R：如何离线安装包和依赖项

Question

首先，我知道关于该主题的其他 post，但它没有解决我的问题。

Offline install of R package and dependencies

我需要在一台离线 Ubuntu 机器上安装一些包，但依赖项总是乱七八糟。

首先，我使用以下代码（在在线 ubuntu 机器上）下载所有包和依赖项：

# Loading library
library(tools)

# Function for downloading packages and dependencies
getPackages <- function(packs){
  packages <- unlist(
    tools::package_dependencies(packs, available.packages(),
                                which=c("Depends", "Imports"), recursive=TRUE)
  )
  packages <- union(packs, packages)
  packages
}

# Determining what packages to download
packages <- getPackages(c("tidyverse", "data.table", "RODBC", "RJDBC", "fasttime", "tidyr", "knitr", "randomForest", "RMySQL", "jsonlite"))

# Downloading packages
download.packages(pkgs = packages, destdir = "/path/to/packages/")

# Writing files such that this folder can be used as a repository
write_PACKAGES("/path/to/packages/")

其次，我使用以下命令在离线机器上安装软件包，如 linked post.

中所见

# Installs local packages
install.packages(c("tidyverse", "data.table", "RODBC", "RJDBC", "fasttime", "tidyr", "knitr", "randomForest", "RMySQL", "jsonlite"), contriburl = "file:///path/to/packages/")

发生的情况是安装在几个包上工作，然后崩溃并显示消息。

ERROR: dependency ‘dplyr’ is not available for package ‘tidyr’
* removing ‘/home/h52z/R/x86_64-pc-linux-gnu-library/3.4/tidyr’
ERROR: dependencies ‘dplyr’, ‘tidyr’ are not available for package ‘tidyverse’
* removing ‘/home/h52z/R/x86_64-pc-linux-gnu-library/3.4/tidyverse’
Warning messages:
1: In install.packages(c("tidyverse", "data.table", "RODBC", "RJDBC",  :
  installation of package ‘dplyr’ had non-zero exit status
2: In install.packages(c("tidyverse", "data.table", "RODBC", "RJDBC",  :
  installation of package ‘tidyr’ had non-zero exit status
3: In install.packages(c("tidyverse", "data.table", "RODBC", "RJDBC",  :
  installation of package ‘tidyverse’ had non-zero exit status

尽管使用 tools 库创建了本地存储库，但安装程序似乎无法处理依赖项。这将是一项繁重的工作，必须自己找出软件包的安装顺序。

你有什么建议吗？我是否必须使用像 miniCRAN 这样的工具，或者我是否需要像另一个 link 中的示例那样下载整个 CRAN 存储库？

Answer 1

我建议 miniCRAN，特别是 pkgDep 函数来处理所有依赖项。例如

library(miniCRAN)

pkgs <- c("tidyverse", "data.table", "RODBC", "RJDBC", "fasttime", "tidyr", 
    "knitr", "randomForest", "RMySQL", "jsonlite")
pkgList <- pkgDep(pkgs, type = "source", suggests = FALSE)
makeRepo(pkgList, path="/path/to/packages/", type = c("source"))

然后您可以使用

从存储库安装

install.packages(pkgs, repos="file://path/to/packages/", type="source")

R：如何离线安装包和依赖项

R: How do I install packages and dependencies offline

r

repository

libraries

offline

cran