Compare commits

...

53 Commits

Author SHA1 Message Date
suyao
76d48f9ccb fix: address PR review issues for Volcengine integration
- Fix region field being ignored: pass user-configured region to listFoundationModels and listEndpoints
- Add user notification before silent fallback when API fails
- Throw error on credential corruption instead of returning null
- Remove redundant credentials (accessKeyId, secretAccessKey) from Redux store (they're securely stored via safeStorage)
- Add warnings field to ListModelsResult for partial API failures
- Fix Redux/IPC order: save to secure storage first, then update Redux on success
- Update related tests

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-29 19:39:15 +08:00
suyao
115cd80432 fix: format 2025-11-27 01:33:17 +08:00
suyao
531101742e feat: add project name support for Volcengine integration 2025-11-27 01:29:02 +08:00
suyao
c3c577dff4 feat: add Volcengine integration with settings and API client
- Implement Volcengine configuration in multiple languages (el-gr, es-es, fr-fr, ja-jp, pt-pt, ru-ru).
- Add Volcengine settings component to manage access key ID, secret access key, and region.
- Create Volcengine service for API interactions, including credential management and model listing.
- Extend OpenAI API client to support Volcengine's signed API for model retrieval.
- Update Redux store to handle Volcengine settings and credentials.
- Implement migration for Volcengine settings in the store.
- Add hooks for accessing and managing Volcengine settings in the application.
2025-11-27 01:15:22 +08:00
Apine
eb4670c22c docs: correct the links on the readme (#11477) 2025-11-26 21:17:25 +08:00
fullex
c0beab0f8a chore: update release notes for v1.7.0-rc.3
- Updated version to 1.7.0-rc.3 in package.json
- Added new features including support for Silicon provider and AIHubMix
- Consolidated bug fixes related to providers, models, UI, and settings
- Improved SDK integration with upgraded dependencies
2025-11-26 21:09:27 +08:00
chenxue
97519d96d7 feat(aihubmix): support nano banana (#11476)
support nano banana
2025-11-26 20:51:52 +08:00
Phantom
cbf1d461f0 fix(i18n): clean up translation tags and untranslated strings (#11471)
fix(i18n): update translation strings in ja-jp and ru-ru files

Remove unnecessary translate_input tags and fix incorrect translations
2025-11-26 20:08:04 +08:00
SuYao
bed55c418d fix: silicon provider code list (#11474) 2025-11-26 19:59:57 +08:00
Copilot
82ef4a32eb Fix Poe API reasoning parameters for GPT-5 and reasoning models (#11379)
* Initial plan

* feat: Add proper Poe API reasoning parameters support for GPT-5 and other models

Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>

* test: Add comprehensive tests for Poe API reasoning support

Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>

* fix: Add missing isGPT5SeriesModel import in reasoning.ts

Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>

* fix: Use correct extra_body format for Poe API reasoning parameters

Per Poe API documentation, custom bot parameters like reasoning_effort
and thinking_budget should be passed directly in extra_body, not as
nested structures.

Changed from:
- reasoning_effort: 'low' -> extra_body: { reasoning_effort: 'low' }
- thinking: { type: 'enabled', budget_tokens: X } -> extra_body: { thinking_budget: X }
- extra_body: { google: { thinking_config: {...} } } -> extra_body: { thinking_budget: X }

Updated tests to match the corrected implementation.

Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>

* fix: Update reasoning parameters and improve type definitions for GPT-5 support

* fix lint

* docs

* fix(reasoning): handle edge cases for models without token limit configuration

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>
Co-authored-by: suyao <sy20010504@gmail.com>
2025-11-26 19:56:31 +08:00
槑囿脑袋
79f75843a7 fix: get quota and quota tips (#11472)
Co-authored-by: Claude <noreply@anthropic.com>
2025-11-26 19:53:59 +08:00
Phantom
91f0c47b33 fix(anthropic): prevent duplicate /v1 in API endpoints (#11467)
* fix(anthropic): prevent duplicate /v1 in API endpoints

Anthropic SDK automatically appends /v1 to endpoints, so we should not add it in our formatting. This change ensures URLs are correctly formatted without duplicate path segments.

* fix(anthropic): strip /v1 suffix in getSdkClient to prevent duplicate in models endpoint

The issue was:
- AI SDK (for chat) needs baseURL with /v1 suffix
- Anthropic SDK (for listModels) automatically appends /v1 to all endpoints

Solution:
- Keep /v1 in formatProviderApiHost for AI SDK compatibility
- Strip /v1 in getSdkClient before passing to Anthropic SDK
- This ensures chat works correctly while preventing /v1/v1/models duplication

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* fix(anthropic): correct preview URL to match actual request behavior

The preview now correctly shows:
- Input: https://api.siliconflow.cn/v2
- Preview: https://api.siliconflow.cn/v2/messages (was incorrectly showing /v2/v1/messages)
- Actual: https://api.siliconflow.cn/v2/messages

This matches the actual behavior where getSdkClient strips /v1 suffix before
passing to Anthropic SDK, which then appends /v1/messages.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* fix(anthropic): strip all API version suffixes, not just /v1

The Anthropic SDK always appends /v1 to endpoints, regardless of the baseURL.
Previously we only stripped /v1 suffix, causing issues with custom versions like /v2.

Now we strip all version suffixes (/v1, /v2, /v1beta, etc.) before passing to Anthropic SDK.

Examples:
- Input: https://api.siliconflow.cn/v2/
- After strip: https://api.siliconflow.cn
- Actual request: https://api.siliconflow.cn/v1/messages 

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* fix(anthropic): correct preview to show AI SDK behavior, not Anthropic SDK

The preview was showing the wrong URL because it was reflecting Anthropic SDK behavior
(which strips versions and uses /v1), but checkApi and chat use AI SDK which preserves
the user's version path.

Now preview correctly shows:
- Input: https://api.siliconflow.cn/v2/
- AI SDK (checkApi/chat): https://api.siliconflow.cn/v2/messages 
- Preview: https://api.siliconflow.cn/v2/messages 

Note: Anthropic SDK (for listModels) still strips versions to use /v1/models,
but this is not shown in preview since it's a different code path.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* refactor(checkApi): remove unnecessary legacy fallback

The legacy fallback logic in checkApi was:
1. Complex and hard to maintain
2. Never actually triggered in practice for Modern SDK supported providers
3. Could cause duplicate API requests

Since Modern AI SDK now handles all major providers correctly,
we can simplify by directly throwing errors instead of falling back.

This also removes unused imports: AiProvider and CompletionsParams.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* fix(anthropic): restore version stripping in getSdkClient for Anthropic SDK

The Anthropic SDK (used for listModels) always appends /v1 to endpoints,
so we need to strip version suffixes from baseURL to avoid duplication.

This only affects Anthropic SDK operations (like listModels).
AI SDK operations (chat/checkApi) use provider.apiHost directly via
providerToAiSdkConfig, which preserves the user's version path.

Examples:
- AI SDK (chat): https://api.siliconflow.cn/v1 -> /v1/messages 
- Anthropic SDK (models): https://api.siliconflow.cn/v1 -> strip v1 -> /v1/models 

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* fix(anthropic): ensure AI SDK gets /v1 in baseURL, strip for Anthropic SDK

The correct behavior is:
1. formatProviderApiHost: Add /v1 to apiHost (for AI SDK compatibility)
2. AI SDK (chat/checkApi): Use apiHost with /v1 -> /v1/messages 
3. Anthropic SDK (listModels): Strip /v1 from baseURL -> SDK adds /v1/models 
4. Preview: Show AI SDK behavior (main use case) -> /v1/messages 

Examples:
- Input: https://api.siliconflow.cn
- Formatted: https://api.siliconflow.cn/v1 (added by formatApiHost)
- AI SDK: https://api.siliconflow.cn/v1/messages 
- Anthropic SDK: https://api.siliconflow.cn (stripped) + /v1/models 
- Preview: https://api.siliconflow.cn/v1/messages 

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* refactor(ai): simplify AiProviderNew initialization and improve docs

Update AiProviderNew constructor to automatically format URLs by default
Add comprehensive documentation explaining constructor behavior and usage

* chore: remove unused play.ts file

* fix(anthropic): strip api version from baseURL to avoid endpoint duplication

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-11-26 19:26:39 +08:00
SuYao
28dff9dfe3 feat: add silicon provider support for Anthropic API compatibility (#11468)
* feat: add silicon provider support for Anthropic API compatibility

* fix: update handling of ANTHROPIC_BASE_URL for silicon provider compatibility

* fix: update anthropicApiHost for silicon provider to use the correct endpoint

* fix: remove silicon from CLAUDE_OFFICIAL_SUPPORTED_PROVIDERS

* chore: add comment to clarify silicon model fallback logic in CLAUDE_OFFICIAL_SUPPORTED_PROVIDERS
2025-11-26 19:19:34 +08:00
fullex
155930ecf4 chore: update @types/react and @types/react-dom to latest versions
- Bumped @types/react from ^19.0.12 to ^19.2.7
- Bumped @types/react-dom from ^19.0.4 to ^19.2.3
- Updated csstype dependency from ^3.0.2 to ^3.2.2 in yarn.lock

These updates ensure compatibility with the latest React types and improve type definitions.
2025-11-26 16:05:40 +08:00
Shuchen Luo
b6b999b635 fix: add claude-opus-4-5 pattern to THINKING_TOKEN_MAP (#11457)
* fix: add claude-opus-4-5 pattern to THINKING_TOKEN_MAP

Adds missing regex pattern for claude-opus-4-5 models (e.g., claude-opus-4-5-20251101)
to the THINKING_TOKEN_MAP configuration. Without this pattern, the model was not
recognized, causing findTokenLimit() to return undefined and leading to an
AI_InvalidArgumentError when using Google Vertex AI Anthropic provider.

The fix adds the pattern 'claude-opus-4-5.*$': { min: 1024, max: 64_000 } to
match the existing claude-4 thinking token configuration.

Fixes AI_InvalidArgumentError: invalid anthropic provider options caused by
budgetTokens receiving NaN instead of a number.

Signed-off-by: Shuchen Luo (personal linux) <nemo0806@gmail.com>

* refactor: make THINKING_TOKEN_MAP constant private

* fix(reasoning): update claude model token limit regex patterns

- Consolidate claude model regex patterns to be more consistent
- Add comprehensive test cases for various claude model variants
- Ensure case insensitivity and proper handling of edge cases

* fix: format

* feat(models): extend claude model regex patterns to support AWS and GCP formats

Update regex patterns in THINKING_TOKEN_MAP to support additional Claude model ID formats used in AWS Bedrock and GCP Vertex AI
Add comprehensive test cases for new model ID formats and reorganize test suite

* fix: format

---------

Signed-off-by: Shuchen Luo (personal linux) <nemo0806@gmail.com>
Co-authored-by: icarus <eurfelux@gmail.com>
2025-11-26 15:47:14 +08:00
SuYao
0d69eeaccf fix: improve Gemini reasoning and message handling (#11439)
* fix: some bug

* fix/test

* fix: lint

* fix: 添加跳过 Gemini3 思考签名的中间件并更新消息转换逻辑

* fix: comment

* fix: js docs

* fix:id bug

* fix: condition

* fix: Update the user's verbosity setting logic to ensure that supported options are prioritized for use.

* fix: Add support for the 'openai-response' provider type.

