* fix: correct meeting attendance calculations
* test: change meetingregistration factory defaults
* test: Setup stats tests to verify counts honor meeting.Attended
But the tests aren't actually looking to see what numbers get generated yet.
* test: add test for attendance cross-talk between meetings
* fix: limit attendance count query to single meeting
* refactor: rename attendance.online to .remote
* fix: only count a given person as onsite or remote, but never both
* test: align tests with cleanup
Co-authored-by: Jennifer Richards <jennifer@painless-security.com>
* refactor: replace datetime.now with timezone.now
* refactor: migrate model fields to use timezone.now as default
* refactor: replace datetime.today with timezone.now
datetime.datetime.today() is equivalent to datetime.datetime.now(); both
return a naive datetime with the current local time.
* refactor: rephrase datetime.now(tz) as timezone.now().astimezone(tz)
This is effectively the same, but is less likely to encourage accidental
use of naive datetimes.
* refactor: revert datetime.today() change to old migrations
* refactor: change a missed datetime.now to timezone.now
* chore: renumber timezone_now migration
* chore: add migration to change timestamps to UTC
* refactor: move tz instantiation/caching from TimeSlot to Meeting
* fix: assume utc if meeting.time_zone is blank
* chore: make datetime.combine() calls tz aware in the meeting app
* ci: correctly use meeting.tz in TimeSlotFactory
* chore: compute TimeSlot utc / local times assuming tz-aware times
* chore: use tzaware math for agenda editor timeslot layout
* chore: fill in Meeting.time_zone where it is blank
Nearly all interim meetings on or before 2016-07-01 have blank
time_zone values. This migration fills these in with PST8PDT.
* chore: disallow blank Meeting.time_zone value
* refactor: no need to handle blank time_zone case in TZ migration
* refactor: remove now-unnecessary checks that meeting has time_zone
* chore: fix timezone handling in agenda.ics and Meeting.updated()
* chore: fix tz handling in interim_request_details, exercise in tests
* chore: fix timezone handling for test_interim_send_announcement
* chore: fix timezone handling in agenda_json()
* chore: fix timezone handling in old agenda
* chore: fix timezone handling for EditTimeslotsTests
* refactor: refactor a few fixes for more consistent timezone handling
* chore: add timezone info to timestamps in fixtures
* chore: remove naive datetime warnings found in meetings.tests_views
* chore: fix a few more test failures in meetings.tests_views
All tests in meetings.tests_views now passing
* chore: remove unused import
* chore: fix timezone handling in test_schedule_generator.py
* chore: fix timezone handling affecting meeting.tests_js
* chore: fix timeslot test bug when local date != UTC date
* test: fix a few failing tests, all meetings tests now pass
(for me, anyway)
* chore: renumber migrations
* chore: update timestamp conversion migration
The django-celery-beat package introduces tables with timestamp
columns. These columns are stored in CELERY_TIMEZONE. Because we run with
this set to UTC, the migration ignores these columns.
* chore: fix pytz-related change in migration
* chore: remove duplicate migrations
* chore: remove CELERY_BEAT_TZ_AWARE setting now that USE_TZ is True
* test: avoid failure in test with bogus timezone
* chore: add migration to change timestamps to UTC
* chore: fill in Meeting.time_zone where it is blank
Nearly all interim meetings on or before 2016-07-01 have blank
time_zone values. This migration fills these in with PST8PDT.
* chore: disallow blank Meeting.time_zone value
* refactor: no need to handle blank time_zone case in TZ migration
* refactor: remove now-unnecessary checks that meeting has time_zone
* chore: renumber migrations
* chore: update timestamp conversion migration
The django-celery-beat package introduces tables with timestamp
columns. These columns are stored in CELERY_TIMEZONE. Because we run with
this set to UTC, the migration ignores these columns.
* chore: fix pytz-related change in migration
* chore: be explicit that Meeting.vtimezone can return None
* refactor: remove unnecessary save()
* refactor: replace datetime.now with timezone.now
* refactor: migrate model fields to use timezone.now as default
* refactor: replace datetime.today with timezone.now
datetime.datetime.today() is equivalent to datetime.datetime.now(); both
return a naive datetime with the current local time.
* refactor: rephrase datetime.now(tz) as timezone.now().astimezone(tz)
This is effectively the same, but is less likely to encourage accidental
use of naive datetimes.
* refactor: revert datetime.today() change to old migrations
* refactor: change a missed datetime.now to timezone.now
* chore: renumber timezone_now migration
* chore: renumber migrations
* chore: remove more tools.ietf.org server only related things.
* chore: remove use of tools.ietf.org floorplans\n\nThe data will move into the FloorPlan models instead.
* feat: add model to track session attendance
* feat: add model to track session attendance
* feat: add api to set session attendees
* fix: use user pk instead off person pk in the attended api.
* feat: calculate three of five from attended
* feat: management utility to populate Attended model history
* docs: document why nomcom calculations don't use Attended yet.
* fix: add migration to add new personalapikey endpoint to choices
* test: verify very old last login prevents api key use,
* chore: address review nits
* chore: comment on some idiosyncracies of the expected input to populate_attended
* fix: add unique_together constraint for the Attended model
* fix: correctly handle empty querysets passed to three_of_five_eligible functions.
* fix: use DST-friendly tz calculation for TimeSlot end times
Around DST changes, localize(time + duration).astimezone() is not
the same as localize(time).astimezone() + duration. The latter allows
tz-aware times to do their magic and give correct clock times.
* fix: refactor/correct calculation of week-view event start/end times
There were bugs in the timestamp calculations around DST switchover.
This corrects those and tries to clarify and improve documentation in
the relevant code.
Adds new filtering tags for sessions which have multiple sessions. I.e., in addition to wg,area,type there is wg-sessa, wg-sessb or wg-sessc etc on the tags, so now you can filter either all wg sessions or only specific sessions. Also removes the unused session.order_number, but not yet the session.order_in_meeting() function (which I think is unused too).
- Legacy-Id: 19080
Note: SVN reference [19074] has been migrated to Git commit ddefdecca7
This change locks down the schedule of any meeting that is fully in the past. It leaves open sessions that have finished for meetings that have not yet finished.
Addresses (partially) issue #3083. Commit ready for merge.
- Legacy-Id: 19030
Allow cancellation of individual sessions of multi-session interim meeting. Fixes#2959.
- Legacy-Id: 18757
Note: SVN reference [18724] has been migrated to Git commit 79971e14c7
