belote-cli 4.9.7__tar.gz → 4.10.0__tar.gz

{belote_cli-4.9.7 → belote_cli-4.10.0}/CHANGELOG.md

@@ -5,6 +5,103 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [4.10.0] - 2026-11-01
+
+Verification pass over two external audit reports (every claim checked
+against the code), then the full follow-up: BelAtro becomes a true expansion
+of the core game, one shared round engine, the god functions split, and three
+features (Settings, achievement toasts, Daily Challenge). Refactors were
+verified with seeded whole-round fingerprints (BelAtro: 800 rounds across all
+bosses/jokers/flags; classic: 450 rounds incl. heavy undo) — byte-identical
+except the intended undo fix. Tests 1228 → 1286, coverage 79% → 83%.
+
+### Added
+- **Settings screen** (main menu → Settings), persisted to `settings.json` in
+  the config dir: AI difficulty (per seat), target, speed, theme, animations,
+  fast redraw, screen-reader hints, replay analysis, and expansion toggles
+  (BelAtro ghost recording). `BELOTE_*` env vars still override while set; the
+  CLI flags now override the saved values for one session.
+- **Achievement toasts:** unlocking an achievement pops a boxed toast (and a
+  screen-reader line) at the end of the round / game.
+- **Daily Challenge:** one date-seeded BelAtro run per day, identical for
+  everyone (fixed deck, starter shop pool, no endless), with a share code and
+  a local best-result record (`daily.json`).
+- **Screen-reader hints now announce declarations** (`a11y.announce_declaration`
+  existed but was never called).
+
+### Fixed
+- **Classic: undo during play re-ran the coinche flow** — the player was
+  re-prompted to coinche and the AI defenders were silently re-polled
+  (shifting their RNG) every time an undo resumed the play phase.
+- **BelAtro collection gallery** pinned the cursor to the top row (the list
+  was sliced from the selected item), hiding every item above it.
+- **BelAtro rules screen** called `hide_cursor()` without writing it (no-op).
+- **Le Divorce / La Trahison** could leave partner trust frozen at 0 if the
+  round aborted; the restore now runs in a `finally`.
+- **BelAtro: Sans Atout bid from the normal prompt became a pass.**
+  `UICallbacks.prompt_bid` filtered UI sentinels with `isinstance(res, str)`,
+  which also matched `SANS_ATOUT_BID` (the plain string `"sans_atout"`).
+- **L'Aristocrate ace payout ignored La Rupture / L'Anarchie.** Aces are now
+  credited via `compute_trick_winners` (new `_aces_won_by_team` helper) instead
+  of raw `trick_winner_seat` with the final trump.
+- **mypy was blind to every absolute `belote.*` import.** A stray empty
+  `src/__init__.py` made mypy name modules `src.belote.*`; with
+  `ignore_missing_imports` all `from belote.… import` resolved to `Any`.
+  Removed it (guarded by `tests/test_repo_layout.py`) and fixed the 9 errors
+  it was hiding (incl. a possible `None.lower()` in the BelAtro rules screen).
+- **Theme file ignored `XDG_CONFIG_HOME` / `%APPDATA%`** (hardcoded
+  `~/.config`, which also let the test suite write the real user config).
+- **Main menu flicker:** the 300ms steam animation no longer clears the whole
+  screen each tick; it repaints in place.
+- **Achievements are now visible** on the Statistics screen (they were
+  evaluated and saved but never shown).
+
+### Changed — architecture
+- **The core never imports BelAtro** (enforced by `tests/test_repo_layout.py`).
+  Boss/deck/voucher rule switches moved from BelAtro's `BossModifiers` +
+  `_joker_state` keys to a core `RuleVariants` (`belote/variants.py`,
+  `GameState.variants`). BelAtro's chips/mult/money/joker state moved off the
+  core `GameState` onto `belatro/core/state.py::BelatroState`, a subclass that
+  core transitions (now generic over `GameState` subclasses) carry along.
+  The stats screen, bid-hint line, I/V keys and main-menu modes are now
+  registration hooks wired from the entry point via `belatro/integration.py`.
+- **One round engine** (`belote/round_engine.py`): classic `gameflow` and
+  BelAtro `round_driver` are two `RoundPolicy` implementations of
+  `play_round`. Coinche level now lives on `state.coinche_level` in both modes.
+- **God functions split:** `process_event` (369 lines) → per-event handler
+  methods; `drive_round` (450) → a round context + phase functions;
+  `_play_blind` (344) → setup / boss / settlement helpers.
+- **Tests:** the 2037-line `test_belatro.py` split by topic (economy, score
+  accumulator, run state, jokers, shop, …); 2 duplicate tests removed; new
+  coverage for the round engine, headless classic rounds, the collection /
+  rules / shop screens, settings and the daily challenge.
+
+### Changed
+- AI / target / speed / theme moved from the main menu into Settings.
+- Shared key-alias table for the Unix and Windows readers; one
+  `ThemeManager.cycle()`; table-driven card art (byte-identical output);
+  `wrap_words` text helper; `BELOTE_NO_DIFF` read once instead of per frame.
+- One `legal_bids(state)` in `game.py` replaces four hand-rolled copies
+  (bid prompt, bid renderer, BelAtro joker hooks, partner legality gate).
+- One `deck.is_trump_card()` replaces the inline "trump suit or Déluge 7/8"
+  predicate in `deck.py`, `game.py`, `ai.py` and `round_driver.py`
+  (`_card_beats` keeps its Tout-Atout-exclusive variant, now commented).
+- New `belote/paths.py` owns data/config locations (was 5 copies; the BelAtro
+  ones fell back to the CWD on Windows without `%APPDATA%`).
+- Dix de Der uses `GLOBAL_CONFIG.LAST_TRICK_BONUS` everywhere (3 hardcoded 10s).
+- BelAtro history entries are built from typed `GameState`/`ScoringBreakdown`
+  instead of `object` + `getattr`.
+
+### Removed
+- `BelAtroAnnounce.score_popup` (slated for removal in 4.8) and its dead
+  overlay flag helpers.
+- `ScoreAccumulator.partner_jokers_double` / `TrustTrack.partner_jokers_double`
+  (deprecated since 3.5.0; `partner_tier` replaces them).
+- Dead code: `Seat.prev_seat`, `GameState.current_bidder`, `set_announced`,
+  `Config.ensure_stats_dir`, legacy `Config.CARD_*` / `MIN_TERM_*` /
+  `THEME_NAME`, `render.CARD_W/CARD_H/CARD_GAP/SIDE_COL_W`,
+  `ansi.save_cursor/restore_cursor/scroll_region`, unused HUD tables.
+
 ## [4.9.7] - 2026-11-01
 
