feedBack/plugins/achievements/engine.py
Byron Gamatos d2569cc2a8
feat(achievements): wall sync drain worker + review fixes (epic PR3) (#592)
* feat(achievements): wall sync drain worker (epic PR3, client side)

Background dead-letter worker that POSTs queued Feat unlocks/removals to the
hosted feedback-achievements wall. Idle unless FEEDBACK_ACHIEVEMENTS_WALL_URL
is set; uses requests + the client-token header (mirrors lyrics_transcribe).

Dead-letter, never drop (pure engine.drain_decision):
  network err / 429 / 5xx -> keep pending (retry)
  other 4xx               -> dead_letter (diagnosable, replayable)
  2xx                     -> delete on server ack
remove-me enqueues a wall removal keyed by the reused player_hash.

Verified by an end-to-end staging round-trip (earn a Feat -> drains onto the
wall with name + short hash -> remove-me -> wall empties) with no IP in tables
or access logs. 42 plugin tests pass (test_sync.py adds the decision table +
ack/retry/dead-letter retention + four-field on-the-wire payload).

The hosted service lives in the new feedback-achievements repo.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* fix(achievements): address local review findings (epic)

Bugs caught in the pre-merge review loop:

- secret_witching Feat was DEAD: post_activity wrote witching_nights_run to the
  DB before snapshotting prev_tiers, so diff_unlocks never saw the fresh unlock.
  Fold the run into the activity delta instead (same asymmetry chart_encore
  uses) so the 7th-night unlock is detected. +regression tests.
- chart_encore broke across restarts: per-chart counter keyed on abs(hash(str)),
  which Python salts per-process (PYTHONHASHSEED). Use a stable sha1 digest so
  the same chart accumulates across sessions. +regression test.
- Bounded the per-activity counter read: _read_counters no longer pulls the
  unbounded chart_plays:* rows (they're bumped/read individually).
- screen.js: gate note:hit/miss on an active-song flag so tuner/calibration note
  events can't inflate Feats or flush a phantom chart:null session.
- screen.js: P-III — prefix the plugin localStorage key (achievements:profile-cat).
- screen.js: extract the duplicated local-ISO-date helper.

45 plugin tests pass (3 new). Wall-side review fixes are in the
feedback-achievements repo.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* feat(achievements): default the drain worker to the hosted wall

Point FEEDBACK_ACHIEVEMENTS_WALL_URL's default at the live got-feedback wall
(https://feedback-achievements.onrender.com) so the drain worker targets it out
of the box; still env-overridable for self-hosting/staging. Nothing publishes
unless the user opted in AND has a profile identity, so a default URL alone
sends nothing.

Tests disable the default (autouse fixture) so no test ever POSTs to production;
drain logic is covered via _drain_once() with an injected poster. 45 pass.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-24 17:01:48 +02:00

