steam-backlog-enforcer/steam_backlog_enforcer/_actions.py
Krzysztof kuhy Rudnicki a191c8e459 Release a game after one new achievement instead of 100% completion
Requiring 100% locked the user onto one game for weeks, which proved
ineffective. The assignment and every manual pick now release as soon as one
achievement is unlocked after the game was assigned (Steam unlock_time vs the
pick's started_at / State.current_assigned_at).

- finished_app_ids still means 100% complete; a game released below 100% gets
  released_at plus a 7-day cooldown and returns to the pool.
- Next pick: least recently assigned first, then shortest HLTB time.
- Releasing never evicts: released/expired picks stay installed and playable
  until the user explicitly chooses another game (new pick-manual, or
  accepting the next assignment); with no replacement the current game stays.
- Tampering detection judges a released game by unlocks after its release.
- The daemon reloads every State field each tick so its sweep save cannot
  overwrite cooldowns or assignment times written by the CLI.
- State.load backfills assignment times once and persists them; load-time
  migrations moved to _state_migrations.py.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PhzaTAWtRfJk8gre6NLSka
2026-09-26 16:46:03 +02:00

156 lines
5.4 KiB
Python

"""Stdout-free, state-only core operations shared by the CLI and MCP server.
Every function here is safe to call from a stdio MCP server, where STDOUT
carries the JSON-RPC protocol and any stray write corrupts the stream. That
means: no ``print``/``_echo``/``sys.stdout`` writes, no ``input()``, and no
``sys.exit()``. The interactive CLI (``main.py``) reuses these same functions so
there is a single tested implementation of the underlying behaviour.
"""
from __future__ import annotations
from datetime import UTC, datetime
from typing import TYPE_CHECKING, Any
from steam_backlog_enforcer._allowed_games import (
active_manual_picks,
allowed_games,
drop_inactive_picks,
)
from steam_backlog_enforcer._assignment_progress import record_assignment
# Marks these two as an intentional re-export (they're imported from
# _allowed_games above, not defined here) -- without this, mypy's
# --no-implicit-reexport flags every downstream `from _actions import
# active_manual_picks` as an error. Everything else in this file is
# accessible normally; this only needs to list names imported-then-reexported.
if TYPE_CHECKING:
from steam_backlog_enforcer.config import Config, State
# How long an abandoned pick stays out of the auto-assignment pool, so that
# ``scan`` does not immediately hand back the game the user just rejected.
ABANDON_COOLDOWN_DAYS = 30
def find_manual_pick(state: State, app_id: int) -> dict[str, Any] | None:
"""Return the active manual pick for *app_id*, or ``None``.
Args:
state: The loaded enforcer state.
app_id: The Steam app id to look for.
Returns:
The matching active pick entry, or ``None`` if *app_id* is not one.
"""
return next(
(p for p in active_manual_picks(state) if p.get("app_id") == app_id),
None,
)
def is_manual_pick_locked(state: State) -> bool:
"""Return ``True`` if any manual pick is currently holding the lock.
With several picks allowed at once the lock releases only when every one
of them is released (one new achievement), finished or expired.
Args:
state: The loaded enforcer state.
Returns:
Whether the manual-pick lock is active right now.
"""
return bool(active_manual_picks(state))
def allowed_app_ids(state: State) -> set[int]:
"""Return every app id the enforcer must leave installed and visible.
This is the single source of truth for "may exist" across the uninstall
guard, the process killer, the library hider and the auto-installer. It is
the union of the active manual picks and the current assignment, so two
manually-picked games survive an enforcement pass while everything else is
still removed.
Args:
state: The loaded enforcer state.
Returns:
The set of allowed app ids (empty when nothing is assigned).
"""
return {app_id for app_id, _ in allowed_games(state)}
def manual_pick_slots_left(state: State, max_picks: int) -> int:
"""Return how many further manual picks fit under *max_picks*.
Args:
state: The loaded enforcer state.
max_picks: Configured cap (``Config.max_manual_picks``).
Returns:
Remaining slots, never negative.
"""
return max(0, max_picks - len(active_manual_picks(state)))
def apply_configured_pick(
config: Config, state: State, app_id: int, game_name: str
) -> str | None:
""":func:`apply_manual_pick` under the cap the configuration sets."""
return apply_manual_pick(
state, app_id, game_name, max_picks=config.max_manual_picks
)
def apply_manual_pick(
state: State,
app_id: int,
game_name: str,
*,
max_picks: int = 1,
) -> str | None:
"""Add *app_id* to the manual picks and persist ``state``.
This is the non-interactive, side-effect-scoped core of the CLI's
``pick-manual`` command. It mutates and saves ``State`` only; it deliberately
does **not** run the destructive post-assignment cascade (uninstalling other
games, installing the pick, hiding the library) that the CLI performs after
its interactive ``YES`` confirmation. Keeping this state-only means an
automated caller (the MCP server) can never wipe installed games.
Picking is an explicit choice of another game, so released, finished and
expired picks are dropped here (and enforcement may then remove them).
Args:
state: The enforcer state to mutate and save.
app_id: The Steam app id to lock in.
game_name: Human-readable name for the picked game.
max_picks: How many picks may be active at once.
Returns:
``None`` on success, or a message explaining why the pick was refused
(already picked, or no slots left), in which case nothing was saved.
"""
active = active_manual_picks(state)
if any(p.get("app_id") == app_id for p in active):
return f"{game_name} (AppID={app_id}) is already one of your manual picks."
if len(active) >= max_picks:
names = ", ".join(f"{p['game_name']} (AppID={p['app_id']})" for p in active)
return (
f"You already have {len(active)} manual pick(s) locked in "
f"(cap is {max_picks}): {names}."
)
now = datetime.now(UTC).isoformat()
drop_inactive_picks(state)
state.manual_picks.append(
{"app_id": app_id, "game_name": game_name, "started_at": now}
)
record_assignment(state, app_id, game_name, at=now)
state.save()
return None