Compare commits

..

33 Commits

Author SHA1 Message Date
Junyan Qin
d568bbedc2 Add OSS and commercial workspace boundaries 2026-05-21 07:25:02 -04:00
Junyan Qin
d78a4fdea4 Document multi-tenant workspace architecture 2026-05-21 07:25:02 -04:00
Dongchuan Fu
894709d577 feat(qrcode-login): enhance WeChat login flow with expiration handlin… (#2212)
* feat(qrcode-login): enhance WeChat login flow with expiration handling and improved session management

* feat(qrcode-login): replace RefreshCw icon with RotateCw for loading state

* feat(qrcode-login): adjust session expiration handling and improve error status management
2026-05-21 14:28:02 +08:00
Rock Chin
6823069103 style(web): format AddModelPopover state initialization 2026-05-20 21:49:16 +08:00
Junyan Qin
699545a196 fix(web): fix models dialog provider type select and split add/scan popovers
1. Fix provider type select showing blank when editing: await
   loadRequesters() before loadProvider() to ensure options are
   populated before setting the selected value.

2. Split 'Add Model' into two separate entries: a '+ Add Model'
   button for manual add and a Radar icon button for scan. Each
   opens its own popover with only one layer of tabs (model type
   for manual, no tabs for scan since types are auto-detected).

3. Fix popover position: side='bottom' instead of 'left'.

4. Fix popover scroll: model type tabs stay fixed at top, content
   area scrolls independently when it overflows.

5. Scan mode now fetches all model types at once (no modelType
   filter), and routes each scanned model to the correct API
   based on its own type field.
2026-05-20 18:21:40 +08:00
Sebastion
f0061817ea fix: remove /debug/exec endpoint that allows authenticated RCE via exec() (#2178)
The /api/v1/system/debug/exec endpoint passes user-supplied HTTP body
directly to Python exec(), enabling arbitrary code execution for any
authenticated user when debug_mode is enabled. This is a critical
security risk (CWE-94): a single misconfiguration or compromised JWT
grants full server-side code execution.

Remove the endpoint entirely. The /debug/plugin/action endpoint (which
does not use exec()) is left intact as it serves a different, scoped
purpose.

Co-authored-by: Junyan Chin <rockchinq@gmail.com>
2026-05-19 00:53:39 +08:00
sheetung
688202e7d1 Merge pull request #2211 from sheetung/feat/aiocqhttp-json-msg
feat(aiocqhttp): handle json type messages in message converter
2026-05-18 15:35:49 +08:00
sheetung
d46b762d03 ci: trigger re-run 2026-05-18 07:32:49 +00:00
sheetung
0963fd5443 feat(aiocqhttp): unify json card message parsing with standard field extraction
Unify JSON card message parsing across mini-program, music, and article/video
types. Extract app, preview, title, and url fields using the standard QQ JSON
card structure (meta.detail_1 / music / news) instead of app-name hardcoding.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-18 07:22:14 +00:00
RockChinQ
6471770737 docs: add practical guide links to localized readmes 2026-05-18 13:18:27 +08:00
RockChinQ
314b7d15bb docs: link practical guides in readme 2026-05-18 13:16:07 +08:00
sheetung
c758908745 feat(aiocqhttp): handle json type messages in message converter
Add support for parsing OneBot JSON message segments (QQ mini-program,
Bilibili share cards, etc.) in the target2yiri converter. Parses the
card metadata and converts it to plain text to avoid silently dropping
these message types.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-18 04:58:48 +00:00
sheetung
767137aaa0 Merge pull request #2209 from sheetung/fix/sidebar-menu-cursor-webui
Fix/sidebar menu cursor webui
2026-05-16 23:34:33 +08:00
sheetung
acb2ce6a40 fix(webui): fix prettier formatting for span with className
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-16 15:21:50 +00:00
sheetung
67784708d6 fix(webui): add cursor-pointer and select-none to sidebar menu text spans
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-16 15:21:50 +00:00
Nody the lobster
1bd9c334aa fix: load persisted plugin config (#2208)
Co-authored-by: RockChinQ <rockchinq@gmail.com>
2026-05-16 15:51:45 +08:00
huanghuoguoguo
17bbc8bf10 Feat/test build (#2174)
* fix(ci): update unit-test workflow paths to match current source layout

Replace stale pkg/** filter with src/langbot/** and add uv.lock.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* docs(tests): update README to reflect current test layout

- Fix stale paths: tests/pipeline → tests/unit_tests/pipeline
- Update CI Python versions: 3.11, 3.12, 3.13
- Add test directory structure for box, config, platform, plugin, provider, storage
- Document pytest markers and uv commands
- Mention planned E2E tests

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add shared test factories package

Create tests/factories/ with reusable test factories:
- FakeApp: mock application with all dependencies
- Message chains: text_chain, mention_chain, image_chain
- Query factories: text_query, group_text_query, command_query, etc.

No test changes - maintains backward compatibility.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add fake provider factory

Add tests/factories/provider.py with:
- FakeProvider: deterministic fake LLM provider
- Error simulation: timeout, auth, rate-limit, malformed
- Request capture for assertions
- fake_model: mock model with attached provider

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add fake platform factory

Add tests/factories/platform.py with:
- FakePlatform: simulated platform adapter
- Inbound message construction: friend/group/image
- Mention-bot flag simulation
- Outbound message capture for assertions
- Streaming output support simulation
- Send failure simulation

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add comprehensive message/query factories

Extend tests/factories/message.py with:
- file_query: file attachment query
- unsupported_query: unknown message segment
- voice_query: audio/voice query
- at_all_query: group @All mention
- query_with_session: query with session object
- query_with_config: query with custom pipeline config

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add fake message flow smoke test

Create tests/smoke/test_fake_message_flow.py:
- TestFakeMessageFlow: factory verification tests
- TestMessageFlowIntegration: minimal flow smoke test
- Tests FakeApp, FakeProvider, FakePlatform, query factories
- Verifies LANGBOT_FAKE_PONG marker response
- Captures outbound messages for assertions

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add developer test-quick command

Add scripts/test-quick.sh and Makefile with:
- test-quick: runs ruff check + unit tests + smoke tests
- No real provider keys or platform accounts required
- Suitable for local branch self-test

Update tests/README.md:
- Document test-quick command
- Document test factories package
- Add smoke tests and factories directory structure

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* fix(test): make test-quick reliable as developer gate

Fixes for D-001验收问题:
1. test-quick.sh: use set -euo pipefail, uv run ruff, no tail pipe
2. Remove unused imports in factories (app.py, platform.py, provider.py)
3. Fix unused variable in smoke test
4. Add noqa: E402 to test_n8nsvapi.py lazy imports
5. Update smoke test docs: "minimal fake flow" not full pipeline

Now test-quick is a reliable gate: lint failures exit 1, test failures propagate.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(unit): add preproc and taskmgr unit tests

U-001: Pipeline Preprocessor tests
- Normal text message processing
- Empty message handling
- Image segment with/without vision model
- Model selection and fallback
- Variable extraction

U-004: Core Task Manager tests (pattern-based)
- Task creation and tracking patterns
- Task cancellation patterns
- Scope-based cancellation
- Task type filtering
- Pruning completed tasks
- Wait all tasks

Taskmgr tests use pattern-based approach to avoid circular import
in source code (taskmgr → app → http_controller → migration → taskmgr).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(unit): add config loader unit tests

U-005: Config Loader tests
- Valid YAML config loading
- Valid JSON config loading
- Invalid YAML/JSON error behavior
- Missing config file creation from template
- Template completion for missing keys
- ConfigManager load/dump operations
- Exists check for both YAML and JSON

All tests use tmp_path fixture, no real project config.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(unit): add chat and command handler pattern tests

U-002: Chat Handler tests (pattern-based)
- Normal message event emission pattern
- prevent_default handling
- User message alteration pattern
- Runner selection pattern
- Streaming/non-streaming response patterns
- Exception handling modes (show-error, show-hint, hide)
- Message history update pattern
- Telemetry payload pattern

U-003: Command Handler tests (pattern-based)
- Command parsing and text extraction
- Event creation pattern
- Privilege/admin check pattern
- Command result handling (text, error, image)
- prevent_default handling
- String truncation helper

Uses pattern-based testing to avoid circular import issues in source code.
Direct imports of handler modules trigger circular import chain.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* style: fix unused imports after ruff auto-fix

Remove unused imports in test files:
- test_config_loader.py: remove unused os
- test_taskmgr.py: remove unused Mock
- test_preproc.py: remove unused unsupported_query, image_chain

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(unit): improve taskmgr tests to test real classes

U-004 improved: Tests now import and test actual classes:
- TaskContext: new(), trace(), to_dict(), placeholder()
- TaskWrapper: task creation, context, exception/result capture, cancel, to_dict
- AsyncTaskManager: create_task, create_user_task, cancel_task, cancel_by_scope
- Task pruning behavior

Uses pre-mocking technique:
- Mock langbot.pkg.core.app before import (breaks circular chain)
- Mock langbot.pkg.core.entities with proper Enum

All 24 tests now test real class behavior, not patterns.
taskmgr.py coverage should improve significantly.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* refactor(test): consolidate FakeApp and add sys.modules isolation utility

- Extract tests/utils/import_isolation.py with isolated_sys_modules context manager
- Extend tests/factories/app.py FakeApp with handler-specific attributes
- Refactor test_chat_handler.py to use centralized FakeApp and cached imports
- Refactor test_command_handler.py with mock_execute_factory fixture
- Refactor test_smoke.py to move import-time sys.modules manipulation into fixture
- Add SQLite migration integration tests (G-002)
- Add HTTP API smoke integration tests (G-005)
- Update CI workflow to call pytest for SQLite migrations (G-004)

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add developer quality gate consolidation (G-007)

- Add scripts/test-integration-fast.sh for fast integration tests
- Add scripts/test-coverage.sh with 12% baseline threshold
- Update Makefile with test-integration-fast, test-coverage, test-all-local
- Update CI workflow with integration and coverage jobs
- Add smoke marker to pytest.ini
- Update tests/README.md with quality gate layers documentation
- Add tests/integration/pipeline/ for pipeline stage-chain tests

Quality gate layers:
- Quick: ruff + unit + smoke (~2 min)
- Fast Integration: SQLite/API/Pipeline (~3 min)
- Coverage: 12% threshold gate (~8 min)
- Full Local: all three combined

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): add PostgreSQL migration slow integration tests (G-003)

- Add tests/integration/persistence/test_migrations_postgres.py
- All tests marked with @pytest.mark.slow
- Tests skip when TEST_POSTGRES_URL is not set (no local PostgreSQL)
- Database isolation via clean_tables and clean_alembic_version fixtures
- Update CI workflow to use pytest instead of inline Python script
- Remove TODO(G-003) comment
- Update tests/README.md with PostgreSQL test documentation

Covered scenarios:
- Baseline stamp sets revision
- Upgrade from baseline to head
- Upgrade idempotent
- Get current on unstamped DB returns None

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* feat(test): Phase 1.5 coverage expansion - COV-001 to COV-013

Coverage baseline raised from 13.65% to 26% (+12.35%)
Gate raised from 12% to 18%

Tasks completed:
- COV-001: Command system unit tests (100% coverage)
- COV-002: API service unit tests batch 1 (user/apikey/model/provider)
- COV-003: Provider model manager unit tests
- COV-004: Pipeline remaining stage tests (aggregator/cntfilter/longtext/msgtrun)
- COV-005: Storage and utils coverage pass
- COV-006: Gate ratchet 12%→15%
- COV-007: Gate ratchet 15%→18%
- COV-008: API service batch 2 (bot/pipeline/webhook/space/maintenance/mcp)
- COV-009: Blocked - API controller circular import issue documented
- COV-010: Plugin runtime unit tests (+0.08%)
- COV-011: RAG and vector unit tests (+0.68%)
- COV-012: Core boot and migration unit tests
- COV-013: Provider requester logic unit tests (+0.62%)

Key additions:
- tests/utils/import_isolation.py: sys.modules isolation for circular imports
- Provider requester mock tests: proved HTTP-dependent code can be tested locally
- Vector filter utilities: 100% coverage on pure functions
- API services: fake persistence pattern for unit testing

Blocked issue COV-009 documented in langbot-test-plan/1.5/issues/

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(phase1): add unit tests for telemetry, plugin, rag, persistence

Add initial unit tests for Phase 1 of test coverage improvement:
- telemetry: test initialization, payload sanitization, early returns (14.3% → 62.9%)
- plugin: test _parse_plugin_id static method
- rag: test _to_i18n_name static method
- persistence: test serialize_model with datetime handling

Overall core coverage: 41.9% → 42.2%

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(phase2): add unit tests for core, persistence, plugin, utils

- Add test_handler_helpers.py for plugin handler helpers (7 tests)
- Add test_mgr_methods.py for persistence manager (5 tests)
- Add test_app_config_validation.py for core app config (12 tests)
- Add test_knowledge_service.py for API knowledge service (22 tests)
- Add test_kbmgr.py for RAG knowledge base manager (39 tests)
- Add test_survey_manager.py for survey manager (22 tests)
- Add test_connector_methods.py for plugin connector (24 tests)
- Add test_funcschema.py for utils function schema (9 tests)
- Add test_platform.py for utils platform detection (7 tests)
- Add test_extract_deps.py for plugin deps extraction (7 tests)
- Add test_database_decorator.py for persistence decorator (7 tests)
- Add test_load_config.py for core config loading (19 tests)
- Add COVERAGE_EXCLUSIONS.md documenting external adapter exclusions
- Fix test_chat_session_limit.py path for portability

Coverage: core 28% → 30%, persistence 24% → 24.4%, plugin 27% → 28%
Total: 1082 tests passed, core module coverage 45.5%

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(integration): add API controller integration tests

- Add test_pipelines.py (10 tests) covering pipelines CRUD operations
  - GET/POST/PUT/DELETE on /api/v1/pipelines
  - Extensions endpoint
  - Metadata endpoint
  - Coverage: pipelines controller 27% → 80%

- Add test_providers.py (10 tests) covering provider/model management
  - Provider CRUD with model counts
  - LLM model CRUD
  - Coverage: providers controller 23% → 81%, models 29% → 45%

Tests use Quart TestClient with mocked services for real HTTP behavior
without external dependencies.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(integration): add knowledge, bots, and model endpoints tests

- Add test_knowledge.py (10 tests) covering knowledge base management
  - CRUD operations on /api/v1/knowledge/bases
  - Files management endpoints
  - Retrieve endpoint with validation
  - Coverage: knowledge/base.py 26% → 91%

- Add test_bots.py (9 tests) covering bot management
  - CRUD operations on /api/v1/platform/bots
  - Logs endpoint
  - Send message endpoint with validation
  - Coverage: platform/bots.py 24% → 87%

- Extend test_providers.py (+4 tests) for embedding/rerank models
  - Embedding models CRUD
  - Rerank models CRUD
  - Coverage: provider/models.py 29% → 60%

Total integration tests: 53 (smoke 12 + pipelines 10 + providers 14 + knowledge 10 + bots 9)

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(integration): add embed and monitoring endpoint tests

Add integration tests for embed widget and monitoring API endpoints:
- test_embed.py: 15 tests for widget.js, logo, turnstile, messages, reset, feedback
- test_monitoring.py: 15 tests for overview, messages, llm-calls, sessions, errors, export

Coverage improvements:
- embed.py: 17% → 56%
- monitoring.py: 17% → 93%

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(e2e): add minimal startup E2E tests