* fix: lint
2025-11-26 15:46:52 +08:00
Phantom
ff48ce0a58 docs: enhance CLAUDE.md with quality guidelines (#11464)
* docs: add linting and testing step to completion guidelines

* docs: update CLAUDE.md with PR template guideline
2025-11-26 15:45:43 +08:00
SuYao
a2de7d48be fix: update Azure provider handling in AI SDK integration (#11465) 2025-11-26 15:43:32 +08:00
fullex
d4396b4890 docs: update links in Chinese contributing guide
- Corrected the paths in the Chinese version of the contributing guide to point to the appropriate documentation locations.
2025-11-26 13:21:10 +08:00
fullex
283519f1fd Merge branch 'main' of github.com:CherryHQ/cherry-studio 2025-11-26 13:17:09 +08:00
fullex
bb41709ce8 docs: update docs directory structure
- Updated links in CONTRIBUTING.md and README.md to point to the correct Chinese documentation paths.
- Removed outdated files including the English and Chinese versions of the branching strategy, contributing guide, and test plan documents.
- Cleaned up references to non-existent documentation in the project structure to streamline the contributor experience.
2025-11-26 13:17:01 +08:00
Copilot
c1f4b5b9b9 Fix: custom parameters for Gemini models (#11456)
* Initial plan

* fix(aiCore): extract AI SDK standard params from custom params for Gemini

Custom parameters like topK, frequencyPenalty, presencePenalty,
stopSequences, and seed should be passed as top-level streamText()
parameters, not in providerOptions. This fixes the issue where these
parameters were being ignored by the AI SDK's @ai-sdk/google module.

Changes:
- Add extractAiSdkStandardParams function to separate standard params
- Update buildProviderOptions to return both providerOptions and standardParams
- Update buildStreamTextParams to spread standardParams into params object
- Update tests to reflect new return structure

Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>

* refactor(aiCore): remove extractAiSdkStandardParams function and its tests, streamline parameter extraction logic

* chore: type

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: DeJeune <67425183+DeJeune@users.noreply.github.com>
Co-authored-by: suyao <sy20010504@gmail.com>
2025-11-26 13:16:58 +08:00
SuYao
5fb59d21ec fix: header merging logic via chore ai-sdk (#11443)
* fix: update provider-utils and add patch for header merging logic

* fix: enhance header merging logic to deduplicate values

* fix: handle null values in header merging logic

* chore: update ai-sdk dependencies and remove obsolete patches

- Updated @ai-sdk/amazon-bedrock from 3.0.56 to 3.0.61
- Updated @ai-sdk/anthropic from 2.0.45 to 2.0.49
- Updated @ai-sdk/gateway from 2.0.13 to 2.0.15
- Updated @ai-sdk/google from 2.0.40 to 2.0.43
- Updated @ai-sdk/google-vertex from 3.0.72 to 3.0.79
- Updated @ai-sdk/openai from 2.0.71 to 2.0.72
- Updated @ai-sdk/provider-utils from patch version to 3.0.17
- Removed obsolete patches for @ai-sdk/openai and @ai-sdk/provider-utils
- Added reasoning_content field to OpenAIChat response and chunk schemas
- Enhanced OpenAIChatLanguageModel to handle reasoning content in responses

* chore
2025-11-26 12:31:55 +08:00
Phantom
e8de31ca64 fix: Groq verbosity setting (#11452)
* feat(settings): show OpenAI settings for supported service tier providers

Add support for displaying OpenAI settings when provider supports service tiers.
This includes refactoring the condition check and fixing variable naming consistency.

* fix(settings): set openAI verbosity to undefined by default

* fix(store): bump version to 178 and disable verbosity for groq provider

Add migration to remove verbosity from groq provider and implement provider utility to check verbosity support
Update provider types to include verbosity support flag

* feat(provider): add verbosity option support for providers

Add verbosity parameter support in provider API options settings

* fix(aiCore): check provider support for verbosity before applying

Add provider validation and check for verbosity support to prevent errors when unsupported providers are used with verbosity settings

* feat(settings): add Groq settings group component and translations

add new GroqSettingsGroup component for managing Groq provider settings
update translations for Groq settings in both zh-cn and en-us locales
refactor OpenAISettingsGroup to separate Groq-specific logic

* feat(i18n): add groq settings and verbosity support translations

add translations for groq settings title and verbosity parameter support in multiple languages

* refactor(settings): simplify service tier mode fallback logic

Remove conditional service tier mode fallback and use provider-specific defaults directly

* fix(provider): remove redundant system provider check in verbosity support

* test(provider): add tests for verbosity support detection

* fix(OpenAISettingsGroup): add endpoint_type check for showSummarySetting condition

Add model.endpoint_type check to properly determine when to show summary setting for OpenAI models

* refactor(selector): simplify selector option types and add utility functions

remove undefined and null from selector option types
add utility functions to convert between option values and real values
update groq and openai settings groups to use new utilities
add new translation for "ignore" option

* fix(ApiOptionsSettings): correct checked state for verbosity toggle

* feat(i18n): add "ignore" translation for multiple languages

* refactor(groq): remove unused model prop and related checks

Clean up GroqSettingsGroup component by removing unused model prop and unnecessary service tier checks
2025-11-25 23:29:03 +08:00
Phantom
69d31a1e2b fix(models): qwen-mt-flash supports text delta (#11448)
refactor(models): improve text delta support check for qwen-mt models

Replace direct qwen-mt model check with regex pattern matching
Add comprehensive test cases for isNotSupportTextDeltaModel
Update all references to use new function name
2025-11-25 22:22:18 +08:00
fullex
fd3b7f717d fix: correct updateAssistantPreset reducer to properly update preset (#11453)
The previous implementation used `a = preset` inside forEach, which only
reassigns the local variable and doesn't actually update the array element.

Changed to use findIndex + direct array assignment to properly update
the preset in the state.

Fixes #11451

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude <noreply@anthropic.com>
2025-11-25 20:59:56 +08:00
LiuVaayne
bcd7bc9f2d ⬆️ chore: upgrade @anthropic-ai/claude-agent-sdk to 0.1.53 (#11444)
- Upgrade from 0.1.30 to 0.1.53
- Re-apply fork() patch for Electron IPC compatibility
2025-11-25 18:46:11 +08:00
kangfenmao
4dd92c3ce1 fix: handle optional provider in isSupportedReasoningEffortGrokModel function 2025-11-25 17:22:54 +08:00
SuYao
dc8df98929 fix: websearch button condition (#11440)
fix: button
2025-11-25 13:24:37 +08:00
fullex
0004a8cafe fix: respect enableMaxTokens setting when maxTokens is not configured (#11438)
* fix: respect enableMaxTokens setting when maxTokens is not configured

When enableMaxTokens is disabled, getMaxTokens() should return undefined
to let the API use its own default value, instead of forcing 4096 tokens.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* fix(modelParameters): handle max tokens when feature is disabled

Check if max tokens feature is enabled before returning undefined to ensure proper API behavior

---------

Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: icarus <eurfelux@gmail.com>
2025-11-25 11:12:50 +08:00
defi-failure
1992363580 chore: bump version to 1.7.0-rc.2 (#11429) 2025-11-24 14:46:10 +08:00
defi-failure
c901771480 chore: update release notes for v1.7.0-rc.2 (#11426) 2025-11-24 11:30:40 +08:00
SuYao
475f718efb fix: improve error handling and display in AiSdkToChunkAdapter (#11423)
* fix: improve error handling and display in AiSdkToChunkAdapter

* fix: test
2025-11-24 10:57:51 +08:00
SuYao
2c3338939e feat: update Google and OpenAI SDKs with new features and fixes (#11395)
* feat: update Google and OpenAI SDKs with new features and fixes

- Updated Google SDK to ensure model paths are correctly formatted.
- Enhanced OpenAI SDK to include support for image URLs in chat responses.
- Added reasoning content handling in OpenAI chat responses and chunks.
- Introduced Azure Anthropic provider configuration for Claude integration.

* fix: azure error

* fix: lint

* fix: test

* fix: test

* fix type

* fix comment

* fix: redundant

* chore resolution

* fix: test

* fix: comment

* fix: comment

* fix

* feat: 添加 OpenRouter 推理中间件以支持内容过滤
2025-11-23 23:18:57 +08:00
槑囿脑袋
64ca3802a4 feat: support gemini 3 pro image preview (#11416)
feat: support gemini 3 pro preview
2025-11-23 21:40:22 +08:00
Phantom
fa361126b8 refactor: aisdk config (#11402)
* refactor: improve model filtering with todo for robust conversion

* refactor(aiCore): add AiSdkConfig type and update provider config handling

- Introduce new AiSdkConfig type in aiCoreTypes for better type safety
- Update provider factory and config to use AiSdkConfig consistently
- Simplify getAiSdkProviderId return type to string
- Add config validation in ModernAiProvider

* refactor(aiCore): move ai core types to dedicated module

Consolidate AI core type definitions into a dedicated module under aiCore/types. This improves code organization by keeping related types together and removes circular dependencies between modules. The change includes:
- Moving AiSdkConfig to aiCore/types
- Updating all imports to reference the new location
- Removing duplicate type definitions

* refactor(provider): add return type to createAiSdkProvider function
2025-11-23 21:12:57 +08:00
SuYao
49903a1567 Test/ai-core (#11307)
* test: 1

* test: 2

* test: 3

* format

* chore: move provider from config to utils

* fix: 4

* test: 5

* chore: redundant logic

* test: add reasoning model tests and improve provider options typings

* chore: format

* test 6

* chore: format

* test: 7

* test: 8

* fix: test

* fix: format and typecheck

* fix error

* test: isClaude4SeriesModel

* fix: test

* fix: test

---------

Co-authored-by: defi-failure <159208748+defi-failure@users.noreply.github.com>
2025-11-23 17:33:27 +08:00
Phantom
086b16a59c ci: update PR title in auto-i18n workflow to be more specific (#11406) 2025-11-23 11:48:44 +08:00
github-actions[bot]
e2562d8224 🤖 Weekly Automated Update: Nov 23, 2025 (#11412)
feat(bot): Weekly automated script run

Co-authored-by: EurFelux <59059173+EurFelux@users.noreply.github.com>
2025-11-23 11:47:54 +08:00
Phantom
c9be949853 fix: adjacent user messages appear when assistant message contains error only (#11390)
* feat(messages): add filter for error-only messages and their related pairs

Add new filter function to remove assistant messages containing only error blocks along with their associated user messages, identified by askId. This improves conversation quality by cleaning up error-only responses.

* refactor(ConversationService): improve message filtering pipeline readability

Break down complex message filtering chain into clearly labeled steps
Add comments explaining each filtering step's purpose
Maintain same functionality while improving code maintainability

* test(messageUtils): add test cases for message filter utilities

* docs(messageUtils): correct jsdoc for filterUsefulMessages

* refactor(ConversationService): extract message filtering logic into pipeline method

Move message filtering steps into a dedicated static method to improve testability and maintainability. Add comprehensive tests to verify pipeline behavior.

* refactor(ConversationService): add logging and improve message filtering readability

Add logger service to track message pipeline output
Split filterUserRoleStartMessages into separate variable for better debugging
2025-11-22 23:00:13 +08:00
defi-failure
ebfb1c5abf fix: add missing execution state for approved tool permissions (#11394) 2025-11-22 21:45:42 +08:00
SuYao
c1f1d7996d test: add thinking budget token test (#11305)
* refactor: add thinking budget token test

* fix comment
2025-11-22 21:43:57 +08:00
Phantom
0a72c613af fix(openai): apply verbosity setting with type safety improvements (#10964)
* refactor(types): consolidate OpenAI types and improve type safety

- Move OpenAI-related types to aiCoreTypes.ts
- Rename FetchChatCompletionOptions to FetchChatCompletionRequestOptions
- Add proper type definitions for service tiers and verbosity
- Improve type guards for service tier checks

* refactor(api): rename options parameter to requestOptions for consistency

Update parameter name across multiple files to use requestOptions instead of options for better clarity and consistency in API calls

* refactor(aiCore): simplify OpenAI summary text handling and improve type safety

- Remove 'off' option from OpenAISummaryText type and use null instead
- Add migration to convert 'off' values to null
- Add utility function to convert undefined to null
- Update Selector component to handle null/undefined values
- Improve type safety in provider options and reasoning params

* fix(i18n): Auto update translations for PR #10964

* feat(utils): add notNull function to convert null to undefined

* refactor(utils): move defined and notNull functions to shared package

Consolidate utility functions into shared package to improve code organization and reuse

* Revert "fix(i18n): Auto update translations for PR #10964"

This reverts commit 68bd7eaac5.