 Repo-wide audit: four parallel reviewers swept the classic engine, the

{belote_cli-4.9.7 → belote_cli-4.10.0}/DEVELOPMENT.md

@@ -171,6 +171,26 @@ PYTHONPATH=src python scripts/benchmark.py
 
 Use these as a regression-detection floor for future changes.
 
+## Architecture: core game vs. BelAtro expansion
+
+- **The core never imports BelAtro.** `tests/test_repo_layout.py` enforces
+  it; only the composition root `belote/main.py` wires the expansion in via
+  `belote/belatro/integration.py` (menu modes, stats section, settings
+  toggle) and BelAtro installs its run-scoped hooks itself (history
+  override, bid hint, extension keys I/V).
+- **One round engine.** `belote/round_engine.py::play_round` owns the phase
+  order (bidding with all-pass handling → contract → tricks). Classic
+  (`gameflow.ClassicRound`) and BelAtro (`belatro/engine/round_driver.py::
+  _RoundContext`) are `RoundPolicy` implementations — change the round's
+  flow there, once.
+- **Rule variants are core.** Boss / deck / voucher effects that change how
+  Belote is played live on `GameState.variants` (`belote/variants.py`);
+  bosses only switch variants on.
+- **Expansion state rides on a subclass.** BelAtro's chips / mult / money /
+  joker state live on `belatro/core/state.py::BelatroState`. Core
+  transitions are generic over `GameState` subclasses and build states with
+  `replace()`, so the subclass survives every move.
+
 ## Accessibility
 
 Set `BELOTE_A11Y=1` to emit one-line plain-text descriptions of card plays,
@@ -196,8 +216,10 @@ run won/lost).
 
 ## Optional Runtime Flags
 
-The following environment variables enable opt-in features. Each is read
-once at startup; toggling mid-run has no effect.
+Players toggle these from the in-game **Settings** screen (`belote/settings.py`
+→ `settings.json`; `settings.apply()` pushes the values into the modules).
+The environment variables below still work and *override* the saved setting
+while set — handy for debugging and CI. Each is read once at startup.
 