Add E2E tests for LangBot startup flow:
- tests/e2e/utils/config_factory.py: minimal config generation
- tests/e2e/utils/process_manager.py: LangBot subprocess management
- tests/e2e/conftest.py: E2E fixtures (session-scoped process)
- tests/e2e/test_startup.py: 12 tests for startup verification

Tests verify:
- boot.py + stages execution
- database initialization (SQLite)
- API availability
- migrations applied

Uses embedded databases (SQLite, Chroma) - no external dependencies.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test(quality): fix fake tests and add missing coverage

P0 fixes:
- telemetry: rewrite fake tests with real behavior verification (25 tests)
- config: delete copied-source tests, use proper imports (2 deleted)
- persistence: fix try-except pass to verify specific errors

P1 fixes:
- pipeline: add real FixedWindowAlgo tests instead of mocks (12 tests)
- provider: add SessionManager and ToolManager tests (25 tests)
- storage: add S3StorageProvider tests with moto mock (16 tests)
- plugin: add handler action tests for setting inheritance (15 tests)
- rag: add file storage and ZIP processing tests (21 tests)
- vector: add VDB filter conversion tests (30 tests)

P2 fixes:
- pipeline/msgtrun: strengthen assertions for exact message count
- api: add response structure validation in integration tests

New test files:
- provider/test_session_manager.py
- provider/test_tool_manager.py
- storage/test_s3storage.py
- plugin/test_handler_actions.py
- rag/test_file_storage.py
- vector/test_vdb_filter_conversion.py

Source code bugs documented:
- provider: TokenManager.next_token() ZeroDivisionError
- telemetry: send_tasks class variable shared state
- command: empty command IndexError, unused parameters
- utils: funcschema KeyError
- entity: vector.py independent declarative_base

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* docs(test): update coverage stats and test structure

- Update coverage from 22% to 30%
- Add new test files to structure:
  - provider: session_manager, tool_manager
  - storage: s3storage
  - plugin: handler_actions
  - rag: file_storage
  - vector: vdb_filter_conversion
  - telemetry: rewritten tests
- Update module coverage percentages

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

* test: add 105 new unit tests for untested core functionality

Add comprehensive tests for B-class issues (core functionality untested):

Pipeline:
- test_pool.py: QueryPool ID generation, caching, async context (12 tests)
- test_ratelimit.py: Fixed timing-sensitive test tolerance
- test_pipelinemgr.py: Use real Pydantic StageProcessResult instead of Mock

Utils:
- test_version.py: Version comparison functions (20 tests)
- test_logcache.py: Log page management and retrieval (18 tests)
- test_httpclient.py: HTTP session pool management (10 tests)
- test_proxy.py: Proxy configuration from env and config (10 tests)
- test_image.py: URL parsing and base64 extraction (12 tests)
- test_pkgmgr.py: Pip command generation (8 tests)

Discover:
- test_engine.py: I18nString, Metadata, Component manifest (15 tests)

Test count: 1193 → 1298 (+105 tests)

Note: Some B-class issues cannot be tested due to circular import bugs
filed as GitHub issues #2175 (pipeline) and #2176 (persistence).

* test: tighten phase 1 coverage contracts

* test: align ci integration isolation

---------

Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-16 12:05:54 +08:00
huanghuoguoguo
4a4c0921a4 fix(plugin): use specific runtime not connected error (#2199) 2026-05-16 11:36:27 +08:00
huanghuoguoguo
e425cf079a fix(pipeline): return query from QueryPool.add_query (#2198) 2026-05-16 11:36:10 +08:00
huanghuoguoguo
245e798b79 fix(pipeline): handle empty longtext response chain (#2197) 2026-05-16 11:35:20 +08:00
huanghuoguoguo
27fdccce16 fix(pipeline): preserve routed flag when aggregating (#2196) 2026-05-16 11:35:00 +08:00
huanghuoguoguo
484643c0ee fix(api): validate api key prefix (#2195) 2026-05-16 11:33:20 +08:00
huanghuoguoguo
ec61459619 fix(api): avoid mutating bot update payload (#2194) 2026-05-16 11:31:59 +08:00
huanghuoguoguo
66ef744447 fix(rag): reject unsafe runtime file paths (#2193) 2026-05-16 11:31:00 +08:00
huanghuoguoguo
10d3a9cc92 fix(api): avoid mutating pipeline update payload (#2192) 2026-05-16 11:30:32 +08:00
huanghuoguoguo
885320e9ae fix(utils): preserve QQ image URL scheme (#2188) 2026-05-16 11:29:31 +08:00
huanghuoguoguo
ed02ac4710 fix(utils): classify runner URLs safely (#2191)
* fix(utils): classify runner URLs safely

* fix(utils): keep runner parse failures unknown
2026-05-16 11:28:34 +08:00
huanghuoguoguo
e4841edbaf fix pkgmgr install requirements default (#2190) 2026-05-16 11:26:49 +08:00
huanghuoguoguo
ef7a06b0db fix telemetry send task isolation (#2187) 2026-05-16 11:26:23 +08:00
huanghuoguoguo
6fe20c1812 fix(core): handle sigint before app startup (#2189) 2026-05-16 11:24:34 +08:00
huanghuoguoguo
9e8c8f79df fix(plugin): validate plugin id format (#2185) 2026-05-16 11:21:58 +08:00
huanghuoguoguo
01d06898fb fix(provider): ignore empty token rotation (#2184) 2026-05-16 11:21:09 +08:00
huanghuoguoguo
0a669c7016 fix(utils): handle missing funcschema parameter docs (#2186) 2026-05-16 11:20:32 +08:00
39 changed files with 1779 additions and 610 deletions

View File

@@ -47,6 +47,8 @@ LangBot is an **open-source, production-grade platform** for building AI-powered
[→ Learn more about all features](https://link.langbot.app/en/docs/features)
📍 Practical guides: [deploy a multi-platform AI bot in 5 minutes](https://blog.langbot.app/en/blog/deploy-ai-bot-in-5-minutes/), [connect DeepSeek to WeChat, Discord, and Telegram](https://blog.langbot.app/en/blog/connect-deepseek-to-wechat/), [run a Dify Agent in Discord, Telegram, and Slack](https://blog.langbot.app/en/blog/dify-agent-discord-telegram-slack/), and [build an n8n-powered chatbot](https://blog.langbot.app/en/blog/n8n-multi-platform-ai-chatbot/).
---
## Quick Start

View File

@@ -47,6 +47,8 @@ LangBot 是一个**开源的生产级平台**,用于构建 AI 驱动的即时
[→ 了解更多功能特性](https://link.langbot.app/zh/docs/features)
📍 实践指南:[5 分钟部署多平台 AI 机器人](https://blog.langbot.app/zh/blog/deploy-ai-bot-in-5-minutes/)、[将 DeepSeek 接入微信、企业微信与 Discord](https://blog.langbot.app/zh/blog/connect-deepseek-to-wechat/)、[让 Dify Agent 跑在 Discord、Telegram 和 Slack 上](https://blog.langbot.app/zh/blog/dify-agent-discord-telegram-slack/),以及[用 n8n 构建多平台 AI 聊天机器人](https://blog.langbot.app/zh/blog/n8n-multi-platform-ai-chatbot/)。
---
## 快速开始

View File

@@ -46,6 +46,8 @@ LangBot es una **plataforma de código abierto y grado de producción** para con
[→ Conocer más sobre todas las funcionalidades](https://link.langbot.app/en/docs/features)
📍 Guías prácticas: [desplegar un bot de IA multiplataforma en 5 minutos](https://blog.langbot.app/en/blog/deploy-ai-bot-in-5-minutes/), [conectar DeepSeek a WeChat, Discord y Telegram](https://blog.langbot.app/en/blog/connect-deepseek-to-wechat/), [ejecutar un Dify Agent en Discord, Telegram y Slack](https://blog.langbot.app/en/blog/dify-agent-discord-telegram-slack/) y [crear un chatbot con n8n](https://blog.langbot.app/en/blog/n8n-multi-platform-ai-chatbot/).
---
## Inicio Rápido

View File

@@ -46,6 +46,8 @@ LangBot est une **plateforme open-source de niveau production** pour créer des
[→ En savoir plus sur toutes les fonctionnalités](https://link.langbot.app/en/docs/features)
📍 Guides pratiques : [déployer un bot IA multiplateforme en 5 minutes](https://blog.langbot.app/en/blog/deploy-ai-bot-in-5-minutes/), [connecter DeepSeek à WeChat, Discord et Telegram](https://blog.langbot.app/en/blog/connect-deepseek-to-wechat/), [exécuter un Dify Agent dans Discord, Telegram et Slack](https://blog.langbot.app/en/blog/dify-agent-discord-telegram-slack/) et [créer un chatbot avec n8n](https://blog.langbot.app/en/blog/n8n-multi-platform-ai-chatbot/).
---
## Démarrage Rapide

View File

@@ -46,6 +46,8 @@ LangBot は、AI搭載のインスタントメッセージングボットを構
[→ すべての機能について詳しく見る](https://link.langbot.app/ja/docs/features)
📍 実践ガイド: [5分でマルチプラットフォームAIボットをデプロイ](https://blog.langbot.app/en/blog/deploy-ai-bot-in-5-minutes/)、[DeepSeekをWeChat・Discord・Telegramに接続](https://blog.langbot.app/en/blog/connect-deepseek-to-wechat/)、[Dify AgentをDiscord・Telegram・Slackで動かす](https://blog.langbot.app/en/blog/dify-agent-discord-telegram-slack/)、[n8n連携チャットボットを構築](https://blog.langbot.app/en/blog/n8n-multi-platform-ai-chatbot/)。
---
## クイックスタート

View File

@@ -46,6 +46,8 @@ LangBot은 AI 기반 인스턴트 메시징 봇을 구축하기 위한 **오픈
[→ 모든 기능 자세히 보기](https://link.langbot.app/en/docs/features)
📍 실전 가이드: [5분 만에 멀티 플랫폼 AI 봇 배포하기](https://blog.langbot.app/en/blog/deploy-ai-bot-in-5-minutes/), [DeepSeek를 WeChat, Discord, Telegram에 연결하기](https://blog.langbot.app/en/blog/connect-deepseek-to-wechat/), [Dify Agent를 Discord, Telegram, Slack에서 실행하기](https://blog.langbot.app/en/blog/dify-agent-discord-telegram-slack/), [n8n 기반 챗봇 만들기](https://blog.langbot.app/en/blog/n8n-multi-platform-ai-chatbot/).
---
## 빠른 시작

View File

@@ -46,6 +46,8 @@ LangBot — это **платформа с открытым исходным к
[→ Подробнее обо всех возможностях](https://link.langbot.app/en/docs/features)
📍 Практические руководства: [развернуть мультиплатформенного ИИ-бота за 5 минут](https://blog.langbot.app/en/blog/deploy-ai-bot-in-5-minutes/), [подключить DeepSeek к WeChat, Discord и Telegram](https://blog.langbot.app/en/blog/connect-deepseek-to-wechat/), [запустить Dify Agent в Discord, Telegram и Slack](https://blog.langbot.app/en/blog/dify-agent-discord-telegram-slack/) и [создать чат-бота на n8n](https://blog.langbot.app/en/blog/n8n-multi-platform-ai-chatbot/).
---
## Быстрый старт

View File

@@ -48,6 +48,8 @@ LangBot 是一個**開源的生產級平台**,用於建構 AI 驅動的即時
[→ 了解更多功能特性](https://link.langbot.app/zh/docs/features)
📍 實踐指南:[5 分鐘部署多平台 AI 機器人](https://blog.langbot.app/zh/blog/deploy-ai-bot-in-5-minutes/)、[將 DeepSeek 接入微信、企業微信與 Discord](https://blog.langbot.app/zh/blog/connect-deepseek-to-wechat/)、[讓 Dify Agent 跑在 Discord、Telegram 和 Slack 上](https://blog.langbot.app/zh/blog/dify-agent-discord-telegram-slack/),以及[用 n8n 建構多平台 AI 聊天機器人](https://blog.langbot.app/zh/blog/n8n-multi-platform-ai-chatbot/)。
---
## 快速開始

View File

@@ -46,6 +46,8 @@ LangBot là một **nền tảng mã nguồn mở, cấp sản xuất** để x
[→ Tìm hiểu thêm về tất cả tính năng](https://link.langbot.app/en/docs/features)
📍 Hướng dẫn thực hành: [triển khai bot AI đa nền tảng trong 5 phút](https://blog.langbot.app/en/blog/deploy-ai-bot-in-5-minutes/), [kết nối DeepSeek với WeChat, Discord và Telegram](https://blog.langbot.app/en/blog/connect-deepseek-to-wechat/), [chạy Dify Agent trên Discord, Telegram và Slack](https://blog.langbot.app/en/blog/dify-agent-discord-telegram-slack/) và [xây dựng chatbot với n8n](https://blog.langbot.app/en/blog/n8n-multi-platform-ai-chatbot/).
---
## Bắt đầu nhanh

