Script for creating Hound config based on public Stash repositories

prepare_hound_config.py

# Copyright (c) 2015 Marcin Mielnicki
# Licensed under the MIT License
#
# Requirements:
# stashy (https://github.com/RisingOak/stashy): pip install stashy
#
# Creates Hound config (https://github.com/etsy/Hound) with public Stash
# repositories
#
# Code formatting:
# autopep8 --in-place --aggressive --aggressive --max-line-length 120 prepare_hound_config.py

import argparse
import itertools
import json
import re
import stashy


def project_repos_iter(stash, project_key):
    repos = stash.projects[project_key].repos.list()
    for repo in repos:
        if repo['project']['public'] or repo['public']:
            slug = repo['slug']
            browse_url = repo['links']['self'][0]['href']
            project = repo['project']['key']
            clone_url = repo['cloneUrl'] if 'cloneUrl' in repo else [
                x for x in repo['links']['clone'] if x['name'] == 'http'][0]['href']
            # remove username from URL
            clone_url = re.sub('://.*?@', '://', clone_url)
            yield (project, slug, browse_url, clone_url)


def all_repos_iter(stash, excluded_projects):
    projects = stash.projects.list()
    for project in projects:
        project_key = project['key']
        repos = project_repos_iter(stash, project_key)
        for repo in repos:
            if project_key not in excluded_projects:
                yield repo


def hound_repos_iter(repos, interval=0):
    for project_key, slug, browse_url, clone_url in repos:
        repo_name = "{}-{}".format(project_key, slug)
        repo_config = {"url": clone_url}
        if interval > 0:
            repo_config["ms-between-poll"] = interval
        base_url = "{}/{{path}}{{anchor}}".format(browse_url)
        repo_config["url-pattern"] = {"base-url": base_url, "anchor": "#{line}"}
        hound_repo = {repo_name: repo_config}
        yield hound_repo


def prepare_parser():
    parser = argparse.ArgumentParser(description='Creates config for Hound based on public Stash repositories.')
    parser.add_argument('stash_url', type=str, help='Stash URL')
    parser.add_argument('stash_username', type=str, help='Stash username')
    parser.add_argument('stash_password', type=str, help='Stash password')
    parser.add_argument('--interval', type=int, help='interval between poll in ms')
    parser.add_argument('--exclude-projects', type=str, nargs='*', default=[],
                        help='list of projects to exclude')
    parser.add_argument('--limit', type=int, help='maximum number of repositories to configure')
    parser.add_argument('--max-concurrent-indexers', type=int, help='max concurrent indexers')
    return parser


def main():
    parser = prepare_parser()
    args = parser.parse_args()

    stash = stashy.connect(args.stash_url, args.stash_username, args.stash_password)
    limit = args.limit
    interval = args.interval
    excluded_projects = args.exclude_projects

    hound_repos = hound_repos_iter(all_repos_iter(stash, excluded_projects), interval)
    if limit > 0:
        hound_repos = itertools.islice(hound_repos, limit)
    hound_repos_dict = {}
    for repo in hound_repos:
        hound_repos_dict = dict(hound_repos_dict.items() + repo.items())

    config = {'dbpath': 'data', 'repos': hound_repos_dict}
    if args.max_concurrent_indexers > 0:
        config['max-concurrent-indexers'] = args.max_concurrent_indexers
    print json.dumps(config, indent=4)

if __name__ == "__main__":
    main()

A rough shell script for doing it without python (just requires curl and jq): https://gist.github.com/l8nite/654567309434e576cdd3