如何在sbt中查看依赖树？

Question

我正在尝试按照documentation：

中的描述检查SBT依赖关系树

sbt inspect tree clean

但是我收到了这个错误：

[error] inspect usage:
[error]   inspect [uses|tree|definitions] <key>   Prints the value for 'key', the defining scope, delegates, related definitions, and dependencies.
[error]
[error] inspect
[error]        ^

有什么问题？为什么没有SBT构建树？

Answer 1

如果你想实际查看库依赖项（就像你使用Maven那样）而不是任务依赖项（inspect tree显示的那个），那么你将需要使用sbt-dependency-graph插件

将以下内容添加到项目/ plugins.sbt（或全局plugins.sbt）。

addSbtPlugin("net.virtual-void" % "sbt-dependency-graph" % "0.9.2")

然后您可以访问dependencyTree命令和其他命令。

Answer 2

从命令行运行时，发送给sbt的每个参数都应该是一个命令，因此sbt inspect tree clean将：

• 运行inspect命令
• 然后运行tree命令
• 然后是clean命令

这显然失败了，因为inspect需要一个参数。这将做你想要的：

sbt "inspect tree clean"

Answer 3

如果您想查看库依赖项，可以使用coursier插件：https://github.com/coursier/coursier/blob/master/doc/FORMER-README.md#printing-trees

输出示例： 文字（没有颜色）：https://gist.github.com/vn971/3086309e5b005576533583915d2fdec4

请注意，插件与打印树的性质完全不同。它专为快速和并发依赖性下载而设计。但它很好，几乎可以添加到任何项目中，所以我认为值得一提。

Answer 4

使用sbt 1.4.0，无需使用插件即可在sbt中使用dependencyTree任务。

sbtdependencyTree

sbt 1.4.0：https://github.com/sbt/sbt/releases/tag/v1.4.0

Answer 5

我尝试使用上面提到的> ## Minimal example with a list of vectors > # Creating list > mylist = rep(list(1:3), 3) > > # Vector on which I will base the replacement > myvec = 1:3 > > # Replacing 1s by NA > for(i in 1:length(mylist)){ + mylist[[i]] = sapply(mylist[[i]], function(x) x[x ==myvec[i]] = NA) + } > > # But I get NAs for every observation > mylist [[1]] [1] NA NA NA [[2]] [1] NA NA NA [[3]] [1] NA NA NA > > # This is what I wanted: > list(c(NA, 2, 3), c(1, NA, 3), c(1, 2, NA)) [[1]] [1] NA 2 3 [[2]] [1] 1 NA 3 [[3]] [1] 1 2 NA > > > ## More complex example, with a list of lists of vectors that more closely approximates my data structure > # Creating list of lists > mynewlist = rep(list(rep(list(1:3), 3)), 3) > > # Replacing 1s by NAs > for(i in 1:length(mynewlist)){ + mynewlist[[i]] = lapply(mynewlist[[i]], function(x) x[x ==myvec[i]] = NA) + } > > # But now each vector becomes a single NA > mynewlist [[1]] [[1]][[1]] [1] NA [[1]][[2]] [1] NA [[1]][[3]] [1] NA [[2]] [[2]][[1]] [1] NA [[2]][[2]] [1] NA [[2]][[3]] [1] NA [[3]] [[3]][[1]] [1] NA [[3]][[2]] [1] NA [[3]][[3]] [1] NA > > # What I wanted: > list(rep(list(c(NA, 2, 3)), 3), rep(list(c(1, NA, 3)), 3), rep(list(c(1, 2, NA)), 3)) [[1]] [[1]][[1]] [1] NA 2 3 [[1]][[2]] [1] NA 2 3 [[1]][[3]] [1] NA 2 3 [[2]] [[2]][[1]] [1] 1 NA 3 [[2]][[2]] [1] 1 NA 3 [[2]][[3]] [1] 1 NA 3 [[3]] [[3]][[1]] [1] 1 2 NA [[3]][[2]] [1] 1 2 NA [[3]][[3]] [1] 1 2 NA 插件，并获得了9K行作为输出（有很多空行和重复项），而输出为〜180行（对于我的项目中的每个依赖项实际上只有一行）在Maven的> # Creating list of lists > mynewestlist = rep(list(rep(list(1:3), 3)), 3) > > # Replacing 1s by NAs > for(i in 1:length(mynewestlist)){ + mynewestlist[[i]] = lapply(mynewestlist[[i]], function(x) ifelse(x ==myvec[i], NA, x)) + } > > # That's better: > mynewestlist [[1]] [[1]][[1]] [1] NA 2 3 [[1]][[2]] [1] NA 2 3 [[1]][[3]] [1] NA 2 3 [[2]] [[2]][[1]] [1] 1 NA 3 [[2]][[2]] [1] 1 NA 3 [[2]][[3]] [1] 1 NA 3 [[3]] [[3]][[1]] [1] 1 2 NA [[3]][[2]] [1] 1 2 NA [[3]][[3]] [1] 1 2 NA > list(rep(list(c(NA, 2, 3), 3), rep(list(1, NA, 3), 3), rep(list(1, 2, NA), 3)) 输出中。所以我为那个Maven目标写了一个sbt包装器task，这是一个丑陋的骇客，但它确实有效：

"net.virtual-void" % "sbt-dependency-graph"