* feat(i18n): add "off" translation and remove "performance" tier

Add "off" translation for multiple languages and remove "performance" service tier option from translations

* Apply suggestion from @EurFelux

* docs(types): clarify handling of undefined and null values

Add comments to explain that undefined is treated as default and null as explicitly off in OpenAIVerbosity and OpenAIServiceTier types. Also update type safety for OpenAIServiceTiers record.

* fix(migration): update migration version from 167 to 171 for removed type

* chore: update store version to 172

* fix(migrate): update migration version number from 171 to 172

* fix(i18n): Auto update translations for PR #10964

* refactor(types): improve type safety for verbosity handling

add NotUndefined and NotNull utility types to better handle null/undefined cases
clarify verbosity types in aiCoreTypes and update related utility functions

* refactor(types): replace null with undefined for verbosity values

Standardize on undefined instead of null for verbosity values to align with OpenAI API docs and improve type consistency

* refactor(aiCore): update OpenAI provider options type import and usage

* fix(openai): change summaryText default from null to 'auto'

Update OpenAI settings to use 'auto' as default summaryText value instead of null for consistency with API behavior. Remove 'off' option and add 'concise' option while maintaining type safety.

* refactor(OpenAISettingsGroup): extract service tier options type for better maintainability

* refactor(types): make SystemProviderIdTypeMap internal type

* docs(provider): clarify OpenAIServiceTier behavior for undefined vs null

Explain that undefined and null values for serviceTier should be treated differently since they affect whether the field appears in the response

* refactor(utils): rename utility functions for clarity

Rename `defined` to `toNullIfUndefined` and `notNull` to `toUndefinedIfNull` to better reflect their functionality

* refactor(aiCore): extract service tier logic and improve type safety

Extract service tier validation logic into separate functions for better reusability
Add proper type annotations for provider options
Pass service tier parameter through provider option builders

* refactor(utils): comment out unused utility functions

Keep commented utility functions for potential future use while cleaning up current codebase

* fix(migration): update migration version number from 172 to 177

* docs(aiCoreTypes): clarify parameter passing behavior in OpenAI API

Update comments to consistently use 'undefined' instead of 'null' when describing parameter passing behavior in OpenAI API requests, as they share the same meaning in this context

---------

Co-authored-by: GitHub Action <action@github.com>
2025-11-22 21:41:12 +08:00
SuYao
a1ac3207f1 fix/anthropic-vertex (#11397)
* 100m

* feat: add web search header for Claude 4 series models

* fix: typo

* fix: identify model

---------

Co-authored-by: defi-failure <159208748+defi-failure@users.noreply.github.com>
2025-11-22 20:56:05 +08:00
Caelan
f98a063a8f Fix the issue where base64 images cannot be saved (#11398) 2025-11-22 20:20:02 +08:00
亢奋猫
1cb2af57ae refactor: optimize DatabaseManager and fix libsql crash issues (#11392)
* refactor: optimize DatabaseManager and fix libsql crash issues

Major improvements:
- Created DatabaseManager singleton to centralize database connection management
- Auto-initialize database in constructor (no manual initialization needed)
- Removed all manual initialize() and ensureInitialized() calls (47 occurrences)
- Simplified initialization logic (removed retry loops that could cause crashes)
- Removed unused close() and reinitialize() methods
- Reduced code from ~270 lines to 172 lines (-36%)

Key changes:
1. DatabaseManager.ts (new file):
   - Singleton pattern with auto-initialization
   - State management (INITIALIZING, INITIALIZED, FAILED)
   - Windows compatibility fixes (empty file detection, intMode: 'number')
   - Simplified waitForInitialization() logic

2. BaseService.ts:
   - Removed static initialize() and ensureInitialized() methods
   - Simplified database/rawClient getters to use DatabaseManager

3. Service classes (AgentService, SessionService, SessionMessageService):
   - Removed all initialize() methods
   - Removed all ensureInitialized() calls
   - Services now work out of the box

4. Main entry points (index.ts, server.ts):
   - Removed explicit database initialization calls
   - Database initializes automatically on first access

Benefits:
- Fixes Windows libsql crashes by removing dangerous retry logic
- Simpler API - no need to remember to call initialize()
- Better separation of concerns
- Cleaner codebase with 36% less code

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* fix: wait for database initialization on app startup

Issue: "Database is still initializing" error on startup
Root cause: Synchronous database getter was called before async initialization completed

Solution:
- Explicitly wait for database initialization in main index.ts
- Import DatabaseManager and call getDatabase() to ensure initialization is complete
- This guarantees database is ready before any service methods are called

Changes:
- src/main/index.ts: Added explicit database initialization wait before API server check

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* refactor: use static import for getDatabaseManager

- Move import to top of file for better code organization
- Remove unnecessary dynamic import

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* refactor: streamline database access in service classes

- Replaced direct database access with asynchronous calls to getDatabase() in various service classes (AgentService, SessionService, SessionMessageService).
- Updated the main index.ts to utilize runAsyncFunction for API server initialization, ensuring proper handling of asynchronous database access.
- Improved code organization and readability by consolidating database access logic.

This change enhances the reliability of database interactions across the application and ensures that services are correctly initialized before use.

* refactor: remove redundant logging in ApiServer initialization

- Removed the logging statement for 'AgentService ready' during server initialization.
- This change streamlines the startup process by eliminating unnecessary log entries.

This update contributes to cleaner logs and improved readability during server startup.

* refactor: change getDatabase method to synchronous return type

- Updated the getDatabase method in DatabaseManager to return a synchronous LibSQLDatabase instance instead of a Promise.
- This change simplifies the database access pattern, aligning with the current initialization logic.

This refactor enhances code clarity and reduces unnecessary asynchronous handling in the database access layer.

* refactor: simplify sessionMessageRepository by removing transaction handling

- Removed transaction handling parameters from message persistence methods in sessionMessageRepository.
- Updated database access to use a direct call to getDatabase() instead of passing a transaction client.
- Streamlined the upsertMessage and persistExchange methods for improved clarity and reduced complexity.

This refactor enhances code readability and simplifies the database interaction logic.

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-11-22 09:12:11 +08:00
fullex
62309ae1bf fix: prevent EventEmitter memory leak in useApiServer hook (#11385)
Implement single instance IPC subscription pattern to resolve MaxListenersExceededWarning. Previously, each component using useApiServer would register a separate 'api-server:ready' listener, and React strict mode double rendering would quickly exceed the 10 listener limit.

Changes:
- Add module-level subscription manager with onReadyCallbacks Set
- Ensure only one IPC listener is registered regardless of component count
- Use useRef to maintain stable callback references
- Properly cleanup subscriptions when all components unmount

This maintains existing behavior while keeping listener count constant at 1.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude <noreply@anthropic.com>
2025-11-21 21:42:34 +08:00
defi-failure
c48f222cdb feat: add endpoint type support for cherryin provider (#11367)
* feat: add endpoint type support for cherryin provider

* chore: bump @cherrystudio/ai-sdk-provider version to 0.1.1

* chore: bump ai-sdk-provider version to 0.1.3
2025-11-21 21:42:08 +08:00
亢奋猫
cea0058f87 refactor: simplify knowledge base creation modal (#11371)
* test(knowledge): fix tests for knowledge base form modal refactoring

Update all test files to match the new vertical layout structure with button-based advanced settings toggle. Remove obsolete tests for deleted features.

Changes:
- Rewrite KnowledgeBaseFormModal.test.tsx for new button-toggle structure
- Remove tests for preprocess and rerank features from GeneralSettingsPanel
- Update AdvancedSettingsPanel tests with required props
- Update all snapshots to reflect new component structure
- Format test files according to biome rules

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* test(knowledge): simplify KnowledgeBaseFormModal button tests

Simplify button interaction tests to avoid text matching issues. Focus on testing behavior rather than implementation details.

Changes:
- Simplify advanced settings toggle test
- Simplify footer buttons test to check button count instead of text content
- Remove fragile text-based button selection

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-11-21 21:34:34 +08:00
beyondkmp
852192dce6 feat: add Git Bash detection and requirement check for Windows agents (#11388)
* feat: add Git Bash detection and requirement check for Windows agents

- Add System_CheckGitBash IPC channel for detecting Git Bash installation
- Implement detection logic checking common installation paths and PATH environment
- Display non-closable error alert in AgentModal when Git Bash is not found
- Disable agent creation/edit button until Git Bash is installed
- Add recheck functionality to verify installation without restarting app

Git Bash is required for agents to function properly on Windows systems.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* i18n: add Git Bash requirement translations for agent modal

- Add English translations for Git Bash detection warnings
- Add Simplified Chinese (zh-cn) translations
- Add Traditional Chinese (zh-tw) translations

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* format code

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-11-21 21:32:53 +08:00
Pleasure1234
eee49d1580 feat: add ChatGPT conversation import feature (#11272)
* feat: add ChatGPT conversation import feature

Introduces a new import workflow for ChatGPT conversations, including UI components, service logic, and i18n support for English, Simplified Chinese, and Traditional Chinese. Adds an import menu to data settings, a popup for file selection and progress, and a service to parse and store imported conversations as topics and messages.

* fix: ci failure

* refactor: import service and add modular importers

Refactored the import service to support a modular importer architecture. Moved ChatGPT import logic to a dedicated importer class and directory. Updated UI components and i18n descriptions for clarity. Removed unused Redux selector in ImportMenuSettings. This change enables easier addition of new importers in the future.

* Apply suggestion from @Copilot

Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>

* fix: improve ChatGPT import UX and set model for assistant

Added a loading state and spinner for file selection in the ChatGPT import popup, with new translations for the 'selecting' state in en-us, zh-cn, and zh-tw locales. Also, set the model property for imported assistant messages to display the GPT-5 logo.

---------

Co-authored-by: SuYao <sy20010504@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
2025-11-21 14:58:47 +08:00
SuYao
dcdd1bf852 refactor: replace renderToolContent function with ToolContent component for improved readability (#11300)
* refactor: replace renderToolContent function with ToolContent component for improved readability

* fix

* fix test
2025-11-21 09:55:46 +08:00
beyondkmp
a12b6bfeca feat: enable native language emoji search with CLDR data format (#11381)
* feat: add i18n support and local data to emoji picker

- Add emoji-picker-element-data package for offline-first emoji data
- Implement i18n translations for emoji picker UI (de, en, es, fr, ja, pt, ru, zh)
- Switch from CDN to local emoji data to improve performance and reliability
- Add locale mapping to match app language with emoji picker data
- Move emoji-picker-element import to EmojiPicker component for better encapsulation
- Use proper TypeScript types instead of 'any' for type safety

This improves user experience by providing localized emoji picker interface
and eliminating dependency on external CDN, ensuring the picker works offline.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

* feat: enable native language emoji search with CLDR data format

Switch from emojibase to CLDR format for emoji-picker-element data to support full multi-language search functionality. Users can now search for emojis in their native language (e.g., German users can search "Herz" for ❤️, Spanish users can search "corazón"). Also improves type safety by using the LanguageVarious type for locale mappings.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-11-20 19:23:27 +08:00
242 changed files with 17478 additions and 4063 deletions

View File

@@ -77,7 +77,7 @@ jobs:
with:
token: ${{ secrets.GITHUB_TOKEN }} # Use the built-in GITHUB_TOKEN for bot actions
commit-message: "feat(bot): Weekly automated script run"
title: "🤖 Weekly Automated Update: ${{ env.CURRENT_DATE }}"
title: "🤖 Weekly Auto I18N Sync: ${{ env.CURRENT_DATE }}"
body: |
This PR includes changes generated by the weekly auto i18n.
Review the changes before merging.