 - `BELOTE_REPLAY=1` — after every Classic round, print a one-line summary
   of how often South's plays matched the Hard-AI's preferred line

{belote_cli-4.9.7 → belote_cli-4.10.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: belote-cli
-Version: 4.9.7
+Version: 4.10.0
 Summary: A 4-player terminal card game
 Project-URL: Homepage, https://github.com/ElysiumDisc/belote
 Project-URL: Repository, https://github.com/ElysiumDisc/belote
@@ -113,14 +113,14 @@ belatro
        .-'``'|-._             )         _.-|
       /  .--.|   `''---...........---''`   |
      /  /    |           BelAtro           |
-     |  |    |       > Start Game <        |
-     |  |    |      AI:     < Hard >       |
-     |  |    |      Target: < 1000 >       |
-     |  |    |     Speed:  < Normal >      |
-     |  |    |  Theme:  < Classic Green >  |
+     |  |    |       Daily Challenge       |
+     |  |    |        > Start Game <       |
+     |  |    |           Settings          |
      |  |    |       Rules & History       |
-      \  \   |         Statistics          |
-       `\ `\ |            Quit             |
+     |  |    |          Statistics         |
+     |  |    |             Quit            |
+      \  \   |                             |
+       `\ `\ |                             |
          `\ `|                             |
          _/ /\                             /
         (__/  \                           /
@@ -166,10 +166,26 @@ belote
 # Play the new Roguelite expansion
 belatro
 
-# Custom settings
+# One-off overrides of the saved settings (not persisted)
 belote --difficulty hard --target 500 --seed 123 --speed fast
 ```
 
+**Settings** live in the main menu's *Settings* screen and are saved to
+`settings.json` in your config directory (`$XDG_CONFIG_HOME/belote`, else
+`~/.config/belote`; `%APPDATA%\belote` on Windows): AI difficulty (per seat
+via Enter), target score, speed, theme, animations, fast redraw,
+screen-reader hints, post-round replay analysis and BelAtro ghost-run
+recording. The old `BELOTE_*` environment variables still work and override
+the saved value while set.
+
+**Daily Challenge** (main menu): one seeded BelAtro run per day, the same for
+everyone (fixed deck, date-derived seed, starter shop pool, no endless).
+The intro shows the day's number and a share code; your best result per day
+is kept locally.
+
+**Achievements** pop a toast the moment you unlock one and are listed on the
+Statistics screen.
+
 
 ## Controls
 
@@ -219,7 +235,7 @@ belote --difficulty hard --target 500 --seed 123 --speed fast
 - **Polished Felt Mat (3.9.4):** The trick mat now has a subtle vignette at its edges, a faint deterministic braille pip-dot texture (fabric-weave feel without intrusive glyphs), and — at standard/spacious terminal sizes — a decorative `╔═══◆═══...═══◆═══╗` outer frame with corner ornaments.
 - **Selection HUD (3.9.4):** Selecting a card in hand now paints a highlighted bar under it AND a centered `► A♠ — Trump ◄` readout below, color-coded by suit / trump / legality.
 - **Hand Sorting:** Strategic "play value" organization (honors grouped together) for better tactical awareness.
-- **Main Menu:** Simple single-player entry point with configurable AI difficulty, Target Score, and Speed.
+- **Main Menu:** Simple single-player entry point; AI difficulty, Target Score, Speed and Theme live in the Settings screen.
 - **Undo/Redo:** Press `Z` to undo your last move during bidding or play.
 - **Statistics:** unified global tracking of games played, win rates, best rounds, and BelAtro expansion milestones.
 - **Responsive Layout (3 tiers):** Three preset layouts — **compact** (80×32, fits 1366×768), **standard** (96×38), **spacious** (120×48+). The game picks the largest preset that fits your terminal on every render, so resizing mid-game adapts automatically; cards, side columns, and HUD verbosity all scale with the preset. Vertical centering pads tall terminals so the game never clings to the top.
@@ -280,7 +296,7 @@ belote/
 PYTHONPATH=src pytest
 ```
 
-Currently **1228 tests** passing with 100% coverage on game-logic modules (4.9.7).
+Currently **1286 tests** passing with 100% coverage on game-logic modules (4.10.0).
 
