$ gem clean and $ gem remove should run in reverse topo sort #112

Closed
zenspider opened this Issue Jun 14, 2011 · 7 comments

Projects

None yet

4 participants

@zenspider zenspider was assigned Jun 14, 2011
@krainboltgreene krainboltgreene changed the title from clean and remove should run in reverse topo sort to $ gem clean and $ gem remove should run in reverse topo sort Jan 1, 2016
@duckinator
Member

rubyforge.org appears to be entirely down, and as such everything related to this issue except the title is missing. Is this still valid? If so, can you provide another source and (preferably) also explain the issue here, so it doesn't disappear again?

@segiddins
Member

It actually seems pretty self-explanatory to me -- the goal is to make sure that you remove the gems with the dependencies first, so you always have a correct setup at every point in the process.

@duckinator
Member

The gem uninstall command contains this code, then loops through and uninstalls deps:

    deplist = Gem::DependencyList.new
# <snip>
    deps = deplist.strongly_connected_components.flatten.reverse

Gem::DependencyList includes the TSort library, and the strongly_connected_components method appears to be defined by said library.

Given that, I'd say this can be closed. (Although I don't have the permissions to close it, so somebody else will need to handle that.)

@segiddins
Member

Thanks for digging into this, @duckinator !

@segiddins segiddins closed this Jan 6, 2016
@zenspider
Contributor

Given that I filed that ticket and some of that code is mine... I'm not so sure.

On Jan 5, 2016, at 19:16, Marie Markwell notifications@github.com wrote:

The gem uninstall command contains this code, then loops through and uninstalls deps:

deplist = Gem::DependencyList.new

deps = deplist.strongly_connected_components.flatten.reverse

Gem::DependencyList includes the TSort library, and the strongly_connected_components method appears to be defined by said library.

Given that, I'd say this can be closed. (Although I don't have the permissions to close it, so somebody else will need to handle that.)


Reply to this email directly or view it on GitHub.

@segiddins
Member

@zenspider the issue is 4.5 years old, and the body is inaccessible. If you happen to remember how to reproduce the issues that prompted the issue, then we can re-open and investigate a fix!

@duckinator
Member

@zenspider the code that is yours showed up 3 months after this issue was opened, and according to your own commit message does what this issue asks for: 2439243

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment