feat: add product expansion suite#302
Closed
InfinityZero3000 wants to merge 20 commits into
Closed
Conversation
Critical authorization bug: /login and /refresh never checked user.is_verified, so accounts with unconfirmed emails (including ones using addresses the registrant doesn't control) could log in and stay logged in indefinitely via refresh-token rotation. Google/ Facebook OAuth is unaffected since it already sets is_verified from the provider's own claim. Also fixes /resend-verification, which checked a nonexistent `email_verified` attribute and never actually fired. Flutter login page now detects the new 403 and routes the user to the existing email-verification-pending screen to resend the link instead of just showing a raw error. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…cal accounts Backfills is_verified=true for users created before commit 41613c4's login/refresh enforcement so existing accounts aren't locked out. Adds is_verified checks to admin login/OTP routes for consistency. Lets Google OAuth link to (and verify) an existing unverified local account when the email matches and Google confirms email_verified=true, instead of always rejecting with "Cannot change provider". Expands auth route test coverage to 56 tests covering Google/Facebook login, admin login/OTP, password change, and resend-verification. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Backend's admin/verification routes return AUTH_FORBIDDEN for inactive/unverified accounts, but the client had no mapping for it and fell through to a generic server-error message. Maps it alongside the other AUTH_* codes so the UI surfaces the real reason. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
story_repository.dart (the domain interface) and presentation widgets imported data/models/* directly, violating Clean Architecture's domain-never-imports-data rule. Adds Story/StoryListItem/TopicSession/ EducationalHints entities under domain/entities/, gives each data model a toEntity() mapper, and updates the repository implementation and all presentation call sites to depend on entities instead of models. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
… service lexi_chat.py mixed route handling with pipeline orchestration and SSE streaming, growing past 1400 lines. Moves the pipeline run, session store/idempotency helpers, and the SSE generator into a new lexi_chat_service.py; the route module now just parses requests, calls the service, and serializes the response. No behavior change — pipeline and streaming tests pass unmodified aside from updated monkeypatch targets. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…min.py main.py imported a 2152-line app/routes/admin.py whose own docstring called it deprecated, while the actual split files (admin_courses.py, admin_gamification.py, admin_system.py) sat unwired and unused — an abandoned refactor that left the "old" file as the only live one. Verified the split files are an exact route-for-route, AST-identical match (45 routes either way) before switching main.py to them and deleting admin.py. Also finishes the original refactor goal for the /admin/seed endpoint: extracts its ~430 lines of literal seed data into app/core/sample_data_catalog.py and the seeding loop into app/services/admin_seed_service.py, deep-copying the course catalog before the seeding loop mutates it in place so repeated seed runs stay idempotent. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…v2.py nodes_v2.py had grown to 2015 lines. Moves retrieve_node (budgeted hybrid retrieval + fusion scoring) into trace_cag/retrieve.py, and generate_node plus its build_generation_prompt/stream_llm_tokens helpers into trace_cag/generate.py, leaving nodes_v2.py at 885 lines. Both are re-exported from nodes_v2.py so existing imports (graph.py, lexi_chat_service.py, tests) keep working unchanged. Pure move — no logic changes; 114/114 trace_cag tests pass. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
_ProfilePageState had grown to 1772 lines covering header, quick actions, level progress, learning stats, weekly activity, badges, and the admin shortcut all in one file. Extracts each section into its own widget under presentation/widgets/profile_page/, dropping profile_page.dart from 2223 to 232 lines. flutter analyze clean, existing profile_page_test.dart (24 tests) passes unmodified. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
_HomePageNewState had grown to 2031 lines. Extracts header, streak card, level/daily-goal row, quick actions, enrolled/featured courses, and the skeleton loading state into widgets under presentation/widgets/home_page/, dropping home_page.dart from 2159 to 222 lines. Also deletes _buildQuickStats and _buildContinueSection — both already carried `// ignore: unused_element` and had zero callers anywhere in the codebase. flutter analyze clean. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…in, tracecag packaging - lexi_chat: thread native_language (ISO 639-1) through request, idempotency hash, and all orchestrator calls (primary/retry/stream) via iso_to_language_name - stt/transcript_normalizer: drop filler-only utterances and dedup leading repeated n-grams (Moonshine hallucination artifact) - groq_key_pool: round-robin fallback over all GROQ_API_KEYS when the Redis-backed pool isn't initialized (standalone CLIs no longer burn one key) - voice_session/metrics/sentence_finalizer/model_registry hardening - trace_cag: cache_utils/l1_state_cache/nodes_v2/retrieve/state refinements; benchmark qa_generation + ranking fixes - tests: native-language hint, benchmark model-strength, normalizer, lexi chat routes, tracecag chat integration - fix(test): sync dependency-constraint expected pins to merged Dependabot bumps (torch 2.12.1, moonshine-voice 0.0.62, sherpa-onnx 1.13.3) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
…ce hardening - shop_catalog: 10 in-game power-up consumables (time freeze, extra time, skip token, reveal/translate hint, shield, extra heart, lucky clover, score multiplier, pair swap) + GAME_POWERUP_ITEM_TYPES; alembic upsert migration - item_effects_service: route power-ups to _handle_instant_game_powerup (echo effects payload, no server state); reject cosmetics from the use endpoint; null-safe hint accounting - DailyChallengeService: idempotent claims via ChallengeRewardClaim unique index + single transaction (add_gems commit=False); guards double-claim - admin shop CRUD now accepts JSON body (ShopItemAdminCreate/Update) with effects/icon_url; list endpoint returns them too - services package: split achievement checker into its own module, lazy __getattr__ keeps imports light and backward compatible - schemas: migrate to Pydantic v2 (ConfigDict/field_validator) + guard test - quota_manager: UTC-keyed quotas, input validation (negative/zero cost, unknown API); content_agent_validation + artifact index; misc cleanup - remove dead check_missing.py / test_vocab_definitions.py - fix(test): admin RBAC create-shop test uses json body to match new contract Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
…up types - adminApi: create/update shop item send JSON body (was query params); add ShopItemEffects type, effects/icon_url fields - ShopPage: item-type list mirrors backend shop_catalog; quick-create templates prefill name/description/price/effects per type - i18n (en/vi) strings for the new shop fields Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
…remony - games: GamePowerupTray widget wired into fill-blank, grammar quiz, hangman, matching, spelling bee, word scramble; game_entities power-up support - lexi_chat: send native_language through datasource/repository/provider/page - gamification: shop_item/wallet entities + power-up provider plumbing; shop item card; league ceremony screen redesign - vocabulary: native-language datasource params; daily review / word-of-day card tweaks; daily challenges widget - tests: power-up provider/tray, wallet entity, shop item, lexi chat provider, vocabulary datasource, user model Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Dependabot bumped kombu to >=5.6.2 (incompatible with the pinned celery>=5.5.3,<5.6, which requires kombu<5.6), making 'pip install -r requirements.txt' fail with ResolutionImpossible in CI. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
…ficiency output) Prevent CI-generated reports from being accidentally committed. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Implemented quiz mode in DailyReviewCard with a modal bottom sheet for mode selection. - Added QuizOptionCard widget for displaying quiz options with different states. - Registered QuizProvider in vocabulary dependency injection setup. - Created unit tests for QuizProvider to ensure quiz functionality works as expected. - Updated various test files to include necessary dependencies and improve structure. - Refactored existing tests for better readability and maintainability.
|
The latest updates on your projects. Learn more about Vercel for GitHub. 2 Skipped Deployments
|
Owner
Author
|
Closed in favor of #303, which is rebuilt from origin/dev and excludes docs/markdown files per request. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Verification
Notes