 ## Technical Integrity
 

{belote_cli-4.9.7 → belote_cli-4.10.0}/README.md

@@ -70,14 +70,14 @@ belatro
        .-'``'|-._             )         _.-|
       /  .--.|   `''---...........---''`   |
      /  /    |           BelAtro           |
-     |  |    |       > Start Game <        |
-     |  |    |      AI:     < Hard >       |
-     |  |    |      Target: < 1000 >       |
-     |  |    |     Speed:  < Normal >      |
-     |  |    |  Theme:  < Classic Green >  |
+     |  |    |       Daily Challenge       |
+     |  |    |        > Start Game <       |
+     |  |    |           Settings          |
      |  |    |       Rules & History       |
-      \  \   |         Statistics          |
-       `\ `\ |            Quit             |
+     |  |    |          Statistics         |
+     |  |    |             Quit            |
+      \  \   |                             |
+       `\ `\ |                             |
          `\ `|                             |
          _/ /\                             /
         (__/  \                           /
@@ -123,10 +123,26 @@ belote
 # Play the new Roguelite expansion
 belatro
 
-# Custom settings
+# One-off overrides of the saved settings (not persisted)
 belote --difficulty hard --target 500 --seed 123 --speed fast
 ```
 
+**Settings** live in the main menu's *Settings* screen and are saved to
+`settings.json` in your config directory (`$XDG_CONFIG_HOME/belote`, else
+`~/.config/belote`; `%APPDATA%\belote` on Windows): AI difficulty (per seat
+via Enter), target score, speed, theme, animations, fast redraw,
+screen-reader hints, post-round replay analysis and BelAtro ghost-run
+recording. The old `BELOTE_*` environment variables still work and override
+the saved value while set.
+
+**Daily Challenge** (main menu): one seeded BelAtro run per day, the same for
+everyone (fixed deck, date-derived seed, starter shop pool, no endless).
+The intro shows the day's number and a share code; your best result per day
+is kept locally.
+
+**Achievements** pop a toast the moment you unlock one and are listed on the
+Statistics screen.
+
 
 ## Controls
 
@@ -176,7 +192,7 @@ belote --difficulty hard --target 500 --seed 123 --speed fast
 - **Polished Felt Mat (3.9.4):** The trick mat now has a subtle vignette at its edges, a faint deterministic braille pip-dot texture (fabric-weave feel without intrusive glyphs), and — at standard/spacious terminal sizes — a decorative `╔═══◆═══...═══◆═══╗` outer frame with corner ornaments.
 - **Selection HUD (3.9.4):** Selecting a card in hand now paints a highlighted bar under it AND a centered `► A♠ — Trump ◄` readout below, color-coded by suit / trump / legality.
 - **Hand Sorting:** Strategic "play value" organization (honors grouped together) for better tactical awareness.
-- **Main Menu:** Simple single-player entry point with configurable AI difficulty, Target Score, and Speed.
+- **Main Menu:** Simple single-player entry point; AI difficulty, Target Score, Speed and Theme live in the Settings screen.
 - **Undo/Redo:** Press `Z` to undo your last move during bidding or play.
 - **Statistics:** unified global tracking of games played, win rates, best rounds, and BelAtro expansion milestones.
 - **Responsive Layout (3 tiers):** Three preset layouts — **compact** (80×32, fits 1366×768), **standard** (96×38), **spacious** (120×48+). The game picks the largest preset that fits your terminal on every render, so resizing mid-game adapts automatically; cards, side columns, and HUD verbosity all scale with the preset. Vertical centering pads tall terminals so the game never clings to the top.
@@ -237,7 +253,7 @@ belote/
 PYTHONPATH=src pytest
 ```
 
-Currently **1228 tests** passing with 100% coverage on game-logic modules (4.9.7).
+Currently **1286 tests** passing with 100% coverage on game-logic modules (4.10.0).
 
 ## Technical Integrity
 

{belote_cli-4.9.7 → belote_cli-4.10.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "belote-cli"
-version = "4.9.7"
+version = "4.10.0"
 description = "A 4-player terminal card game"
 readme = "README.md"
 requires-python = ">=3.10"

{belote_cli-4.9.7 → belote_cli-4.10.0}/src/belote/__init__.py

@@ -1,3 +1,3 @@
-__version__ = "4.9.7"
+__version__ = "4.10.0"
 
 __all__ = ["__version__"]

{belote_cli-4.9.7 → belote_cli-4.10.0}/src/belote/a11y.py

@@ -55,6 +55,17 @@ def is_enabled() -> bool:
     return _ENABLED
 
 
+def set_enabled(on: bool) -> None:
+    """Switch screen-reader hints on/off (the Settings screen)."""
+    global _ENABLED, _sink
+    if on == _ENABLED:
+        return
+    _ENABLED = on
+    if _sink is not None and _sink is not sys.stderr:
+        _sink.close()
+    _sink = None
+
+
 def _refresh_enabled_from_env() -> None:
     """Re-read BELOTE_A11Y from the live environment. Public for tests."""
     global _ENABLED, _sink

