如何基于var_x与对应is_deleted列提取首个符合条件的var_x值？

Tidyverse 实现 #

利用dplyr的数据操作能力和tidyr的重塑功能，先将宽表转为长表，匹配对应状态列后筛选出符合条件的记录，再取每行首个有效值合并回原数据：

library(tidyverse)

# 提取所有var_x的编号（自动适配1-9的范围）
var_nums <- str_extract(names(df), "(?<=var_)\\d+") %>% 
  unique() %>% 
  na.omit() %>% 
  as.integer()

df_result <- df %>%
  rownames_to_column("row_id") %>%
  # 展开var_x列
  pivot_longer(
    cols = all_of(paste0("var_", var_nums)),
    names_to = "var_num",
    values_to = "var_value",
    names_prefix = "var_"
  ) %>%
  # 展开对应is_deleted列并匹配
  left_join(
    df %>%
      rownames_to_column("row_id") %>%
      pivot_longer(
        cols = all_of(paste0("var_", var_nums, "_is_deleted")),
        names_to = "del_num",
        values_to = "is_deleted",
        names_pattern = "(\\d+)_is_deleted"
      ) %>%
      select(row_id, del_num, is_deleted),
    by = c("row_id", "var_num" = "del_num")
  ) %>%
  # 筛选状态为No的记录，取每行第一个
  filter(is_deleted == "No") %>%
  group_by(row_id) %>%
  slice_head(n = 1) %>%
  ungroup() %>%
  select(row_id, var = var_value) %>%
  # 合并回原数据
  right_join(df %>% rownames_to_column("row_id"), by = "row_id") %>%
  select(-row_id, everything(), var)

print(df_result)

Base R 实现 #

通过遍历每行数据，依次检查每个var_x对应的状态列，找到首个标记为"No"的var_x值：

# 提取所有var_x的编号
var_nums <- unique(na.omit(as.integer(sub("var_(\\d+).*", "\\1", names(df)))))

# 定义单行处理函数
get_first_valid <- function(row) {
  for(num in var_nums) {
    del_col <- paste0("var_", num, "_is_deleted")
    var_col <- paste0("var_", num)
    # 跳过NA状态，找到第一个"No"对应的var值
    if(!is.na(row[del_col]) && row[del_col] == "No") {
      return(row[var_col])
    }
  }
  return(NA) # 无符合条件时返回NA
}

# 应用到每行生成新列
df$var <- apply(df, 1, get_first_valid)

print(df)

两种方法都能自动适配1-9个var_x列的场景，处理后即可得到预期的var列。

内容的提问来源于stack exchange，提问作者pure_func