election/openstack_election/cmds/setup_election_config.py
Ghanshyam Mann 3021ebe674 Fix the tag to be used for setting election config
When we start the election configuration for dates and
other deadlines, we use the dummy tag generated with election
month and which gets updated once governance release the
actual tag of semver. It is confusing that initial tag
generated by setup_election_config cmd is actual tag or just
a placeholder.

Governance repo has switched to semver type tag since 2018
- https://github.com/openstack/governance/tags?after=oct-2017-elections

Election repo started the below process since 2019:
1. Add placeholder/dummy tag in election config (election month tag generated by script)
- https://review.opendev.org/c/openstack/election/+/661673
2. On email_deadline, release a new tag of governance repo
- https://review.opendev.org/c/openstack/releases/+/680691
3. Update the released tag in election config
- https://review.opendev.org/c/openstack/election/+/681274

This commit does not change the existing process and make the placeholder tag more
clear so that we know that we need to release a new tag in governance at the email_deadline
and update here.

Change-Id: Id8d5e4689ad2374a197e63e7fb41c6b5c265ff5c
2023-06-22 21:40:46 +00:00

291 lines
12 KiB
Python
Executable File

# Licensed under the Apache License, Version 2.0 (the "License"); you may
# not use this file except in compliance with the License. You may obtain
# a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# License for the specific language governing permissions and limitations
# under the License.
import argparse
import datetime
import pytz
import sys
import yaml
from collections import OrderedDict
from openstack_election.config import ISO_FMT
from openstack_election import utils
ONE_WEEK = datetime.timedelta(weeks=1)
TWO_WEEK = datetime.timedelta(weeks=2)
election_parameters = {
'PTL': {
'milestone': 'Release',
'weeks_prior': 2,
'events': ['Election', 'Nominations', ],
},
'TC': {
'milestone': 'Release',
'weeks_prior': 2,
'events': ['Election', 'Campaigning', 'Nominations', ],
},
'combined': {
'milestone': 'Release',
'weeks_prior': 2,
'events': ['Election', 'Campaigning', 'Nominations'],
},
}
def _dict_representer(dumper, data):
return dumper.represent_dict(data.items())
def valid_version(value):
try:
value = float(value)
except ValueError:
msg = ("{} is not an valid release version. Pass release version, for"
" example 2023.1".format(value))
raise Exception(msg)
return value
def valid_date(opt):
try:
d = datetime.datetime.strptime(opt, "%Y-%m-%d")
except ValueError:
msg = "Not a valid date: '{0}'.".format(opt)
raise argparse.ArgumentTypeError(msg)
return d.replace(tzinfo=pytz.UTC)
def find_previous_wednesday(date):
# The are smarter ways to do this
while date.weekday() != 2:
date -= datetime.timedelta(days=1)
return date
def iso_fmt(d):
return d.strftime(ISO_FMT)
def validate_tc_charter(election_type, release_schedule,
selected_start, selected_end):
# NOTE (gmann): This function will validate the selected start and
# end date by this script against what TC charter says.
# As per the translation from current charter(as of 2023-03-03):
# - PTL election needs to be held(start and end) on or before R-2 week
# - TC election needs to be held(start and end) in between of R-8 to R-2
# week.
expected_start_date = 0
expected_end_date = 0
for week in release_schedule.get('cycle', []):
if election_type == 'PTL':
expected_start_date = datetime.datetime.strptime(
str(release_schedule['start-week']),
"%Y-%m-%d").replace(tzinfo=pytz.UTC)
else:
if week.get('name') == 'R-8':
expected_start_date = datetime.datetime.strptime(
week['start'], "%Y-%m-%d").replace(tzinfo=pytz.UTC)
if week.get('name') == 'R-2':
expected_end_date = datetime.datetime.strptime(
week['end'], "%Y-%m-%d").replace(tzinfo=pytz.UTC)
if expected_start_date and expected_end_date:
break
if (selected_start < expected_start_date or
selected_end > expected_end_date):
print("Error: generated start and end date as per given dates in\n"
" parameter is not matching with the TC charter,\n"
" please select the release date correctly.")
exit(1)
def select_release_end_date(release_name, release_schedule):
date = None
event = '%s-final' % (release_name[0:1])
for week in release_schedule.get('cycle', []):
if event in week.get('x-project', []):
date = valid_date(week['end'])
return date
def main():
parser = argparse.ArgumentParser(description=('Given a release '
'date pick some dates for '
'the election'))
parser.add_argument('release', type=valid_version,
help='release version. Example 2023.2')
parser.add_argument('type', choices=['TC', 'PTL', 'combined'])
parser.add_argument('--tc-seats', default=4, choices=['4', '5'],
help='number of TC seats up for election')
# If date is not passed in command line then, this script will
# automatically pick the release end date.
parser.add_argument('--date', default=None, type=valid_date,
help='Date from release schedule in the form '
'YYYY-MM-DD')
args = parser.parse_args()
params = election_parameters[args.type]
offset = datetime.timedelta(weeks=params['weeks_prior'])
# We need to know the releases in order. Fortunately this data exists
# in the releases repo in a really easy format to understand.
series_data = utils.get_series_data()
