Skip to content

Conversation

@AkihiroSuda
Copy link
Member

discussed in #67

Signed-off-by: Akihiro Suda suda.akihiro@lab.ntt.co.jp

Signed-off-by: Akihiro Suda <suda.akihiro@lab.ntt.co.jp>
@dmcgowan
Copy link
Member

dmcgowan commented Jun 7, 2017

I see the original statement about switching to relative paths, but not sure I agree. Unlike tars, continuity manifests are intended to describe a complete volume. When applied to a context directory, it is not intended that the manifest will merge with the existing system. On creation, it is not expected that only a subset of directories will be chosen or a random collection of files will be included, rather the context directory is intended to be preserved as is by the manifest. To this end, you can think of a manifest more as describing something that could be treated as a mount, that mount has its own root and that root will be treated relative to wherever it is mounted, but the mount has a root. By the same logic, the manifest has a root and it is appropriate to describe the files in that manifest as absolute paths with that root.

@AkihiroSuda
Copy link
Member Author

@stevvooe WDYT?

@stevvooe
Copy link
Member

@dmcgowan's argument is fairly convincing. It looks like the existing code treats everything as absolute paths, which seems to have been my intention at the time.

I'm sorry for sending you down this route.

What do we need to do to standardize on absolute paths?

@AkihiroSuda
Copy link
Member Author

sure, opened issue #84 toward standardizing absolute path spec

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants