Unfetched vs zero issues

[gordongli]

Since it is difficult to track whether ghcrawler has completed crawling for a repository's data, it would be nice for Measure to determine whether there are actually no issues for a repository vs not having fetched them yet. This would help with automating the currently manual commands to queue the unfetched repositories and start ghcrawler.

One approach could be to use the Github API v3 to GET /repos/:owner/:repo's open_issues_count field. (See https://developer.github.com/v3/repos/#get)

WARNING: there are no issues recorded for the ${owner}/${repo} 
  repository. This may just be because we haven't fetched that data 
  yet.

[jeremy-lq]

+1 to the desire for automation. We're looking into a couple solutions now, and will update when we have a solid to implement.