{belote_cli-4.9.7 → belote_cli-4.10.0}/src/belote/ai.py

@@ -4,7 +4,7 @@ import random
 from collections import Counter
 from enum import Enum
 
-from .deck import _NONTRUMP_ORDER, Card, Contract, Rank, Suit, trick_rank
+from .deck import _NONTRUMP_ORDER, Card, Contract, Rank, Suit, is_trump_card, trick_rank
 from .game import (
     SANS_ATOUT_BID,
     BidValue,
@@ -70,7 +70,7 @@ class AIPlayer:
         # unseeded Random() for legacy callers that construct an AIPlayer
         # directly (e.g. test fixtures).
         self._rng = rng if rng is not None else random.Random()
-        # Set per decide_card() call from state.boss_modifiers.seven_eight_trump.
+        # Set per decide_card() call from state.variants.seven_eight_trump.
         # All ranking helpers in this class read it via self._se.
         self._se = False
 
@@ -140,7 +140,7 @@ class AIPlayer:
         # set at round start in current code, but the assumption is
         # defensive — a future boss or voucher that toggles visibility
         # mid-round would otherwise see stale cached partner cards.
-        hide = bool(state.boss_modifiers.hide_partner_hand)
+        hide = bool(state.variants.hide_partner_hand)
         partner_key: tuple[int, int, bool] = (completed_count, current_count, hide)
         if self.memory.last_partner_hand_key != partner_key:
             p = partner(self.seat)
@@ -199,9 +199,9 @@ class AIPlayer:
         or None (pass). TA and SA are only considered in round 2; round 1 is
         "take the up-card suit at the standard contract" only.
         """
-        self._se = state.boss_modifiers.seven_eight_trump
+        self._se = state.variants.seven_eight_trump
         # Boss: La Solitude (Partner forced pass)
-        if state.boss_modifiers.partner_forced_pass and self.seat == partner(Seat.SOUTH):
+        if state.variants.partner_forced_pass and self.seat == partner(Seat.SOUTH):
             return None
 
         hand = state.hand_of(self.seat)
@@ -329,7 +329,7 @@ class AIPlayer:
         jack-heavy hand under Le Sauvage even though those jacks would score
         zero in actual play.
         """
-        bm = state.boss_modifiers
+        bm = state.variants
         ta_pts = sum(card_points_with_modifiers(c, Suit.TOUT_ATOUT, bm) for c in hand)
         # The Jack-count bonus is the AI's TA strength signal — drop it if
         # jacks are suppressed.
@@ -364,17 +364,13 @@ class AIPlayer:
         # see deck.py::trick_rank and ::card_points). Every ranking / point
         # read in this method must thread `_se` through or the AI under-
         # values trump cards that should beat them.
-        self._se = state.boss_modifiers.seven_eight_trump
+        self._se = state.variants.seven_eight_trump
 
         # Boss: L'Agent Double (Partner sabotages on 3 random tricks)
-        if state.boss_modifiers.agent_double_active and self.seat == partner(Seat.SOUTH):
+        if state.variants.agent_double_active and self.seat == partner(Seat.SOUTH):
             current_trick = len(state.completed_tricks) + 1
-            raw_tricks = state._joker_state.get("agent_double_tricks", frozenset())
-            sabotage_tricks: frozenset[int] = (
-                raw_tricks if isinstance(raw_tricks, frozenset) else frozenset()
-            )
             trump = state.trump
-            if current_trick in sabotage_tricks and trump is not None:
+            if current_trick in state.variants.sabotage_tricks and trump is not None:
                 return min(legal, key=lambda c: trick_rank(c, trump, self._se))
 
         if not hand:
@@ -459,7 +455,7 @@ class AIPlayer:
         # aces_zero, jacks_zero, tens_zero, ban_clubs). Pre-4.1.1 medium AI used
         # raw card_points and would "preserve" a K of trump under kings_zero as
         # if it still scored 20 — picking it as the highest-value card to keep.
-        bm = state.boss_modifiers
+        bm = state.variants
 
         # Update void inferences
         self._update_voids(state)
@@ -588,7 +584,7 @@ class AIPlayer:
             if c.suit in suit_cards:
                 suit_cards[c.suit].append(c)
 