View File

@@ -0,0 +1,858 @@
# LangBot 多租户与多用户改造方案
## 目标
本方案面向 LangBot 从“单实例单管理员”演进到 SaaS 友好的“多 workspace、多账户、多权限”架构。
核心定义:
- Account登录主体。一个自然人或服务账号可加入多个 workspace。
- Workspace租户边界。一个 workspace 内可拥有多个用户、机器人、流水线、模型、知识库、扩展、监控数据与 API Key。
- Membership账户与 workspace 的关系,承载角色与权限。
- Role/Permissionworkspace 内权限,不再用“是否是当前唯一用户”来决定访问能力。
目标体验:
- 新用户登录后可以创建 workspace、加入 workspace、切换 workspace。
- 同一个账户可加入多个 workspace每个 workspace 权限不同。
- 一个 workspace 可邀请多个用户协作,并分别设置 owner/admin/editor/viewer 等权限。
- 所有业务资源默认属于某个 workspace所有 API 默认在当前 workspace 下工作。
- Plugin SDK、MCP、知识库、模型调用、监控日志都能拿到稳定的 workspace 上下文,并且不跨租户泄露数据。
## 调研结论
### 当前 LangBot 的单用户假设
LangBot 现在已经有 `users` 表和 JWT 登录,但仍是单用户/单租户模型:
- `src/langbot/pkg/entity/persistence/user.py``User` 只保存 `user/password/account_type/space_*`没有角色、状态、workspace 关系。
- `src/langbot/pkg/api/http/service/user.py` 通过 `is_initialized()` 判断系统是否已有用户;`create_or_update_space_user()` 在系统已初始化且邮箱不匹配时拒绝新用户,这直接限制了多用户登录。
- `src/langbot/pkg/api/http/controller/group.py``AuthType.USER_TOKEN` 验证后只向 handler 注入 `user_email`JWT payload 也只有 `user`,没有 `account_id``workspace_id``role``permissions`
- `src/langbot/pkg/api/http/service/apikey.py` 的 API Key 只验证 key 是否存在,没有 owner、scope、workspace。
- `web/src/app/infra/http/BaseHttpClient.ts``localStorage.token` 读取单个 token并在所有请求上加 `Authorization`;前端没有 workspace selector也没有当前 workspace 上下文。
当前登录流程更像“初始化一个本地管理账号”,而不是 SaaS 账户体系。要支持多用户,必须把“初始化状态”和“首个 workspace 创建”拆开。
### 业务资源当前都是全局资源
主要持久化表没有租户字段:
- Bot`bots`
- Pipeline`legacy_pipelines``pipeline_run_records`
- Model`model_providers``llm_models``embedding_models``rerank_models`
- Plugin`plugin_settings`
- MCP`mcp_servers`
- RAG`knowledge_bases``knowledge_base_files``knowledge_base_chunks`
- Monitoring`monitoring_messages``monitoring_llm_calls``monitoring_sessions``monitoring_errors``monitoring_embedding_calls``monitoring_feedback`
- API Key`api_keys`
- Webhook`webhooks`
- Metadata`metadata`
- Binary storage`binary_storages`
对应服务也直接 select 全表,例如:
- `BotService.get_bots()` 返回所有 bot。
- `PipelineService.get_pipelines()` 返回所有 pipeline。
- `ModelProviderService.get_providers()` 返回所有 provider。
- `MCPService.get_mcp_servers()` 返回所有 MCP server。
- 插件和二进制存储没有 workspace 维度,插件 workspace storage 在 SDK 里还硬编码为 `default`
所以改造重点不是只给用户表加字段,而是给资源访问层统一加入 workspace scope。
### 运行时也存在全局单例假设
`src/langbot/pkg/core/stages/build_app.py` 初始化的是一个全局 `Application`,其中包含单例:
- `platform_mgr`
- `pipeline_mgr`
- `model_mgr`
- `tool_mgr`
- `plugin_connector`
- `sess_mgr`
- `rag_mgr`
- `vector_db_mgr`
当前运行时把所有 bot、pipeline、model、plugin、MCP 都加载到同一套内存管理器。多租户改造需要决定:是共享运行时并在对象上带 workspace 过滤,还是每个 workspace 拆 runtime shard。第一阶段建议共享进程、强制 workspace-aware等规模变大后再演进为按 workspace 分片。
### Plugin SDK 对 workspace 的假设
SDK 当前只认识 bot/pipeline/query/session不认识租户
- `src/langbot_plugin/api/entities/builtin/pipeline/query.py``Query``query_id/launcher_type/launcher_id/sender_id/bot_uuid/pipeline_uuid`,没有 `workspace_id/account_id`
- `src/langbot_plugin/api/entities/builtin/provider/session.py``Session` 只按 `launcher_type + launcher_id` 表达会话。
- `src/langbot_plugin/api/proxies/langbot_api.py` 暴露 `get_bots/get_llm_models/invoke_llm/list_tools/vector_*` 等 Host API都是全局语义。
- `src/langbot_plugin/runtime/io/handlers/plugin.py``set_workspace_storage/get_workspace_storage``owner_type` 设为 `workspace`,但 `owner` 固定为 `default`
- LangBot 侧 `src/langbot/pkg/plugin/handler.py` 处理插件请求时,会把 `GET_BOTS``GET_LLM_MODELS``VECTOR_*` 等转到全局服务。
这意味着多租户落地时,不能只在 Web API 层过滤;插件可以通过 Host API 访问全局资源,所以 SDK/Runtime 通信也必须传递 workspace context。
## 开源版与商业版产品边界
LangBot 是开源软件,但多 workspace 能力本质上是 SaaS 控制面能力。如果把完整多 workspace、成员协作、订阅权益和配额代码都放进开源仓库只靠本地 feature flag 或本地 license check无法有效避免第三方 fork 后自建 SaaS。因此建议采用 open-core 架构:开源版保留单 workspace 执行能力,账户、订阅、权益和多 workspace 协作能力放到 LangBot Space/Cloud Control Plane 和商业模块中。
### 版本边界
推荐拆成三层:
- `LangBot Core OSS`:开源,自托管,默认只有一个隐式 `default workspace`。它可以在数据结构上兼容 workspace但产品能力上不提供创建多个 workspace、切换 workspace、成员邀请、成员权限管理、审计和多租户配额。
- `LangBot Space / Cloud Control Plane`:托管控制面,负责 account、workspace、membership、subscription、billing、entitlement、license token、workspace quota、marketplace 权益等能力。
- `LangBot Commercial Module`:商业闭源或私有包,承载多 workspace、团队协作、RBAC、自定义角色、审计日志、SAML/SSO、企业私有化授权等能力。
企业私有化版本可以采用 `LangBot Core + Commercial Module + License Token` 的形式交付。开源 Core 仍然可独立运行,但只能作为单 workspace 自托管产品,不提供 SaaS 多租户控制面。
### OSS 中如何保留兼容但不开放多 workspace
为了让后续商业版复用同一套资源隔离模型OSS 代码里可以保留 `workspace_uuid` 相关字段和默认 workspace 迁移,但应限制为单 workspace
- 首次初始化时创建一个 `Default Workspace`
- 所有资源自动归属这个 default workspace。
- 不暴露 `POST /api/v1/workspaces`
- 不暴露 workspace switcher。
- 不暴露成员邀请和成员角色管理。
- 不支持一个 account 加入多个 workspace。
- 不支持 workspace 数量大于 1。
- 前端不展示 workspace selector。
- API 层如果收到非 default workspace 的 `X-Workspace-Id`,直接拒绝。
也就是说OSS 可以是 workspace-aware但不是 multi-workspace-enabled。这样做的价值是代码结构提前适配租户隔离未来商业版不用重写所有资源模型同时开源版用户无法直接通过 UI/API 获得 SaaS 型多租户能力。
### 账户、订阅和权益抽到 Space
账户和订阅体系建议从 LangBot Core 中抽出,交给 Space 控制面:
```text
LangBot Space
-> Account
-> Workspace
-> Membership
-> Subscription
-> Entitlement
-> License Token
LangBot Core
-> Validate entitlement / license
-> Run bots, pipelines, plugins, MCP, RAG
-> Enforce local resource scope
-> Report usage
```
这样做有几个原因:
- 账号体系如果完全在本地,第三方容易直接改库创建 workspace/membership。
- 订阅、配额和商业权益如果完全在本地,容易绕过。
- Space 可以统一处理 OAuth、组织、邀请、付款、发票、套餐、权益、Marketplace 分发权限。
- LangBot Core 只作为执行面消费 Space 下发的 entitlement减少商业规则暴露。
### Entitlement 设计
Space 向 LangBot Core 下发签名权益,可以是在线校验,也可以为企业版提供短期/长期离线 license token。
示例:
```json
{
"edition": "oss",
"workspace_limit": 1,
"member_limit": 1,
"multi_workspace": false,
"rbac": false,
"audit_log": false,
"custom_roles": false,
"sso": false,
"commercial_use": false,
"expires_at": 1893456000
}
```
OSS 默认权益:
- `workspace_limit = 1`
- `member_limit = 1`
- `multi_workspace = false`
- `rbac = false`
- `audit_log = false`
- `sso = false`
Cloud/Pro/Enterprise 权益:
- `workspace_limit > 1`
- `member_limit > 1`
- `multi_workspace = true`
- `rbac = true`
- 可按套餐打开 audit、custom roles、SSO、usage reporting、enterprise support 等能力。
Core 执行面需要在关键入口强制校验 entitlement
- 创建 workspace。
- 邀请成员。
- 修改成员角色。
- 切换 workspace。
- 创建超过 quota 的资源。
- 开启商业模块功能。
### 商业模块边界
以下能力不建议进入 OSS 仓库的完整实现:
- 多 workspace 创建和切换。
- Workspace 成员邀请。
- Workspace RBAC 和自定义角色。
- Workspace 审计日志。
- Workspace 级用量和配额管理。
- 订阅、账单、发票。
- 企业 SSO/SAML/OIDC。
- 在线/离线 license 管理。
- 多租户 SaaS 运营控制台。
OSS 仓库可以保留接口占位、默认 workspace 兼容和必要的数据隔离字段,但完整交互、管理 UI、权益校验器和多 workspace policy engine 应由 Space 或商业模块提供。
### 防自建 SaaS 的现实边界
技术上无法 100% 阻止别人 fork 开源代码后自行改造。更可靠的策略是组合:
- 不把完整商业多 workspace 实现放进 OSS。
- Space 控制面提供账号、订阅、权益、Marketplace 和官方托管能力。
- 商业模块闭源或私有分发。
- 使用商标、云服务条款和商业 license 限制“自称官方 LangBot SaaS”或未经授权商用托管。
- 如果当前开源 license 对托管商用限制不足,需要单独评估 license 策略,必要时引入 open-core license 或新增商业附加条款。具体 license 调整需要法律评审。
结论:多 workspace 的底层 schema 可以在 OSS 中以 default workspace 兼容方式铺路,但多 workspace 产品能力、账户订阅权益、协作管理和 SaaS 控制面应放到 Space/商业模块,不作为开源版可直接使用的能力。
## 推荐总体架构
采用“单实例多 workspace资源行级隔离运行时上下文隔离”的架构
```mermaid
flowchart TD
A["Account"] --> B["WorkspaceMembership"]
B --> C["Workspace"]
C --> D["Bots"]
C --> E["Pipelines"]
C --> F["Models & Providers"]
C --> G["Knowledge Bases"]
C --> H["Extensions: Plugins / MCP"]
C --> I["API Keys & Webhooks"]
C --> J["Monitoring"]
D --> K["Runtime Query"]
E --> K
K --> L["Plugin Runtime"]
K --> M["MCP Runtime"]
L --> N["Workspace-scoped Host APIs"]
```
原则:
- 账户全局唯一workspace 是所有业务资源的归属边界。
- 所有 HTTP handler 在进入业务服务前解析出 `RequestContext(account, workspace, membership, permissions)`
- 所有 service 方法显式接收 `ctx``workspace_id`,禁止在业务服务里无条件 select 全表。
- 运行时对象的 key 从 `uuid` 扩展为 `(workspace_id, uuid)` 或使用全局唯一 uuid 但必须记录 workspace_id 并校验。
- 插件/MCP/知识库/模型调用都按 query 所属 workspace 过滤可用资源。
## 数据模型设计
### Account
替代现有 `users` 的语义,建议保留表名但升级字段,避免过大迁移:
字段建议:
- `id`
- `uuid`
- `email`
- `password_hash`
- `display_name`
- `avatar_url`
- `account_type`: `local | space`
- `status`: `active | disabled | deleted`
- `space_account_uuid`
- `space_access_token`
- `space_refresh_token`
- `space_access_token_expires_at`
- `space_api_key`
- `created_at`
- `updated_at`
兼容策略:
- 旧字段 `user` 迁移为 `email`,可以短期保留 alias。
-`password` 迁移为 `password_hash`也可先保持列名不变service 层改命名。
- JWT 中不要继续只放 email应放 `sub=account_uuid`
### Workspace
新增 `workspaces`
- `uuid`
- `name`
- `slug`
- `avatar_url`
- `type`: `personal | team`
- `status`: `active | suspended | deleted`
- `default_language`
- `created_by_account_uuid`
- `created_at`
- `updated_at`
每个账户首次登录时自动创建一个 personal workspace。旧单用户实例迁移时创建一个 `Default Workspace`
### WorkspaceMembership
新增 `workspace_memberships`
- `workspace_uuid`
- `account_uuid`
- `role`: `owner | admin | developer | operator | viewer`
- `status`: `active | invited | disabled`
- `invited_by_account_uuid`
- `joined_at`
- `created_at`
- `updated_at`
唯一索引:
- `(workspace_uuid, account_uuid)`
### WorkspaceInvitation
新增 `workspace_invitations`
- `uuid`
- `workspace_uuid`
- `email`
- `role`
- `token_hash`
- `expires_at`
- `accepted_at`
- `created_by_account_uuid`
- `created_at`
用于邀请外部用户加入 workspace。Space OAuth 登录时可以根据 email 自动匹配未接受邀请。
### Role 与 Permission
先用固定角色,后续再做自定义角色。
建议权限:
- `workspace.manage`
- `member.view`
- `member.invite`
- `member.update_role`
- `member.remove`
- `bot.view`
- `bot.manage`
- `pipeline.view`
- `pipeline.manage`
- `model.view`
- `model.manage`
- `knowledge.view`
- `knowledge.manage`
- `extension.view`
- `extension.manage`
- `monitoring.view`
- `apikey.manage`
- `webhook.manage`
- `billing.view`
- `billing.manage`
角色映射:
| Role | 说明 | 权限 |
| --- | --- | --- |
| owner | workspace 拥有者 | 全部权限;可转让 owner不可被其他角色移除 |
| admin | 管理员 | 除 owner 转让和删除 workspace 外的全部权限 |
| developer | 构建者 | 管理 bot、pipeline、model、knowledge、extension、webhook可看监控 |
| operator | 运营者 | 查看和启停 bot、查看监控、查看配置不可改密钥和删除资源 |
| viewer | 只读成员 | 只读资源和监控 |
### 业务资源加 workspace_uuid
以下表需要新增 `workspace_uuid`
- `bots`
- `legacy_pipelines`
- `pipeline_run_records`
- `model_providers`
- `llm_models`
- `embedding_models`
- `rerank_models`
- `plugin_settings`
- `mcp_servers`
- `knowledge_bases`
- `knowledge_base_files`
- `knowledge_base_chunks`
- `monitoring_*`
- `api_keys`
- `webhooks`
- `binary_storages`
- `metadata`
索引建议:
- 所有资源表加 `(workspace_uuid, created_at)``(workspace_uuid, updated_at)`
- 资源唯一键从单列改为 workspace 复合唯一:
- `bots.uuid` 可保持全局唯一,但查询仍必须带 workspace。
- `plugin_settings` 主键从 `(plugin_author, plugin_name)` 改为 `(workspace_uuid, plugin_author, plugin_name)`
- `mcp_servers.name` 如果未来要求唯一,必须是 `(workspace_uuid, name)`
- `metadata.key` 改为 `(workspace_uuid, key)`,系统级 metadata 单独放 `system_metadata` 或使用 `workspace_uuid=NULL`
- `binary_storages.unique_key` 建议改为 `workspace_uuid + owner_type + owner + key` 的 hash。
### API Key
API Key 必须归属于 workspace
