R语言调试：如何跳过特定语句？含耗时函数场景的技术咨询

Hey there! I totally get the frustration of waiting around for super slow functions when you just want to test the rest of your code. Let's tackle your two questions one by one.

1. 是否可以跳过function1？ #

Absolutely! Looking at your code, function1() is called but its return value isn't assigned to any variable, and there's no indication it modifies global state that the rest of f1() depends on. That means you can safely skip it without breaking subsequent logic.

Here's a clean way to do it with a toggle switch, so you can easily turn it back on when you need to run the full function:

# Add a toggle to control heavy function execution
run_heavy_steps <- FALSE

f1 = function(){
  if(run_heavy_steps){
    function1() # Takes 1 hour - skipped when toggle is FALSE
  }
  
  b = function2() # Takes 2 hours
  c = function3(b)
  statement1
  statement2
  ...
}

Alternatively, you could just comment out the function1() line temporarily, but using a toggle is more reusable if you need to switch back and forth often.

2. 是否可以跳过function2？ #

This one's trickier because b (the output of function2()) is a critical dependency for function3() and the rest of your code. You can't skip function2() entirely, but you can replace it with precomputed or simulated data to avoid waiting 2 hours every time you test.

Here are two practical approaches:

Approach 1: Use simulated test data #

Create a mock version of b that matches the structure and data type of what function2() returns. This lets you test the rest of the function without running the slow step:

f1 = function(){
  # Skip function2() and use simulated data instead
  # b = function2() # Takes 2 hours
  b <- create_mock_b() # Write this function to generate realistic test data
  
  c = function3(b)
  statement1
  statement2
  ...
}

# Example mock function (adjust based on function2's actual output)
create_mock_b <- function(){
  # If function2 returns a data frame, simulate one:
  data.frame(
    col1 = rnorm(100),
    col2 = sample(c("A", "B", "C"), 100, replace = TRUE)
  )
}

Approach 2: Cache the result of function2() #

Run function2() once, save its output to a file, and load that cached value every subsequent time you run f1(). This way you only wait 2 hours once:

f1 = function(){
  # Check for cached version of b
  if(file.exists("b_cache.RData")){
    load("b_cache.RData") # Load precomputed b
  } else {
    b = function2() # Run once and cache
    save(b, file = "b_cache.RData")
  }
  
  c = function3(b)
  statement1
  statement2
  ...
}

For a more robust caching solution, use the memoise package, which handles caching automatically:

library(memoise)

# Create a memoised version of function2()
memoised_function2 <- memoise(function2)

f1 = function(){
  # First run executes function2(), subsequent runs use cached value
  b <- memoised_function2()
  
  c = function3(b)
  statement1
  statement2
  ...
}

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