-        bm = state.boss_modifiers
+        bm = state.variants
         # Honors are J/9/A; drop a rank from the count if it's zeroed by a
         # boss flag (3.9.3 — honor-counting was previously boss-blind).
         # 4.6.4: hoisted out of the per-suit loop. The closure was being
@@ -716,7 +712,7 @@ class AIPlayer:
             if lead_suit_for_ranks is not None and c.suit != lead_suit_for_ranks:
                 if is_sa or trump is Suit.TOUT_ATOUT:
                     return -1
-                if c.suit != trump and not (se and c.rank in (Rank.SEVEN, Rank.EIGHT)):
+                if not is_trump_card(c, trump, se):
                     return -1
             return _NONTRUMP_ORDER[c.rank] if is_sa else trick_rank(c, trump, se)
 
@@ -852,7 +848,7 @@ class AIPlayer:
         # (kings_zero / aces_zero / jacks_zero / tens_zero / ban_clubs).
         # Propagates to _score_discarding_strategy / _score_winning_strategy
         # via the `points` parameter and to _score_leading_strategy via `bm`.
-        bm = state.boss_modifiers
+        bm = state.variants
         points = card_points_with_modifiers(card, trump, bm)
         # Small per-card tiebreaker: when win/loss heuristics are otherwise
         # equal, slightly bias toward playing the higher-value card. The
@@ -1124,7 +1120,7 @@ class AIPlayer:
         # Le Républicain (or any deck/voucher that sets the flag): 7s and 8s
         # are wild and may be played on any suit, so an off-suit 7/8 doesn't
         # prove void in lead suit.
-        wild_active = bool(state._joker_state.get("republicain_wild"))
+        wild_active = state.variants.wild_seven_eight
 
         # 1. Process new completed tricks
         trump = state.trump

{belote_cli-4.9.7 → belote_cli-4.10.0}/src/belote/ansi.py

@@ -236,6 +236,11 @@ def clear_screen() -> str:
     return "\x1b[H\x1b[2J"
 
 
+def clear_below() -> str:
+    """Erase from the cursor to the end of the screen."""
+    return "\x1b[J"
+
+
 def clear_line() -> str:
     return "\x1b[2K"
 
@@ -260,18 +265,6 @@ def alt_screen_off() -> str:
     return "\x1b[?1049l"
 
 
-def save_cursor() -> str:
-    return "\x1b7"
-
-
-def restore_cursor() -> str:
-    return "\x1b8"
-
-
-def scroll_region(top: int, bottom: int) -> str:
-    return f"\x1b[{top};{bottom}r"
-
-
 # Palette accessors. Read precomputed escape strings from the active palette
 # (rebuilt by `_refresh_theme_cache()` via theme_manager's change callback).
 def felt_bg() -> str:

{belote_cli-4.9.7 → belote_cli-4.10.0}/src/belote/belatro/core/round_ledger.py

@@ -1,10 +1,10 @@
 """4.6.2 — RoundLedger.
 
 A mutable side-channel that lives for one BelAtro round and replaces the
-per-event `dataclasses.replace(GameState, ...)` pattern in
+per-event `dataclasses.replace(BelatroState, ...)` pattern in
 `ScoreAccumulator.update_state`.
 
-Frozen-GameState contract still holds at round boundaries: `seal_round()` does
+Frozen-BelatroState contract still holds at round boundaries: `seal_round()` does
 the single `replace()` that produces the final canonical state at round end.
 
 Why this design:
@@ -31,7 +31,7 @@ from contextlib import contextmanager
 from dataclasses import dataclass, field, replace
 from typing import Any
 
-from belote.game import GameState
+from .state import BelatroState
 
 
 @dataclass(slots=True)
@@ -55,8 +55,8 @@ class RoundLedger:
     # updates.
     joker_state: dict[str, Any] = field(default_factory=dict)
 
-    def seal_round(self, state: GameState) -> GameState:
-        """Produce the canonical sealed GameState for the end of the round.
+    def seal_round(self, state: BelatroState) -> BelatroState:
+        """Produce the canonical sealed BelatroState for the end of the round.
 
         Single `replace()` that stamps the ledger's chips/mult/money into the
         frozen state. `joker_state` is already live on `state` (shared dict)

{belote_cli-4.9.7 → belote_cli-4.10.0}/src/belote/belatro/core/run_state.py

@@ -95,6 +95,9 @@ class BelAtroRun:
 
     # ── Determinism ────────────────────────────────────────
     seed: int | None = None
+    # ISO date when this is a Daily Challenge run (see belatro/daily.py):
+    # starter shop pool, no endless offer.
+    daily: str | None = None
     _rng: Any = None
 
     # ── Idempotency guard for voucher.apply() ──────────────