View File

@@ -1,152 +0,0 @@
diff --git a/dist/index.js b/dist/index.js
index c2ef089c42e13a8ee4a833899a415564130e5d79..75efa7baafb0f019fb44dd50dec1641eee8879e7 100644
--- a/dist/index.js
+++ b/dist/index.js
@@ -471,7 +471,7 @@ function convertToGoogleGenerativeAIMessages(prompt, options) {
// src/get-model-path.ts
function getModelPath(modelId) {
- return modelId.includes("/") ? modelId : `models/${modelId}`;
+ return modelId.includes("models/") ? modelId : `models/${modelId}`;
}
// src/google-generative-ai-options.ts
diff --git a/dist/index.mjs b/dist/index.mjs
index d75c0cc13c41192408c1f3f2d29d76a7bffa6268..ada730b8cb97d9b7d4cb32883a1d1ff416404d9b 100644
--- a/dist/index.mjs
+++ b/dist/index.mjs
@@ -477,7 +477,7 @@ function convertToGoogleGenerativeAIMessages(prompt, options) {
// src/get-model-path.ts
function getModelPath(modelId) {
- return modelId.includes("/") ? modelId : `models/${modelId}`;
+ return modelId.includes("models/") ? modelId : `models/${modelId}`;
}
// src/google-generative-ai-options.ts
diff --git a/dist/internal/index.js b/dist/internal/index.js
index 277cac8dc734bea2fb4f3e9a225986b402b24f48..bb704cd79e602eb8b0cee1889e42497d59ccdb7a 100644
--- a/dist/internal/index.js
+++ b/dist/internal/index.js
@@ -432,7 +432,15 @@ function prepareTools({
var _a;
tools = (tools == null ? void 0 : tools.length) ? tools : void 0;
const toolWarnings = [];
- const isGemini2 = modelId.includes("gemini-2");
+ // These changes could be safely removed when @ai-sdk/google v3 released.
+ const isLatest = (
+ [
+ 'gemini-flash-latest',
+ 'gemini-flash-lite-latest',
+ 'gemini-pro-latest',
+ ]
+ ).some(id => id === modelId);
+ const isGemini2OrNewer = modelId.includes("gemini-2") || modelId.includes("gemini-3") || isLatest;
const supportsDynamicRetrieval = modelId.includes("gemini-1.5-flash") && !modelId.includes("-8b");
const supportsFileSearch = modelId.includes("gemini-2.5");
if (tools == null) {
@@ -458,7 +466,7 @@ function prepareTools({
providerDefinedTools.forEach((tool) => {
switch (tool.id) {
case "google.google_search":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({ googleSearch: {} });
} else if (supportsDynamicRetrieval) {
googleTools2.push({
@@ -474,7 +482,7 @@ function prepareTools({
}
break;
case "google.url_context":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({ urlContext: {} });
} else {
toolWarnings.push({
@@ -485,7 +493,7 @@ function prepareTools({
}
break;
case "google.code_execution":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({ codeExecution: {} });
} else {
toolWarnings.push({
@@ -507,7 +515,7 @@ function prepareTools({
}
break;
case "google.vertex_rag_store":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({
retrieval: {
vertex_rag_store: {
diff --git a/dist/internal/index.mjs b/dist/internal/index.mjs
index 03b7cc591be9b58bcc2e775a96740d9f98862a10..347d2c12e1cee79f0f8bb258f3844fb0522a6485 100644
--- a/dist/internal/index.mjs
+++ b/dist/internal/index.mjs
@@ -424,7 +424,15 @@ function prepareTools({
var _a;
tools = (tools == null ? void 0 : tools.length) ? tools : void 0;
const toolWarnings = [];
- const isGemini2 = modelId.includes("gemini-2");
+ // These changes could be safely removed when @ai-sdk/google v3 released.
+ const isLatest = (
+ [
+ 'gemini-flash-latest',
+ 'gemini-flash-lite-latest',
+ 'gemini-pro-latest',
+ ]
+ ).some(id => id === modelId);
+ const isGemini2OrNewer = modelId.includes("gemini-2") || modelId.includes("gemini-3") || isLatest;
const supportsDynamicRetrieval = modelId.includes("gemini-1.5-flash") && !modelId.includes("-8b");
const supportsFileSearch = modelId.includes("gemini-2.5");
if (tools == null) {
@@ -450,7 +458,7 @@ function prepareTools({
providerDefinedTools.forEach((tool) => {
switch (tool.id) {
case "google.google_search":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({ googleSearch: {} });
} else if (supportsDynamicRetrieval) {
googleTools2.push({
@@ -466,7 +474,7 @@ function prepareTools({
}
break;
case "google.url_context":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({ urlContext: {} });
} else {
toolWarnings.push({
@@ -477,7 +485,7 @@ function prepareTools({
}
break;
case "google.code_execution":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({ codeExecution: {} });
} else {
toolWarnings.push({
@@ -499,7 +507,7 @@ function prepareTools({
}
break;
case "google.vertex_rag_store":
- if (isGemini2) {
+ if (isGemini2OrNewer) {
googleTools2.push({
retrieval: {
vertex_rag_store: {
@@ -1434,9 +1442,7 @@ var googleTools = {
vertexRagStore
};
export {
- GoogleGenerativeAILanguageModel,
getGroundingMetadataSchema,
- getUrlContextMetadataSchema,
- googleTools
+ getUrlContextMetadataSchema, GoogleGenerativeAILanguageModel, googleTools
};
//# sourceMappingURL=index.mjs.map
\ No newline at end of file

View File

@@ -0,0 +1,26 @@
diff --git a/dist/index.js b/dist/index.js
index 51ce7e423934fb717cb90245cdfcdb3dae6780e6..0f7f7009e2f41a79a8669d38c8a44867bbff5e1f 100644
--- a/dist/index.js
+++ b/dist/index.js
@@ -474,7 +474,7 @@ function convertToGoogleGenerativeAIMessages(prompt, options) {
// src/get-model-path.ts
function getModelPath(modelId) {
- return modelId.includes("/") ? modelId : `models/${modelId}`;
+ return modelId.includes("models/") ? modelId : `models/${modelId}`;
}
// src/google-generative-ai-options.ts
diff --git a/dist/index.mjs b/dist/index.mjs
index f4b77e35c0cbfece85a3ef0d4f4e67aa6dde6271..8d2fecf8155a226006a0bde72b00b6036d4014b6 100644
--- a/dist/index.mjs
+++ b/dist/index.mjs
@@ -480,7 +480,7 @@ function convertToGoogleGenerativeAIMessages(prompt, options) {
// src/get-model-path.ts
function getModelPath(modelId) {
- return modelId.includes("/") ? modelId : `models/${modelId}`;
+ return modelId.includes("models/") ? modelId : `models/${modelId}`;
}
// src/google-generative-ai-options.ts

View File

@@ -1,131 +0,0 @@
diff --git a/dist/index.mjs b/dist/index.mjs
index b3f018730a93639aad7c203f15fb1aeb766c73f4..ade2a43d66e9184799d072153df61ef7be4ea110 100644
--- a/dist/index.mjs
+++ b/dist/index.mjs
@@ -296,7 +296,14 @@ var HuggingFaceResponsesLanguageModel = class {
metadata: huggingfaceOptions == null ? void 0 : huggingfaceOptions.metadata,
instructions: huggingfaceOptions == null ? void 0 : huggingfaceOptions.instructions,
...preparedTools && { tools: preparedTools },
- ...preparedToolChoice && { tool_choice: preparedToolChoice }
+ ...preparedToolChoice && { tool_choice: preparedToolChoice },
+ ...(huggingfaceOptions?.reasoningEffort != null && {
+ reasoning: {
+ ...(huggingfaceOptions?.reasoningEffort != null && {
+ effort: huggingfaceOptions.reasoningEffort,
+ }),
+ },
+ }),
};
return { args: baseArgs, warnings };
}
@@ -365,6 +372,20 @@ var HuggingFaceResponsesLanguageModel = class {
}
break;
}
+ case 'reasoning': {
+ for (const contentPart of part.content) {
+ content.push({
+ type: 'reasoning',
+ text: contentPart.text,
+ providerMetadata: {
+ huggingface: {
+ itemId: part.id,
+ },
+ },
+ });
+ }
+ break;
+ }
case "mcp_call": {
content.push({
type: "tool-call",
@@ -519,6 +540,11 @@ var HuggingFaceResponsesLanguageModel = class {
id: value.item.call_id,
toolName: value.item.name
});
+ } else if (value.item.type === 'reasoning') {
+ controller.enqueue({
+ type: 'reasoning-start',
+ id: value.item.id,
+ });
}
return;
}
@@ -570,6 +596,22 @@ var HuggingFaceResponsesLanguageModel = class {
});
return;
}
+ if (isReasoningDeltaChunk(value)) {
+ controller.enqueue({
+ type: 'reasoning-delta',
+ id: value.item_id,
+ delta: value.delta,
+ });
+ return;
+ }
+
+ if (isReasoningEndChunk(value)) {
+ controller.enqueue({
+ type: 'reasoning-end',
+ id: value.item_id,
+ });
+ return;
+ }
},
flush(controller) {
controller.enqueue({
@@ -593,7 +635,8 @@ var HuggingFaceResponsesLanguageModel = class {
var huggingfaceResponsesProviderOptionsSchema = z2.object({
metadata: z2.record(z2.string(), z2.string()).optional(),
instructions: z2.string().optional(),
- strictJsonSchema: z2.boolean().optional()
+ strictJsonSchema: z2.boolean().optional(),
+ reasoningEffort: z2.string().optional(),
});
var huggingfaceResponsesResponseSchema = z2.object({
id: z2.string(),
@@ -727,12 +770,31 @@ var responseCreatedChunkSchema = z2.object({
model: z2.string()
})
});
+var reasoningTextDeltaChunkSchema = z2.object({
+ type: z2.literal('response.reasoning_text.delta'),
+ item_id: z2.string(),
+ output_index: z2.number(),
+ content_index: z2.number(),
+ delta: z2.string(),
+ sequence_number: z2.number(),
+});
+
+var reasoningTextEndChunkSchema = z2.object({
+ type: z2.literal('response.reasoning_text.done'),
+ item_id: z2.string(),
+ output_index: z2.number(),
+ content_index: z2.number(),
+ text: z2.string(),
+ sequence_number: z2.number(),
+});
var huggingfaceResponsesChunkSchema = z2.union([
responseOutputItemAddedSchema,
responseOutputItemDoneSchema,
textDeltaChunkSchema,
responseCompletedChunkSchema,
responseCreatedChunkSchema,
+ reasoningTextDeltaChunkSchema,
+ reasoningTextEndChunkSchema,
z2.object({ type: z2.string() }).loose()
// fallback for unknown chunks
]);
@@ -751,6 +813,12 @@ function isResponseCompletedChunk(chunk) {
function isResponseCreatedChunk(chunk) {
return chunk.type === "response.created";
}
+function isReasoningDeltaChunk(chunk) {
+ return chunk.type === 'response.reasoning_text.delta';
+}
+function isReasoningEndChunk(chunk) {
+ return chunk.type === 'response.reasoning_text.done';
+}
// src/huggingface-provider.ts
function createHuggingFace(options = {}) {

