Phase D: presence intent comment + sendZoneCombatMessages contract

D1: single block comment above the presence.go best-effort QueryRow
cluster (lines 149-198). Zero-value fallbacks are correct for new users
with no activity rows yet.

D2: contract comment on sendZoneCombatMessages — returns a done channel,
callers chaining post-flush work MUST consume it; in-flow handlers with
nothing to chain MUST explicitly discard. Blocking on flush would stall
command handlers behind the 2-3s/message pacing, defeating streaming.
Two fire-and-forget sites (adventure_arena.go round-advance,
dnd_zone_cmd.go streamFlow) now use `_ =` + comment.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
prosolis
2026-05-15 08:03:42 -07:00
parent 2907a8ac1d
commit 9aca6b202d
4 changed files with 20 additions and 2 deletions

View File

@@ -429,7 +429,10 @@ func (p *AdventurePlugin) resolveArenaSurvival(ctx MessageContext, run *ArenaRun
finalMessage += "`!arena fight` — Face this opponent"
}
p.sendZoneCombatMessages(ctx.Sender, phaseMessages, finalMessage)
// Fire-and-forget: no post-flush work; blocking would stall the
// round-advance handler behind streamed pacing. Contract is honored
// by the explicit discard — see sendZoneCombatMessages comment.
_ = p.sendZoneCombatMessages(ctx.Sender, phaseMessages, finalMessage)
return nil
}

View File

@@ -244,6 +244,13 @@ func (p *AdventurePlugin) sendCombatMessages(userID id.UserID, phaseMessages []s
// sendZoneCombatMessages is the zone/expedition variant — same staging as
// arena, but tighter pacing (23s) so dungeon advances feel snappier.
//
// Contract: returns a done channel that closes after the final DM has
// been sent. Callers that need to chain post-flush work (death follow-ups,
// hospital ads, tier countdowns) MUST consume it. Callers in the middle
// of a command-handler flow that have no post-flush work should explicitly
// discard with `_ = ...` — blocking on flush would stall the handler
// behind the 23s/message pacing, which is the whole point of streaming.
func (p *AdventurePlugin) sendZoneCombatMessages(userID id.UserID, phaseMessages []string, finalMessage string) <-chan struct{} {
return p.sendCombatMessagesWithDelay(userID, phaseMessages, finalMessage, 2, 2) // 23s
}

View File

@@ -652,7 +652,10 @@ func (p *AdventurePlugin) streamFlow(userID id.UserID, phaseMessages []string, f
if len(phaseMessages) == 0 {
return p.SendDM(userID, finalMessage)
}
p.sendZoneCombatMessages(userID, phaseMessages, finalMessage)
// Fire-and-forget: streamFlow's whole job is to hand off to the
// streamer and return — no post-flush chaining. Explicit discard
// honors the sendZoneCombatMessages contract.
_ = p.sendZoneCombatMessages(userID, phaseMessages, finalMessage)
return nil
}

View File

@@ -146,6 +146,11 @@ func (p *PresencePlugin) handleWhois(ctx MessageContext) error {
displayName = string(targetUser)
}
// best-effort: activity tracking is non-critical; missing rows are
// expected for new users. The cluster of `_ = QueryRow(...).Scan(...)`
// calls below all share this rationale — zero values are the correct
// fallback (no messages yet, no streak, no rep, no presence row).
// Get stats
var totalMessages, totalWords int
_ = db.Get().QueryRow(