- `workspace_uuid`
- `created_by_account_uuid`
- `scopes`
- `expires_at`
- `last_used_at`
- `status`
验证 API Key 后生成 `RequestContext`
- `account_uuid=None` 或 service account uuid
- `workspace_uuid=key.workspace_uuid`
- `permissions=key.scopes`
这样 `/api/v1/platform/bots/<uuid>/send_message` 之类接口不会跨 workspace 操作 bot。
## 后端改造方案
### RequestContext
新增统一上下文对象,例如:
```python
class RequestContext:
account_uuid: str | None
workspace_uuid: str
role: str | None
permissions: set[str]
auth_type: Literal["user_token", "api_key"]
```
改造 `RouterGroup.route()`
- `USER_TOKEN`:验证 JWT读取 `account_uuid`,再从 header/query/cookie 中解析 current workspace。
- `API_KEY`:验证 API Key直接得到 workspace。
- `USER_TOKEN_OR_API_KEY`:两者都返回同一种 `RequestContext`
- handler 参数从可选 `user_email` 升级为可选 `ctx`;兼容期同时支持 `user_email`
当前 workspace 传递方式:
- 推荐 header`X-Workspace-Id: <workspace_uuid>`
- Web 前端同时把当前 workspace 存在 localStorage。
- 如果未传,后端用账户最近使用 workspace 或第一个 active membership。
JWT payload
```json
{
"sub": "account_uuid",
"email": "user@example.com",
"iss": "LangBot-...",
"exp": 1234567890
}
```
不要把 workspace 写死在 JWT 里,否则切换 workspace 需要刷新 token。可以额外支持短 TTL workspace token但第一阶段不必。
### 服务层改造模式
所有 service 方法增加 `ctx``workspace_uuid`
```python
async def get_bots(self, ctx: RequestContext, include_secret: bool = True):
require(ctx, "bot.view")
query = sqlalchemy.select(Bot).where(Bot.workspace_uuid == ctx.workspace_uuid)
```
需要改造的服务:
- `UserService`:拆成 AccountService + WorkspaceService 更清晰。
- `ApiKeyService`:按 workspace 管理 key。
- `BotService`:所有 bot 查询/创建/更新/删除按 workspace。
- `PipelineService`:所有 pipeline 查询/默认 pipeline 按 workspace。
- `ModelProviderService` 和 model services按 workspace 隔离 provider 和 model。
- `MCPService`:按 workspace 管理 MCP server运行时按 workspace host。
- `KnowledgeService/RAGRuntimeService`:按 workspace 管理 KB、文件、collection。
- `MonitoringService`:记录和查询都带 workspace。
- `WebhookService`:按 workspace 管理 webhook。
- `PluginRuntimeConnector`:插件安装、设置、配置按 workspace。
### HTTP API 形态
保留现有路径,靠 `X-Workspace-Id` 表示当前 workspace可减少前端和 SDK 破坏:
- `GET /api/v1/workspaces`
- `POST /api/v1/workspaces`
- `GET /api/v1/workspaces/current`
- `PUT /api/v1/workspaces/current`
- `GET /api/v1/workspaces/<workspace_uuid>/members`
- `POST /api/v1/workspaces/<workspace_uuid>/invitations`
- `PUT /api/v1/workspaces/<workspace_uuid>/members/<account_uuid>`
- `DELETE /api/v1/workspaces/<workspace_uuid>/members/<account_uuid>`
现有资源 API
- `/api/v1/platform/bots`
- `/api/v1/pipelines`
- `/api/v1/provider/*`
- `/api/v1/plugins`
- `/api/v1/mcp`
- `/api/v1/knowledge`
继续保留,但必须从 `RequestContext.workspace_uuid` 过滤。
对外 API Key 也使用相同路径,只是由 key 决定 workspace。
### 初始化流程
现有 `/api/v1/user/init` 含义改为“创建首个账号和首个 workspace”
1. 如果系统没有任何 account
- 创建 account。
- 创建 personal/team workspace。
- 创建 owner membership。
- 创建默认 pipeline。
- 标记 wizard status 到该 workspace metadata。
2. 如果系统已有 account
- 禁止无邀请注册,除非配置允许公开注册。
- Space OAuth 登录后,如果没有 membership引导创建 workspace 或接受邀请。
`/api/v1/user/account-info` 不应再只返回 first user应返回
- `initialized`
- `registration_mode`
- `space_enabled`
- `default_login_methods`
登录成功后前端调用 `/api/v1/workspaces` 选择 workspace。
### 运行时隔离
第一阶段采用共享进程 + workspace-aware runtime
- `RuntimeBot` 增加 `workspace_uuid`
- `RuntimePipeline` 增加 `workspace_uuid`
- `Query` 增加 `workspace_uuid`,从 bot/pipeline 派生。
- `SessionManager.get_session()` key 从 `(launcher_type, launcher_id)` 改为 `(workspace_uuid, bot_uuid, launcher_type, launcher_id)`
- `PipelineManager.pipeline_dict` key 可保持 pipeline uuid但所有 load/get 都校验 workspace如果 uuid 不是全局唯一则改为 `(workspace_uuid, pipeline_uuid)`
- `ModelManager` provider/model 加 workspace 过滤;`get_model_by_uuid` 必须确保 query workspace 可访问。
- `ToolManager` 中 MCP tools、plugin tools 按 query workspace 过滤。
后续规模化时可演进:
- workspace runtime shard每个 workspace 一套 plugin runtime/MCP runtime。
- 大客户独立进程或独立数据库。
## Plugin SDK 与 Runtime 改造
### Query/Event 增加 workspace context
SDK `Query` 增加:
- `workspace_uuid: str`
- `workspace_slug: str | None`
- `account_uuid: str | None`,仅 Web/API 触发时可能有,聊天平台消息通常为空。
Event 模型通过 `event.query.workspace_uuid` 可拿到租户上下文;序列化时也应包含这些字段。
向后兼容:
- 字段可选,默认 `None`
- 老插件不感知这些字段也能跑。
- 新插件可通过 `ctx.event.query.workspace_uuid` 或新增 `ctx.get_workspace()` 访问。
### Host API 默认按当前 workspace 限制
`LangBotAPIProxy` 的以下方法必须由 Host 端按 workspace 过滤:
- `get_bots`
- `get_bot_info`
- `send_message`
- `get_llm_models`
- `invoke_llm`
- `list_plugins_manifest`
- `list_commands`
- `list_tools`
- `call_tool`
- `invoke_embedding`
- `vector_*`
- `list_knowledge_bases`
- `retrieve_knowledge`
建议新增显式方法:
- `get_workspace_info()`
- `get_current_account()`
- `get_workspace_storage(...)`
但不要让插件传入任意 workspace id 来越权访问。插件请求的 workspace 应由 Runtime 根据当前 query/plugin connection 填充。
### Workspace storage 修复
当前 SDK runtime 中:
```python
data["owner_type"] = "workspace"
data["owner"] = "default"
```
必须改为:
- 如果请求来自 query/eventowner 为 `workspace_uuid`
- 如果请求来自后台插件任务owner 为 plugin 安装所属 workspace。
- Host 侧 `binary_storages``workspace_uuid`,并在 unique key 中包含 workspace。
Plugin storage 建议也同时加 workspace
- 现在 plugin storage owner 是 `author/name`,这会导致同一插件在不同 workspace 的私有数据冲突。
- 应改为 `(workspace_uuid, plugin_id, key)`
### 插件安装与配置
`plugin_settings` 从全局变为 workspace-scoped
- 同一个插件可安装到多个 workspace。
- 每个 workspace 有自己的 enabled、priority、config、install_source、install_info。
- 插件 runtime 列表需要能按 workspace 过滤。
实现路线有两种:
1. 共享插件进程,插件代码只加载一份,设置和调用时附带 workspace。
2. 每个 workspace 一个插件容器实例,隔离最彻底但资源占用更高。
推荐第一阶段采用方案 1但要求
- 所有 RuntimeToLangBot/PluginToRuntime action 都能携带 `workspace_uuid`
- 插件 config 获取时按 workspace 返回。
- 插件 page API 请求必须校验当前用户在该 workspace 有访问权限。
### MCP
MCP server 是租户资源:
- `mcp_servers.workspace_uuid`
- MCP session key 从 `server_name` 改为 `(workspace_uuid, server_name)` 或使用全局 uuid。
- Pipeline extension preferences 中绑定 MCP server uuid 时,只能绑定同 workspace 的 server。
- MCP tool 列表在 query 执行时按 query.workspace_uuid + pipeline 绑定关系过滤。
## 前端改造
### Workspace selector
Home layout 顶部或 sidebar 增加 workspace selector
- 当前 workspace 名称和头像。
- 切换 workspace 后写入 `localStorage.currentWorkspaceId`
- 所有请求自动带 `X-Workspace-Id`
- 切换后刷新 sidebar 数据和页面缓存。
`BaseHttpClient` request interceptor 增加:
```ts
const workspaceId = localStorage.getItem("currentWorkspaceId");
if (workspaceId) config.headers["X-Workspace-Id"] = workspaceId;
```
### 用户与成员管理页面
新增页面:
- `/home/workspace/settings`
- `/home/workspace/members`
- `/home/workspace/invitations`
能力:
- owner/admin 邀请成员。
- owner/admin 修改成员角色。
- owner 移除成员、转让 owner。
- 所有人可切换 workspace。
- viewer/operator 在 UI 上隐藏不可操作按钮,但后端仍做权限校验。
### 登录与注册
登录后流程:
1. `authUser` 拿 token。
2. `initializeUserInfo()` 获取 account info。
3. `GET /api/v1/workspaces`
4. 如果没有 workspace进入创建 workspace 向导。
5. 如果有多个 workspace默认进入最近使用 workspace可切换。
注册页不再表达“初始化管理员账号”,而是:
- 首次系统启动:创建首个 owner + default workspace。
- 后续:根据配置允许公开注册,或只能接受邀请。
### 旧页面影响
需要逐个检查这些页面的数据加载是否都依赖当前 workspace
- Bots
- Pipelines
- Plugins/Market/MCP
- Knowledge
- Monitoring
- Models dialog
- API integration dialog
- Wizard
## 迁移方案
### 迁移阶段 0准备
- 引入 `workspaces``workspace_memberships``workspace_invitations`
-`users` 增加 `uuid/status/display_name` 等字段。
- 创建 `RequestContext`,但先不强制所有服务改完。
### 迁移阶段 1默认 workspace
对现有实例执行迁移:
1. 创建 `Default Workspace`
2. 找到现有第一个 user设为 owner。
3. 所有已有资源写入 `workspace_uuid=default_workspace_uuid`
4. `metadata` 迁入 default workspace确实全局的配置放到 `system_metadata`
5. `binary_storages``owner_type=workspace, owner=default` 改为 owner 为 default workspace uuid。
6. 插件 `plugin_settings` 归入 default workspace。
### 迁移阶段 2服务层强制 scope
- 改所有 service 查询,必须要求 `workspace_uuid`
- API Key 迁移为 workspace key。
- 所有写操作必须检查权限。
- 监控和任务查询按 workspace 过滤。
### 迁移阶段 3运行时上下文
- `Query``Session``RuntimeBot``RuntimePipeline` 增加 workspace。
- Plugin/MCP/Model/RAG runtime 全部按 workspace 过滤。
- 修复 SDK workspace storage。
### 迁移阶段 4前端多 workspace
- 登录后 workspace 选择。
- Header/sidebar workspace switcher。
- 成员和邀请管理。
- 所有 API 请求带 `X-Workspace-Id`
### 迁移阶段 5安全收敛
- 添加跨 workspace 越权测试。
- 添加 API Key scope 测试。
- 添加插件 Host API 过滤测试。
- 添加 MCP 和 RAG 隔离测试。
## 安全边界
必须防的场景:
- 用户 A 修改 URL 中 bot uuid访问用户 B workspace 的 bot。
- API Key 来自 workspace A但调用 workspace B 的 bot。
- 插件通过 `get_bots()` 枚举所有 workspace 的 bot。
- 插件通过 `workspace_storage` 读取其它 workspace 的数据。
- MCP server 名称相同导致 session 复用。
- monitoring session_id 相同导致数据串租户。
- Space OAuth 登录时,同 email 账户被错误绑定到已有本地 account。
建议策略:
- 所有资源访问都使用 `workspace_uuid + resource_id`
- 所有 service 方法入口做权限检查。
- 插件 Host API 的 workspace 不信任插件入参,只信任 query/runtime connection 上下文。
- API Key 只授予最小 scope默认不允许成员管理。
- owner 角色不能被普通 admin 移除或降权。
## 实施优先级
### P0基础租户骨架
- Account uuid/status。
- Workspace / Membership / Invitation。
- RequestContext。
- JWT 改为 account uuid。
- 前端 current workspace header。
### P1资源行级隔离
- Bots、Pipelines、Models、MCP、Plugins、Knowledge、Monitoring、API Keys 全部加 workspace_uuid。
- service 查询统一加 workspace filter。
- 权限矩阵落地。
### P2运行时隔离
- Query、Session、RuntimeBot、RuntimePipeline 加 workspace。
- Plugin Host API 和 MCP tools 按 workspace 过滤。
- SDK workspace storage 从 `default` 改为真实 workspace。
### P3协作体验
- 邀请成员。
- 成员列表和角色管理。
- workspace switcher。
- 最近使用 workspace。
### P4SaaS 运维增强
- Workspace 级用量统计。
- Workspace 级限额max_bots/max_pipelines/max_extensions/tokens/storage。
- 审计日志。
- workspace suspend/delete。
- 可选自定义角色。
## 测试计划
后端测试:
- 账户可加入多个 workspace。
- 同账户在不同 workspace 权限不同。
- viewer 不能创建/修改资源。
- API Key 只能访问所属 workspace。
- 所有资源 list/get/update/delete 都不能跨 workspace。
- 默认 workspace 迁移后旧数据可用。
运行时测试:
- 两个 workspace 使用相同 `launcher_id` 不共享 session。
- 两个 workspace 使用相同 MCP server name 不共享 MCP session。
- 插件 `get_bots()` 只能看到当前 workspace bot。
- 插件 `workspace_storage` 在不同 workspace 读写隔离。
- Pipeline 只调用当前 workspace 绑定的插件和 MCP tools。
前端测试:
- 登录后自动进入最近 workspace。
- 切换 workspace 后列表数据变化。
- 无权限按钮隐藏,直接调用 API 也被后端拒绝。
- 邀请成员流程完整。
迁移测试:
- SQLite 老实例迁移。
- PostgreSQL 老实例迁移。
- 已有 local account 迁移为 default workspace owner。
- 已有 Space account token 和 Space model provider API key 不丢失。
## 关键实现注意事项
- 不建议在第一版做数据库 schema-per-tenant。LangBot 当前 ORM 和运行时均以单库单表为主,先做 shared schema + workspace_uuid 成本更低。
- 不建议每个 workspace 立即启动独立 plugin runtime。先共享 runtime强制 action 带 workspace大客户隔离可作为后续部署形态。
- 不要只在前端过滤 workspace。插件、API Key、MCP、RAG 都能绕过前端,必须在后端和运行时层过滤。
- `metadata` 要拆清楚wizard status 属于 workspace系统版本/迁移信息属于 system。
- `users.user` 用 email 当主键语义不稳,应尽快引入 `account_uuid` 并让 JWT 以 uuid 为准。
- `plugin_settings` 当前主键没有 workspace改造时要先改主键/唯一约束,否则同插件无法在多个 workspace 配不同配置。
## 建议落地顺序
1. 新增 workspace/account/membership 表和 RequestContext。
2. 迁移旧数据到 default workspace。
3. 改 auth 和前端请求头,让每个请求都有 current workspace。
4. 从最核心资源开始逐个加 scopebot -> pipeline -> provider/model -> plugin/MCP -> knowledge -> monitoring。
5. 改 SDK Query/Event 和 runtime storage。
6. 上成员管理 UI 和邀请。
7. 做越权测试和迁移测试。
这个顺序的好处是可以较早让主 UI 在一个 workspace 下继续工作,同时把最危险的跨租户泄露面逐步收紧。

