当某些行包含逗号作为千位分隔符和 " 标志并且没有小数的行没有标志时如何在 R 中读取数据
How to read data in R when some rows contain commas as thousand separator and " flag and the rows without decimals don´t have flag
我正在 R 中加载一个 csv（以逗号分隔），其中包含引号 " 包装每一行，其中包含一个带有小数的值的列，并且特定值用双引号 """ ，没有这个问题的行，没有 "" 换行
csv 文件如下所示：
YEAR,COUNTRY,VALUE_A,VALUE_B
2019,SPAIN, 2000, 300
"2019,SPAIN,""2000,54"",300"
"2014,SPAIN,""2003,223"",""125,057"""
2018,FRANCE,1900,280
这太蠢了...我能想到的最好办法是将其作为文本阅读，然后使用 gsub 清除千位标记和双引号。
# Read the csv as text, so we can run it through gsub
#
file_connection <- file("path_to_csv.csv")
text <- readLines(file_connection)
close(file_connection)
读取csv的字符串内容后我们可以处理文本"formatting"
# 1. Remove the comma as thousand mark
# There HAS to be a better way to do this regex but I couldn't remember
#
sanitized_mark <- gsub('(\"\"[0-9]+),([0-9]+\"\")', '\1\2', text)
# 2. Remove all double quotes
#
sanitized_quotes <- gsub('\"', '', sanitized_mark)
# Paste it all together adding a newline character after each element
#
sanitized <- paste0(sanitized_quotes, collapse="\n")
可以使用 text 参数
读取生成的字符串，就好像它是 .csv 的内容一样
df <- read.csv(text=sanitized)
我正在 R 中加载一个 csv（以逗号分隔），其中包含引号 " 包装每一行，其中包含一个带有小数的值的列，并且特定值用双引号 """ ，没有这个问题的行，没有 "" 换行
csv 文件如下所示：
YEAR,COUNTRY,VALUE_A,VALUE_B
2019,SPAIN, 2000, 300
"2019,SPAIN,""2000,54"",300"
"2014,SPAIN,""2003,223"",""125,057"""
2018,FRANCE,1900,280
这太蠢了...我能想到的最好办法是将其作为文本阅读，然后使用 gsub 清除千位标记和双引号。
# Read the csv as text, so we can run it through gsub
#
file_connection <- file("path_to_csv.csv")
text <- readLines(file_connection)
close(file_connection)
读取csv的字符串内容后我们可以处理文本"formatting"
# 1. Remove the comma as thousand mark
# There HAS to be a better way to do this regex but I couldn't remember
#
sanitized_mark <- gsub('(\"\"[0-9]+),([0-9]+\"\")', '\1\2', text)
# 2. Remove all double quotes
#
sanitized_quotes <- gsub('\"', '', sanitized_mark)
# Paste it all together adding a newline character after each element
#
sanitized <- paste0(sanitized_quotes, collapse="\n")
可以使用 text 参数
df <- read.csv(text=sanitized)