Ping @stevvooe!

Comments are there to help people understand the behavior of a function. This PR just re-states the type signature and seems to be just a find/replace to make lint pass. This is really not in the spirit of effort.

The graph package is subtle and full of archaeology that should be reflected in the godoc. It is going to take time to get proper doc comments in.

@stevvooe I have tried my best to address your concerns. Please review and let me know. I have not tried to just get the lint pass but when adding comments to the code the context that I have is very limited. We do the best we can and then we need expertise to fix godocs and some point in time, either we do it as part of this effort or run the godocs with doc experts afterwards.

@brahmaroutu I am not really talking about documentation expertise. We need the expertise of those who have a deep understanding of this code base. Right now, these comments aren't providing any value other than to pass golint. Most additions are still tautological.

This isn't background that either you or I possess. It will take some time, study and consideration to get this right, in the absence of tracking down the original authors of this code. It is better to be failing the linter than to have useless method commentary.

LGTM

@aaronlehmann PTAL

LGTM

LGTM

@icecrime @jfrazelle Would like to merge but the CI system seems to be failing on something unrelated. Any thoughts?

@stevvooe For these golint PRs, I just merge nevertheless, after testing locally.

LGTM

@tiborvass There are a few symbol changes that need to be checked on build.

@stevvooe what do you mean?

@tiborvass graph.DEFAULTTAG -> graph.DefaultTag should be checked by CI.

@stevvooe I checked locally because CI was down