如何基于行号条件更新DataFrame（R语言dplyr实现）

我们以mpg数据集为例，先添加行号作为ID列：

DATA1 <- mpg
DATA2 <- dplyr::mutate(DATA1, ID = row_number())

需求是基于ID值仅更新DATA2中的部分列，比如：

IF ID1 then manufacturer="Honda", cyl=6, year=2005,
IF ID2 then manufacturer="Toyota", class="sedan"

原来的方法通过多次嵌套mutate和ifelse实现，但代码过于冗长，不适用于涉及多行修改的大型数据集：

DATA3 <- DATA2 %>%
  mutate(manufacturer = ifelse(ID == 1, "Honda", manufacturer))  %>% 
  mutate(cyl = ifelse(ID == 1, 6, cyl)) %>%
  mutate(year = ifelse(ID == 1, 2005, year))
DATA3 <- DATA2 %>%
  mutate(manufacturer = ifelse(ID == 2, "Toyota", manufacturer))  %>%
  mutate(class = ifelse(ID == 2, "sedan", class)) 

方法1：用case_when在单个mutate中批量处理 #

把同一ID的多列更新逻辑集中到一个mutate里，用case_when替代重复的ifelse，代码更简洁易维护：

library(dplyr)

DATA3 <- DATA2 %>%
  mutate(
    manufacturer = case_when(
      ID == 1 ~ "Honda",
      ID == 2 ~ "Toyota",
      TRUE ~ manufacturer  # 其他ID保持原数值
    ),
    cyl = case_when(
      ID == 1 ~ 6,
      TRUE ~ cyl
    ),
    year = case_when(
      ID == 1 ~ 2005,
      TRUE ~ year
    ),
    class = case_when(
      ID == 2 ~ "sedan",
      TRUE ~ class
    )
  )

方法2：创建更新规则表，通过连接实现批量更新（适合大量ID场景） #

如果要更新的ID和列数量很多，先做一个更新规则表，再通过连接合并，用coalesce自动替换需要更新的值，未指定的列保留原值：

library(dplyr)

# 定义更新规则：只填需要修改的ID和对应列，NA表示不修改该列
update_rules <- tibble(
  ID = c(1, 2),
  manufacturer = c("Honda", "Toyota"),
  cyl = c(6, NA),
  year = c(2005, NA),
  class = c(NA, "sedan")
)

# 合并规则表并替换数值
DATA3 <- DATA2 %>%
  left_join(update_rules, by = "ID") %>%
  mutate(
    manufacturer = coalesce(manufacturer.y, manufacturer.x),
    cyl = coalesce(cyl.y, cyl.x),
    year = coalesce(year.y, year.x),
    class = coalesce(class.y, class.x)
  ) %>%
  select(-ends_with(".x"), -ends_with(".y"))  # 删除合并产生的冗余列

内容的提问来源于stack exchange，提问作者user2596306