R Studio读取335.1MB CSV文件报错，求替代解决方案

我有一个335.1MB的CSV文件，用R Studio读取时遇到问题，两种readr函数返回不同错误：

• 使用read_csv时：

> Mydata <- read_csv("P:/Projects/Project/Data_folder/mydata/mydata.csv")
Rows: 2068023 Columns: 1                                                                                                                
Error in nchar(x, "width") : invalid multibyte string, element 1
In addition: Warning message:
One or more parsing issues, see `problems()` for details 

• 使用read_csv2时：

> Mydata <- read_csv2("P:/Projects/Project/Data_folder/mydata/mydata.csv")
i Using "','" as decimal and "'.'" as grouping mark. Use `read_delim()` for more control.
Rows: 2045785 Columns: 25                                                                                                                
Error in nchar(x, "width") : invalid multibyte string, element 1
In addition: Warning message:
One or more parsing issues, see `problems()` for details 

read_csv2能正确识别列数，但仍报错，请问我应该换其他方法读取这个文件吗？

1. 优先排查编码问题
   错误提示里的invalid multibyte string几乎都是编码不匹配导致的。可以先指定编码尝试读取，比如：

   # 先检测文件编码
   readr::guess_encoding("P:/Projects/Project/Data_folder/mydata/mydata.csv")
   # 根据检测结果指定编码，比如UTF-8或GBK
   Mydata <- read_csv2("P:/Projects/Project/Data_folder/mydata/mydata.csv", locale = locale(encoding = "UTF-8"))
   

2. 用read_delim()手动控制格式
   read_csv2默认用分号做分隔符，你的文件可能是逗号分隔但小数位用逗号，直接手动指定规则避免自动识别错误：

   Mydata <- read_delim("P:/Projects/Project/Data_folder/mydata/mydata.csv", 
                        delim = ",", 
                        locale = locale(decimal_mark = ",", grouping_mark = "."))
   

3. 优化大文件读取方式
   335MB的文件虽不算超大，但如果内存紧张，可分块读取：

   # 按10万行分块读取
   chunk_reader <- read_csv2("P:/Projects/Project/Data_folder/mydata/mydata.csv", chunk_size = 100000)
   Mydata <- data.frame()
   for(chunk in chunk_reader) {
       Mydata <- rbind(Mydata, chunk)
   }
   

   更高效的方式是用vroom包，专门针对大CSV做了优化，速度快且内存占用低：

   install.packages("vroom")
   library(vroom)
   Mydata <- vroom("P:/Projects/Project/Data_folder/mydata/mydata.csv", delim = ",", locale = locale(decimal_mark = ","))
   

4. 定位具体解析错误
   如果能读取部分数据，执行problems(Mydata)查看具体哪一行哪一列出现问题，针对性清理脏数据。

内容的提问来源于stack exchange，提问作者Espejito