使用rlang s sym在功能中并获取错误

Question

我具有以下功能：

library(tidyverse)
library(dplyr)
library(rlang)

custom_filter <- function(
  df,
  pos_filter_1 = NULL,
  ct_filter_1,
  filter_1
){
  pos_filter_1 <- paste0("df$", pos_filter_1)
  
  position_filter_1 <- if (!is.null(pos_filter_1)) (match.fun(filter_1)(!!rlang::sym(pos_filter_1), ct_filter_1)) else TRUE

  fin_df <- df %>%
    dplyr::filter(position_filter_1)
  
  return(fin_df)
}

但是，当我尝试在数据框架上使用它时，我会在！rlang :: Sym（pos_filter_1）中获得错误错误：无效参数类型。这是一个示例，我尝试在行上过滤列qb_count小于或等于5：

df <- tibble(x1 = letters[1:9],
             qb_count = 1:9,
             rb_count = 10:18)

custom_filter(df, 'qb_count', 5, "<=")

我缺少什么？如果我在功能之外进行操作，这很好。感谢任何帮助！

Answer 1

或更轻松的选项，我们可以创建，

custom_filter <- function(
  df,
  pos_filter_1 = NULL,
  ct_filter_1,
  filter_1
){
  
 
if(is.null(pos_filter_1)) {
 df

} else {
  df %>%
    
    dplyr::filter(match.fun(filter_1)(!! rlang::ensym(pos_filter_1),
         ct_filter_1) )
  
  
  }
  }

如果/else 为-testing

>   custom_filter(df, NULL, 5, "<=")
# A tibble: 9 × 3
  x1    qb_count rb_count
  <chr>    <int>    <int>
1 a            1       10
2 b            2       11
3 c            3       12
4 d            4       13
5 e            5       14
6 f            6       15
7 g            7       16
8 h            8       17
9 i            9       18
> custom_filter(df, 'qb_count', 5, "<=")
# A tibble: 5 × 3
  x1    qb_count rb_count
  <chr>    <int>    <int>
1 a            1       10
2 b            2       11
3 c            3       12
4 d            4       13
5 e            5       14

---

而无需使用检查/else将代码包装在if_any/if_All （它应该与一个或多个列一起使用）并使用any_vars（如果不存在列，则不会评估）

custom_filter <- function(
  df,
  pos_filter_1 = NULL,
  ct_filter_1,
  filter_1
  )
  {



  df %>%    
    dplyr::filter(if_any(any_of(pos_filter_1), 
          ~ match.fun(filter_1)(.x,
                ct_filter_1) ))
  
  
  
  }

- 测试

>   custom_filter(df, NULL, 5, "<=")
# A tibble: 9 × 3
  x1    qb_count rb_count
  <chr>    <int>    <int>
1 a            1       10
2 b            2       11
3 c            3       12
4 d            4       13
5 e            5       14
6 f            6       15
7 g            7       16
8 h            8       17
9 i            9       18

>   custom_filter(df, 'qb_count', 5, "<=")
# A tibble: 5 × 3
  x1    qb_count rb_count
  <chr>    <int>    <int>
1 a            1       10
2 b            2       11
3 c            3       12
4 d            4       13
5 e            5       14