如何扩展grepl多条件匹配，实现SIC编码到文字分类的转换？

我有一个如下结构的数据框：

tibble [9 x 2] (S3: tbl_df/tbl/data.frame)
 $ Date: chr [1:9] "Tuesday 4 October 2022" "Wednesday 5 October 2022" "Thursday 6 October 2022" "Friday 7 October 2022:"
 $ SIC : chr [1:9] "01500" "01610" "01629" "01630"

我希望通过grepl将'SIC'列转换为对应的文字分类，生成名为"Type"的新列，对应关系如下：

• 01500 = "Mixed farming"
• 01610 = "Support activities for crop production"
• 01629 = " Support activities for animal production (other than farm animal boarding and care) n.e.c."
• 01630 = "Post-harvest crop activities"

目前我只实现了单条件匹配：

df$Type <- ifelse(grepl("01500", df$SIC), "Mixed farming", "Other")

这段代码只能处理单个情况，请问该如何扩展以支持更多条件？

方法1：嵌套ifelse #

直接嵌套多个ifelse依次匹配条件，适合条件数量较少的场景：

df$Type <- ifelse(grepl("01500", df$SIC), "Mixed farming",
                  ifelse(grepl("01610", df$SIC), "Support activities for crop production",
                         ifelse(grepl("01629", df$SIC), "    Support activities for animal production (other than farm animal boarding and care) n.e.c.",
                                ifelse(grepl("01630", df$SIC), "Post-harvest crop activities", "Other"))))

注：如果你的SIC列是精确编码（无多余字符），用df$SIC == "01500"这种精确匹配的效率会比grepl更高。

方法2：使用dplyr::case_when（推荐） #

如果用tidyverse工具链，case_when的语法更清晰，可读性更强：

library(dplyr)

df <- df %>%
  mutate(Type = case_when(
    grepl("01500", SIC) ~ "Mixed farming",
    grepl("01610", SIC) ~ "Support activities for crop production",
    grepl("01629", SIC) ~ "    Support activities for animal production (other than farm animal boarding and care) n.e.c.",
    grepl("01630", SIC) ~ "Post-harvest crop activities",
    TRUE ~ "Other" # 匹配不到的情况统一归为Other
  ))

方法3：构建匹配表关联（更易维护） #

如果后续需要添加更多SIC编码对应关系，建议先做一个匹配表，再通过left_join关联，方便后续管理：

library(dplyr)

# 构建SIC与分类的匹配表
sic_mapping <- tibble(
  SIC = c("01500", "01610", "01629", "01630"),
  Type = c("Mixed farming", 
           "Support activities for crop production",
           "    Support activities for animal production (other than farm animal boarding and care) n.e.c.",
           "Post-harvest crop activities")
)

# 关联匹配表生成Type列
df <- df %>%
  left_join(sic_mapping, by = "SIC") %>%
  mutate(Type = ifelse(is.na(Type), "Other", Type)) # 未匹配到的编码设为Other

这种方法的优势是，后续新增或修改编码对应关系时，只需要调整sic_mapping表，无需修改逻辑代码，适合编码数量较多的场景。

内容的提问来源于stack exchange，提问作者alec22