使用R语言for循环生成图表时遇文件打开错误的求助

问题描述 #

我用R语言的for循环从大型数据集Tot-P.csv生成多张图表，数据集包含A-E共5台仪器的多日期数据，目标是给每台仪器的每个日期生成对应图表。

使用的代码如下：

Phos =  read.csv("Tot-P.csv", header = TRUE)

InstrumentA = subset(Phos, Phos$Instrument == "A") 
InstrumentB = subset(Phos, Phos$Instrument == "B") 
InstrumentC = subset(Phos, Phos$Instrument == "C") 
InstrumentD = subset(Phos, Phos$Instrument == "D") 
InstrumentE = subset(Phos, Phos$Instrument == "E") 


AllDatesA = unique(InstrumentA$Date)
for (i in 1:length(AllDatesA)) {
  Date.i = AllDatesA[i]
  InstrumentA.i = InstrumentA[InstrumentA$Date == Date.i, ]
  YourFileName = paste(Date.i, ".jpg", sep = "")
  jpeg(file = YourFileName)
  plot(x = InstrumentA.i$Sample_ID, y = InstrumentA.i$Results, xlab = "Sample ID", 
       ylab = "Concentration (mg P/l)", main = Date.i)
  dev.off()
}

AllDatesB = unique(InstrumentB$Date)
for (i in 1:length(AllDatesB)) {
  Date.i = AllDatesB[i]
  InstrumentB.i = InstrumentB[InstrumentB$Date == Date.i, ]
  YourFileName = paste(Date.i, ".jpg", sep = "")
  jpeg(file = YourFileName)
  plot(x = InstrumentB.i$Sample_ID, y = InstrumentB.i$Results, xlab = "Sample ID", 
       ylab = "Concentration (mg P/l)", main = Date.i)
  dev.off()
}

AllDatesC = unique(InstrumentC$Date)
for (i in 1:length(AllDatesC)) {
  Date.i = AllDatesC[i]
  InstrumentC.i = InstrumentC[InstrumentC$Date == Date.i, ]
  YourFileName = paste(Date.i, ".jpg", sep = "")
  jpeg(file = YourFileName)
  plot(x = InstrumentC.i$Sample_ID, y = InstrumentC.i$Results, xlab = "Sample ID", 
       ylab = "Concentration (mg P/l)", main = Date.i)
  dev.off()
}

AllDatesD = unique(InstrumentD$Date)
for (i in 1:length(AllDatesD)) {
  Date.i = AllDatesD[i]
  InstrumentD.i = InstrumentD[InstrumentD$Date == Date.i, ]
  YourFileName = paste(Date.i, ".jpg", sep = "")
  jpeg(file = YourFileName)
  plot(x = InstrumentA.i$Sample_ID, y = InstrumentA.i$Results, xlab = "Sample ID", 
       ylab = "Concentration (mg P/l)", main = Date.i)
  dev.off()
}

AllDatesE = unique(InstrumentE$Date)
for (i in 1:length(AllDatesE)) {
  Date.i = AllDatesE[i]
  InstrumentE.i = InstrumentE[InstrumentE$Date == Date.i, ]
  YourFileName = paste(Date.i, ".jpg", sep = "")
  jpeg(file = YourFileName)
  plot(x = InstrumentE.i$Sample_ID, y = InstrumentE.i$Results, xlab = "Sample ID", 
       ylab = "Concentration (mg P/l)", main = Date.i)
  dev.off()
}

运行第一个for循环时，收到错误信息：

Error in dev.off():
QuartzBitmap_Output = unable to open file '9/23/19.jpg'

我猜测错误和文件路径有关，但不知道怎么修复，求帮助。

解决方案 #

1. 核心问题：文件名中的非法路径字符 #

错误原因是日期字符串里的/是系统的路径分隔符，程序会把9/23/19.jpg解析成9/23/19.jpg这个路径，而非单个文件名，但你的工作目录下不存在9/23这个子文件夹，因此无法创建文件。

修复方法是把日期中的/替换成合法的文件名字符（比如-或_），修改生成文件名的代码：

# 将日期中的/替换为-，生成合法文件名
YourFileName = paste(gsub("/", "-", Date.i), ".jpg", sep = "")

2. 优化重复代码（推荐） #

你当前的代码重复了5次几乎相同的循环，不仅冗余，还容易出现笔误（比如InstrumentD的循环里误用了InstrumentA.i的变量）。可以用split按仪器分组，通过嵌套循环简化代码，同时给文件名加入仪器名，避免不同仪器同一天的文件互相覆盖：

Phos <- read.csv("Tot-P.csv", header = TRUE)
# 按仪器名称拆分数据集
instrument_groups <- split(Phos, Phos$Instrument)

# 遍历每个仪器组
for (inst_name in names(instrument_groups)) {
  inst_data <- instrument_groups[[inst_name]]
  all_dates <- unique(inst_data$Date)
  
  # 遍历当前仪器的所有日期
  for (date_i in all_dates) {
    date_data <- inst_data[inst_data$Date == date_i, ]
    # 生成包含仪器名和合法日期的文件名
    file_name <- paste(inst_name, gsub("/", "-", date_i), ".jpg", sep = "_")
    
    jpeg(file = file_name)
    plot(x = date_data$Sample_ID, y = date_data$Results, 
         xlab = "Sample ID", ylab = "Concentration (mg P/l)", 
         main = paste(inst_name, "-", date_i))
    dev.off()
  }
}

内容的提问来源于stack exchange，提问作者ktarr24