View File

@@ -0,0 +1,140 @@
diff --git a/dist/index.js b/dist/index.js
index 73045a7d38faafdc7f7d2cd79d7ff0e2b031056b..8d948c9ac4ea4b474db9ef3c5491961e7fcf9a07 100644
--- a/dist/index.js
+++ b/dist/index.js
@@ -421,6 +421,17 @@ var OpenAICompatibleChatLanguageModel = class {
text: reasoning
});
}
+ if (choice.message.images) {
+ for (const image of choice.message.images) {
+ const match1 = image.image_url.url.match(/^data:([^;]+)/)
+ const match2 = image.image_url.url.match(/^data:[^;]*;base64,(.+)$/);
+ content.push({
+ type: 'file',
+ mediaType: match1 ? (match1[1] ?? 'image/jpeg') : 'image/jpeg',
+ data: match2 ? match2[1] : image.image_url.url,
+ });
+ }
+ }
if (choice.message.tool_calls != null) {
for (const toolCall of choice.message.tool_calls) {
content.push({
@@ -598,6 +609,17 @@ var OpenAICompatibleChatLanguageModel = class {
delta: delta.content
});
}
+ if (delta.images) {
+ for (const image of delta.images) {
+ const match1 = image.image_url.url.match(/^data:([^;]+)/)
+ const match2 = image.image_url.url.match(/^data:[^;]*;base64,(.+)$/);
+ controller.enqueue({
+ type: 'file',
+ mediaType: match1 ? (match1[1] ?? 'image/jpeg') : 'image/jpeg',
+ data: match2 ? match2[1] : image.image_url.url,
+ });
+ }
+ }
if (delta.tool_calls != null) {
for (const toolCallDelta of delta.tool_calls) {
const index = toolCallDelta.index;
@@ -765,6 +787,14 @@ var OpenAICompatibleChatResponseSchema = import_v43.z.object({
arguments: import_v43.z.string()
})
})
+ ).nullish(),
+ images: import_v43.z.array(
+ import_v43.z.object({
+ type: import_v43.z.literal('image_url'),
+ image_url: import_v43.z.object({
+ url: import_v43.z.string(),
+ })
+ })
).nullish()
}),
finish_reason: import_v43.z.string().nullish()
@@ -795,6 +825,14 @@ var createOpenAICompatibleChatChunkSchema = (errorSchema) => import_v43.z.union(
arguments: import_v43.z.string().nullish()
})
})
+ ).nullish(),
+ images: import_v43.z.array(
+ import_v43.z.object({
+ type: import_v43.z.literal('image_url'),
+ image_url: import_v43.z.object({
+ url: import_v43.z.string(),
+ })
+ })
).nullish()
}).nullish(),
finish_reason: import_v43.z.string().nullish()
diff --git a/dist/index.mjs b/dist/index.mjs
index 1c2b9560bbfbfe10cb01af080aeeed4ff59db29c..2c8ddc4fc9bfc5e7e06cfca105d197a08864c427 100644
--- a/dist/index.mjs
+++ b/dist/index.mjs
@@ -405,6 +405,17 @@ var OpenAICompatibleChatLanguageModel = class {
text: reasoning
});
}
+ if (choice.message.images) {
+ for (const image of choice.message.images) {
+ const match1 = image.image_url.url.match(/^data:([^;]+)/)
+ const match2 = image.image_url.url.match(/^data:[^;]*;base64,(.+)$/);
+ content.push({
+ type: 'file',
+ mediaType: match1 ? (match1[1] ?? 'image/jpeg') : 'image/jpeg',
+ data: match2 ? match2[1] : image.image_url.url,
+ });
+ }
+ }
if (choice.message.tool_calls != null) {
for (const toolCall of choice.message.tool_calls) {
content.push({
@@ -582,6 +593,17 @@ var OpenAICompatibleChatLanguageModel = class {
delta: delta.content
});
}
+ if (delta.images) {
+ for (const image of delta.images) {
+ const match1 = image.image_url.url.match(/^data:([^;]+)/)
+ const match2 = image.image_url.url.match(/^data:[^;]*;base64,(.+)$/);
+ controller.enqueue({
+ type: 'file',
+ mediaType: match1 ? (match1[1] ?? 'image/jpeg') : 'image/jpeg',
+ data: match2 ? match2[1] : image.image_url.url,
+ });
+ }
+ }
if (delta.tool_calls != null) {
for (const toolCallDelta of delta.tool_calls) {
const index = toolCallDelta.index;
@@ -749,6 +771,14 @@ var OpenAICompatibleChatResponseSchema = z3.object({
arguments: z3.string()
})
})
+ ).nullish(),
+ images: z3.array(
+ z3.object({
+ type: z3.literal('image_url'),
+ image_url: z3.object({
+ url: z3.string(),
+ })
+ })
).nullish()
}),
finish_reason: z3.string().nullish()
@@ -779,6 +809,14 @@ var createOpenAICompatibleChatChunkSchema = (errorSchema) => z3.union([
arguments: z3.string().nullish()
})
})
+ ).nullish(),
+ images: z3.array(
+ z3.object({
+ type: z3.literal('image_url'),
+ image_url: z3.object({
+ url: z3.string(),
+ })
+ })
).nullish()
}).nullish(),
finish_reason: z3.string().nullish()

View File

@@ -1,5 +1,5 @@
diff --git a/dist/index.js b/dist/index.js
index 992c85ac6656e51c3471af741583533c5a7bf79f..83c05952a07aebb95fc6c62f9ddb8aa96b52ac0d 100644
index bf900591bf2847a3253fe441aad24c06da19c6c1..c1d9bb6fefa2df1383339324073db0a70ea2b5a2 100644
--- a/dist/index.js
+++ b/dist/index.js
@@ -274,6 +274,7 @@ var openaiChatResponseSchema = (0, import_provider_utils3.lazyValidator)(
@@ -18,7 +18,7 @@ index 992c85ac6656e51c3471af741583533c5a7bf79f..83c05952a07aebb95fc6c62f9ddb8aa9
tool_calls: import_v42.z.array(
import_v42.z.object({
index: import_v42.z.number(),
@@ -785,6 +787,13 @@ var OpenAIChatLanguageModel = class {
@@ -795,6 +797,13 @@ var OpenAIChatLanguageModel = class {
if (text != null && text.length > 0) {
content.push({ type: "text", text });
}
@@ -32,7 +32,7 @@ index 992c85ac6656e51c3471af741583533c5a7bf79f..83c05952a07aebb95fc6c62f9ddb8aa9
for (const toolCall of (_a = choice.message.tool_calls) != null ? _a : []) {
content.push({
type: "tool-call",
@@ -866,6 +875,7 @@ var OpenAIChatLanguageModel = class {
@@ -876,6 +885,7 @@ var OpenAIChatLanguageModel = class {
};
let metadataExtracted = false;
let isActiveText = false;
@@ -40,7 +40,7 @@ index 992c85ac6656e51c3471af741583533c5a7bf79f..83c05952a07aebb95fc6c62f9ddb8aa9
const providerMetadata = { openai: {} };
return {
stream: response.pipeThrough(
@@ -923,6 +933,21 @@ var OpenAIChatLanguageModel = class {
@@ -933,6 +943,21 @@ var OpenAIChatLanguageModel = class {
return;
}
const delta = choice.delta;
@@ -62,7 +62,7 @@ index 992c85ac6656e51c3471af741583533c5a7bf79f..83c05952a07aebb95fc6c62f9ddb8aa9
if (delta.content != null) {
if (!isActiveText) {
controller.enqueue({ type: "text-start", id: "0" });
@@ -1035,6 +1060,9 @@ var OpenAIChatLanguageModel = class {
@@ -1045,6 +1070,9 @@ var OpenAIChatLanguageModel = class {
}
},
flush(controller) {

View File

@@ -1,8 +1,8 @@
diff --git a/sdk.mjs b/sdk.mjs
index 8cc6aaf0b25bcdf3c579ec95cde12d419fcb2a71..3b3b8beaea5ad2bbac26a15f792058306d0b059f 100755
index bf429a344b7d59f70aead16b639f949b07688a81..f77d50cc5d3fb04292cb3ac7fa7085d02dcc628f 100755
--- a/sdk.mjs
+++ b/sdk.mjs
@@ -6213,7 +6213,7 @@ function createAbortController(maxListeners = DEFAULT_MAX_LISTENERS) {
@@ -6250,7 +6250,7 @@ function createAbortController(maxListeners = DEFAULT_MAX_LISTENERS) {
}
// ../src/transport/ProcessTransport.ts
@@ -11,16 +11,20 @@ index 8cc6aaf0b25bcdf3c579ec95cde12d419fcb2a71..3b3b8beaea5ad2bbac26a15f79205830
import { createInterface } from "readline";
// ../src/utils/fsOperations.ts
@@ -6505,14 +6505,11 @@ class ProcessTransport {
@@ -6619,18 +6619,11 @@ class ProcessTransport {
const errorMessage = isNativeBinary(pathToClaudeCodeExecutable) ? `Claude Code native binary not found at ${pathToClaudeCodeExecutable}. Please ensure Claude Code is installed via native installer or specify a valid path with options.pathToClaudeCodeExecutable.` : `Claude Code executable not found at ${pathToClaudeCodeExecutable}. Is options.pathToClaudeCodeExecutable set?`;
throw new ReferenceError(errorMessage);
}
- const isNative = isNativeBinary(pathToClaudeCodeExecutable);
- const spawnCommand = isNative ? pathToClaudeCodeExecutable : executable;
- const spawnArgs = isNative ? [...executableArgs, ...args] : [...executableArgs, pathToClaudeCodeExecutable, ...args];
- this.logForDebugging(isNative ? `Spawning Claude Code native binary: ${spawnCommand} ${spawnArgs.join(" ")}` : `Spawning Claude Code process: ${spawnCommand} ${spawnArgs.join(" ")}`);
+ this.logForDebugging(`Forking Claude Code Node.js process: ${pathToClaudeCodeExecutable} ${args.join(" ")}`);
const stderrMode = env.DEBUG || stderr ? "pipe" : "ignore";
- const spawnMessage = isNative ? `Spawning Claude Code native binary: ${spawnCommand} ${spawnArgs.join(" ")}` : `Spawning Claude Code process: ${spawnCommand} ${spawnArgs.join(" ")}`;
- logForSdkDebugging(spawnMessage);
- if (stderr) {
- stderr(spawnMessage);
- }
+ logForSdkDebugging(`Forking Claude Code Node.js process: ${pathToClaudeCodeExecutable} ${args.join(" ")}`);
const stderrMode = env.DEBUG_CLAUDE_AGENT_SDK || stderr ? "pipe" : "ignore";
- this.child = spawn(spawnCommand, spawnArgs, {
+ this.child = fork(pathToClaudeCodeExecutable, args, {
cwd,

View File

@@ -10,7 +10,9 @@ This file provides guidance to AI coding assistants when working with code in th
- **Log centrally**: Route all logging through `loggerService` with the right context—no `console.log`.
- **Research via subagent**: Lean on `subagent` for external docs, APIs, news, and references.
- **Always propose before executing**: Before making any changes, clearly explain your planned approach and wait for explicit user approval to ensure alignment and prevent unwanted modifications.
- **Lint, test, and format before completion**: Coding tasks are only complete after running `yarn lint`, `yarn test`, and `yarn format` successfully.
- **Write conventional commits**: Commit small, focused changes using Conventional Commit messages (e.g., `feat:`, `fix:`, `refactor:`, `docs:`).
- **Follow PR template**: When submitting pull requests, follow the template in `.github/pull_request_template.md` to ensure complete context and documentation.
## Development Commands

