R 在 filter_at 中开始_with 错误（`starts_with()` 必须在 *selecting* 函数中使用。）

Question

使用类似于此的数据集：

id <- c("john","paul","george","ringo","mick","dozy","beaky","tich")
allow_a <- c("Y","N","Y","Y","","","Y","")
allow_b <- c("N","","N","","N","N","N","Y")
generic_a <- c("N","","","Y","Y","Y","N","")
generic_b <- c("N","","","N","Y","N","","Y")

df <- data.frame(id, allow_a, allow_b, generic_a, generic_b)

如果我用starts_with函数运行一些东西：

df %>%
select(starts_with("all"))

它工作正常并选择正确的列。 但是，如果我尝试：

filter_at(df,
          vars(starts_with("all")),
          any_vars(. == "Y"))

我收到错误Error: `starts_with()` must be used within a *selecting* function. 如果我使用rlang::last_error()来跟踪我得到的错误：

Backtrace:
  1. dplyr::filter_at(...)
 11. tidyselect::starts_with("all")
 13. tidyselect::peek_vars(fn = "starts_with")

我想这是已安装软件包的问题，​​但不确定如何修复它。 软件包版本是：

dplyr_0.8.4
tidyselect_1.0.0

任何人的想法？

Answer 1

我不完全确定发生了什么，但经过今天的一些测试，它与expss包有冲突。 我不知道为什么starts_with这里挑出来，但contains提到装载时expss到库：

Loading required package: expss

Attaching package: ‘expss’

The following objects are masked from ‘package:tidyr’:

    contains, nest

The following objects are masked from ‘package:data.table’:

    copy, like

The following objects are masked from ‘package:dplyr’:

    between, compute, contains, first, last, na_if, recode, vars

不确定如何追溯错误，但如果其他用户遇到相同的错误，这可能很有用...