Notify the requesting profile once its media request is actually present in the catalog (roadmap 06, item 2). Completion transitions stay notification-agnostic; a presence-gated pass at the end of each reconcile run fires the notice, so it means "watchable in Silo", not "download finished". - New System.DispatchOperational: delivery insert + webhook/web-push outbox enqueue in one transaction, post-commit multi-dispatch. The webhook auto-disable notice now rides the same path (replacing its hand-rolled hub publish and the now-removed InsertOperational), which also delivers auto-disable notices over web push. - At-most-once delivery: partial unique index on (profile_id, reason_flags->>'request_id') plus a fulfilled_notified_at marker on media_requests, backfilled for pre-existing completed requests so deploys never flood. - Per-webhook notify_requests toggle (default on) through repo, service, API, and settings UI; gated independently of the episode reason flags. - request.fulfilled rendering in web inbox, realtime toast, web push payload, and Discord/generic webhook payloads, deep-linking to the matched catalog item. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
100 lines
3.4 KiB
Go
100 lines
3.4 KiB
Go
package notifications
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
|
|
"github.com/oklog/ulid/v2"
|
|
)
|
|
|
|
// OperationalDispatch describes how one operational delivery (a non-fanout
|
|
// notice such as webhook.auto_disabled or request.fulfilled) reaches the
|
|
// per-target channels. WebhookFilter selects which of the profile's enabled
|
|
// webhooks receive it; nil means the type must not reach webhooks at all
|
|
// (e.g. the auto-disable notice's loop guard). Web push has no per-type
|
|
// filter: profile-level gating happens before dispatch.
|
|
type OperationalDispatch struct {
|
|
WebhookFilter func(Webhook) bool
|
|
}
|
|
|
|
// DispatchOperational durably creates one operational delivery. The inbox row
|
|
// and the per-target webhook / web push outbox rows commit in a single
|
|
// transaction — a crash afterwards delays channel sends instead of dropping
|
|
// them, because the retry workers recover pending outbox rows — then realtime
|
|
// and channel dispatch run post-commit. Returns nil when the delivery deduped
|
|
// away (the partial unique indexes make operational notices idempotent).
|
|
func (s *System) DispatchOperational(ctx context.Context, delivery Delivery, opts OperationalDispatch) (*InsertedDelivery, error) {
|
|
if s == nil {
|
|
return nil, nil
|
|
}
|
|
tx, err := s.pool.Begin(ctx)
|
|
if err != nil {
|
|
return nil, fmt.Errorf("begin operational dispatch tx: %w", err)
|
|
}
|
|
defer func() { _ = tx.Rollback(ctx) }()
|
|
|
|
inserted, err := s.Deliveries.BulkInsert(ctx, tx, []Delivery{delivery})
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if len(inserted) == 0 {
|
|
return nil, nil
|
|
}
|
|
row := inserted[0]
|
|
|
|
if opts.WebhookFilter != nil && s.webhookRepo != nil && s.Settings.WebhooksEnabled(ctx) {
|
|
hooksByProfile, err := s.webhookRepo.ListEnabledByProfiles(ctx, tx, []string{delivery.ProfileID})
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
attempts := make([]DeliveryAttempt, 0, 2)
|
|
for _, hook := range hooksByProfile[delivery.ProfileID] {
|
|
if !opts.WebhookFilter(hook) {
|
|
continue
|
|
}
|
|
attempts = append(attempts, DeliveryAttempt{
|
|
ID: ulid.Make().String(),
|
|
NotificationDeliveryID: row.ID,
|
|
TargetID: hook.ID,
|
|
})
|
|
}
|
|
if err := s.webhookRepo.EnqueueAttempts(ctx, tx, attempts); err != nil {
|
|
return nil, err
|
|
}
|
|
}
|
|
if s.webPushRepo != nil && s.Settings.WebPushEnabled(ctx) {
|
|
subsByProfile, err := s.webPushRepo.ListEnabledByProfiles(ctx, tx, []string{delivery.ProfileID})
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
attempts := make([]DeliveryAttempt, 0, 2)
|
|
for _, sub := range subsByProfile[delivery.ProfileID] {
|
|
attempts = append(attempts, DeliveryAttempt{
|
|
ID: ulid.Make().String(),
|
|
NotificationDeliveryID: row.ID,
|
|
TargetID: sub.ID,
|
|
})
|
|
}
|
|
if err := s.webPushRepo.EnqueueAttempts(ctx, tx, attempts); err != nil {
|
|
return nil, err
|
|
}
|
|
}
|
|
if err := tx.Commit(ctx); err != nil {
|
|
return nil, fmt.Errorf("commit operational dispatch: %w", err)
|
|
}
|
|
|
|
// Post-commit dispatch is best-effort: the durable inbox row covers
|
|
// websocket reconnect, and the retry workers recover the outbox rows.
|
|
full, err := s.Deliveries.GetRowByID(ctx, row.ID)
|
|
if err != nil || full == nil {
|
|
s.logger.Warn("operational delivery reload failed",
|
|
"delivery_id", row.ID, "error", err)
|
|
return &row, nil
|
|
}
|
|
if err := s.dispatcher.Dispatch(ctx, *full); err != nil {
|
|
s.logger.Warn("operational delivery dispatch failed",
|
|
"delivery_id", row.ID, "error", err)
|
|
}
|
|
return &row, nil
|
|
}
|