View File

@@ -1,4 +1,4 @@
[中文](docs/CONTRIBUTING.zh.md) | [English](CONTRIBUTING.md)
[中文](docs/zh/guides/contributing.md) | [English](CONTRIBUTING.md)
# Cherry Studio Contributor Guide
@@ -32,7 +32,7 @@ To help you get familiar with the codebase, we recommend tackling issues tagged
### Testing
Features without tests are considered non-existent. To ensure code is truly effective, relevant processes should be covered by unit tests and functional tests. Therefore, when considering contributions, please also consider testability. All tests can be run locally without dependency on CI. Please refer to the "Testing" section in the [Developer Guide](docs/dev.md).
Features without tests are considered non-existent. To ensure code is truly effective, relevant processes should be covered by unit tests and functional tests. Therefore, when considering contributions, please also consider testability. All tests can be run locally without dependency on CI. Please refer to the "Testing" section in the [Developer Guide](docs/zh/guides/development.md).
### Automated Testing for Pull Requests
@@ -60,7 +60,7 @@ Maintainers are here to help you implement your use case within a reasonable tim
### Participating in the Test Plan
The Test Plan aims to provide users with a more stable application experience and faster iteration speed. For details, please refer to the [Test Plan](docs/testplan-en.md).
The Test Plan aims to provide users with a more stable application experience and faster iteration speed. For details, please refer to the [Test Plan](docs/en/guides/test-plan.md).
### Other Suggestions

View File

@@ -34,7 +34,7 @@
</a>
</h1>
<p align="center">English | <a href="./docs/README.zh.md">中文</a> | <a href="https://cherry-ai.com">Official Site</a> | <a href="https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us">Documents</a> | <a href="./docs/dev.md">Development</a> | <a href="https://github.com/CherryHQ/cherry-studio/issues">Feedback</a><br></p>
<p align="center">English | <a href="./docs/zh/README.md">中文</a> | <a href="https://cherry-ai.com">Official Site</a> | <a href="https://docs.cherry-ai.com/cherry-studio-wen-dang/en-us">Documents</a> | <a href="./docs/en/guides/development.md">Development</a> | <a href="https://github.com/CherryHQ/cherry-studio/issues">Feedback</a><br></p>
<div align="center">
@@ -67,7 +67,7 @@ Cherry Studio is a desktop client that supports multiple LLM providers, availabl
👏 Join [Telegram Group](https://t.me/CherryStudioAI)[Discord](https://discord.gg/wez8HtpxqQ) | [QQ Group(575014769)](https://qm.qq.com/q/lo0D4qVZKi)
❤️ Like Cherry Studio? Give it a star 🌟 or [Sponsor](docs/sponsor.md) to support the development!
❤️ Like Cherry Studio? Give it a star 🌟 or [Sponsor](docs/zh/guides/sponsor.md) to support the development!
# 🌠 Screenshot
@@ -175,7 +175,7 @@ We welcome contributions to Cherry Studio! Here are some ways you can contribute
6. **Community Engagement**: Join discussions and help users.
7. **Promote Usage**: Spread the word about Cherry Studio.
Refer to the [Branching Strategy](docs/branching-strategy-en.md) for contribution guidelines
Refer to the [Branching Strategy](docs/en/guides/branching-strategy.md) for contribution guidelines
## Getting Started

View File

@@ -14,7 +14,7 @@
}
},
"enabled": true,
"includes": ["**/*.json", "!*.json", "!**/package.json"]
"includes": ["**/*.json", "!*.json", "!**/package.json", "!coverage/**"]
},
"css": {
"formatter": {
@@ -23,7 +23,7 @@
},
"files": {
"ignoreUnknown": false,
"includes": ["**", "!**/.claude/**"],
"includes": ["**", "!**/.claude/**", "!**/.vscode/**"],
"maxSize": 2097152
},
"formatter": {

81
docs/README.md Normal file
View File

@@ -0,0 +1,81 @@
# Cherry Studio Documentation / 文档
This directory contains the project documentation in multiple languages.
本目录包含多语言项目文档。
---
## Languages / 语言
- **[中文文档](./zh/README.md)** - Chinese Documentation
- **English Documentation** - See sections below
---
## English Documentation
### Guides
| Document | Description |
|----------|-------------|
| [Development Setup](./en/guides/development.md) | Development environment setup |
| [Branching Strategy](./en/guides/branching-strategy.md) | Git branching workflow |
| [i18n Guide](./en/guides/i18n.md) | Internationalization guide |
| [Logging Guide](./en/guides/logging.md) | How to use the logger service |
| [Test Plan](./en/guides/test-plan.md) | Test plan and release channels |
### References
| Document | Description |
|----------|-------------|
| [App Upgrade Config](./en/references/app-upgrade.md) | Application upgrade configuration |
| [CodeBlockView Component](./en/references/components/code-block-view.md) | Code block view component |
| [Image Preview Components](./en/references/components/image-preview.md) | Image preview components |
---
## 中文文档
### 指南 (Guides)
| 文档 | 说明 |
|------|------|
| [开发环境设置](./zh/guides/development.md) | 开发环境配置 |
| [贡献指南](./zh/guides/contributing.md) | 如何贡献代码 |
| [分支策略](./zh/guides/branching-strategy.md) | Git 分支工作流 |
| [测试计划](./zh/guides/test-plan.md) | 测试计划和发布通道 |
| [国际化指南](./zh/guides/i18n.md) | 国际化开发指南 |
| [日志使用指南](./zh/guides/logging.md) | 如何使用日志服务 |
| [中间件开发](./zh/guides/middleware.md) | 如何编写中间件 |
| [记忆功能](./zh/guides/memory.md) | 记忆功能使用指南 |
| [赞助信息](./zh/guides/sponsor.md) | 赞助相关信息 |
### 参考 (References)
| 文档 | 说明 |
|------|------|
| [消息系统](./zh/references/message-system.md) | 消息系统架构和 API |
| [数据库结构](./zh/references/database.md) | 数据库表结构 |
| [服务](./zh/references/services.md) | 服务层文档 (KnowledgeService) |
| [代码执行](./zh/references/code-execution.md) | 代码执行功能 |
| [应用升级配置](./zh/references/app-upgrade.md) | 应用升级配置 |
| [CodeBlockView 组件](./zh/references/components/code-block-view.md) | 代码块视图组件 |
| [图像预览组件](./zh/references/components/image-preview.md) | 图像预览组件 |
---
## Missing Translations / 缺少翻译
The following documents are only available in Chinese and need English translations:
以下文档仅有中文版本,需要英文翻译:
- `guides/contributing.md`
- `guides/memory.md`
- `guides/middleware.md`
- `guides/sponsor.md`
- `references/message-system.md`
- `references/database.md`
- `references/services.md`
- `references/code-execution.md`

View File

Before

Width:  |  Height:  |  Size: 150 KiB

After

Width:  |  Height:  |  Size: 150 KiB

View File

Before

Width:  |  Height:  |  Size: 40 KiB

After

Width:  |  Height:  |  Size: 40 KiB

View File

Before

Width:  |  Height:  |  Size: 35 KiB

After

Width:  |  Height:  |  Size: 35 KiB

View File

Before

Width:  |  Height:  |  Size: 563 KiB

After

Width:  |  Height:  |  Size: 563 KiB

View File

@@ -16,7 +16,7 @@ Cherry Studio implements a structured branching strategy to maintain code qualit
- Only accepts documentation updates and bug fixes
- Thoroughly tested before production deployment
For details about the `testplan` branch used in the Test Plan, please refer to the [Test Plan](testplan-en.md).
For details about the `testplan` branch used in the Test Plan, please refer to the [Test Plan](./test-plan.md).
## Contributing Branches

View File

@@ -18,11 +18,11 @@ The plugin has already been configured in the project — simply install it to g
### Demo
![demo-1](./.assets.how-to-i18n/demo-1.png)
![demo-1](../../assets/images/i18n/demo-1.png)
![demo-2](./.assets.how-to-i18n/demo-2.png)
![demo-2](../../assets/images/i18n/demo-2.png)
![demo-3](./.assets.how-to-i18n/demo-3.png)
![demo-3](../../assets/images/i18n/demo-3.png)
## i18n Conventions

View File

@@ -19,7 +19,7 @@ Users are welcome to submit issues or provide feedback through other channels fo
### Participating in the Test Plan
Developers should submit `PRs` according to the [Contributor Guide](../CONTRIBUTING.md) (and ensure the target branch is `main`). The repository maintainers will evaluate whether the `PR` should be included in the Test Plan based on factors such as the impact of the feature on the application, its importance, and whether broader testing is needed.
Developers should submit `PRs` according to the [Contributor Guide](../../CONTRIBUTING.md) (and ensure the target branch is `main`). The repository maintainers will evaluate whether the `PR` should be included in the Test Plan based on factors such as the impact of the feature on the application, its importance, and whether broader testing is needed.
If the `PR` is added to the Test Plan, the repository maintainers will:

View File

@@ -85,7 +85,7 @@ Main responsibilities:
- **SvgPreview**: SVG image preview
- **GraphvizPreview**: Graphviz diagram preview
All special view components share a common architecture for consistent user experience and functionality. For detailed information about these components and their implementation, see [Image Preview Components Documentation](./ImagePreview-en.md).
All special view components share a common architecture for consistent user experience and functionality. For detailed information about these components and their implementation, see [Image Preview Components Documentation](./image-preview.md).
#### StatusBar

View File

@@ -192,4 +192,4 @@ Image Preview Components integrate seamlessly with CodeBlockView:
- Shared state management
- Responsive layout adaptation
For more information about the overall CodeBlockView architecture, see [CodeBlockView Documentation](./CodeBlockView-en.md).
For more information about the overall CodeBlockView architecture, see [CodeBlockView Documentation](./code-block-view.md).

View File

@@ -1,3 +0,0 @@
# 消息的生命周期
![image](./message-lifecycle.png)

View File

@@ -1,11 +0,0 @@
# 数据库设置字段
此文档包含部分字段的数据类型说明。
## 字段
| 字段名 | 类型 | 说明 |
| ------------------------------ | ------------------------------ | ------------ |
| `translate:target:language` | `LanguageCode` | 翻译目标语言 |
| `translate:source:language` | `LanguageCode` | 翻译源语言 |
| `translate:bidirectional:pair` | `[LanguageCode, LanguageCode]` | 双向翻译对 |

