r - 如何将 difftime 的值从小时数转换为天数？

Question

我有看起来像的数据框。

structure(list(format_date = c("Feb-21 (W5)", "Mar-21 (W1)", 
"Mar-21 (W2)"), Total_count = c(53L, 1079L, 1288L), Count_Diff_1 = c(29L, 
508L, 519L), Count_Diff_2 = c(22L, 505L, 624L), Count_Diff_3 = c(2L, 
66L, 145L), Average_Diff_1 = structure(c(10.3464846743295, 6.03846784776903, 
2.75151412973667), class = "difftime", units = "hours"), Average_Diff_2 = structure(c(6.46958333333333, 
7.78101430143014, 2.58788906695157), class = "difftime", units = "hours"), 
    Average_Diff_3 = structure(c(162.45419129418, 133.452629846369, 
    54.3061989570153), class = "difftime", units = "hours"), 
    Per_Diff_1 = c(0.547169811320755, 0.470806302131603, 0.402950310559006
    ), Per_Diff_2 = c(0.415094339622642, 0.468025949953661, 0.484472049689441
    ), Per_Diff_3 = c(0.0377358490566038, 0.0611677479147359, 
    0.112577639751553)), row.names = c(NA, -3L), groups = structure(list(
    format_date = c("Feb-21 (W5)", "Mar-21 (W1)", "Mar-21 (W2)"
    ), .rows = structure(list(1L, 2L, 3L), ptype = integer(0), class = c("vctrs_list_of", 
    "vctrs_vctr", "list"))), row.names = c(NA, 3L), class = c("tbl_df", 
"tbl", "data.frame"), .drop = TRUE), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"))

如果在 Average_1、Average_2 和 Average_3 列中超过 24 小时，我想以天为单位转换值并隐藏 Per_Diff_1、Per_Diff_2 和 Per_Diff_3，百分比保留 2 个小数点，带 % 符号。

另外，如果我能设法以降序排列 format_date 就好了。现在它是 Feb-21 (W5)、Mar-21 (W1) 和 Mar-21 (W2) 它应该是Mar-21 (W2)、Mar-21 (W1) 和 Feb-21 (W5)。

必需的 df<-

format_date   Total_count Count_Diff_1 Count_Diff_2 Count_Diff_3 Average_1    Average_2    Average_3 Per_Diff_1 Per_Diff_2 Per_Diff_3
Feb-21 (W5)    53         29           22           2            10.34 hours  6.46 hours   6.7 Days  54.71%     41.50%     3.77%

Answer 1

我们可以在 starts_with 'Average' 的列上使用 across 来转换值，如果它大于 24 然后除以 24，附加 'Days' 作为后缀，或者通过附加“小时”作为后缀返回。类似地，starts_with 'Per' 的列可以使用来自 scales 的 label_percent 格式化为百分比。然后，在转换为 yearmon 类(来自 zoo::as.yearmon)

library(dplyr)
library(stringr)
library(scales)
df1 %>%
   ungroup %>%
   mutate(across(starts_with('Average'), 
     ~ case_when(as.numeric(.) > 24 ~ str_c(round(as.numeric(.)/24, 2),
    ' Days'),
      TRUE ~ str_c(as.character(round(., 2)), ' hours'))),
      across(starts_with('Per'), label_percent())) %>%
   arrange(desc(zoo::as.yearmon(str_remove(format_date, '\\s+.*'), '%b-%y')),
          desc(readr::parse_number(str_remove_all(format_date, '.*\\(|\\)'))))

-输出

# A tibble: 3 x 11
#  format_date Total_count Count_Diff_1 Count_Diff_2 Count_Diff_3 Average_Diff_1 Average_Diff_2 Average_Diff_3 Per_Diff_1 Per_Diff_2 Per_Diff_3
#  <chr>             <int>        <int>        <int>        <int> <chr>          <chr>          <chr>          <chr>      <chr>      <chr>     
#1 Mar-21 (W2)        1288          519          624          145 2.75 hours     2.59 hours     2.26 Days      40.3%      48.4%      11.3%     
#2 Mar-21 (W1)        1079          508          505           66 6.04 hours     7.78 hours     5.56 Days      47.1%      46.8%      6.1%      
#3 Feb-21 (W5)          53           29           22            2 10.35 hours    6.47 hours     6.77 Days      54.7%      41.5%      3.8%