View File

@@ -179,8 +179,6 @@ class AdaptersRouterGroup(group.RouterGroup):
"""Start WeChat QR code login. Returns session_id + QR code data URL."""
import uuid
import time
import io
import base64
from langbot.libs.openclaw_weixin_api.client import OpenClawWeixinClient, DEFAULT_BASE_URL
@@ -208,60 +206,32 @@ class AdaptersRouterGroup(group.RouterGroup):
async def run_login():
try:
import qrcode as qr_lib
for _attempt in range(3):
qr_resp = await client.fetch_qrcode()
if not qr_resp.qrcode or not qr_resp.qrcode_img_content:
raise Exception('Failed to get QR code from server')
# Generate QR code image locally
qr = qr_lib.QRCode(error_correction=qr_lib.constants.ERROR_CORRECT_L)
qr.add_data(qr_resp.qrcode_img_content)
qr.make(fit=True)
img = qr.make_image(fill_color='black', back_color='white')
buf = io.BytesIO()
img.save(buf, format='PNG')
b64 = base64.b64encode(buf.getvalue()).decode('utf-8')
data_url = f'data:image/png;base64,{b64}'
def _update_qr():
session['qr_data_url'] = data_url
session['expire_at'] = time.time() + 480 # 8 minutes
def on_qrcode(qr_data_url: str, _qr_url: str):
def _update():
session['qr_data_url'] = qr_data_url
session['expire_at'] = time.time() + 180
session['status'] = 'waiting'
loop.call_soon_threadsafe(_update_qr)
# Poll for scan status
deadline = loop.time() + 180
while loop.time() < deadline:
try:
status_resp = await client.poll_qrcode_status(qr_resp.qrcode)
except Exception:
await asyncio.sleep(2)
continue
if status_resp.status == 'confirmed' and status_resp.bot_token:
session['status'] = 'success'
session['token'] = status_resp.bot_token
session['base_url'] = status_resp.baseurl or client.base_url
session['account_id'] = status_resp.ilink_bot_id or ''
return
if status_resp.status == 'expired':
break # retry with new QR code
await asyncio.sleep(1)
else:
pass # timeout, retry
# All retries exhausted
session['status'] = 'error'
session['error'] = 'QR code login failed: max retries exceeded'
loop.call_soon_threadsafe(_update)
result = await client.login(
max_retries=1,
poll_timeout_ms=180_000,
on_qrcode=on_qrcode,
)
session['status'] = 'success'
session['token'] = result.token
session['base_url'] = result.base_url
session['account_id'] = result.account_id
except Exception as e:
session['status'] = 'error'
session['error'] = str(e)
error_message = str(e)
if 'expired' in error_message.lower() or 'max retries exceeded' in error_message.lower():
session['status'] = 'expired'
session['error'] = 'QR code expired'
else:
session['status'] = 'error'
session['error'] = error_message
finally:
await client.close()
@@ -295,7 +265,11 @@ class AdaptersRouterGroup(group.RouterGroup):
if not session:
return self.http_status(404, -1, 'Session not found')
data = {'status': session['status']}
data = {
'status': session['status'],
'qr_data_url': session['qr_data_url'],
'expire_at': session['expire_at'],
}
if session['status'] == 'success':
data['token'] = session['token']
@@ -305,6 +279,9 @@ class AdaptersRouterGroup(group.RouterGroup):
elif session['status'] == 'error':
data['error'] = session['error']
_weixin_login_sessions.pop(session_id, None)
elif session['status'] == 'expired':
data['error'] = session['error']
_weixin_login_sessions.pop(session_id, None)
return self.success(data=data)

View File

@@ -7,8 +7,10 @@ import httpx
import uuid
import os
import posixpath
import sqlalchemy
from .....core import taskmgr
from .....entity.persistence import plugin as persistence_plugin
from .. import group
from langbot_plugin.runtime.plugin.mgr import PluginInstallSource
@@ -148,7 +150,15 @@ class PluginsRouterGroup(group.RouterGroup):
return self.http_status(404, -1, 'plugin not found')
if quart.request.method == 'GET':
return self.success(data={'config': plugin['plugin_config']})
result = await self.ap.persistence_mgr.execute_async(
sqlalchemy.select(persistence_plugin.PluginSetting.config)
.where(persistence_plugin.PluginSetting.plugin_author == author)
.where(persistence_plugin.PluginSetting.plugin_name == plugin_name)
)
persisted_config = result.scalar_one_or_none()
config = persisted_config if persisted_config is not None else plugin['plugin_config']
return self.success(data={'config': config})
elif quart.request.method == 'PUT':
data = await quart.request.json

View File

@@ -140,17 +140,6 @@ class SystemRouterGroup(group.RouterGroup):
async def _() -> str:
return self.success(data=await self.ap.maintenance_service.get_storage_analysis())
@self.route('/debug/exec', methods=['POST'], auth_type=group.AuthType.USER_TOKEN)
async def _() -> str:
if not constants.debug_mode:
return self.http_status(403, 403, 'Forbidden')
py_code = await quart.request.data
ap = self.ap
return self.success(data=exec(py_code, {'ap': ap}))
@self.route(
'/debug/plugin/action',
methods=['POST'],

View File

@@ -3,6 +3,7 @@ import typing
import asyncio
import traceback
import datetime
import json
import aiocqhttp
import pydantic
@@ -293,6 +294,29 @@ class AiocqhttpMessageConverter(abstract_platform_adapter.AbstractMessageConvert
elif msg.type == 'dice':
face_id = msg.data['result']
yiri_msg_list.append(platform_message.Face(face_type='dice', face_id=int(face_id), face_name='骰子'))
elif msg.type == 'json':
try:
raw = msg.data.get('data', {})
if isinstance(raw, str):
raw = json.loads(raw)
if isinstance(raw, dict):
_meta = raw.get('meta', {}) or {}
if isinstance(_meta, dict):
_detail = _meta.get('detail_1') or _meta.get('music') or _meta.get('news') or {}
else:
_detail = {}
if isinstance(_detail, dict):
preview = _detail.get('preview', '')
title = _detail.get('desc', '') or _detail.get('title', '')
url = _detail.get('qqdocurl', '') or _detail.get('jumpUrl', '')
else:
preview = title = url = ''
text = ' '.join([f'[{raw.get("app", "")}]', preview, title, url]).strip()
yiri_msg_list.append(platform_message.Plain(text=text or '[收到一张JSON卡片]'))
else:
yiri_msg_list.append(platform_message.Plain(text=str(raw)))
except Exception:
yiri_msg_list.append(platform_message.Plain(text='[收到一张JSON卡片]'))
chain = platform_message.MessageChain(yiri_msg_list)

View File

@@ -0,0 +1,28 @@
from __future__ import annotations
import pytest
def dedupe_preregistered_groups() -> None:
"""Keep API integration route registration isolated across test modules."""
from langbot.pkg.api.http.controller import group
seen: set[tuple[str, str]] = set()
unique_groups = []
for group_cls in group.preregistered_groups:
key = (group_cls.name, group_cls.path)
if key in seen:
continue
seen.add(key)
unique_groups.append(group_cls)
group.preregistered_groups[:] = unique_groups
@pytest.fixture(scope='module')
def http_controller_cls(mock_circular_import_chain):
"""Import HTTPController under each module's circular-import isolation."""
from langbot.pkg.api.http.controller.main import HTTPController
dedupe_preregistered_groups()
return HTTPController

View File

@@ -102,11 +102,9 @@ def fake_bot_app():
@pytest.fixture(scope='module')
async def quart_test_client(fake_bot_app):
async def quart_test_client(fake_bot_app, http_controller_cls):
"""Create Quart test client (module scope to avoid route re-registration)."""
from langbot.pkg.api.http.controller.main import HTTPController
controller = HTTPController(fake_bot_app)
controller = http_controller_cls(fake_bot_app)
await controller.initialize()
client = controller.quart_app.test_client()

View File

@@ -101,11 +101,9 @@ def fake_embed_app():
@pytest.fixture(scope='module')
async def quart_test_client(fake_embed_app):
async def quart_test_client(fake_embed_app, http_controller_cls):
"""Create Quart test client (module scope)."""
from langbot.pkg.api.http.controller.main import HTTPController
controller = HTTPController(fake_embed_app)
controller = http_controller_cls(fake_embed_app)
await controller.initialize()
client = controller.quart_app.test_client()
@@ -299,4 +297,4 @@ class TestEmbedFeedbackEndpoint:
json={'message_id': 'msg-123', 'feedback_type': 99}
)
assert response.status_code == 400
assert response.status_code == 400

View File

@@ -107,11 +107,9 @@ def fake_knowledge_app():
@pytest.fixture(scope='module')
async def quart_test_client(fake_knowledge_app):
async def quart_test_client(fake_knowledge_app, http_controller_cls):
"""Create Quart test client (module scope to avoid route re-registration)."""
from langbot.pkg.api.http.controller.main import HTTPController
controller = HTTPController(fake_knowledge_app)
controller = http_controller_cls(fake_knowledge_app)
await controller.initialize()
client = controller.quart_app.test_client()

View File

@@ -113,11 +113,9 @@ def fake_monitoring_app():
@pytest.fixture(scope='module')
async def quart_test_client(fake_monitoring_app):
async def quart_test_client(fake_monitoring_app, http_controller_cls):
"""Create Quart test client (module scope)."""
from langbot.pkg.api.http.controller.main import HTTPController
controller = HTTPController(fake_monitoring_app)
controller = http_controller_cls(fake_monitoring_app)
await controller.initialize()
client = controller.quart_app.test_client()
@@ -329,4 +327,4 @@ class TestMonitoringExportEndpoint:
headers={'Authorization': 'Bearer test_token'}
)
assert response.status_code == 200
assert response.status_code == 200

View File

@@ -119,11 +119,9 @@ def fake_pipeline_app():
@pytest.fixture(scope='module')
async def quart_test_client(fake_pipeline_app):
async def quart_test_client(fake_pipeline_app, http_controller_cls):
"""Create Quart test client (module scope to avoid route re-registration)."""
from langbot.pkg.api.http.controller.main import HTTPController
controller = HTTPController(fake_pipeline_app)
controller = http_controller_cls(fake_pipeline_app)
await controller.initialize()
client = controller.quart_app.test_client()

View File

@@ -116,11 +116,9 @@ def fake_provider_app():
@pytest.fixture(scope='module')
async def quart_test_client(fake_provider_app):
async def quart_test_client(fake_provider_app, http_controller_cls):
"""Create Quart test client (module scope to avoid route re-registration)."""
from langbot.pkg.api.http.controller.main import HTTPController
controller = HTTPController(fake_provider_app)
controller = http_controller_cls(fake_provider_app)
await controller.initialize()
client = controller.quart_app.test_client()

View File

@@ -119,15 +119,13 @@ def fake_api_app():
# ============== QUART TEST CLIENT FIXTURE ==============
@pytest.fixture
async def quart_test_client(fake_api_app):
async def quart_test_client(fake_api_app, http_controller_cls):
"""
Create Quart test client with real HTTPController and route registration.
Requires mock_circular_import_chain fixture to run first (usefixtures).
"""
from langbot.pkg.api.http.controller.main import HTTPController
controller = HTTPController(fake_api_app)
controller = http_controller_cls(fake_api_app)
await controller.initialize()
client = controller.quart_app.test_client()
@@ -344,4 +342,4 @@ class TestRealImports:
break
assert user_group is not None
assert user_group.path == '/api/v1/user'
assert user_group.path == '/api/v1/user'

View File

@@ -0,0 +1,66 @@
"""
PoC test for CWE-94: Authenticated RCE via exec() on user-supplied Python code.
The /api/v1/system/debug/exec endpoint passes raw HTTP body to exec(),
allowing arbitrary code execution when debug_mode is True.
This test verifies that:
1. The exec() endpoint is removed from the codebase entirely.
2. No route matches /api/v1/system/debug/exec.
"""
import ast
import pathlib
# Resolve project root (one level up from tests/)
_PROJECT_ROOT = pathlib.Path(__file__).resolve().parent.parent
VULN_FILE = (
_PROJECT_ROOT
/ "src"
/ "langbot"
/ "pkg"
/ "api"
/ "http"
/ "controller"
/ "groups"
/ "system.py"
)
def test_no_exec_call_in_system_controller():
"""Verify there is no exec() call in system.py that takes user input."""
with open(VULN_FILE, "r") as f:
source = f.read()
tree = ast.parse(source)
exec_calls = []
for node in ast.walk(tree):
if isinstance(node, ast.Call):
func = node.func
# Match bare exec() call
if isinstance(func, ast.Name) and func.id == "exec":
exec_calls.append(node.lineno)
assert len(exec_calls) == 0, (
f"Found exec() call(s) at line(s) {exec_calls} in system.py. "
"User-supplied code must never be passed to exec()."
)
def test_no_debug_exec_route():
"""Verify the /debug/exec route is not registered."""
with open(VULN_FILE, "r") as f:
source = f.read()
assert "debug/exec" not in source, (
"The /debug/exec route still exists in system.py. "
"This endpoint allows arbitrary code execution and must be removed."
)
if __name__ == "__main__":
test_no_exec_call_in_system_controller()
test_no_debug_exec_route()
print("All tests passed!")

View File

@@ -479,6 +479,47 @@ class TestMessageAggregatorMerge:
assert "hello" in merged_str
assert "world" in merged_str
def test_merge_messages_preserves_routed_by_rule_if_any_input_matches(self):
"""Merged PendingMessage should keep routed_by_rule when any input was rule-routed."""
aggregator = get_aggregator_module()
app = make_aggregator_app()
agg = aggregator.MessageAggregator(app)
chain1 = text_chain("first")
chain2 = text_chain("second")
event = friend_message_event(chain1)
adapter = mock_adapter()
pending1 = aggregator.PendingMessage(
bot_uuid='test-bot',
launcher_type=provider_session.LauncherTypes.PERSON,
launcher_id=12345,
sender_id=12345,
message_event=event,
message_chain=chain1,
adapter=adapter,
pipeline_uuid='test-pipeline-uuid',
routed_by_rule=False,
)
pending2 = aggregator.PendingMessage(
bot_uuid='test-bot',
launcher_type=provider_session.LauncherTypes.PERSON,
launcher_id=12345,
sender_id=12345,
message_event=event,
message_chain=chain2,
adapter=adapter,
pipeline_uuid='test-pipeline-uuid',
routed_by_rule=True,
)
merged = agg._merge_messages([pending1, pending2])
assert merged.routed_by_rule is True
assert str(merged.message_chain) == 'first\nsecond'
class TestMessageAggregatorFlush:
"""Tests for buffer flush behavior."""
@@ -594,44 +635,3 @@ class TestMessageAggregatorFlushAll:
# Both buffers should be flushed
assert len(agg.buffers) == 0
assert app.query_pool.add_query.call_count == 2
class TestMessageAggregatorMergeRoutedFlag:
"""Tests for preserving routed message state during merge."""
def test_merge_messages_preserves_routed_by_rule_if_any_input_matches(self):
"""Merged PendingMessage keeps routed_by_rule when any input was rule-routed."""
aggregator = get_aggregator_module()
agg = aggregator.MessageAggregator(ap=None)
chain1 = text_chain("first")
chain2 = text_chain("second")
event = friend_message_event(chain1)
adapter = mock_adapter()
pending1 = aggregator.PendingMessage(
bot_uuid='test-bot',
launcher_type=provider_session.LauncherTypes.PERSON,
launcher_id=12345,
sender_id=12345,
message_event=event,
message_chain=chain1,
adapter=adapter,
pipeline_uuid='test-pipeline',
routed_by_rule=False,
)
pending2 = aggregator.PendingMessage(
bot_uuid='test-bot',
launcher_type=provider_session.LauncherTypes.PERSON,
launcher_id=12345,
sender_id=12345,
message_event=event,
message_chain=chain2,
adapter=adapter,
pipeline_uuid='test-pipeline',
routed_by_rule=True,
)
merged = agg._merge_messages([pending1, pending2])
assert merged.routed_by_rule is True
assert str(merged.message_chain) == 'first\nsecond'

View File

@@ -119,10 +119,10 @@ class TestContentFilterStageInit:
await stage.initialize(pipeline_config)
assert [filter_impl.name for filter_impl in stage.filter_chain] == [
assert {filter_impl.name for filter_impl in stage.filter_chain} == {
'ban-word-filter',
'content-ignore',
]
}
class TestPreContentFilter:

