如何在purrr::map中跳过无数据eventid并继续处理其余条目？

我有一个返回eventid列表的函数，列表内容如下：

[1] "314cdd9cecb9b66219c944996f0249b2" "ab26545fc28c693c52329db2a68a06a9" 
"818b7fece8a6f82cecf0b4ef38f903d3"
 [4] "9427475e5b454a44a4d9cc365d72e416" "3ee7569ac3f38bd5d62ca03b13ee1344" 
"35c8c545c966fd2ad738dcccd02a6d8f"
 [7] "f2aa3876acabca48a66ea6eae3d9f1b9" "63308e00869e2009bc6597edbdaf0c99" 
"96e9ac4414cf1fb0ef4164d710c420c8"
[10] "e6b22267452d3dd4d99fa62ff71f7fcc" "8d78600b0a7ab6a7f15c5f935cbace68" 
"80d27fa88ccd30a960caacce5dfac049"
[13] "c95982844161e998ab02c1eab506902d"

随后我使用以下代码，基于上述列表中的eventid执行自定义函数my_func：

purrr::map(event_ids, ~my_func(sport = "basketball_nba", eventId = .x))

正常情况下，该代码会返回类似如下的tibble结果：

my_func('basketball_nba', '314cdd9cecb9b66219c944996f0249b2')
# A tibble: 475 × 13
id                               sport_key  sport…¹ comme…² home_…³ away_…⁴ bookm…⁵ title key   
last_…⁶ name  price point
<chr>                            <chr>      <chr>   <chr>   <chr>   <chr>   <chr>   <chr> 
<chr> <chr>   <chr> <int> <dbl>
1 314cdd9cecb9b66219c944996f0249b2 basketbal… NBA     2023-0… Charlo… Miami … draftk… Draf… 
alte… 2023-0… Char…  -475  15.5

但当某个eventid无对应数据时，会抛出如下错误：

> nba_alt_lines <- purrr::map(event_ids, ~my_func(sport = "basketball_nba", eventId = .x))
Error in `purrr::map()`:
ℹ In index: 7.
Caused by error in `rename()`:
! Can't rename columns that don't exist.
✖ Column `key` doesn't exist.
Run `rlang::last_error()` to see where the error occurred.
> rlang::last_error()
<error/purrr_error_indexed>
Error in `purrr::map()`:
ℹ In index: 7.
Caused by error in `rename()`:
! Can't rename columns that don't exist.
✖ Column `key` doesn't exist.
---
Backtrace:
1. purrr::map(event_ids, ~my_func(sport = "basketball_nba", eventId = .x))
12. dplyr:::rename.data.frame(., bookmaker_key = "key")

请问是否可以通过编程方式跳过无数据的eventid，继续处理其余有效的eventid？

解决方案 #

方法1：使用purrr::safely()包装函数 #

safely()会为每个函数调用返回包含result和error的列表，出错时result为NULL，error存储错误详情。可以借此过滤出成功的结果：

# 用safely包装my_func，保留错误信息
safe_my_func <- purrr::safely(my_func)

# 批量执行所有eventid
results <- purrr::map(event_ids, ~safe_my_func(sport = "basketball_nba", eventId = .x))

# 提取成功的结果，过滤掉NULL值
successful_results <- purrr::keep(results, ~!is.null(.x$result)) %>% 
  purrr::map("result")

# 合并所有成功结果为单个tibble（可选）
combined_tibble <- dplyr::bind_rows(successful_results)

方法2：使用purrr::possibly()返回默认值 #

possibly()可以指定函数出错时返回的默认值，比如空tibble，后续直接合并即可：

# 定义出错时返回空tibble（建议和正常结果列结构一致，避免合并报错）
possibly_my_func <- purrr::possibly(
  my_func, 
  otherwise = tibble::tibble(
    id = character(), sport_key = character(), sport_title = character(),
    commence_time = character(), home_team = character(), away_team = character(),
    bookmaker_key = character(), bookmaker_title = character(), market_key = character(),
    market_last_update = character(), outcome_name = character(), price = integer(),
    point = numeric()
  )
)

# 批量执行，出错时返回预设空tibble
results <- purrr::map(event_ids, ~possibly_my_func(sport = "basketball_nba", eventId = .x))

# 合并结果并过滤空行
combined_tibble <- dplyr::bind_rows(results) %>% 
  dplyr::filter(!dplyr::if_all(dplyr::everything(), is.na))

方法3：手动用tryCatch捕获错误 #

如果不想依赖purrr的包装函数，可以在匿名函数中直接用tryCatch捕获错误，跳过出错的eventid：

results <- purrr::map(event_ids, function(id) {
  tryCatch(
    # 尝试执行函数
    my_func(sport = "basketball_nba", eventId = id),
    # 捕获错误时执行的逻辑
    error = function(e) {
      # 可选：打印错误信息，方便排查
      message(sprintf("跳过eventid %s: %s", id, e$message))
      # 返回NULL，后续过滤
      NULL
    }
  )
})

# 过滤NULL结果并合并为单个tibble
combined_tibble <- dplyr::bind_rows(purrr::compact(results))

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