View File

@@ -1,127 +0,0 @@
# messageBlock.ts 使用指南
该文件定义了用于管理应用程序中所有 `MessageBlock` 实体的 Redux Slice。它使用 Redux Toolkit 的 `createSlice``createEntityAdapter` 来高效地处理规范化的状态,并提供了一系列 actions 和 selectors 用于与消息块数据交互。
## 核心目标
- **状态管理**: 集中管理所有 `MessageBlock` 的状态。`MessageBlock` 代表消息中的不同内容单元(如文本、代码、图片、引用等)。
- **规范化**: 使用 `createEntityAdapter``MessageBlock` 数据存储在规范化的结构中(`{ ids: [], entities: {} }`),这有助于提高性能和简化更新逻辑。
- **可预测性**: 提供明确的 actions 来修改状态,并通过 selectors 安全地访问状态。
## 关键概念
- **Slice (`createSlice`)**: Redux Toolkit 的核心 API用于创建包含 reducer 逻辑、action creators 和初始状态的 Redux 模块。
- **Entity Adapter (`createEntityAdapter`)**: Redux Toolkit 提供的工具,用于简化对规范化数据的 CRUD创建、读取、更新、删除操作。它会自动生成 reducer 函数和 selectors。
- **Selectors**: 用于从 Redux store 中派生和计算数据的函数。Selectors 可以被记忆化memoized以提高性能。
## State 结构
`messageBlocks` slice 的状态结构由 `createEntityAdapter` 定义,大致如下:
```typescript
{
ids: string[]; // 存储所有 MessageBlock ID 的有序列表
entities: { [id: string]: MessageBlock }; // 按 ID 存储 MessageBlock 对象的字典
loadingState: 'idle' | 'loading' | 'succeeded' | 'failed'; // (可选) 其他状态,如加载状态
error: string | null; // (可选) 错误信息
}
```
## Actions
该 slice 导出以下 actions (由 `createSlice``createEntityAdapter` 自动生成或自定义)
- **`upsertOneBlock(payload: MessageBlock)`**:
- 添加一个新的 `MessageBlock` 或更新一个已存在的 `MessageBlock`。如果 payload 中的 `id` 已存在,则执行更新;否则执行插入。
- **`upsertManyBlocks(payload: MessageBlock[])`**:
- 添加或更新多个 `MessageBlock`。常用于批量加载数据(例如,加载一个 Topic 的所有消息块)。
- **`removeOneBlock(payload: string)`**:
- 根据提供的 `id` (payload) 移除单个 `MessageBlock`
- **`removeManyBlocks(payload: string[])`**:
- 根据提供的 `id` 数组 (payload) 移除多个 `MessageBlock`。常用于删除消息或清空 Topic 时清理相关的块。
- **`removeAllBlocks()`**:
- 移除 state 中的所有 `MessageBlock` 实体。
- **`updateOneBlock(payload: { id: string; changes: Partial<MessageBlock> })`**:
- 更新一个已存在的 `MessageBlock``payload` 需要包含块的 `id` 和一个包含要更改的字段的 `changes` 对象。
- **`setMessageBlocksLoading(payload: 'idle' | 'loading')`**:
- (自定义) 设置 `loadingState` 属性。
- **`setMessageBlocksError(payload: string)`**:
- (自定义) 设置 `loadingState``'failed'` 并记录错误信息。
**使用示例 (在 Thunk 或其他 Dispatch 的地方):**
```typescript
import { upsertOneBlock, removeManyBlocks, updateOneBlock } from './messageBlock'
import store from './store' // 假设这是你的 Redux store 实例
// 添加或更新一个块
const newBlock: MessageBlock = {
/* ... block data ... */
}
store.dispatch(upsertOneBlock(newBlock))
// 更新一个块的内容
store.dispatch(updateOneBlock({ id: blockId, changes: { content: 'New content' } }))
// 删除多个块
const blockIdsToRemove = ['id1', 'id2']
store.dispatch(removeManyBlocks(blockIdsToRemove))
```
## Selectors
该 slice 导出由 `createEntityAdapter` 生成的基础 selectors并通过 `messageBlocksSelectors` 对象访问:
- **`messageBlocksSelectors.selectIds(state: RootState): string[]`**: 返回包含所有块 ID 的数组。
- **`messageBlocksSelectors.selectEntities(state: RootState): { [id: string]: MessageBlock }`**: 返回块 ID 到块对象的映射字典。
- **`messageBlocksSelectors.selectAll(state: RootState): MessageBlock[]`**: 返回包含所有块对象的数组。
- **`messageBlocksSelectors.selectTotal(state: RootState): number`**: 返回块的总数。
- **`messageBlocksSelectors.selectById(state: RootState, id: string): MessageBlock | undefined`**: 根据 ID 返回单个块对象,如果找不到则返回 `undefined`
**此外,还提供了一个自定义的、记忆化的 selector**
- **`selectFormattedCitationsByBlockId(state: RootState, blockId: string | undefined): Citation[]`**:
- 接收一个 `blockId`
- 如果该 ID 对应的块是 `CITATION` 类型,则提取并格式化其包含的引用信息(来自网页搜索、知识库等),进行去重和重新编号,最后返回一个 `Citation[]` 数组,用于在 UI 中显示。
- 如果块不存在或类型不匹配,返回空数组 `[]`
- 这个 selector 封装了处理不同引用来源Gemini, OpenAI, OpenRouter, Zhipu 等)的复杂逻辑。
**使用示例 (在 React 组件或 `useSelector` 中):**
```typescript
import { useSelector } from 'react-redux'
import { messageBlocksSelectors, selectFormattedCitationsByBlockId } from './messageBlock'
import type { RootState } from './store'
// 获取所有块
const allBlocks = useSelector(messageBlocksSelectors.selectAll)
// 获取特定 ID 的块
const specificBlock = useSelector((state: RootState) => messageBlocksSelectors.selectById(state, someBlockId))
// 获取特定引用块格式化后的引用列表
const formattedCitations = useSelector((state: RootState) => selectFormattedCitationsByBlockId(state, citationBlockId))
// 在组件中使用引用数据
// {formattedCitations.map(citation => ...)}
```
## 集成
`messageBlock.ts` slice 通常与 `messageThunk.ts` 中的 Thunks 紧密协作。Thunks 负责处理异步逻辑(如 API 调用、数据库操作),并在需要时 dispatch `messageBlock` slice 的 actions 来更新状态。例如,当 `messageThunk` 接收到流式响应时,它会 dispatch `upsertOneBlock``updateOneBlock` 来实时更新对应的 `MessageBlock`。同样,删除消息的 Thunk 会 dispatch `removeManyBlocks`
理解 `messageBlock.ts` 的职责是管理**状态本身**,而 `messageThunk.ts` 负责**触发状态变更**的异步流程,这对于维护清晰的应用架构至关重要。

View File

@@ -1,105 +0,0 @@
# messageThunk.ts 使用指南
该文件包含用于管理应用程序中消息流、处理助手交互以及同步 Redux 状态与 IndexedDB 数据库的核心 Thunk Action Creators。主要围绕 `Message``MessageBlock` 对象进行操作。
## 核心功能
1. **发送/接收消息**: 处理用户消息的发送,触发助手响应,并流式处理返回的数据,将其解析为不同的 `MessageBlock`
2. **状态管理**: 确保 Redux store 中的消息和消息块状态与 IndexedDB 中的持久化数据保持一致。
3. **消息操作**: 提供删除、重发、重新生成、编辑后重发、追加响应、克隆等消息生命周期管理功能。
4. **Block 处理**: 动态创建、更新和保存各种类型的 `MessageBlock`(文本、思考过程、工具调用、引用、图片、错误、翻译等)。
## 主要 Thunks
以下是一些关键的 Thunk 函数及其用途:
1. **`sendMessage(userMessage, userMessageBlocks, assistant, topicId)`**
- **用途**: 发送一条新的用户消息。
- **流程**:
- 保存用户消息 (`userMessage`) 及其块 (`userMessageBlocks`) 到 Redux 和 DB。
- 检查 `@mentions` 以确定是单模型响应还是多模型响应。
- 创建助手消息(们)的存根 (Stub)。
- 将存根添加到 Redux 和 DB。
- 将核心处理逻辑 `fetchAndProcessAssistantResponseImpl` 添加到该 `topicId` 的队列中以获取实际响应。
- **Block 相关**: 主要处理用户消息的初始 `MessageBlock` 保存。
2. **`fetchAndProcessAssistantResponseImpl(dispatch, getState, topicId, assistant, assistantMessage)`**
- **用途**: (内部函数) 获取并处理单个助手响应的核心逻辑,被 `sendMessage`, `resend...`, `regenerate...`, `append...` 等调用。
- **流程**:
- 设置 Topic 加载状态。
- 准备上下文消息。
- 调用 `fetchChatCompletion` API 服务。
- 使用 `createStreamProcessor` 处理流式响应。
- 通过各种回调 (`onTextChunk`, `onThinkingChunk`, `onToolCallComplete`, `onImageGenerated`, `onError`, `onComplete` 等) 处理不同类型的事件。
- **Block 相关**:
- 根据流事件创建初始 `UNKNOWN` 块。
- 实时创建和更新 `MAIN_TEXT``THINKING` 块,使用 `throttledBlockUpdate``throttledBlockDbUpdate` 进行节流更新。
- 创建 `TOOL`, `CITATION`, `IMAGE`, `ERROR` 等类型的块。
- 在事件完成时(如 `onTextComplete`, `onToolCallComplete`)将块状态标记为 `SUCCESS``ERROR`,并使用 `saveUpdatedBlockToDB` 保存最终状态。
- 使用 `handleBlockTransition` 管理非流式块(如 `TOOL`, `CITATION`)的添加和状态更新。
3. **`loadTopicMessagesThunk(topicId, forceReload)`**
- **用途**: 从数据库加载指定主题的所有消息及其关联的 `MessageBlock`
- **流程**:
- 从 DB 获取 `Topic` 及其 `messages` 列表。
- 根据消息 ID 列表从 DB 获取所有相关的 `MessageBlock`
- 使用 `upsertManyBlocks` 将块更新到 Redux。
- 将消息更新到 Redux。
- **Block 相关**: 负责将持久化的 `MessageBlock` 加载到 Redux 状态。
4. **删除 Thunks**
- `deleteSingleMessageThunk(topicId, messageId)`: 删除单个消息及其所有 `MessageBlock`
- `deleteMessageGroupThunk(topicId, askId)`: 删除一个用户消息及其所有相关的助手响应消息和它们的所有 `MessageBlock`
- `clearTopicMessagesThunk(topicId)`: 清空主题下的所有消息及其所有 `MessageBlock`
- **Block 相关**: 从 Redux 和 DB 中移除指定的 `MessageBlock`
5. **重发/重新生成 Thunks**
- `resendMessageThunk(topicId, userMessageToResend, assistant)`: 重发用户消息。会重置(清空 Block 并标记为 PENDING所有与该用户消息关联的助手响应然后重新请求生成。
- `resendUserMessageWithEditThunk(topicId, originalMessage, mainTextBlockId, editedContent, assistant)`: 用户编辑消息内容后重发。先更新用户消息的 `MAIN_TEXT` 块内容,然后调用 `resendMessageThunk`
- `regenerateAssistantResponseThunk(topicId, assistantMessageToRegenerate, assistant)`: 重新生成单个助手响应。重置该助手消息(清空 Block 并标记为 PENDING然后重新请求生成。
- **Block 相关**: 删除旧的 `MessageBlock`,并在重新生成过程中创建新的 `MessageBlock`
6. **`appendAssistantResponseThunk(topicId, existingAssistantMessageId, newModel, assistant)`**
- **用途**: 在已有的对话上下文中,针对同一个用户问题,使用新选择的模型追加一个新的助手响应。
- **流程**:
- 找到现有助手消息以获取原始 `askId`
- 创建使用 `newModel` 的新助手消息存根(使用相同的 `askId`)。
- 添加新存根到 Redux 和 DB。
-`fetchAndProcessAssistantResponseImpl` 添加到队列以生成新响应。
- **Block 相关**: 为新的助手响应创建全新的 `MessageBlock`
7. **`cloneMessagesToNewTopicThunk(sourceTopicId, branchPointIndex, newTopic)`**
- **用途**: 将源主题的部分消息(及其 Block克隆到一个**已存在**的新主题中。
- **流程**:
- 复制指定索引前的消息。
- 为所有克隆的消息和 Block 生成新的 UUID。
- 正确映射克隆消息之间的 `askId` 关系。
- 复制 `MessageBlock` 内容,更新其 `messageId` 指向新的消息 ID。
- 更新文件引用计数(如果 Block 是文件或图片)。
- 将克隆的消息和 Block 保存到新主题的 Redux 状态和 DB 中。
- **Block 相关**: 创建 `MessageBlock` 的副本,并更新其 ID 和 `messageId`
8. **`initiateTranslationThunk(messageId, topicId, targetLanguage, sourceBlockId?, sourceLanguage?)`**
- **用途**: 为指定消息启动翻译流程,创建一个初始的 `TRANSLATION` 类型的 `MessageBlock`
- **流程**:
- 创建一个状态为 `STREAMING``TranslationMessageBlock`
- 将其添加到 Redux 和 DB。
- 更新原消息的 `blocks` 列表以包含新的翻译块 ID。
- **Block 相关**: 创建并保存一个占位的 `TranslationMessageBlock`。实际翻译内容的获取和填充需要后续步骤。
## 内部机制和注意事项
- **数据库交互**: 通过 `saveMessageAndBlocksToDB`, `updateExistingMessageAndBlocksInDB`, `saveUpdatesToDB`, `saveUpdatedBlockToDB`, `throttledBlockDbUpdate` 等辅助函数与 IndexedDB (`db`) 交互,确保数据持久化。
- **状态同步**: Thunks 负责协调 Redux Store 和 IndexedDB 之间的数据一致性。
- **队列 (`getTopicQueue`)**: 使用 `AsyncQueue` 确保对同一主题的操作(尤其是 API 请求)按顺序执行,避免竞态条件。
- **节流 (`throttle`)**: 对流式响应中频繁的 Block 更新(文本、思考)使用 `lodash.throttle` 优化性能,减少 Redux dispatch 和 DB 写入次数。
- **错误处理**: `fetchAndProcessAssistantResponseImpl` 内的回调函数(特别是 `onError`)处理流处理和 API 调用中可能出现的错误,并创建 `ERROR` 类型的 `MessageBlock`
开发者在使用这些 Thunks 时,通常需要提供 `dispatch`, `getState` (由 Redux Thunk 中间件注入),以及如 `topicId`, `assistant` 配置对象, 相关的 `Message``MessageBlock` 对象/ID 等参数。理解每个 Thunk 的职责和它如何影响消息及块的状态至关重要。

