运行文件夹中的每个文件

Can I write one R script to open and run every R file in a folder?

我可以写一个R脚本来打开并运行文件夹中的每个R文件吗?

I know how to check for the presence of a file in a folder, how to read every file in a folder as a text connection and how to read every data file in a folder.

我知道如何检查文件夹中文件的存在，如何读取文件夹中的每个文件作为文本连接，以及如何读取文件夹中的每个数据文件。

However, I want to execute every R script in a folder one at a time, ideally using a single R script and the default R gui installed on a Windows desktop during installation.

但是，我希望每次都在一个文件夹中执行每个R脚本，理想情况下，在安装过程中使用一个R脚本和默认的R gui安装在Windows桌面上。

I suspect I might need to run R from the command line instead and write some sort of batch file to do this. I have rarely run R from the command line and never written a batch file for R.

我怀疑我可能需要从命令行运行R，并编写某种批处理文件来实现这一点。我很少从命令行运行R，也从不为R编写批处理文件。

Here are some example R scripts all stored in a folder named run_all_these:

下面是一些示例R脚本，它们都存储在名为run_all_these的文件夹中:

The file run.one.r contains:

文件run.one。r包含:

a <- 10
b <- 20
c <- a+b
c

The file run.two.r contains:

文件run.two。r包含:

a <- 10
b <- 20
c <- a-b
c

The file run.three.r contains:

文件run.three。r包含:

a <- 10
b <- 20
c <- a*b
c

The file run.four.r contains:

文件run.four。r包含:

a <- 10
b <- 20
c <- a/b
c

I found virtually nothing on this topic using Google. Although, I did find a little on batch files here:

使用谷歌，我几乎找不到这个主题。虽然，我在这里找到了一些批次文件:

http://cran.r-project.org/bin/windows/base/rw-FAQ.html

My actually R scripts will each create their own output files when run. So, I am mostly concerned with running the R scripts right now. Although the next step would be to open each R script, change a from 10 to 100 and run them again. Perhaps that should be a follow-up post.

实际上，我的R脚本在运行时将各自创建自己的输出文件。所以，我现在最关心的是运行R脚本。尽管下一步是打开每个R脚本，但是将a从10改为100，然后再次运行它们。也许这应该是一个后续的帖子。

Thank you for any suggestions.

谢谢你的建议。

EDIT Nov 20, 2013:

编辑2013年11月20日:

After discussion with Ricardo Saporta below I changed the four input files to:

在与下面的Ricardo Saporta讨论后，我将四个输入文件改为:

File run.one.r:

文件run.one.r:

a <- 10
b <- 20
c <- a+b
print(c)

File run.two.r:

文件run.two.r:

a <- 10
b <- 20
c <- a-b
print(c)

File run.three.r:

文件run.three.r:

a <- 10
b <- 20
c <- a*b
print(c)

File run.four.r:

文件run.four.r:

a <- 10
b <- 20
c <- a/b
print(c)

3 个解决方案

#1

I have the following function in my utils file:

我的utils文件中有以下功能:

## finds all .R files within a folder and soruces them
sourceEntireFolder <- function(folderName, verbose=FALSE, showWarnings=TRUE) { 
  files <- list.files(folderName, full.names=TRUE)

  # Grab only R files
  files <- files[ grepl("\\.[rR]$", files) ]

  if (!length(files) && showWarnings)
    warning("No R files in ", folderName)

  for (f in files) {
    if (verbose)
      cat("sourcing: ", f, "\n")
    ## TODO:  add caught whether error or not and return that
    try(source(f, local=FALSE, echo=FALSE), silent=!verbose)
  }
  return(invisible(NULL))
}

#2

sapply( list.files(run_all_these, full.names=TRUE), source )

You would need to make sure that run_all_these was a valid Windows directory specification.

您需要确保run_all_these是有效的Windows目录规范。

#3

If you don't have subfolders:

如果你没有子文件夹:

library(tidyverse)

list.files("name_folder", full.names = TRUE) %>% walk(source)

#1