The fa_lock_week config existed but was never enforced. Now /dropadd blocks
adding players FROM Free Agency when current_week >= fa_lock_week (14).
Dropping players TO FA remains allowed after the deadline.
Also consolidates two league_service.get_current_state() calls into one
shared fetch at the top of add_move() to eliminate a redundant API round-trip.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- custom_commands_service: return POST response directly from create_command()
instead of a follow-up GET by_name
- custom_commands_service: return PUT response directly from update_command()
instead of a follow-up GET by_name
- custom_commands_service: avoid GET after PUT in get_or_create_creator() by
constructing updated creator from model_copy()
- custom_commands_service: return POST response directly from get_or_create_creator()
creator creation instead of a follow-up GET
- custom_commands_service: parallelize all 9 sequential API calls in
get_statistics() with asyncio.gather()
- help_commands_service: return POST response directly from create_help()
instead of a follow-up GET by_name
- help_commands_service: return PUT response directly from update_help()
instead of a follow-up GET by_name
- tests: update test_update_help_success to mock PUT returning dict data
Closes#95
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Add is_past_trade_deadline property to Current model and guard /trade initiate,
submit, and finalize flows. All checks fail-closed (block if API unreachable).
981 tests pass.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
PR #106 added await to scorecard_tracker calls but the tracker
methods were still sync, causing TypeError in production:
- /scorebug: "object NoneType can't be used in 'await' expression"
- live_scorebug_tracker: "object list can't be used in 'await' expression"
Also fixes 5 missing awaits in cleanup_service.py and updates tests.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace custom _make_game/_make_team helpers with existing test
factories for consistency with the rest of the test suite.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Closes#88
Replaced sequential for-loops in get_team_schedule(), get_recent_games(),
and get_upcoming_games() with asyncio.gather() to fire all per-week HTTP
requests concurrently. Also adds import asyncio which was missing.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Closes#99
- Add module-level `_user_team_cache` dict with 60-second TTL so
`get_user_major_league_team` is called at most once per minute per
user instead of on every keystroke.
- Reduce `search_players(limit=50)` to `limit=25` to match Discord's
25-choice display cap and avoid fetching unused results.
- Add `TestGetCachedUserTeam` class covering cache hit, TTL expiry, and
None caching; add `clear_user_team_cache` autouse fixture to prevent
test interference via module-level state.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Keep both the type: ignore annotation and the logger.info call
in admin_maintenance.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The previous code attempted to register a maintenance mode gate via
@self.tree.interaction_check inside setup_hook. That pattern is invalid
in discord.py — interaction_check is an overridable method on CommandTree,
not a decorator. The assignment was silently dropped, making maintenance
mode a no-op and producing a RuntimeWarning about an unawaited coroutine.
Changes:
- Add MaintenanceAwareTree(discord.app_commands.CommandTree) that overrides
interaction_check: blocks non-admins when bot.maintenance_mode is True,
always passes admins through, no-op when maintenance mode is off
- Pass tree_cls=MaintenanceAwareTree to super().__init__() in SBABot.__init__
- Add self.maintenance_mode: bool = False to SBABot.__init__
- Update /admin-maintenance command to actually toggle bot.maintenance_mode
- Add tests/test_bot_maintenance_tree.py with 8 unit tests covering all
maintenance mode states, admin pass-through, DM context, and missing attr
Closes#82
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Ensures all client.post() calls to collection endpoints include
trailing slashes, matching the standardized database API routes.
Covers BaseService.create(), TransactionService, InjuryService,
and DraftListService POST calls.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Reverts universal trailing slash in _build_url which broke custom_commands
endpoints (401 on /execute/). Instead, add trailing slashes only to the
two batch POST endpoints (plays/, decisions/) that need them to avoid
307 redirects dropping request bodies.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The FastAPI server returns 307 redirects for URLs without trailing slashes.
aiohttp follows these redirects but converts POST to GET, silently dropping
the request body. This caused play-by-play and decision data from
/submit-scorecard to never be persisted to the database despite the API
returning success.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add centralized `is_admin(interaction)` helper that includes the
`isinstance(interaction.user, discord.Member)` guard, preventing
AttributeError in DM contexts.
Use it in `can_edit_player_image()` which previously accessed
`guild_permissions.administrator` directly without the isinstance
guard. Update the corresponding test to mock the user with
`spec=discord.Member` so the isinstance check passes.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Roster-level errors, warnings, and suggestions now display as
"[BSG] Too many ML players" so users know which team each issue
applies to.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Removes the @pytest.mark.skip and implements the test using a
patched RosterValidation that raises on the first call, triggering
the except clause in validate_transaction. Verifies the method
returns is_legal=False with a descriptive error message.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
TransactionBuilder cached roster data indefinitely via _roster_loaded flag,
causing validation to use stale counts when builders persisted across multiple
/ilmove invocations. Now validate_transaction() always fetches fresh roster
data. Also adds dependency injection for roster_service to improve testability.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Codebase audit identified ~50 lazy imports. Moved 42 unnecessary ones to
top-level imports — only keeping those justified by circular imports,
init-order dependencies, or optional dependency guards. Updated test mock
patch targets where needed. See #57 for remaining DI candidates.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Prevents users from managing injuries for players not on their team.