View File

@@ -11,7 +11,7 @@ Tests cover:
from __future__ import annotations
import pytest
from unittest.mock import Mock
from unittest.mock import AsyncMock, Mock
from importlib import import_module
from tests.factories import (
@@ -166,29 +166,6 @@ class TestLongTextProcessStageProcess:
assert isinstance(components[0], platform_message.Plain)
assert components[0].text == 'short response'
@pytest.mark.asyncio
async def test_empty_response_message_chain_continues_without_processing(self):
"""Empty response chains should be a no-op for long text processing."""
longtext = get_longtext_module()
entities = get_entities_module()
app = FakeApp()
stage = longtext.LongTextProcessStage(app)
pipeline_config = make_longtext_config(strategy='forward', threshold=1)
await stage.initialize(pipeline_config)
query = text_query("hello")
query.pipeline_config = pipeline_config
query.resp_message_chain = []
result = await stage.process(query, 'LongTextProcessStage')
assert result.result_type == entities.ResultType.CONTINUE
assert result.new_query is query
assert query.resp_message_chain == []
@pytest.mark.asyncio
async def test_non_plain_component_skips(self):
"""resp_message_chain with non-Plain components should skip processing."""
@@ -223,6 +200,48 @@ class TestLongTextProcessStageProcess:
assert components[0].text == 'short'
assert components[1].url == 'https://example.com/img.png'
@pytest.mark.asyncio
async def test_empty_resp_message_chain(self):
"""Empty resp_message_chain should be handled gracefully."""
longtext = get_longtext_module()
entities = get_entities_module()
app = FakeApp()
stage = longtext.LongTextProcessStage(app)
pipeline_config = make_longtext_config(strategy='forward')
await stage.initialize(pipeline_config)
query = text_query("hello")
query.pipeline_config = pipeline_config
query.resp_message_chain = []
result = await stage.process(query, 'LongTextProcessStage')
assert result.result_type == entities.ResultType.CONTINUE
assert result.new_query is query
@pytest.mark.asyncio
async def test_empty_response_message_chain_does_not_call_strategy(self):
"""Empty response chains should be a no-op for long text processing."""
longtext = get_longtext_module()
entities = get_entities_module()
app = FakeApp()
stage = longtext.LongTextProcessStage(app)
stage.strategy_impl = AsyncMock()
query = text_query("hello")
query.pipeline_config = make_longtext_config(strategy='forward', threshold=1)
query.resp_message_chain = []
result = await stage.process(query, 'LongTextProcessStage')
assert result.result_type == entities.ResultType.CONTINUE
assert result.new_query is query
stage.strategy_impl.process.assert_not_called()
class TestForwardStrategy:
"""Tests for ForwardComponentStrategy."""

View File

@@ -223,10 +223,8 @@ def test_token_manager_next_token_empty():
"""Test TokenManager.next_token with empty tokens doesn't error."""
mgr = token.TokenManager(name='test', tokens=[])
mgr.next_token()
assert mgr.next_token() is None
assert mgr.get_token() == ''
assert mgr.using_token_index == 0
# ============================================================================

View File

@@ -434,6 +434,23 @@ class TestRAGRuntimeServiceGetFileStream:
assert result == b''
@pytest.mark.asyncio
async def test_get_file_stream_normalizes_safe_path(self):
"""Safe relative paths are normalized before loading."""
mock_app = self._create_mock_app()
mocks = self._make_rag_import_mocks()
with isolated_sys_modules(mocks):
from langbot.pkg.rag.service.runtime import RAGRuntimeService
service = RAGRuntimeService(mock_app)
result = await service.get_file_stream('knowledge/./files/doc.pdf')
assert result == b'file content'
mock_app.storage_mgr.storage_provider.load.assert_called_once_with('knowledge/files/doc.pdf')
@pytest.mark.asyncio
async def test_get_file_stream_path_traversal_blocked(self):
"""Path traversal attacks are blocked."""
@@ -454,6 +471,37 @@ class TestRAGRuntimeServiceGetFileStream:
with pytest.raises(ValueError, match='Invalid storage path'):
await service.get_file_stream('knowledge/../../../etc/passwd')
@pytest.mark.asyncio
@pytest.mark.parametrize(
'storage_path',
[
'',
'../secret.txt',
'/absolute/path.txt',
'..\\secret.txt',
'nested\\..\\secret.txt',
'%2e%2e/secret.txt',
'nested/%2e%2e/secret.txt',
'C:\\secret.txt',
'safe/\x00file.txt',
],
)
async def test_get_file_stream_rejects_unsafe_paths(self, storage_path: str):
"""Unsafe runtime file paths are rejected before storage load."""
mock_app = self._create_mock_app()
mocks = self._make_rag_import_mocks()
with isolated_sys_modules(mocks):
from langbot.pkg.rag.service.runtime import RAGRuntimeService
service = RAGRuntimeService(mock_app)
with pytest.raises(ValueError, match='Invalid storage path'):
await service.get_file_stream(storage_path)
mock_app.storage_mgr.storage_provider.load.assert_not_called()
@pytest.mark.asyncio
async def test_get_file_stream_normalizes_path(self):
"""Valid paths with .. in filename (not traversal) should work."""
@@ -472,50 +520,3 @@ class TestRAGRuntimeServiceGetFileStream:
# Let's test a simple valid path
await service.get_file_stream('knowledge/files/test.pdf')
mock_app.storage_mgr.storage_provider.load.assert_called()
@pytest.mark.asyncio
async def test_get_file_stream_normalizes_safe_relative_path(self):
"""Safe relative paths are normalized before loading."""
mock_app = self._create_mock_app()
mocks = self._make_rag_import_mocks()
with isolated_sys_modules(mocks):
from langbot.pkg.rag.service.runtime import RAGRuntimeService
service = RAGRuntimeService(mock_app)
await service.get_file_stream('knowledge/./files/doc.pdf')
mock_app.storage_mgr.storage_provider.load.assert_called_once_with('knowledge/files/doc.pdf')
@pytest.mark.asyncio
@pytest.mark.parametrize(
"storage_path",
[
"",
"../secret.txt",
"/absolute/path.txt",
"..\\secret.txt",
"nested\\..\\secret.txt",
"%2e%2e/secret.txt",
"nested/%2e%2e/secret.txt",
"C:\\secret.txt",
"safe/\x00file.txt",
],
)
async def test_get_file_stream_rejects_unsafe_paths(self, storage_path):
"""Traversal, absolute, encoded, and Windows-style paths are rejected."""
mock_app = self._create_mock_app()
mocks = self._make_rag_import_mocks()
with isolated_sys_modules(mocks):
from langbot.pkg.rag.service.runtime import RAGRuntimeService
service = RAGRuntimeService(mock_app)
with pytest.raises(ValueError, match='Invalid storage path'):
await service.get_file_stream(storage_path)
mock_app.storage_mgr.storage_provider.load.assert_not_called()

View File

@@ -191,19 +191,18 @@ class TestGetFuncSchema:
assert result['parameters']['properties']['param_name']['description'] == 'This is the param description.'
def test_missing_parameter_doc_uses_empty_description(self):
"""Test that undocumented parameters do not crash schema generation."""
"""Undocumented parameters should not break schema generation."""
funcschema = get_funcschema_module()
def partially_documented_func(documented: str, undocumented: int):
"""Function with one undocumented param.
def sample_function(documented: str, undocumented: int):
"""Sample function.
Args:
documented: Documented parameter.
documented(str): documented parameter description
"""
pass
result = funcschema.get_func_schema(partially_documented_func)
result = funcschema.get_func_schema(sample_function)
props = result['parameters']['properties']
assert props['documented']['description'] == 'Documented parameter.'
assert props['undocumented']['description'] == ''
assert result['parameters']['properties']['documented']['description'] == 'documented parameter description'
assert result['parameters']['properties']['undocumented']['description'] == ''

View File

@@ -56,21 +56,29 @@ class TestGetQQImageDownloadableUrl:
# Fragment is not included in query string parsing
assert "http://example.com/image.jpg" in result_url
def test_https_url_preserves_scheme(self):
"""Parse HTTPS URL without downgrading the scheme."""
def test_https_url(self):
"""Parse HTTPS URL and preserve its scheme."""
url = "https://example.com/image.jpg"
result_url, query = get_qq_image_downloadable_url(url)
assert result_url == "https://example.com/image.jpg"
assert query == {}
def test_missing_scheme_defaults_to_http(self):
"""Parse scheme-less URL with the existing HTTP default."""
url = "example.com/image.jpg?param=value"
result_url, query = get_qq_image_downloadable_url(url)
def test_preserves_qq_https_scheme_and_query(self):
"""QQ image URLs keep HTTPS and query parameters."""
result_url, query = get_qq_image_downloadable_url(
'https://gchat.qpic.cn/gchatpic_new/abc/0?term=2&is_origin=1'
)
assert result_url == "http://example.com/image.jpg"
assert query == {"param": ["value"]}
assert result_url == 'https://gchat.qpic.cn/gchatpic_new/abc/0'
assert query == {'term': ['2'], 'is_origin': ['1']}
def test_defaults_missing_scheme_to_http(self):
"""Scheme-less image URLs default to HTTP."""
result_url, query = get_qq_image_downloadable_url('gchat.qpic.cn/gchatpic_new/abc/0?term=2')
assert result_url == 'http://gchat.qpic.cn/gchatpic_new/abc/0'
assert query == {'term': ['2']}
class TestExtractB64AndFormat:

View File

@@ -75,6 +75,61 @@ class TestPkgMgr:
]
mock_pipmain.assert_called_once_with(expected_args)
def test_install_requirements_defaults_extra_params_to_none(self):
"""install_requirements should not use a mutable default for extra_params."""
signature = inspect.signature(pkgmgr.install_requirements)
assert signature.parameters['extra_params'].default is None
def test_install_requirements_omitted_extra_params_uses_independent_base_commands(self, monkeypatch):
"""Omitted extra_params should not share mutable state across calls."""
calls = []
monkeypatch.setattr(pkgmgr, 'pipmain', calls.append)
pkgmgr.install_requirements('requirements.txt')
pkgmgr.install_requirements('requirements-dev.txt')
assert calls == [
[
'install',
'-r',
'requirements.txt',
'-i',
'https://pypi.tuna.tsinghua.edu.cn/simple',
'--trusted-host',
'pypi.tuna.tsinghua.edu.cn',
],
[
'install',
'-r',
'requirements-dev.txt',
'-i',
'https://pypi.tuna.tsinghua.edu.cn/simple',
'--trusted-host',
'pypi.tuna.tsinghua.edu.cn',
],
]
def test_install_requirements_preserves_explicit_extra_params(self, monkeypatch):
"""Explicit extra_params should be appended to the generated pip command."""
calls = []
monkeypatch.setattr(pkgmgr, 'pipmain', calls.append)
pkgmgr.install_requirements('requirements.txt', extra_params=['--no-deps'])
assert calls == [
[
'install',
'-r',
'requirements.txt',
'-i',
'https://pypi.tuna.tsinghua.edu.cn/simple',
'--trusted-host',
'pypi.tuna.tsinghua.edu.cn',
'--no-deps',
]
]
def test_install_requirements_with_extra_params(self):
"""install_requirements handles extra params."""
with patch('langbot.pkg.utils.pkgmgr.pipmain') as mock_pipmain:
@@ -100,34 +155,3 @@ class TestPkgMgr:
call_args = mock_pipmain.call_args[0][0]
assert '--no-cache-dir' in call_args
assert '--verbose' in call_args
def test_install_requirements_defaults_extra_params_to_none(self):
"""install_requirements does not use a mutable list default."""
signature = inspect.signature(pkgmgr.install_requirements)
assert signature.parameters['extra_params'].default is None
def test_install_requirements_omitted_extra_params_are_isolated(self):
"""Repeated calls without extra_params use independent base commands."""
with patch('langbot.pkg.utils.pkgmgr.pipmain') as mock_pipmain:
pkgmgr.install_requirements('requirements.txt')
pkgmgr.install_requirements('requirements-dev.txt')
assert mock_pipmain.call_args_list[0].args[0] == [
'install',
'-r',
'requirements.txt',
'-i',
'https://pypi.tuna.tsinghua.edu.cn/simple',
'--trusted-host',
'pypi.tuna.tsinghua.edu.cn',
]
assert mock_pipmain.call_args_list[1].args[0] == [
'install',
'-r',
'requirements-dev.txt',
'-i',
'https://pypi.tuna.tsinghua.edu.cn/simple',
'--trusted-host',
'pypi.tuna.tsinghua.edu.cn',
]

View File

@@ -87,6 +87,22 @@ class TestGetRunnerCategory:
assert get_runner_category("test", "https://example.com") == RunnerCategory.CLOUD
assert get_runner_category("test", "https://myserver.example.org") == RunnerCategory.CLOUD
@pytest.mark.parametrize(
'runner_url',
[
'api.dify.ai/v1',
'localhost:7860',
'https:///v1',
'https://',
'https://exa mple.com',
'http://[::1',
'http://localhost:bad',
],
)
def test_invalid_urls_return_unknown(self, runner_url):
"""Invalid or incomplete URLs should return UNKNOWN."""
assert get_runner_category("test", runner_url) == RunnerCategory.UNKNOWN
def test_urlparse_exception_returns_unknown(self):
"""Exception during URL parsing should return UNKNOWN."""
# Test by mocking urlparse to raise an exception
@@ -99,49 +115,36 @@ class TestGetRunnerCategory:
result = runner.get_runner_category("test", "http://example.com")
assert result == RunnerCategory.UNKNOWN
@pytest.mark.parametrize(
"runner_url",
[
"api.dify.ai/v1",
"localhost:7860",
"https:///v1",
"https://",
"https://exa mple.com",
"http://[::1",
"http://localhost:bad",
],
)
def test_invalid_urls_return_unknown(self, runner_url):
"""Invalid or scheme-less URLs should not default to CLOUD."""
assert get_runner_category("test", runner_url) == RunnerCategory.UNKNOWN
def test_url_without_scheme_returns_unknown(self):
"""URL without scheme should return UNKNOWN."""
assert get_runner_category("test", "example.com") == RunnerCategory.UNKNOWN
@pytest.mark.parametrize(
"runner_url",
'runner_url',
[
"http://localhost:7860",
"http://127.0.0.1:7860",
"http://10.0.0.1:7860",
"http://172.16.0.1:7860",
"http://172.31.255.255:7860",
"http://192.168.1.20:7860",
"http://[::1]:7860",
'http://localhost:7860',
'http://127.0.0.1:7860',
'http://10.0.0.1:7860',
'http://172.16.0.1:7860',
'http://172.31.255.255:7860',
'http://192.168.1.20:7860',
'http://[::1]:7860',
],
)
def test_local_hosts_are_detected_with_ipaddress(self, runner_url):
"""Loopback/private IP addresses and localhost should be LOCAL."""
assert get_runner_category("test", runner_url) == RunnerCategory.LOCAL
def test_detects_local_hosts_with_ipaddress(self, runner_url):
"""Local hostnames and private IPs should be categorized as LOCAL."""
assert get_runner_category('langflow-api', runner_url) == RunnerCategory.LOCAL
@pytest.mark.parametrize(
"runner_url",
'runner_url',
[
"http://10.evil.com",
"http://192.168.example.com",
'http://10.evil.com',
'http://192.168.example.com',
],
)
def test_private_ip_prefix_domains_are_not_local(self, runner_url):
"""Domain names that only look like private IP prefixes should not be LOCAL."""
assert get_runner_category("test", runner_url) == RunnerCategory.CLOUD
assert get_runner_category('langflow-api', runner_url) == RunnerCategory.CLOUD
class TestIsCloudRunner:
"""Test is_cloud_runner helper function."""

