    summarize(n_nam = n(),
              n_item = sum(n_item))%>%
    mutate(pct = round(n_nam/N_nam*100,digits = 1),
           txt = paste0( n_nam, " (", pct, "%)"),
           Day_n = (paste0("Day ", Day," (N=",  N_nam, ")")))%>%
    ungroup %>% 
    select(Day_n , Store, Item, txt) %>%
    group_by(Store, Item, txt) %>%
    summarise(Day_n = if(n() > 1) 
     sprintf('Day %s %s', paste(range(readr::parse_number(unique(Day_n))), 
     collapse=' - '), 
       str_remove(first(Day_n), '^[^(]+')) else Day_n) %>%
    pivot_wider(values_from = txt, names_from = Day_n) %>%
    mutate_at(vars(starts_with(c("Day"))), ~if_else(is.na(.), "", .)) %>%
    arrange(Store, Item) %>% 
    group_by(store2 = Store) %>% 
    mutate(Store = if_else(row_number() != 1, "", Store))%>%
    ungroup() %>%
    select(Store, Item, str_sort(names(.)[-(1:2)], numeric = TRUE), -store2)

}

-测试

Shop_day <- Shop_list %>%
  bind_rows(Shop_list) %>%
  Shop_fcn ()

flextable(Shop_day)

-输出