Admins bypass the check; org affiliates (MiL/IL) are recognized.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The else branch in descriptive_text() caught both "away leading" and
"tied" cases, always overwriting the tied text. Changed to if/elif/else
so tied scores display correctly.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
load_existing_transactions only queried for the base team abbreviation
(e.g. "POR"), missing trades involving MiL/IL affiliates ("PORMIL",
"PORIL"). This caused false "too many players" errors when a pending
trade would have cleared a roster spot.
- get_team_transactions now accepts Union[str, List[str]] for team_abbrev
- load_existing_transactions queries all org affiliates [BASE, BASEMiL, BASEIL]
- Added 5 tests covering the fix and backwards compatibility
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The production container has ambiguous timezone config — /etc/localtime
points to Etc/UTC but date reports CST. The transaction freeze/thaw task
used datetime.now() (naive, relying on OS timezone), causing scheduling
to fire at unpredictable wall-clock times.
- Add utils/timezone.py with centralized Chicago timezone helpers
- Fix tasks/transaction_freeze.py to use now_chicago() for scheduling
- Fix utils/logging.py timestamp to use proper UTC-aware datetime
- Add 14 timezone utility tests
- Update freeze task tests to mock now_chicago instead of datetime
Closes#43
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
#40: ScorecardTracker cached data in memory at startup — background task never
saw newly published scorecards. Fixed by reloading from disk on every read.
#39: Win percentage defaulted to 50% when unavailable, showing a misleading
50/50 bar. Now defaults to None with "unavailable" message in embed. Parsing
handles decimal (0.75), percentage string, and empty values. Also fixed
orientation bug where win% was always shown as home team's even when the
sheet reports the away team as the leader.
Additionally: live scorebug tracker now distinguishes between "all games
confirmed final" and "sheet read failures" — transient Google Sheets errors
no longer hide the live scores channel.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- #37: Fix stale comment in transaction_freeze.py referencing wrong moveid format
- #27: Change config.testing default from True to False (was masking prod behavior)
- #25: Replace deprecated asyncio.get_event_loop() with get_running_loop()
- #38: Replace naive datetime.now() with timezone-aware datetime.now(UTC) across
7 source files and 4 test files to prevent subtle timezone bugs
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Ran `ruff check --select F401 --fix` to auto-remove 221 unused imports,
manually removed 4 unused `import discord` from package __init__.py files,
and fixed test import for DISAPPOINTMENT_TIERS to reference canonical location.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The Stadium Image field was added to the weather embed but the
test_full_weather_workflow assertion wasn't updated to match.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
RosterValidation used total_wara instead of total_sWAR, causing /legal
to silently fail. Transaction embed and submit validation now pass
next_week to validate_transaction() so pending trades are included in
roster count projections. Moved lazy imports to top-level in
transaction_embed.py. Fixed dropadd integration test fixtures that
exceeded sWAR cap.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The API expects 'demotion_week' as the query parameter name, not 'dem_week'.
Updated service to send correct parameter name and tests to verify.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Use channel.send() instead of followup.send() for custom command output
(webhook-based followup messages don't trigger mention notifications)
- Add ephemeral "Sending..." confirmation to satisfy interaction response
- Add utils/mentions.py for converting text @mentions to Discord format
- Add tests for mention conversion utility
Co-Authored-By: Claude <noreply@anthropic.com>
- PlayerService.search_players() now supports all_seasons=True to search across all 13 seasons
- Autocomplete shows unique player names (most recent season's team) instead of duplicates
- Command defaults to most recent season when no season parameter specified
- Users can specify season parameter for historical data
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
Bug Fix:
- Fixed /dropadd transactions being marked frozen=True during thaw period
- Now uses current_state.freeze to set frozen flag correctly
- Transactions entered Sat-Sun are now unfrozen and execute Monday
New Feature - Transaction Thaw Report:
- Added data structures for thaw reporting (ThawReport, ThawedMove,
CancelledMove, ConflictResolution, ConflictContender)
- Modified resolve_contested_transactions() to return conflict details
- Added _post_thaw_report() to post formatted report to admin channel
- Report shows thawed moves, cancelled moves, and conflict resolution
- Handles Discord's 2000 char limit with _send_long_report()
Tests:
- Updated test_views_transaction_embed.py for frozen flag behavior
- Added test for thaw period (freeze=False) scenario
- Updated test_tasks_transaction_freeze.py for new return values
- All tests passing
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add sWAR cap validation to TransactionBuilder.validate_transaction()
- Use team-specific salary_cap from Team.salary_cap field
- Fall back to config.swar_cap_limit (32.0) if team has no custom cap
- Add major_league_swar_cap field to RosterValidationResult
- Update major_league_swar_status to show ✅/❌ with cap limit
- Add 4 new tests for sWAR cap validation
This fixes a bug where IL moves could put a team over their sWAR cap
because the validation only checked roster counts, not sWAR limits.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Prevents players who are already claimed in another team's pending
transaction (frozen=false, cancelled=false) from being added to a new
transaction for the same week.
