R语言函数可选参数：两种写法差异及代码优化咨询

Hey there! Let's break down exactly why your second createTibble function is throwing an error, then walk through a few ways to eliminate that duplicate code without sacrificing readability.

The Root of the Error #

Your mistake comes from mixing up two very different tools:

• if_else() (and dplyr::if_else()) is a vector-wise function. It’s designed to pick values for individual elements based on a logical vector of the same length. For example, if_else(c(TRUE, FALSE), 1, 0) returns c(1,0).
• But you’re trying to use it as a flow control tool (like base R’s if/else) to choose between two entire data frames. The missing(wk) check gives a single TRUE/FALSE value, but the branches return full tbl_df objects—this type mismatch is why you get the error: condition must be a logical vector, not a tbl_df/tbl/data.frame.

Fixes to Avoid Duplicate Code #

Here are three clean ways to rewrite your function without repeating that long pipeline:

1. Extract Repeated Code to a Middle Variable #

This is the most straightforward approach—handle all the shared processing first, then add the optional filter only if wk is provided:

createTibble <- function(i, yr, wk){
  # Do all the shared processing once
  processed_data <- list_sheets[[i]] %>% 
    filter(LASTUSER %in% users) %>%
    mutate(
      year = as.numeric(format(DATEMODIFI, "%Y")), 
      month = as.numeric(format(DATEMODIFI, "%m")), 
      week = week(DATEMODIFI), 
      day = as.numeric(format(DATEMODIFI, "%d"))
    ) %>%
    select(-DATEMODIFI) %>%
    filter(year == yr)
  
  # Add week filter only if wk is supplied
  if (!missing(wk)) {
    processed_data <- processed_data %>% filter(week == wk)
  }
  
  processed_data
}

2. Combine Filters in a Single filter() Call #

You can also wrap the week condition in a base R if() directly inside the filter() function. If wk is missing, we just return TRUE (which doesn’t filter any rows):

createTibble <- function(i, yr, wk){
  list_sheets[[i]] %>% 
    filter(LASTUSER %in% users) %>%
    mutate(
      year = as.numeric(format(DATEMODIFI, "%Y")), 
      month = as.numeric(format(DATEMODIFI, "%m")), 
      week = week(DATEMODIFI), 
      day = as.numeric(format(DATEMODIFI, "%d"))
    ) %>%
    select(-DATEMODIFI) %>%
    filter(
      year == yr,
      # Only check week if wk is provided
      if (missing(wk)) TRUE else week == wk
    )
}

3. Use purrr::when() for Chained Branching #

If you’re comfortable with the tidyverse’s purrr package, when() lets you write sequential condition checks right in your pipe:

library(purrr)

createTibble <- function(i, yr, wk){
  list_sheets[[i]] %>% 
    filter(LASTUSER %in% users) %>%
    mutate(
      year = as.numeric(format(DATEMODIFI, "%Y")), 
      month = as.numeric(format(DATEMODIFI, "%m")), 
      week = week(DATEMODIFI), 
      day = as.numeric(format(DATEMODIFI, "%d"))
    ) %>%
    select(-DATEMODIFI) %>%
    filter(year == yr) %>%
    when(
      !missing(wk) ~ filter(., week == wk),
      . ~ .  # Return the data as-is if wk is missing
    )
}

All three of these approaches eliminate duplicate code while keeping your logic clear—no more repeating that entire pipeline twice!

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