# TODO(gmann): release versions (release-id) is introduced from 2023.1
# (Antelope) cycle and before that we have only release 'name'. We fetch
# last 3 release data here so until 2023.1 we can fecth release-ids and
# rest with 'name'.
# We will use release version for election tooling and release repo is not
# yet fully moved to the release version so continue using the release
# name to fetch the data from release repo.
release_data = []
names = []
for x in series_data:
names.append(x['name'])
if 'release-id' in x:
release_data.append(str(x['release-id']))
else:
release_data.append(x['name'])
# Find where in the list the requested release sits. This will typically
# be very early but don't assume that.
idx = (release_data.index(str(args.release)) if str(args.release)
in release_data else -1)
# If date is not passed to this script then autimatically
# select the release end date.
if (args.date is None):
schedule = utils.get_schedule_data(names[idx+1])
args.date = select_release_end_date(names[idx+1], schedule)
if args.date is None:
print("Error: no end date found in series data")
exit(1)
# Given the release history:
# Stein, Rocky[0], Queens[1], Pike[2], Ocata[3]
# For the Stein elections candidates come from the previous 2 cycles so:
# Rocky and Queens.
timeframe_name = '%s-%s' % (release_data[idx+2].capitalize(),
release_data[idx+1].capitalize())
# The Queens development cycle begins when we branch Pike-RC1, so collect
# that date from the releases data.
schedule = utils.get_schedule_data(names[idx+3])
event = '%s-rc1' % (names[idx+3][0:1])
for week in schedule.get('cycle', []):
if event in week.get('x-project', []):
timeframe_start = valid_date(week['end'])
print('Setting %s Election\n%s is at: %s' % (args.type,
params['milestone'],
args.date.date()))
end = args.date - offset
print('Latest possible completion is at: %s' % (end.date()))
end = find_previous_wednesday(end)
print('Moving back to Wednesday: %s' % (end.date()))
end = end.replace(hour=23, minute=45)
events = []
for event in params['events']:
e_types = [args.type]
if args.type == 'combined':
e_types = ['PTL', 'TC']
if event == 'Campaigning':
e_types = ['TC']
start = end - TWO_WEEK
# For a TC or combined election we want the email deadline to match the
# beginning of the Campaigning period, which gives the officials time
# to validate the rolls
if args.type in ['TC', 'combined'] and event == 'Campaigning':
# NOTE(gmann): As per TC charter, TC campaigning duration is
# one week.
start = end - ONE_WEEK
email_deadline = start.replace(hour=0, minute=0)
# Otherwise for a PTL election we want to set the email deadline to the
# begining of the Nomination period, again to give officials time to
# validate the rolls
elif args.type == 'PTL' and event == 'Nominations':
email_deadline = start.replace(hour=0, minute=0)
if event == 'Election':
schedule = utils.get_schedule_data(names[idx+1])
validate_tc_charter(args.type, schedule, start, end)
for e_type in e_types:
name = '%s %s' % (e_type, event)
events.insert(0, OrderedDict(name=name,
start=iso_fmt(start),
end=iso_fmt(end)))
print('%s from %s to %s' % (name, iso_fmt(start), iso_fmt(end)))
end = start
print('Set email_deadline to %s' % (iso_fmt(email_deadline)))
if args.type == 'PTL':
# For a PTL election we haven't closed the current cycle so we set the
# timeframe right up to the beginning of the nomination period.
print('Setting PTL timeframe end to email_deadline')
timeframe_end = email_deadline
else:
# For a TC election we should have completed the previous cycle so grab
# the release date for it. It is however possible that the gap between
# that release and the summit doesn't allow for an election. In that
# case we need to use the email_deadline for timeframe_end
# Grab the rlease data and fromvert it to a datetime
timeframe_end = series_data[idx+1]['initial-release']
timeframe_end = datetime.datetime.combine(timeframe_end,
datetime.time(0, 0))
timeframe_end = timeframe_end.replace(tzinfo=pytz.UTC)
if timeframe_end < email_deadline:
print('Setting TC timeframe end to %s Release date %s' %
(series_data[idx+1]['name'], iso_fmt(timeframe_end)))
else:
timeframe_end = email_deadline
print('Setting TC timeframe end to email_deadline')
print('Begining of %s Cycle @ %s' % (release_data[idx+2].capitalize(),
timeframe_start))
print('End of %s cycle @ %s' % (release_data[idx+1].capitalize(),
timeframe_end))
timeframe_span = timeframe_end - timeframe_start
timeframe_span_ok = (datetime.timedelta(11*365/12) <=
timeframe_span <=
datetime.timedelta(13*365/12))
print('Election timeframe: %ss' % (timeframe_span))
if not timeframe_span_ok:
print('Looks like election timespan is outside of \'normal\'')
print('Minimum: %s' % (datetime.timedelta(11*365/12)))
print('Current: %s' % (timeframe_span))
print('Maximum: %s' % (datetime.timedelta(13*365/12)))
configuration = OrderedDict(
release=args.release,
election_type=args.type.lower(),
tag='to-be-released',
tc_seats=int(args.tc_seats),
timeframe=OrderedDict(name=timeframe_name,
start=iso_fmt(timeframe_start),
end=iso_fmt(timeframe_end),
email_deadline=iso_fmt(email_deadline)
),
timeline=events,
)
yaml.Dumper.add_representer(OrderedDict, _dict_representer)
print(yaml.dump(configuration, default_flow_style=False,
default_style='', explicit_start=True))
return 0 if timeframe_span_ok else 1
if __name__ == '__main__':
sys.exit(main())