Changes:
- Add is_player_in_pending_transaction() to TransactionService
- Make TransactionBuilder.add_move() async with validation
- Add check_pending_transactions flag (default True for /dropadd)
- Skip validation for /ilmove and trades (check_pending_transactions=False)
- Add tests/conftest.py for proper test isolation
- Add 4 new tests for pending transaction validation
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
During the draft, teams draft 32 players and then drop to 26. The cap
calculation must account for remaining draft picks:
- max_zeroes = 32 - projected_roster_size (remaining draft picks)
- players_counted = 26 - max_zeroes (how many current players count)
This allows teams to draft expensive players mid-draft knowing they'll
drop cheap ones later. Previously the code was using min(roster_size, 26)
which didn't account for future picks, causing false cap violations.
Example: WAI with 18 players drafting 19th:
- Old (broken): players_counted = 19, sum all players
- New (fixed): max_zeroes = 13, players_counted = 13, only cheapest 13 count
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
The get_recent_picks service method was subtracting 1 from overall_end,
but callers were already passing currentpick - 1. This caused the "Last 5
picks" list on the OnTheClock embed to skip the most recently completed
pick (showing picks 2-6 before current instead of 1-5).
Removed the extra subtraction in the service method since callers already
handle the exclusivity of the overall_end parameter.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add paused field to DraftData model
- Add pause_draft() and resume_draft() methods to DraftService
- Add /draft-admin pause and /draft-admin resume commands
- Block picks in /draft command when draft is paused
- Skip auto-draft in draft_monitor when draft is paused
- Update status embeds to show paused state
- Add comprehensive tests for pause/resume
When paused:
- Timer is stopped (set to False)
- Deadline is set far in future
- All /draft picks are blocked
- Auto-draft monitor skips processing
When resumed:
- Timer is restarted with fresh deadline
- Picks are allowed again
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
DraftList.from_api_data() now properly calls Player.from_api_data()
for nested player objects, ensuring player.team_id is correctly
extracted from the nested team object. Also fixed validate_cap_space()
unpacking to accept all 3 return values.
Bug: Auto-draft was marking all players as "not available" because
player.team_id was None (None != 547 always True).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Add DraftSheetService with write_pick(), write_picks_batch(),
clear_picks_range(), and get_sheet_url() methods
- Integrate sheet writes in /draft command (fire-and-forget pattern)
- Integrate sheet writes in draft_monitor.py for auto-draft picks
- Add /draft-admin resync-sheet command for bulk recovery
- Add sheet link to /draft-status embed
- Add draft_sheet_keys config with env var overrides per season
- Add get_picks_with_players() to draft_pick_service for resync
- Add 13 unit tests for DraftSheetService (all passing)
- Update CLAUDE.md documentation files
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Teams can now make up missed draft picks when not on the clock:
- Added get_skipped_picks_for_team() to draft_pick_service
- /draft checks for skipped picks when user isn't on the clock
- Uses earliest (lowest overall) skipped pick first
- Shows footer noting "Making up skipped pick" on success
- Does NOT advance draft when using skipped pick
- Fixed duplicate embed when command run in ping channel
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Root Cause Fixes:
- Add _extract_items_and_count_from_response() override to DraftPickService
to handle API returning 'picks' key instead of 'draftpicks'
- Add custom from_api_data() to DraftPick model to handle API field mapping
(origowner/owner/player -> origowner_id/owner_id/player_id)
Enhancements:
- Add timer status to /draft-admin set-pick success message
- Shows relative deadline timestamp when timer active
- Shows "Timer Inactive" when timer not running
Also includes related draft module improvements from prior work.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Major fixes:
- Rename test_url_accessibility() to check_url_accessibility() in
commands/profile/images.py to prevent pytest from detecting it as a test
- Rewrite test_services_injury.py to use proper client mocking pattern
(mock service._client directly instead of HTTP responses)
- Fix Giphy API response structure in test_commands_soak.py
(data.images.original.url not data.url)
- Update season config from 12 to 13 across multiple test files
- Fix decorator mocking patterns in transaction/dropadd tests
- Skip integration tests that require deep decorator mocking
Test patterns applied:
- Use AsyncMock for service._client instead of aioresponses for service tests
- Mock at the service level rather than HTTP level for better isolation
- Use explicit call assertions instead of exact parameter matching
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Fix DraftPickService to send full model body on PATCH (API requirement)
- Fix DraftListService to use client-side sorting (API doesn't support sort param)
- Fix parameter names: round_start/end -> pick_round_start/end
- Add 53 tests covering DraftService, DraftPickService, DraftListService
- Add draft model tests for DraftData, DraftPick, DraftList
- Add OpenAPI spec URL to CLAUDE.md for API reference
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Update SOAK listener tests to match refactored simple string detection
(removed SOAK_PATTERN regex import, now uses ' soak' in text.lower())
- Fix DraftList model tests to provide nested Team/Player objects
(model requires full objects, not just IDs)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
