Live crons

Two functions operate on live DynamoDB rather than the weekly S3 dump. They run on prod schedules and mutate game/tournament/challenge state directly.

`starttournaments`

Schedule: 10:00 and 22:00 UTC daily
Source: src/functions/starttournaments.ts

Purpose

Automates the tournament lifecycle:

Find tournaments that are ready to start (enough registered players, scheduled start time reached)
Cancel tournaments that cannot start (insufficient players)
Create initial games for started tournaments via GameFactory
Send notification emails through SES

DynamoDB access

Queries and updates TOURNAMENT records on abstract-play-{stage}. Uses retry logic for throttling (ThrottlingException, etc.).

Gameslib integration

gameinfo — tournament-capable meta games and variant validation
GameFactory — instantiate games from tournament configuration
GameBase / GameBaseSimultaneous — create first moves for simultaneous-start games

Email / i18n

Uses i18next with the apback namespace. Locale files in src/locales/ (en, fr, it). Email templates reference tournament name, meta game, and player lists.

Related backend docs

Tournaments subsystem
Database schema — TOURNAMENT record shape

`standingchallenges`

Schedule: 00:00 and 12:00 UTC daily
Source: src/functions/standingchallenges.ts

Purpose

Processes preset standing challenge requests stored as REALSTANDING records. When conditions are met (matching players online, preset rules satisfied), the cron creates challenge records in DynamoDB so the normal challenge flow in node-backend can pick them up.

DynamoDB access

Queries REALSTANDING and related user/challenge records. Does not use gameslib — pure DynamoDB document operations.

Related backend docs

Challenges subsystem

Dev vs prod

Both functions deploy to dev stacks but EventBridge schedules are disabled on dev. Test by invoking manually:

serverless invoke -f starttournaments --stage prod
serverless invoke -f standingchallenges --stage prod

Use prod with care — these mutate live data.