View File

@@ -338,7 +338,9 @@ function NavItems({
tooltip={config.name}
>
{config.icon}
<span>{config.name}</span>
<span className="cursor-pointer select-none">
{config.name}
</span>
</SidebarMenuButton>
</SidebarMenuItem>
);
@@ -728,7 +730,9 @@ function NavItems({
}}
>
{config.icon}
<span>{config.name}</span>
<span className="cursor-pointer select-none">
{config.name}
</span>
<div className="ml-auto flex items-center gap-0.5 -mr-1">
{canCreate &&
(isPlugin ? (
@@ -1108,7 +1112,7 @@ function PluginPagesNav() {
className="select-none"
>
{pluginIcon}
<span>{page.name}</span>
<span className="cursor-pointer">{page.name}</span>
</SidebarMenuButton>
</SidebarMenuItem>
);
@@ -1128,7 +1132,7 @@ function PluginPagesNav() {
className="select-none"
>
{pluginIcon}
<span>{label}</span>
<span className="cursor-pointer">{label}</span>
<ChevronRight className="ml-auto size-4 transition-transform duration-200 group-data-[state=open]/collapsible:rotate-90" />
</SidebarMenuButton>
</CollapsibleTrigger>
@@ -1144,7 +1148,9 @@ function PluginPagesNav() {
onClick={() => navigate(route)}
className="select-none"
>
<span>{page.name}</span>
<span className="cursor-pointer">
{page.name}
</span>
</SidebarMenuSubButton>
</SidebarMenuSubItem>
);

View File

@@ -295,7 +295,7 @@ export default function ModelsDialog({
async function handleScanModels(
providerUuid: string,
modelType: ModelType,
modelType?: ModelType,
): Promise<ScanModelsResult> {
try {
const resp = await httpClient.scanProviderModels(providerUuid, modelType);
@@ -319,19 +319,26 @@ export default function ModelsDialog({
setIsSubmitting(true);
try {
for (const item of models) {
if (modelType === 'llm') {
const effectiveType = item.model.type || modelType;
if (effectiveType === 'llm') {
await httpClient.createProviderLLMModel({
name: item.model.name,
provider_uuid: providerUuid,
abilities: item.abilities,
extra_args: {},
} as never);
} else {
} else if (effectiveType === 'embedding') {
await httpClient.createProviderEmbeddingModel({
name: item.model.name,
provider_uuid: providerUuid,
extra_args: {},
} as never);
} else {
await httpClient.createProviderRerankModel({
name: item.model.name,
provider_uuid: providerUuid,
extra_args: {},
} as never);
}
}
setAddModelPopoverOpen(null);

View File

@@ -73,10 +73,13 @@ export default function ProviderForm({
>([]);
useEffect(() => {
loadRequesters();
if (providerId) {
loadProvider(providerId);
async function init() {
await loadRequesters();
if (providerId) {
await loadProvider(providerId);
}
}
init();
}, [providerId]);
async function loadRequesters() {

View File

@@ -8,7 +8,6 @@ import {
Wrench,
Check,
RefreshCw,
Search,
} from 'lucide-react';
import { Button } from '@/components/ui/button';
import { Input } from '@/components/ui/input';
@@ -33,6 +32,8 @@ import ExtraArgsEditor from './ExtraArgsEditor';
interface AddModelPopoverProps {
isOpen: boolean;
initialMode?: 'manual' | 'scan';
trigger?: React.ReactNode;
onOpen: () => void;
onClose: () => void;
onAddModel: (
@@ -41,7 +42,7 @@ interface AddModelPopoverProps {
abilities: string[],
extraArgs: ExtraArg[],
) => Promise<void>;
onScanModels: (modelType: ModelType) => Promise<ScanModelsResult>;
onScanModels: (modelType?: ModelType) => Promise<ScanModelsResult>;
onAddScannedModels: (
modelType: ModelType,
models: SelectedScannedModel[],
@@ -60,6 +61,8 @@ interface AddModelPopoverProps {
export default function AddModelPopover({
isOpen,
initialMode = 'manual',
trigger,
onOpen,
onClose,
onAddModel,
@@ -92,7 +95,7 @@ export default function AddModelPopover({
const wasOpen = prevIsOpenRef.current;
if (isOpen && !wasOpen) {
setTab('llm');
setMode('manual');
setMode(initialMode);
setName('');
setAbilities([]);
setExtraArgs([]);
@@ -101,8 +104,12 @@ export default function AddModelPopover({
setSelectedScannedModels({});
setScanQuery('');
onResetTestResult();
if (initialMode === 'scan') {
handleScan();
}
}
prevIsOpenRef.current = isOpen;
// eslint-disable-next-line react-hooks/exhaustive-deps
}, [isOpen, onResetTestResult]);
useEffect(() => {
@@ -122,9 +129,8 @@ export default function AddModelPopover({
const handleScan = async () => {
setScanLoading(true);
try {
const result = await onScanModels(tab);
const result = await onScanModels(trigger ? undefined : tab);
// Enrich abilities from debug.response.data (e.g. features.tools.function_calling)
const debugData = (
result.debug?.response as { data?: Record<string, unknown>[] }
)?.data;
@@ -143,9 +149,9 @@ export default function AddModelPopover({
| undefined;
const tools = features?.tools as Record<string, unknown> | undefined;
if (tools?.function_calling === true) {
const abilities = new Set(model.abilities || []);
abilities.add('func_call');
model.abilities = [...abilities];
const nextAbilities = new Set(model.abilities || []);
nextAbilities.add('func_call');
model.abilities = [...nextAbilities];
}
}
}
@@ -247,305 +253,321 @@ export default function AddModelPopover({
onOpenChange={(open) => (open ? onOpen() : onClose())}
>
<PopoverTrigger asChild>
<Button
variant="ghost"
size="sm"
className="h-6 text-xs"
onClick={(e) => e.stopPropagation()}
>
<Plus className="h-3 w-3 mr-1" />
{t('models.addModel')}
</Button>
{trigger || (
<Button
variant="ghost"
size="sm"
className="h-6 text-xs"
onClick={(e) => e.stopPropagation()}
>
<Plus className="h-3 w-3 mr-1" />
{t('models.addModel')}
</Button>
)}
</PopoverTrigger>
<PopoverContent
className="w-[min(24rem,calc(100vw-2rem))] max-h-[70vh] overflow-y-auto overscroll-none focus:outline-none focus-visible:outline-none focus-visible:ring-0"
style={{
maxHeight: 'min(70vh, var(--radix-popover-content-available-height))',
}}
className="w-[min(24rem,calc(100vw-2rem))] max-h-[calc(100vh-8rem)] flex flex-col overflow-hidden"
align="end"
side="left"
side="bottom"
sideOffset={8}
collisionPadding={16}
onWheel={(e) => e.stopPropagation()}
onTouchMove={(e) => e.stopPropagation()}
onClick={(e) => e.stopPropagation()}
>
<Tabs value={tab} onValueChange={(v) => setTab(v as ModelType)}>
<TabsList className="grid w-full grid-cols-3">
<TabsTrigger value="llm">
<MessageSquareText className="h-4 w-4 mr-1" />
{t('models.chat')}
</TabsTrigger>
<TabsTrigger value="embedding">
<Cpu className="h-4 w-4 mr-1" />
{t('models.embedding')}
</TabsTrigger>
<TabsTrigger value="rerank">
<ArrowUpDown className="h-4 w-4 mr-1" />
{t('models.rerank')}
</TabsTrigger>
</TabsList>
<Tabs
value={tab}
onValueChange={(v) => setTab(v as ModelType)}
className="flex flex-col min-h-0 flex-1"
>
<div className="flex-shrink-0">
{!(trigger && initialMode === 'scan') && (
<TabsList className="grid w-full grid-cols-3">
<TabsTrigger value="llm">
<MessageSquareText className="h-4 w-4 mr-1" />
{t('models.chat')}
</TabsTrigger>
<TabsTrigger value="embedding">
<Cpu className="h-4 w-4 mr-1" />
{t('models.embedding')}
</TabsTrigger>
<TabsTrigger value="rerank">
<ArrowUpDown className="h-4 w-4 mr-1" />
{t('models.rerank')}
</TabsTrigger>
</TabsList>
)}
</div>
<Tabs
value={mode}
onValueChange={(v) => setMode(v as 'manual' | 'scan')}
>
<TabsList className="grid w-full grid-cols-2 mt-3">
<TabsTrigger value="manual">{t('models.manualAdd')}</TabsTrigger>
<TabsTrigger value="scan">{t('models.scanAdd')}</TabsTrigger>
</TabsList>
<div className="overflow-y-auto flex-1 min-h-0">
<Tabs
value={mode}
onValueChange={(v) => setMode(v as 'manual' | 'scan')}
>
{!trigger && (
<TabsList className="grid w-full grid-cols-2 mt-3">
<TabsTrigger value="manual">
{t('models.manualAdd')}
</TabsTrigger>
<TabsTrigger value="scan">{t('models.scanAdd')}</TabsTrigger>
</TabsList>
)}
<TabsContent value="manual" className="mt-3">
<div className="space-y-3">
<div className="space-y-2">
<Label>{t('models.modelName')}</Label>
<Input
placeholder={t('models.modelName')}
value={name}
onChange={(e) => setName(e.target.value)}
/>
</div>
{tab === 'llm' && (
<TabsContent value="manual" className="mt-3">
<div className="space-y-3">
<div className="space-y-2">
<Label>{t('models.abilities')}</Label>
<div className="flex gap-4">
<div className="flex items-center gap-2">
<Checkbox
id="add-vision"
checked={abilities.includes('vision')}
onCheckedChange={(checked) =>
toggleAbility('vision', checked as boolean)
}
/>
<Label htmlFor="add-vision" className="text-sm">
<Eye className="h-3 w-3 inline mr-1" />
{t('models.visionAbility')}
</Label>
</div>
<div className="flex items-center gap-2">
<Checkbox
id="add-func-call"
checked={abilities.includes('func_call')}
onCheckedChange={(checked) =>
toggleAbility('func_call', checked as boolean)
}
/>
<Label htmlFor="add-func-call" className="text-sm">
<Wrench className="h-3 w-3 inline mr-1" />
{t('models.functionCallAbility')}
</Label>
<Label>{t('models.modelName')}</Label>
<Input
placeholder={t('models.modelName')}
value={name}
onChange={(e) => setName(e.target.value)}
/>
</div>
{tab === 'llm' && (
<div className="space-y-2">
<Label>{t('models.abilities')}</Label>
<div className="flex gap-4">
<div className="flex items-center gap-2">
<Checkbox
id="add-vision"
checked={abilities.includes('vision')}
onCheckedChange={(checked) =>
toggleAbility('vision', checked as boolean)
}
/>
<Label htmlFor="add-vision" className="text-sm">
<Eye className="h-3 w-3 inline mr-1" />
{t('models.visionAbility')}
</Label>
</div>
<div className="flex items-center gap-2">
<Checkbox
id="add-func-call"
checked={abilities.includes('func_call')}
onCheckedChange={(checked) =>
toggleAbility('func_call', checked as boolean)
}
/>
<Label htmlFor="add-func-call" className="text-sm">
<Wrench className="h-3 w-3 inline mr-1" />
{t('models.functionCallAbility')}
</Label>
</div>
</div>
</div>
)}
<ExtraArgsEditor
args={extraArgs}
onChange={setExtraArgs}
modelType={tab}
/>
<div className="flex gap-2">
<Button
className="flex-1"
size="sm"
onClick={handleAdd}
disabled={isSubmitting || isTesting}
>
{isSubmitting ? t('common.saving') : t('common.add')}
</Button>
<Button
className="flex-1"
size="sm"
variant="outline"
onClick={handleTest}
disabled={isSubmitting || isTesting}
>
{isTesting ? (
t('common.loading')
) : testResult?.success ? (
<>
<Check className="h-4 w-4 mr-1 text-green-500" />
{(testResult.duration / 1000).toFixed(1)}s
</>
) : (
t('common.test')
)}
</Button>
</div>
</div>
</TabsContent>
<TabsContent value="scan" className="space-y-2 mt-0 pt-0">
{scanLoading ? (
<div className="flex items-center justify-center py-4">
<RefreshCw className="h-4 w-4 mr-2 animate-spin text-muted-foreground" />
<span className="text-sm text-muted-foreground">
{t('models.scanModels')}...
</span>
</div>
) : (
<>
<div className="space-y-2">
<Input
placeholder={t('models.searchScannedModels')}
value={scanQuery}
onChange={(e) => setScanQuery(e.target.value)}
disabled={scannedModels.length === 0}
/>
{selectableModels.length > 0 && (
<div className="flex items-center gap-2 pt-1">
<Checkbox
id="scan-select-all"
checked={allSelected}
onCheckedChange={toggleSelectAll}
/>
<Label
htmlFor="scan-select-all"
className="text-sm font-medium"
>
{t('models.selectAll')}
<span className="text-muted-foreground ml-1">
({Object.keys(selectedScannedModels).length}/
{selectableModels.length})
</span>
</Label>
</div>
)}
</div>
<div
className="h-64 overflow-y-auto overscroll-contain rounded-md border"
onWheel={(e) => e.stopPropagation()}
>
<div className="p-3 space-y-2">
{filteredScannedModels.length === 0 ? (
<p className="text-sm text-muted-foreground">
{scannedModels.length === 0
? t('models.noScannedModels')
: t('models.noScannedModelsMatch')}
</p>
) : (
filteredScannedModels.map((model) => {
const isSelected = Boolean(
selectedScannedModels[model.id],
);
const selectedAbilities =
selectedScannedModels[model.id]?.abilities || [];
return (
<div
key={model.id}
className="rounded-md border p-3 space-y-2"
>
<div className="flex items-start gap-3">
<Checkbox
checked={isSelected || model.already_added}
disabled={model.already_added}
onCheckedChange={(checked) =>
toggleScannedModel(
model,
checked as boolean,
)
}
/>
<div className="min-w-0 flex-1">
<div className="text-sm font-medium break-all">
{model.name}
</div>
<div className="text-xs text-muted-foreground">
{model.already_added
? t('models.alreadyAdded')
: model.type === 'llm'
? t('models.chat')
: model.type === 'embedding'
? t('models.embedding')
: t('models.rerank')}
</div>
</div>
</div>
{model.type === 'llm' &&
isSelected &&
!model.already_added && (
<div className="flex gap-4 pl-7">
<div className="flex items-center gap-2">
<Checkbox
id={`scan-vision-${model.id}`}
checked={selectedAbilities.includes(
'vision',
)}
onCheckedChange={(checked) =>
toggleScannedModelAbility(
model.id,
'vision',
checked as boolean,
)
}
/>
<Label
htmlFor={`scan-vision-${model.id}`}
className="text-sm"
>
<Eye className="h-3 w-3 inline mr-1" />
{t('models.visionAbility')}
</Label>
</div>
<div className="flex items-center gap-2">
<Checkbox
id={`scan-func-${model.id}`}
checked={selectedAbilities.includes(
'func_call',
)}
onCheckedChange={(checked) =>
toggleScannedModelAbility(
model.id,
'func_call',
checked as boolean,
)
}
/>
<Label
htmlFor={`scan-func-${model.id}`}
className="text-sm"
>
<Wrench className="h-3 w-3 inline mr-1" />
{t('models.functionCallAbility')}
</Label>
</div>
</div>
)}
</div>
);
})
)}
</div>
</div>
</>
)}
<ExtraArgsEditor
args={extraArgs}
onChange={setExtraArgs}
modelType={tab}
/>
<div className="flex gap-2">
<Button
className="flex-1"
size="sm"
onClick={handleAdd}
disabled={isSubmitting || isTesting}
onClick={handleAddScanned}
disabled={
isSubmitting ||
scanLoading ||
Object.keys(selectedScannedModels).length === 0
}
>
{isSubmitting ? t('common.saving') : t('common.add')}
{isSubmitting
? t('common.saving')
: t('models.addSelectedModels')}
</Button>
<Button
className="flex-1"
size="sm"
variant="outline"
onClick={handleTest}
disabled={isSubmitting || isTesting}
size="sm"
onClick={handleScan}
disabled={scanLoading || isSubmitting}
>
{isTesting ? (
t('common.loading')
) : testResult?.success ? (
<>
<Check className="h-4 w-4 mr-1 text-green-500" />
{(testResult.duration / 1000).toFixed(1)}s
</>
) : (
t('common.test')
)}
<RefreshCw
className={`h-3.5 w-3.5 ${scanLoading ? 'animate-spin' : ''}`}
/>
</Button>
</div>
</div>
</TabsContent>
<TabsContent value="scan" className="space-y-3 mt-3">
<div className="text-xs text-muted-foreground">
{t('models.scanModelsHint')}
</div>
<div className="flex gap-2">
<Button
className="flex-1"
size="sm"
variant="outline"
onClick={handleScan}
disabled={scanLoading || isSubmitting}
>
{scanLoading ? (
<RefreshCw className="h-4 w-4 mr-1 animate-spin" />
) : (
<Search className="h-4 w-4 mr-1" />
)}
{t('models.scanModels')}
</Button>
<Button
className="flex-1"
size="sm"
onClick={handleAddScanned}
disabled={
isSubmitting ||
scanLoading ||
Object.keys(selectedScannedModels).length === 0
}
>
{isSubmitting
? t('common.saving')
: t('models.addSelectedModels')}
</Button>
</div>
<div className="space-y-2">
<Label>{t('models.scannedModels')}</Label>
<Input
placeholder={t('models.searchScannedModels')}
value={scanQuery}
onChange={(e) => setScanQuery(e.target.value)}
disabled={scannedModels.length === 0}
/>
{selectableModels.length > 0 && (
<div className="flex items-center gap-2 pt-1">
<Checkbox
id="scan-select-all"
checked={allSelected}
onCheckedChange={toggleSelectAll}
/>
<Label
htmlFor="scan-select-all"
className="text-sm font-medium"
>
{t('models.selectAll')}
<span className="text-muted-foreground ml-1">
({Object.keys(selectedScannedModels).length}/
{selectableModels.length})
</span>
</Label>
</div>
)}
</div>
<div
className="h-64 overflow-y-auto overscroll-none rounded-md border"
onWheel={(e) => e.stopPropagation()}
>
<div className="p-3 space-y-2">
{filteredScannedModels.length === 0 ? (
<p className="text-sm text-muted-foreground">
{scannedModels.length === 0
? t('models.noScannedModels')
: t('models.noScannedModelsMatch')}
</p>
) : (
filteredScannedModels.map((model) => {
const isSelected = Boolean(
selectedScannedModels[model.id],
);
const selectedAbilities =
selectedScannedModels[model.id]?.abilities || [];
return (
<div
key={model.id}
className="rounded-md border p-3 space-y-2"
>
<div className="flex items-start gap-3">
<Checkbox
checked={isSelected || model.already_added}
disabled={model.already_added}
onCheckedChange={(checked) =>
toggleScannedModel(model, checked as boolean)
}
/>
<div className="min-w-0 flex-1">
<div className="text-sm font-medium break-all">
{model.name}
</div>
<div className="text-xs text-muted-foreground">
{model.already_added
? t('models.alreadyAdded')
: model.type === 'llm'
? t('models.chat')
: model.type === 'embedding'
? t('models.embedding')
: t('models.rerank')}
</div>
</div>
</div>
{tab === 'llm' &&
isSelected &&
!model.already_added && (
<div className="flex gap-4 pl-7">
<div className="flex items-center gap-2">
<Checkbox
id={`scan-vision-${model.id}`}
checked={selectedAbilities.includes(
'vision',
)}
onCheckedChange={(checked) =>
toggleScannedModelAbility(
model.id,
'vision',
checked as boolean,
)
}
/>
<Label
htmlFor={`scan-vision-${model.id}`}
className="text-sm"
>
<Eye className="h-3 w-3 inline mr-1" />
{t('models.visionAbility')}
</Label>
</div>
<div className="flex items-center gap-2">
<Checkbox
id={`scan-func-${model.id}`}
checked={selectedAbilities.includes(
'func_call',
)}
onCheckedChange={(checked) =>
toggleScannedModelAbility(
model.id,
'func_call',
checked as boolean,
)
}
/>
<Label
htmlFor={`scan-func-${model.id}`}
className="text-sm"
>
<Wrench className="h-3 w-3 inline mr-1" />
{t('models.functionCallAbility')}
</Label>
</div>
</div>
)}
</div>
);
})
)}
</div>
</div>
</TabsContent>
</Tabs>
</TabsContent>
</Tabs>
</div>
</Tabs>
</PopoverContent>
</Popover>

