如何在R中获得带有group_by子句的变量列表？

Question

我试图在R中使用group_by()子句来获取字符串值的列表，请在下面找到一个示例数据。这是我试过的。

result <- data %>%
  group_by(station) %>%
  summarise(values = list(variable))

measurement_vars <- c("PRCP", "SNOW", "SNWD", "TMAX", "TMIN")

在这种情况下，values列将是一个列表。我想通过使用values函数来检查measurement_vars列是否包含特定的字符串，例如measurement_vars。%in%函数不检查列表中的所有值。因此，我尝试unlist() values；但是，它没有工作。我的问题类似于this one，但它是在SQL中。这是我在结果中所期望的(values不会是一个列表)。

​

​

我想过滤包括所有measurement_vars的电台。

data <- structure(list(station = c("ACW00011604", "ACW00011604", "ACW00011604", 
"ACW00011604", "ACW00011604", "ACW00011604", "ACW00011604", "ACW00011604", 
"ACW00011604", "ACW00011604", "ACW00011604", "ACW00011647", "ACW00011647", 
"ACW00011647", "ACW00011647", "ACW00011647", "ACW00011647", "ACW00011647", 
"AE000041196", "AE000041196", "AE000041196", "AE000041196", "AEM00041194", 
"AEM00041194", "AEM00041194", "AEM00041194", "AEM00041217", "AEM00041217", 
"AEM00041217", "AEM00041217"), lat = c(17.1167, 17.1167, 17.1167, 
17.1167, 17.1167, 17.1167, 17.1167, 17.1167, 17.1167, 17.1167, 
17.1167, 17.1333, 17.1333, 17.1333, 17.1333, 17.1333, 17.1333, 
17.1333, 25.333, 25.333, 25.333, 25.333, 25.255, 25.255, 25.255, 
25.255, 24.433, 24.433, 24.433, 24.433), lon = c(-61.7833, -61.7833, 
-61.7833, -61.7833, -61.7833, -61.7833, -61.7833, -61.7833, -61.7833, 
-61.7833, -61.7833, -61.7833, -61.7833, -61.7833, -61.7833, -61.7833, 
-61.7833, -61.7833, 55.517, 55.517, 55.517, 55.517, 55.364, 55.364, 
55.364, 55.364, 54.651, 54.651, 54.651, 54.651), variable = c("TMAX", 
"TMIN", "PRCP", "SNOW", "SNWD", "PGTM", "WDFG", "WSFG", "WT03", 
"WT08", "WT16", "TMAX", "TMIN", "PRCP", "SNOW", "SNWD", "WT03", 
"WT16", "TMAX", "TMIN", "PRCP", "TAVG", "TMAX", "TMIN", "PRCP", 
"TAVG", "TMAX", "TMIN", "PRCP", "TAVG"), start = c(1949, 1949, 
1949, 1949, 1949, 1949, 1949, 1949, 1949, 1949, 1949, 1961, 1961, 
1957, 1957, 1957, 1961, 1961, 1944, 1944, 1944, 1944, 1983, 1983, 
1983, 1983, 1983, 1983, 1984, 1983), end = c(1949, 1949, 1949, 
1949, 1949, 1949, 1949, 1949, 1949, 1949, 1949, 1961, 1961, 1970, 
1970, 1970, 1961, 1966, 2022, 2022, 2022, 2022, 2022, 2022, 2022, 
2022, 2022, 2022, 2020, 2022)), row.names = c(NA, -30L), class = c("tbl_df", 
"tbl", "data.frame"))

Answer 1

我们需要遍历list。要么使用lapply/sapply，要么与purrr::map一起使用

library(dplyr)
library(purrr)
result %>% 
  filter(map_lgl(values, ~ all(measurement_vars %in% .x)))

-output

# A tibble: 2 × 2
  station     values    
  <chr>       <list>    
1 ACW00011604 <chr [11]>
2 ACW00011647 <chr [7]> 

如果我们想要filter原始数据，也可以这样做。

data %>%
   group_by(station) %>% 
   filter(all(measurement_vars %in% variable)) %>%
   ungroup

-output

# A tibble: 18 × 6
   station       lat   lon variable start   end
   <chr>       <dbl> <dbl> <chr>    <dbl> <dbl>
 1 ACW00011604  17.1 -61.8 TMAX      1949  1949
 2 ACW00011604  17.1 -61.8 TMIN      1949  1949
 3 ACW00011604  17.1 -61.8 PRCP      1949  1949
 4 ACW00011604  17.1 -61.8 SNOW      1949  1949
 5 ACW00011604  17.1 -61.8 SNWD      1949  1949
 6 ACW00011604  17.1 -61.8 PGTM      1949  1949
 7 ACW00011604  17.1 -61.8 WDFG      1949  1949
 8 ACW00011604  17.1 -61.8 WSFG      1949  1949
 9 ACW00011604  17.1 -61.8 WT03      1949  1949
10 ACW00011604  17.1 -61.8 WT08      1949  1949
11 ACW00011604  17.1 -61.8 WT16      1949  1949
12 ACW00011647  17.1 -61.8 TMAX      1961  1961
13 ACW00011647  17.1 -61.8 TMIN      1961  1961
14 ACW00011647  17.1 -61.8 PRCP      1957  1970
15 ACW00011647  17.1 -61.8 SNOW      1957  1970
16 ACW00011647  17.1 -61.8 SNWD      1957  1970
17 ACW00011647  17.1 -61.8 WT03      1961  1961
18 ACW00011647  17.1 -61.8 WT16      1961  1966

或在base R中

Filter(\(x) all(measurement_vars %in% x), 
    with(data, split(variable, station)))
$ACW00011604
 [1] "TMAX" "TMIN" "PRCP" "SNOW" "SNWD" "PGTM" "WDFG" "WSFG" "WT03" "WT08" "WT16"

$ACW00011647
[1] "TMAX" "TMIN" "PRCP" "SNOW" "SNWD" "WT03" "WT16"