refactor repo_stats to use os.pipe

[techknowlogick]

credit to @zeripath for discovering inefficiency and proposing solution.

[zeripath]

You've done the pipe bit correctly as far as I can see. The only thing is that you're throwing away the stderr which would be useful if there's an error so you need to keep it and merge it with the error when there's an error.

Given we have to keep the authors names and email addresses in memory I worry that there is still a potential memory problem although this is a lot better. It may be that we have to run stats by periods and then progressively update otherwise the load on initial import could be bad.

[stale]

This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs during the next 2 months. Thank you for your contributions.

[6543]

jup