遇到tidyr与rlang报错，如何解决该问题？

Hey there! Since you mentioned hitting errors with tidyr and rlang, let’s walk through the most common pitfalls and fixes—these two packages work closely together, especially with tidy evaluation, so it’s easy to trip up here.

1. Error: "Object 'x' not found" when using dynamic column names #

This is the most frequent issue when you’re trying to use a variable to specify column names (in select(), pivot_longer(), etc.) but aren’t leveraging rlang’s tidy evaluation tools correctly.

Example Scenario: #

You try to run this code and get an object-not-found error:

library(tidyr)
library(rlang)

target_col <- "mpg"
mtcars %>% select(target_col)

Fix: #

Use the {{ }} (curly-curly) operator to unquote the variable—this tells tidyr to evaluate it as a column name:

mtcars %>% select({{ target_col }})

If you’re working with a character vector of multiple columns, use all_of() instead:

target_cols <- c("mpg", "cyl")
mtcars %>% select(all_of(target_cols))

2. Error: "Can't convert a symbol to a string" in custom functions #

When writing your own functions that wrap tidyr tools, misusing expression capture (enquo()) or forgetting to unquote can trigger this error.

Example Scenario: #

This function fails because col is treated as a raw symbol instead of a column reference:

my_pivot <- function(data, col) {
  data %>% pivot_longer(cols = col, names_to = "variable", values_to = "value")
}

my_pivot(mtcars, mpg)

Fix: #

Either capture the expression with enquo() and unquote with !! (bang-bang), or use the simpler curly-curly shorthand:

# Option 1: enquo() + !!
my_pivot <- function(data, col) {
  col_quo <- enquo(col)
  data %>% pivot_longer(cols = !!col_quo, names_to = "variable", values_to = "value")
}

# Option 2: Curly-curly (cleaner for most cases)
my_pivot <- function(data, col) {
  data %>% pivot_longer(cols = {{ col }}, names_to = "variable", values_to = "value")
}

my_pivot(mtcars, mpg) # Now works!

3. Error: "Evaluation error: ..." from rlang::eval_tidy() #

This usually happens when you pass an invalid expression to a tidyr function that relies on rlang’s evaluation—like misformatting a values_fn in pivot_wider().

Example Scenario: #

This code fails because mean(mpg) isn’t properly passed as a function:

mtcars %>% 
  pivot_wider(names_from = cyl, values_from = mpg, values_fn = mean(mpg))

Fix: #

Pass a function reference or a formula to define the aggregation correctly:

# Basic function reference
mtcars %>% 
  pivot_wider(names_from = cyl, values_from = mpg, values_fn = list(mean))

# With additional arguments (like handling NAs)
mtcars %>% 
  pivot_wider(names_from = cyl, values_from = mpg, values_fn = ~mean(.x, na.rm = TRUE))

4. Version Mismatch Errors #

Weird, unexplainable errors often pop up when your tidyr and rlang versions are out of sync—tidyr depends heavily on rlang, so old versions can clash.

Fix: #

Update both packages to their latest stable versions:

install.packages(c("tidyr", "rlang"))

If none of these match your specific error, share the exact error message and a minimal reproducible snippet of your code—that’ll help zero in on the exact fix!

内容的提问来源于stack exchange，提问作者user8486156