编写R函数：自动识别dta/sav格式并转换为CSV文件

需求背景 #

现有手动读取.dta和.sav格式文件并转换为CSV的R脚本，希望重构为可接收输入文件路径和输出路径的函数，通过格式判断自动调用对应读取函数，最终导出CSV。已有初步函数框架，需补充完善逻辑。

原始手动脚本示例 #

##### Reading in the data ####

#Sometimes you may need to load in multiple files and merge them. Loading in second 
# dta and merging process has been commented out
dta_data <- read_dta("/ihme/limited_use/IDENT/PROJECT_FOLDERS/UNICEF_MICS/UZB/2021_2022/UZB_MICS6_2021_2022_HH_Y2022M12D27.DTA")
#dta_data_2 <- read_dta("/ihme/limited_use/IDENT/PROJECT_FOLDERS/WB_LSMS_ISA/MWI/2019_2020/MWI_LSMS_ISA_IHS5_2019_2020_HH_GEOVARIABLES_Y2022M10D26.DTA")
#merged_MWI <- merge(dta_data, dta_data_2, by = "ea_id", all.x = TRUE)
write_csv(x=dta_data, path = "UZB_MICS6_2021_2022_HH_Y2022M12D27.csv")

#This loads in SAV files, as well as checking where the output will be saved. 
sav_data <- read_sav("/home/j/DATA/WB_LSMS/JAM/2012/JAM_JSLC_2012_ANNUAL_Y2022M11D29.SAV")
getwd() # this is the folder it will save into unless you specify otherwise in the path below
write_csv(x=sav_data, path="JAM_JSLC_2012_ANNUAL_Y2022M11D29.csv")

完善后的函数实现 #

# 先确保加载所需依赖包
library(haven)
library(readr)

convert_to_csv <- function(input_file, output_path) {
  # 提取文件扩展名并转为小写，兼容大小写不同的格式（如.DTA/.SAV）
  file_ext <- tolower(tools::file_ext(input_file))
  
  # 根据扩展名选择对应读取函数
  if (file_ext == "dta") {
    data <- haven::read_dta(input_file)
  } else if (file_ext == "sav") {
    data <- haven::read_sav(input_file)
  } else {
    # 对不支持的格式抛出明确错误
    stop("仅支持.dta或.sav格式的文件，请检查输入路径")
  }
  
  # 将读取的数据写入CSV
  readr::write_csv(data, path = output_path)
  
  # 输出转换成功提示
  message(paste("文件已转换完成，保存路径：", output_path))
}

关键细节说明 #

• 依赖包明确：提前加载haven（负责读取dta/sav格式）和readr（负责写入CSV），避免环境依赖问题
• 扩展名兼容：用tools::file_ext提取扩展名并转小写，解决不同大小写格式的识别问题
• 错误处理：对非dta/sav格式直接抛出错误，避免无意义的运行报错
• 命名空间规范：调用函数时加上包名前缀，避免和其他包的同名函数冲突
• 用户反馈：添加message输出保存路径，让用户直观确认转换结果

使用示例 #

# 转换dta格式文件
convert_to_csv(
  input_file = "/ihme/limited_use/IDENT/PROJECT_FOLDERS/UNICEF_MICS/UZB/2021_2022/UZB_MICS6_2021_2022_HH_Y2022M12D27.DTA",
  output_path = "UZB_MICS6_2021_2022_HH_Y2022M12D27.csv"
)

# 转换sav格式文件
convert_to_csv(
  input_file = "/home/j/DATA/WB_LSMS/JAM/2012/JAM_JSLC_2012_ANNUAL_Y2022M11D29.SAV",
  output_path = "JAM_JSLC_2012_ANNUAL_Y2022M11D29.csv"
)

内容的提问来源于stack exchange，提问作者sgy0003