192 lines
6.9 KiB
Python
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

"""Achievements & Feats of Power — pure evaluation helpers.
This module holds the side-effect-free core of the engine so it is unit
testable (constitution P-V): no IO, no SQLite, no clock. `routes.py` owns the
storage/HTTP shell and calls into these functions.
**Integration law (structural):** Feats are evaluated from *activity counters*
only (`evaluate_feats` / `apply_activity`); competency Achievements are recorded
from *competency events* the source reports (`report-unlock`). Nothing here ever
converts an activity count into a competency unlock or vice versa.
"""
from __future__ import annotations
# Counter keys the activity model owns. `*_max` keys take the running maximum;
# everything else is a cumulative running total. Kept here (not in routes) so a
# test can assert the contract without standing up a DB.
MAX_COUNTERS = frozenset({"notes_session_max", "streak_insong_max", "chart_encore_max"})
def tier_index_for(tiers, value):
"""Highest 0-based tier index whose threshold is met by ``value``.
Returns -1 when no tier is reached. Tiers are assumed ascending; we scan
all of them rather than short-circuit so an out-of-order catalogue still
resolves to the largest satisfied tier.
"""
idx = -1
for i, threshold in enumerate(tiers or []):
try:
if value >= threshold:
idx = i
except TypeError:
continue
return idx
def feat_counter_value(feat, counters):
"""Activity-counter value backing a Feat definition (0 when absent)."""
key = feat.get("counter")
if not key:
return 0
try:
return int(counters.get(key, 0) or 0)
except (TypeError, ValueError):
return 0
def evaluate_feats(feat_defs, counters):
"""Map ``feat_id -> highest reached tier index`` for all satisfied Feats.
A Feat with no tiers, or whose counter hasn't reached tier 0, is omitted.
Pure: takes the current counters snapshot, returns a plain dict.
"""
out = {}
for feat in feat_defs or []:
fid = feat.get("id")
if not fid:
continue
tiers = feat.get("tiers") or []
if not tiers:
continue
ti = tier_index_for(tiers, feat_counter_value(feat, counters))
if ti >= 0:
out[fid] = ti
return out
def apply_activity(counters, delta):
"""Return a NEW counters dict after folding in one activity ``delta``.
Cumulative keys add; ``*_max`` keys keep the running maximum. The caller
(routes.py) is responsible for the only stateful bit — the per-chart play
count — and passes the post-increment value as ``delta['chart_play_count']``
so this function stays pure.
Recognised delta fields (all optional, default 0):
notes -> notes_total (+=)
song_done -> songs_done (+=)
seconds -> time_total_seconds (+=)
session_notes -> notes_session_max (max)
in_song_streak -> streak_insong_max (max)
chart_play_count -> chart_encore_max (max)
"""
out = dict(counters or {})
def _cur(key):
try:
return int(out.get(key, 0) or 0)
except (TypeError, ValueError):
return 0
def _int(v):
try:
return int(v or 0)
except (TypeError, ValueError):
return 0
out["notes_total"] = _cur("notes_total") + _int(delta.get("notes"))
out["songs_done"] = _cur("songs_done") + _int(delta.get("song_done"))
out["time_total_seconds"] = _cur("time_total_seconds") + _int(delta.get("seconds"))
out["notes_session_max"] = max(_cur("notes_session_max"), _int(delta.get("session_notes")))
out["streak_insong_max"] = max(_cur("streak_insong_max"), _int(delta.get("in_song_streak")))
if delta.get("chart_play_count") is not None:
out["chart_encore_max"] = max(_cur("chart_encore_max"), _int(delta.get("chart_play_count")))
return out
def consecutive_run_length(dates):
"""Longest run of consecutive calendar dates in ``dates`` (ISO 'YYYY-MM-DD').
Used by the `secret_witching` Feat (practise in the 25am window on N
consecutive nights). Pure date arithmetic so it's unit-testable; routes.py
feeds it the distinct night-dates recorded in `comp_ledger`.
"""
from datetime import date
parsed = []
for d in dates or []:
try:
y, m, dd = (int(x) for x in str(d).split("-"))
parsed.append(date(y, m, dd))
except (ValueError, TypeError):
continue
if not parsed:
return 0
parsed = sorted(set(parsed))
best = run = 1
for prev, cur in zip(parsed, parsed[1:]):
if (cur - prev).days == 1:
run += 1
best = max(best, run)
else:
run = 1
return best
# ── Data-minimization contract (binding, code-enforced) ──────────────────────
# The wall payload key-set is frozen here and asserted by a unit test. The
# serializer below is the ONLY way outbound data is built — never dict(row) or
# **model — so a stray field cannot leak. Adding a key makes the test go red.
WALL_PAYLOAD_KEYS = ("display_name", "player_hash", "achievement_id", "unlocked_at")
def build_wall_payload(display_name, player_hash, achievement_id, unlocked_at):
"""Build the EXACT four-field wall payload. ``achievement_id`` must always be
a Feat id (the caller only ever invokes this for Feat unlocks — competency
never syncs). Explicit literal dict on purpose; do not refactor into a
row/model splat."""
return {
"display_name": display_name,
"player_hash": player_hash,
"achievement_id": achievement_id,
"unlocked_at": unlocked_at,
}
def drain_decision(status):
"""Dead-letter state machine for one wall-sync attempt (pure).
``status`` is the HTTP status code, or ``None`` for a network error.
Returns one of:
'ack' → server accepted; delete the row.
'retry' → keep it pending (network error, 429 backoff, or 5xx).
'dead' → any other 4xx; move to dead_letter (diagnosable/replayable).
A row is NEVER silently dropped — it leaves the queue only on 'ack' (or a
user opt-out wiping it).
"""
if status is None:
return "retry"
if 200 <= status < 300:
return "ack"
if status == 429:
return "retry"
if 400 <= status < 500:
return "dead"
return "retry" # 5xx — transient server-side, try again later
def diff_unlocks(prev_tiers, new_tiers):
"""Feat ids whose tier advanced (incl. first unlock).
``prev_tiers`` / ``new_tiers`` are ``feat_id -> tier_index`` maps as
returned by :func:`evaluate_feats`. Returns the ids that are newly present
or moved to a higher tier — i.e. the Feats to record + announce this round.
"""
out = []
for fid, tier in (new_tiers or {}).items():
if tier > prev_tiers.get(fid, -1):
out.append(fid)
return out