#!/usr/bin/python

"""
This script queries PDC for all the packages in Fedora, it then go through
each of them and determines if they are inactive on all branches or not.
If they are, it will then query dist-git to find out which ones have
maintainers in addition to `orphan` listed there.

"""

import collections
import logging
import sys
import time

import requests
from requests.adapters import HTTPAdapter
from requests.packages.urllib3.util.retry import Retry


_log = logging.getLogger(__name__)
pdc_active_branches = "https://pdc.fedoraproject.org/extras/active_branches.json"
distgit_api_base_url = "https://src.fedoraproject.org/api/0/"

pdc_namespace_to_dist_git = {
    "rpm": "rpms",
    "container": "container",
    "flatpak": "flatpaks",
    "module": "modules",
}


def retry_session():
    session = requests.Session()
    retry = Retry(
        total=5,
        read=5,
        connect=5,
        backoff_factor=0.3,
        status_forcelist=(500, 502, 504),
    )
    adapter = HTTPAdapter(max_retries=retry)
    session.mount("http://", adapter)
    session.mount("https://", adapter)
    return session


def main():
    """ Queries PDC to find packages retired on all branches.
    From there queries dist-git to find if they still have maintainers.
    """

    _log.info("Querying PDC")
    session = retry_session()

    branch_info = session.get(pdc_active_branches).json()

    fully_retired = collections.defaultdict(list)
    for namespace in sorted(branch_info):
        _log.info("Processing: %s", namespace)
        for package in sorted(branch_info[namespace]):
            _log.info("    %s/%s", namespace, package)
            if True not in [el[1] for el in branch_info[namespace][package]]:
                fully_retired[namespace].append(package)

    for namespace in fully_retired:
        _log.info(
            "%s full retired packages found in %s",
            len(fully_retired[namespace]),
            namespace,
        )

    for namespace in fully_retired:
        _log.info("Processing: %s", namespace)
        for package in fully_retired[namespace]:
            _log.info("    %s/%s", namespace, package)
            ns = pdc_namespace_to_dist_git[namespace]
            url = f"{distgit_api_base_url}{ns}/{package}"

            cnt = 0
            while 1:
                try:
                    data = session.get(url).json()
                    break
                except Exception:
                    if cnt == 4:
                        raise

                    cnt += 1
                    time.sleep(30)

            packagers = set()
            for lvl in data["access_users"]:
                packagers.update([u for u in data["access_users"][lvl]])
            if "orphan" in packagers:
                packagers.remove("orphan")
            else:
                print(
                    f"{ns}/{package} is retired but does not list 'orphan' in its users"
                )

            groups = set()
            for lvl in data["access_groups"]:
                groups.update([u for u in data["access_groups"][lvl]])

            if packagers or groups:
                string = f"{ns}/{package} is retired and has the following:"
                if packagers:
                    string += f" 'maintainers': {(', ').join(packagers)}"
                if groups:
                    string += f" 'groups': @{(', @').join(groups)}"
                print(string)


if __name__ == "__main__":
    logging.basicConfig(
        stream=sys.stderr, format="%(message)s", level=logging.INFO,
    )
    main()