操作 data.frame 列表时出现意外错误

Question

我将 data.frame 列表作为自定义函数的输出，因此我打算将每个 data.frame 按其最后一列拆分，其中给出了阈值。但是，我很好地处理了这两个列表，并将它们组合起来只得到一个表。但是在操作这个新表时出现错误。我无法弄清楚问题来自哪里。我该如何解决这个错误？谁能指出我可能修复此错误？如果可以修复此错误，我想实现包装器。如何轻松操作 data.frame 列表？有更好的方法来调试错误吗？

小例子：

savedDF <- list(
  bar = data.frame(.start=c(12,21,37), .stop=c(14,29,45), .score=c(5,9,4)),
  cat = data.frame(.start=c(18,42,18,42,81), .stop=c(27,46,27,46,114), .score=c(10,5,10,5,34)),
  foo = data.frame(.start=c(3,3,33,3,33,91), .stop=c(24,24,10,24,10,17), .score=c(22,22,6,22,6,7))
)

discardedDF <- list(
  bar = data.frame(.start=c(16,29), .stop=c(20,37), .score=c(2,11)),
  cat = data.frame(.start=c(21,31), .stop=c(23,43), .score=c(1,9)),
  foo = data.frame(.start=c(54, 79), .stop=c(71,93), .score=c(3,8))
)

我可以这样操作：

both <- do.call("rbind", c(savedDF, discardedDF))
cn <- c("letter", "seq")
# FIXME : 
DF <- cbind(
  read.table(text = chartr("_", ".", rownames(both)), header=T, sep = ".", col.names = cn), 
  both)
DF <- transform(DF, isPassed = ifelse(.score > 8, "Pass", "Fail"))

by(DF, DF[c("letter", "isPassed")], 
   function(x) write.csv(x[-(1:length(savedDF))], 
                         sprintf("%s_%s_%s.csv", x$letter[1], x$isPassed[1])))

但我有一个错误

Error in scan(file = file, what = what, sep = sep, quote = quote, dec = dec,  : 
  line 15 did not have 2 elements

为什么我有这个错误？谁能指出我如何解决这个问题？

我想要的输出是 CSV 文件列表，如下所示：

bar.saved.Pass.csv
bar.saved.Fail.csv
bar.discarded.Pass.csv
bar.discarded.Fail.csv

cat.saved.Pass.csv
cat.saved.Fail.csv
cat.discarded.Pass.csv
cat.discarded.Fail.csv

foo.saved.Pass.csv
foo.saved.Fail.csv
foo.discarded.Pass.csv
foo.discarded.Fail.csv

但我认为仍然不需要控制导出的 CSV 文件。如何改进此包装器的功能？我打算让使用自定义选择输出目录，或者更动态会很好。任何想法 ？非常感谢

Answer 1

这是你要找的吗？

library(tidyverse)
library(magrittr)

both <- do.call("rbind", c(savedDF, discardedDF))
both %<>% rownames_to_column(var = "cn")
both %<>% separate(cn, c("letters", "seq"), sep = "\\.")
both %<>% mutate(isPassed = ifelse(.score > 8, "Passed", "Failed"),
                 isDiscard = ifelse(is.na(seq), "Saved", "Discarded"))

list_of_dfs <- both %>% split(list(.$letters, .$isPassed, .$isDiscard))
csv_names <- paste0("/Users/nathanday/Desktop/", names(list_of_dfs), ".csv") # change this path
mapply(write.csv, list_of_dfs, csv_names)

%<>% 运算符是简写，因此 both %<>% rownames_to_columm(var = "cn") 与 both <- rownames_to_column(both, var = "cn") 相同

为了使其更“动态”以允许输出路径输入，您可以将其包装在您已经拥有的函数结构中，如下所示：

output_where <- function(output_path, list1, list2) {
    if (!dir.exists(output_path)) {
        dir.create(file.path(output_path))
    }
    both <- do.call(rbind, c(list1, list2))
    both %<>% rownames_to_column(var = "cn")
    both %<>% separate(cn, c("letters", "seq"), sep = "\\.")
    both %<>% mutate(isPassed = ifelse(.score > 8, "Passed", "Failed"), isDiscard = ifelse(is.na(seq), "Saved", "Discarded"))

    list_of_dfs <- both %>% split(list(.$letters, .$isPassed, .$isDiscard))
    csv_names <- paste0(output_path, names(list_of_dfs), ".csv")
    return(mapply(write.csv, list_of_dfs, csv_names))
}

output_where("~/Desktop/", savedDF, discardedDF)

更多动态：

output_where <- function(output_path, list1, list2) {
    if (!dir.exists(output_path)) {
        dir.create(file.path(output_path))
    }
    names(list1) <- paste("list1", names(list1), sep = ".")
    names(list2) <- paste("list2", names(list2), sep = ".")
    both <- do.call(rbind, c(list1, list2))
    both %<>% rownames_to_column(var = "cn")
    both %<>% separate(cn, c("original_list", "letters", "seq"), sep = "\\.")
    both %<>% mutate(isPassed = ifelse(.score > 8, "Passed", "Failed"))

    list_of_dfs <- both %>% split(list(.$letters, .$isPassed, .$original_list))
    csv_names <- paste0(output_path, names(list_of_dfs), ".csv")
    return(mapply(write.csv, list_of_dfs, csv_names))
}