View File

@@ -6,6 +6,7 @@ import {
Trash2,
Settings,
LogIn,
Radar,
} from 'lucide-react';
import { httpClient, systemInfo } from '@/app/infra/http/HttpClient';
import { ModelProvider } from '@/app/infra/entities/api';
@@ -60,7 +61,7 @@ interface ProviderCardProps {
abilities: string[],
extraArgs: ExtraArg[],
) => Promise<void>;
onScanModels: (modelType: ModelType) => Promise<ScanModelsResult>;
onScanModels: (modelType?: ModelType) => Promise<ScanModelsResult>;
onAddScannedModels: (
modelType: ModelType,
models: SelectedScannedModel[],
@@ -130,6 +131,7 @@ export default function ProviderCard({
const { t } = useTranslation();
const [deleteProviderConfirmOpen, setDeleteProviderConfirmOpen] =
useState(false);
const [addModelMode, setAddModelMode] = useState<'manual' | 'scan'>('manual');
const canDelete =
!isLangBotModels &&
@@ -310,19 +312,75 @@ export default function ProviderCard({
<div />
)}
{!isLangBotModels && (
<AddModelPopover
isOpen={addModelPopoverOpen === provider.uuid}
onOpen={onOpenAddModel}
onClose={onCloseAddModel}
onAddModel={onAddModel}
onScanModels={onScanModels}
onAddScannedModels={onAddScannedModels}
onTestModel={onTestModel}
isSubmitting={isSubmitting}
isTesting={isTesting}
testResult={testResult}
onResetTestResult={onResetTestResult}
/>
<div className="flex items-center gap-1">
<AddModelPopover
isOpen={
addModelPopoverOpen === provider.uuid &&
addModelMode === 'manual'
}
initialMode="manual"
trigger={
<Button
variant="ghost"
size="sm"
className="h-6 text-xs"
onClick={(e) => {
e.stopPropagation();
setAddModelMode('manual');
}}
>
<Plus className="h-3 w-3 mr-1" />
{t('models.addModel')}
</Button>
}
onOpen={() => {
setAddModelMode('manual');
onOpenAddModel();
}}
onClose={onCloseAddModel}
onAddModel={onAddModel}
onScanModels={onScanModels}
onAddScannedModels={onAddScannedModels}
onTestModel={onTestModel}
isSubmitting={isSubmitting}
isTesting={isTesting}
testResult={testResult}
onResetTestResult={onResetTestResult}
/>
<AddModelPopover
isOpen={
addModelPopoverOpen === provider.uuid &&
addModelMode === 'scan'
}
initialMode="scan"
trigger={
<Button
variant="ghost"
size="icon"
className="h-6 w-6"
onClick={(e) => {
e.stopPropagation();
setAddModelMode('scan');
}}
>
<Radar className="h-3 w-3" />
</Button>
}
onOpen={() => {
setAddModelMode('scan');
onOpenAddModel();
}}
onClose={onCloseAddModel}
onAddModel={onAddModel}
onScanModels={onScanModels}
onAddScannedModels={onAddScannedModels}
onTestModel={onTestModel}
isSubmitting={isSubmitting}
isTesting={isTesting}
testResult={testResult}
onResetTestResult={onResetTestResult}
/>
</div>
)}
</div>
</CardHeader>

View File

@@ -90,7 +90,7 @@ export interface ProviderCardProps {
abilities: string[],
extraArgs: ExtraArg[],
) => Promise<void>;
onScanModels: (modelType: ModelType) => Promise<ScanModelsResult>;
onScanModels: (modelType?: ModelType) => Promise<ScanModelsResult>;
onAddScannedModels: (
modelType: ModelType,
models: SelectedScannedModel[],

View File

@@ -4,11 +4,16 @@ import {
DialogContent,
DialogHeader,
DialogTitle,
DialogFooter,
} from '@/components/ui/dialog';
import { Button } from '@/components/ui/button';
import { useTranslation } from 'react-i18next';
import { Loader2, RefreshCw, CheckCircle2, XCircle } from 'lucide-react';
import {
Loader2,
RefreshCw,
RotateCw,
CheckCircle2,
XCircle,
} from 'lucide-react';
import QRCode from 'qrcode';
export type QrLoginPlatform = 'feishu' | 'weixin' | 'dingtalk' | 'wecombot';
@@ -96,7 +101,7 @@ interface QrCodeLoginDialogProps {
onSuccess: (credentials: Record<string, string>) => void;
}
type DialogState = 'connecting' | 'waiting' | 'success' | 'error';
type DialogState = 'connecting' | 'waiting' | 'expired' | 'success' | 'error';
const POLL_INTERVAL_MS = 3000;
@@ -115,8 +120,10 @@ export default function QrCodeLoginDialog({
const [errorMessage, setErrorMessage] = useState('');
const pollTimerRef = useRef<ReturnType<typeof setInterval> | null>(null);
const countdownRef = useRef<ReturnType<typeof setInterval> | null>(null);
const checkExpiredRef = useRef<ReturnType<typeof setInterval> | null>(null);
const abortRef = useRef<AbortController | null>(null);
const sessionIdRef = useRef<string | null>(null);
const baseUrlRef = useRef('');
const cleanedRef = useRef(false);
const onSuccessRef = useRef(onSuccess);
@@ -140,11 +147,14 @@ export default function QrCodeLoginDialog({
clearInterval(countdownRef.current);
countdownRef.current = null;
}
if (checkExpiredRef.current) {
clearInterval(checkExpiredRef.current);
checkExpiredRef.current = null;
}
if (abortRef.current) {
abortRef.current.abort();
abortRef.current = null;
}
// Cancel backend session
if (sessionIdRef.current) {
const token = localStorage.getItem('token');
const baseUrl =
@@ -171,6 +181,7 @@ export default function QrCodeLoginDialog({
const token = localStorage.getItem('token');
const baseUrl = import.meta.env.VITE_API_BASE_URL || window.location.origin;
baseUrlRef.current = baseUrl;
const cfg = platformConfigRef.current;
try {
@@ -191,8 +202,6 @@ export default function QrCodeLoginDialog({
const { session_id, qr_data_url, qr_url, expire_at } = json.data;
sessionIdRef.current = session_id;
// qr_data_url is a pre-rendered data URL (WeChat);
// qr_url is a plain URL string (Feishu) that needs local QR generation.
if (qr_data_url) {
setQrDataUrl(qr_data_url);
} else if (qr_url) {
@@ -204,11 +213,9 @@ export default function QrCodeLoginDialog({
}
setState('waiting');
// Calculate remaining seconds
const remaining = Math.max(0, Math.floor(expire_at - Date.now() / 1000));
setExpireIn(remaining);
// Start countdown
countdownRef.current = setInterval(() => {
setExpireIn((prev) => {
if (prev <= 1) {
@@ -222,7 +229,35 @@ export default function QrCodeLoginDialog({
});
}, 1000);
// Start polling
// When countdown hits 0, stop polling and show expired state
checkExpiredRef.current = setInterval(() => {
setExpireIn((current) => {
if (current <= 0) {
if (checkExpiredRef.current) {
clearInterval(checkExpiredRef.current);
checkExpiredRef.current = null;
}
if (pollTimerRef.current) {
clearInterval(pollTimerRef.current);
pollTimerRef.current = null;
}
if (sessionIdRef.current) {
fetch(
`${baseUrlRef.current}${cfg.apiBase}/${sessionIdRef.current}`,
{
method: 'DELETE',
headers: { Authorization: `Bearer ${token}` },
keepalive: true,
},
).catch(() => {});
sessionIdRef.current = null;
}
setState('expired');
}
return current;
});
}, 500);
pollTimerRef.current = setInterval(async () => {
try {
const pollRes = await fetch(
@@ -237,7 +272,7 @@ export default function QrCodeLoginDialog({
const { status, error, ...rest } = pollJson.data;
if (status === 'success') {
sessionIdRef.current = null; // backend already cleaned up
sessionIdRef.current = null;
cleanup();
setState('success');
setTimeout(() => {
@@ -249,9 +284,14 @@ export default function QrCodeLoginDialog({
cleanup();
setState('error');
setErrorMessage(error || tRef.current(cfg.failedKey));
} else if (status === 'expired') {
sessionIdRef.current = null;
cleanup();
setExpireIn(0);
setState('expired');
}
} catch {
// ignore poll errors, will retry next interval
// ignore poll errors
}
}, POLL_INTERVAL_MS);
} catch (err: unknown) {
@@ -323,6 +363,31 @@ export default function QrCodeLoginDialog({
</div>
)}
{/* QR code expired — click overlay to refresh */}
{state === 'expired' && qrDataUrl && (
<div className="flex flex-col items-center space-y-3">
<p className="text-sm text-muted-foreground text-center">
{t(platformConfig.scanQRCodeKey)}
</p>
<button
type="button"
className="relative border rounded-lg p-2 bg-white cursor-pointer group"
onClick={() => startLogin()}
>
<img
src={qrDataUrl}
alt="QR Code"
className="w-56 h-56 opacity-40"
/>
<div className="absolute inset-0 flex items-center justify-center bg-white/60 rounded-lg group-hover:bg-white/70 transition-colors">
<div className="flex items-center justify-center w-16 h-16 rounded-full bg-black/5 group-hover:bg-black/10 transition-colors">
<RotateCw className="h-8 w-8 text-muted-foreground" />
</div>
</div>
</button>
</div>
)}
{/* Success */}
{state === 'success' && (
<div className="flex flex-col items-center space-y-3 py-8">
@@ -350,7 +415,7 @@ export default function QrCodeLoginDialog({
</div>
{state === 'error' && (
<DialogFooter>
<div className="flex justify-end gap-2">
<Button variant="outline" onClick={() => handleOpenChange(false)}>
{t('common.cancel')}
</Button>
@@ -358,7 +423,7 @@ export default function QrCodeLoginDialog({
<RefreshCw className="h-4 w-4 mr-1.5" />
{t(platformConfig.retryKey)}
</Button>
</DialogFooter>
</div>
)}
</DialogContent>
</Dialog>