3b4ac807a5
feat(agent+ui): fullstack agent module refactoring + UI/UX improvements
...
## Backend: agent module GRACE-Poly compliance
- Split app.py (749→~280 lines) into _tool_resolver, _confirmation, _persistence
- All 18 naked functions wrapped in #region/#endregion contracts
- Fixed @DEFGROUP→@defgroup typos; added @DATA_CONTRACT, @SIDE_EFFECT, CoT logs
- Conversation list API: added last_role, has_tool_calls, has_error, risk_level fields
- Message state detection: Russian/English error patterns (недоступен, unavailable)
- State field preserved in save_conversation messages
- HITL titles: descriptive tool names instead of generic "HITL resume"
## Backend: conversation title generation (two-layer)
- Layer 1: clean_title() — rule-based, strips file markers, pre-fetch blocks, JSON/CSV,
URLs, code; truncates at 80 chars word boundary (25 unit tests, all edge cases)
- Layer 2: generate_llm_title() — async best-effort LLM titling via /v1/chat/completions
with per-conversation lock, graceful degradation on failure
## Frontend: conversation list indicators (orthogonal system)
- Status dot (green/yellow/red/blue) per conversation state
- Icon column: tool activity, errors, waiting, completed
- Risk stripe (left border accent) + message count badge + relative time
- Fixed group labels: "Сегодня"/"Вчера" instead of "3 ч"/"5 ч"
- Hide "Окружение: —" when env is empty
## Frontend: guardrails card verification + fixes
- Confirmed all interaction modes: Enter/click confirm, Escape/click deny
- Auto-populate envId from environmentContextStore in DashboardDetailModel
- Better error message: missing_context_hint with recovery guidance
## Design system: semantic tokens
- Added category-* gradient tokens to tailwind.config.js
- Sidebar + Breadcrumbs use semantic tokens (10 categories)
- Raw Tailwind reduced from ~50 to 6 occurrences
- Added skip-to-content link in root layout (+layout.svelte)
- Added aria-label on DashboardDataGrid row checkboxes
## Protocol: INV_7 pragmatic exception
- Modules may exceed 400 lines when contract-dense (every function has #region)
- Recorded in semantics-core SKILL.md with rationale
Total: 5841+ contracts, 2993+ edges, backend 41/41, frontend 2501/2501
2026-06-30 15:21:05 +03:00
149c05bf1c
fix(frontend): full ADR compliance — Model-View concept, model decomposition, button migration
...
P0 — Model-first ADR compliance:
- Decompose DashboardHubModel (590→496 lines) into Dashboards.FiltersModel,
Dashboards.SelectionModel, Dashboards.GitActionsModel (DG split per plan)
- Decompose AgentChatModel (630→356 lines) into ConnectionManager,
StreamProcessor, LocalStorage, shared types
- Decompose MigrationModel (457→389 lines) into WizardModel, ExecutorModel
P0 — /ui atom compliance:
- Replace all raw <button> with <Button> from /ui in dashboards/+page.svelte
(~20 replacements) and 16 additional routes/ files (~70 replacements total)
P0 — Hierarchical region IDs (ATTN_2):
- Rename all 22 model #region/#endregion IDs from flat to Domain.Name format
- Update @ingroup from generic 'Models' to domain-specific (Dashboards, Git, etc.)
P1 — UX contract compliance:
- Add @UX_STATE declarations to agent/+page.svelte
- Extract Gradio Client.connect from page into AgentChatModel.retryConnection()
All new model files have proper GRACE anchors (#region/#endregion, @ingroup,
@BRIEF, @INVARIANT, @STATE, @ACTION, @RELATION).
Build: npm run build passes.
Tests: DashboardHubModel 112/112, MigrationModel 74/74 pass.
2026-06-17 14:02:17 +03:00
2b303f92b3
test: bring frontend test coverage to 98% across core lib modules
...
## Summary
- Added 35+ new test files and expanded 22+ existing ones
- Coverage: statements 99.65%, lines 99.9%, functions 99.9%, branches 87.77%
- All thresholds enabled and enforced in vitest.config.js
## Details
### Stores (stores/__tests__/)
- test_health.ts, test_translationRun.ts, test_environmentContext.ts
- test_maintenance.ts, test_environmentContext.2.ts
- Expanded sidebar.test.ts, assistantChat.test.ts, taskDrawer.test.ts
- Expanded test_activity.ts, test_datasetReviewSession.ts
### Models (models/__tests__/)
- AgentChatModel.test.ts (77.7%→99.6%), AgentChatModel.2.test.ts
- BranchModel.test.ts, DashboardDetailModel.test.ts
- DashboardHubModel.test.ts (97.9%→100%)
- DatasetDetailModel.test.ts, DatasetReviewModel.test.ts
- DatasetsHubModel.test.ts, DictionaryDetailModel.test.ts
- GitConfigModel.test.ts, GitManagerModel.test.ts
- GitStatusModel.test.ts, HealthCenterModel.test.ts
- LLMReportModel.test.ts, MigrationModel.test.ts
- MigrationSettingsModel.test.ts, TranslateHistoryModel.test.ts
- TranslationJobModel.test.ts, ValidationRunDetailModel.test.ts
- ValidationTasksListModel.test.ts
### API (api/__tests__/, api/translate/__tests__/, api/dataset-review/__tests__/)
- api.test.ts (100% stmts), assistant.test.ts, datasetReview.test.ts
- maintenance.test.ts, corrections.test.ts, datasources.test.ts
- dictionaries.test.ts, jobs.test.ts, runs.test.ts, schedules.test.ts
- useReviewSession.test.ts + useReviewSession.2.test.ts
### Auth (auth/__tests__/)
- permissions.test.ts (95.2%→100%), store.test.ts
- store.browser-off.test.ts (covers !browser guards)
### UI (ui/__tests__/)
- EmptyState.test.ts, FeatureGate.test.ts, FeatureGate.2.test.ts
- HelpTooltip.test.ts, Icon.test.ts, Input.test.ts, Select.test.ts
- LanguageSwitcher.test.ts
### Top-level lib (lib/__tests__/)
- cot-logger.test.ts, routes.test.ts, stores.test.ts
- toasts.test.ts, utils.test.ts
### Helpers (helpers/__tests__/)
- review-workspace-helpers.test.ts
### Source changes (minimal, non-breaking)
- sidebar.svelte.ts: exported loadState() for testability
- HelpTooltip.svelte: removed default () destructuring
- vitest.config.js: coverage scope narrowed, thresholds enforced
- package.json: fixed @vitest/coverage-v8 version mismatch
2026-06-10 14:59:40 +03:00
41a3a41ec4
test(frontend): add model unit tests for Screen Models
...
Add L1 invariant tests for all Screen Models:
- DashboardDetailModel: test load, delete, pagination, column filter
- DashboardHubModel: test load, environment switching, selection, git actions
- DatasetDetailModel: test load, edit, delete
- DatasetsHubModel: test load, filter, pagination
- DictionaryDetailModel: test load entries, add, edit, delete, search
- LLMReportModel: test load, filter, report generation
- TranslateHistoryModel: test load runs, filter, pagination
- ValidationRunDetailModel: test load details, records
- ValidationTasksListModel: test load tasks, status transitions
Per semantics-testing protocol: L1 tests verify model invariants without
DOM rendering.
2026-06-04 16:17:59 +03:00