View File

@@ -1,156 +0,0 @@
# useMessageOperations.ts 使用指南
该文件定义了一个名为 `useMessageOperations` 的自定义 React Hook。这个 Hook 的主要目的是为 React 组件提供一个便捷的接口用于执行与特定主题Topic相关的各种消息操作。它封装了调用 Redux Thunks (`messageThunk.ts`) 和 Actions (`newMessage.ts`, `messageBlock.ts`) 的逻辑,简化了组件与消息数据交互的代码。
## 核心目标
- **封装**: 将复杂的消息操作逻辑(如删除、重发、重新生成、编辑、翻译等)封装在易于使用的函数中。
- **简化**: 让组件可以直接调用这些操作函数,而无需直接与 Redux `dispatch` 或 Thunks 交互。
- **上下文关联**: 所有操作都与传入的 `topic` 对象相关联,确保操作作用于正确的主题。
## 如何使用
在你的 React 函数组件中,导入并调用 `useMessageOperations` Hook并传入当前活动的 `Topic` 对象。
```typescript
import React from 'react';
import { useMessageOperations } from '@renderer/hooks/useMessageOperations';
import type { Topic, Message, Assistant, Model } from '@renderer/types';
interface MyComponentProps {
currentTopic: Topic;
currentAssistant: Assistant;
}
function MyComponent({ currentTopic, currentAssistant }: MyComponentProps) {
const {
deleteMessage,
resendMessage,
regenerateAssistantMessage,
appendAssistantResponse,
getTranslationUpdater,
createTopicBranch,
// ... 其他操作函数
} = useMessageOperations(currentTopic);
const handleDelete = (messageId: string) => {
deleteMessage(messageId);
};
const handleResend = (message: Message) => {
resendMessage(message, currentAssistant);
};
const handleAppend = (existingMsg: Message, newModel: Model) => {
appendAssistantResponse(existingMsg, newModel, currentAssistant);
}
// ... 在组件中使用其他操作函数
return (
<div>
{/* Component UI */}
<button onClick={() => handleDelete('some-message-id')}>Delete Message</button>
{/* ... */}
</div>
);
}
```
## 返回值
`useMessageOperations(topic)` Hook 返回一个包含以下函数和值的对象:
- **`deleteMessage(id: string)`**:
- 删除指定 `id` 的单个消息。
- 内部调用 `deleteSingleMessageThunk`
- **`deleteGroupMessages(askId: string)`**:
- 删除与指定 `askId` 相关联的一组消息(通常是用户提问及其所有助手回答)。
- 内部调用 `deleteMessageGroupThunk`
- **`editMessage(messageId: string, updates: Partial<Message>)`**:
- 更新指定 `messageId` 的消息的部分属性。
- **注意**: 目前主要用于更新 Redux 状态
- 内部调用 `newMessagesActions.updateMessage`
- **`resendMessage(message: Message, assistant: Assistant)`**:
- 重新发送指定的用户消息 (`message`),这将触发其所有关联助手响应的重新生成。
- 内部调用 `resendMessageThunk`
- **`resendUserMessageWithEdit(message: Message, editedContent: string, assistant: Assistant)`**:
- 在用户消息的主要文本块被编辑后,重新发送该消息。
- 会先查找消息的 `MAIN_TEXT` 块 ID然后调用 `resendUserMessageWithEditThunk`
- **`clearTopicMessages(_topicId?: string)`**:
- 清除当前主题(或可选的指定 `_topicId`)下的所有消息。
- 内部调用 `clearTopicMessagesThunk`
- **`createNewContext()`**:
- 发出一个全局事件 (`EVENT_NAMES.NEW_CONTEXT`),通常用于通知 UI 清空显示,准备新的上下文。不直接修改 Redux 状态。
- **`displayCount`**:
- (非操作函数) 从 Redux store 中获取当前的 `displayCount` 值。
- **`pauseMessages()`**:
- 尝试中止当前主题中正在进行的消息生成(状态为 `processing``pending`)。
- 通过查找相关的 `askId` 并调用 `abortCompletion` 来实现。
- 同时会 dispatch `setTopicLoading` action 将加载状态设为 `false`
- **`resumeMessage(message: Message, assistant: Assistant)`**:
- 恢复/重新发送一个用户消息。目前实现为直接调用 `resendMessage`
- **`regenerateAssistantMessage(message: Message, assistant: Assistant)`**:
- 重新生成指定的**助手**消息 (`message`) 的响应。
- 内部调用 `regenerateAssistantResponseThunk`
- **`appendAssistantResponse(existingAssistantMessage: Message, newModel: Model, assistant: Assistant)`**:
- 针对 `existingAssistantMessage` 所回复的**同一用户提问**,使用 `newModel` 追加一个新的助手响应。
- 内部调用 `appendAssistantResponseThunk`
- **`getTranslationUpdater(messageId: string, targetLanguage: string, sourceBlockId?: string, sourceLanguage?: string)`**:
- **用途**: 获取一个用于逐步更新翻译块内容的函数。
- **流程**:
1. 内部调用 `initiateTranslationThunk` 来创建或获取一个 `TRANSLATION` 类型的 `MessageBlock`,并获取其 `blockId`
2. 返回一个**异步更新函数**。
- **返回的更新函数 `(accumulatedText: string, isComplete?: boolean) => void`**:
- 接收累积的翻译文本和完成状态。
- 调用 `updateOneBlock` 更新 Redux 中的翻译块内容和状态 (`STREAMING``SUCCESS`)。
- 调用 `throttledBlockDbUpdate` 将更新(节流地)保存到数据库。
- 如果初始化失败Thunk 返回 `undefined`),则此函数返回 `null`
- **`createTopicBranch(sourceTopicId: string, branchPointIndex: number, newTopic: Topic)`**:
- 创建一个主题分支,将 `sourceTopicId` 主题中 `branchPointIndex` 索引之前的消息克隆到 `newTopic` 中。
- **注意**: `newTopic` 对象必须是调用此函数**之前**已经创建并添加到 Redux 和数据库中的。
- 内部调用 `cloneMessagesToNewTopicThunk`
## 依赖
- **`topic: Topic`**: 必须传入当前操作上下文的主题对象。Hook 返回的操作函数将始终作用于这个主题的 `topic.id`
- **Redux `dispatch`**: Hook 内部使用 `useAppDispatch` 获取 `dispatch` 函数来调用 actions 和 thunks。
## 相关 Hooks
在同一文件中还定义了两个辅助 Hook
- **`useTopicMessages(topic: Topic)`**:
- 使用 `selectMessagesForTopic` selector 来获取并返回指定主题的消息列表。
- **`useTopicLoading(topic: Topic)`**:
- 使用 `selectNewTopicLoading` selector 来获取并返回指定主题的加载状态。
这些 Hook 可以与 `useMessageOperations` 结合使用,方便地在组件中获取消息数据、加载状态,并执行相关操作。

View File

@@ -34,7 +34,7 @@
</a>
</h1>
<p align="center">
<a href="https://github.com/CherryHQ/cherry-studio">English</a> | 中文 | <a href="https://cherry-ai.com">官方网站</a> | <a href="https://docs.cherry-ai.com/cherry-studio-wen-dang/zh-cn">文档</a> | <a href="./dev.md">开发</a> | <a href="https://github.com/CherryHQ/cherry-studio/issues">反馈</a><br>
<a href="https://github.com/CherryHQ/cherry-studio">English</a> | 中文 | <a href="https://cherry-ai.com">官方网站</a> | <a href="https://docs.cherry-ai.com/cherry-studio-wen-dang/zh-cn">文档</a> | <a href="./guides/development.md">开发</a> | <a href="https://github.com/CherryHQ/cherry-studio/issues">反馈</a><br>
</p>
<!-- 题头徽章组合 -->
@@ -70,7 +70,7 @@ Cherry Studio 是一款支持多个大语言模型LLM服务商的桌面客
👏 欢迎加入 [Telegram 群组](https://t.me/CherryStudioAI)[Discord](https://discord.gg/wez8HtpxqQ) | [QQ群(575014769)](https://qm.qq.com/q/lo0D4qVZKi)
❤️ 喜欢 Cherry Studio? 点亮小星星 🌟 或 [赞助开发者](sponsor.md)! ❤️
❤️ 喜欢 Cherry Studio? 点亮小星星 🌟 或 [赞助开发者](./guides/sponsor.md)! ❤️
# 📖 使用教程
@@ -181,7 +181,7 @@ https://docs.cherry-ai.com
6. **社区参与**:加入讨论并帮助用户
7. **推广使用**:宣传 Cherry Studio
参考[分支策略](branching-strategy-zh.md)了解贡献指南
参考[分支策略](./guides/branching-strategy.md)了解贡献指南
## 入门
@@ -190,7 +190,7 @@ https://docs.cherry-ai.com
3. **提交更改**:提交并推送您的更改
4. **打开 Pull Request**:描述您的更改和原因
有关更详细的指南,请参阅我们的 [贡献指南](CONTRIBUTING.zh.md)
有关更详细的指南,请参阅我们的 [贡献指南](./guides/contributing.md)
感谢您的支持和贡献!

View File

@@ -16,7 +16,7 @@ Cherry Studio 采用结构化的分支策略来维护代码质量并简化开发
- 只接受文档更新和 bug 修复
- 经过完整测试后可以发布到生产环境
关于测试计划所使用的`testplan`分支,请查阅[测试计划](testplan-zh.md)。
关于测试计划所使用的`testplan`分支,请查阅[测试计划](./test-plan.md)。
## 贡献分支

View File

@@ -1,6 +1,6 @@
# Cherry Studio 贡献者指南
[**English**](../CONTRIBUTING.md) | [**中文**](CONTRIBUTING.zh.md)
[**English**](../../../CONTRIBUTING.md) | **中文**
欢迎来到 Cherry Studio 的贡献者社区!我们致力于将 Cherry Studio 打造成一个长期提供价值的项目,并希望邀请更多的开发者加入我们的行列。无论您是经验丰富的开发者还是刚刚起步的初学者,您的贡献都将帮助我们更好地服务用户,提升软件质量。
@@ -24,7 +24,7 @@
## 开始之前
请确保阅读了[行为准则](../CODE_OF_CONDUCT.md)和[LICENSE](../LICENSE)。
请确保阅读了[行为准则](../../../CODE_OF_CONDUCT.md)和[LICENSE](../../../LICENSE)。
## 开始贡献
@@ -32,7 +32,7 @@
### 测试
未经测试的功能等同于不存在。为确保代码真正有效,应通过单元测试和功能测试覆盖相关流程。因此,在考虑贡献时,也请考虑可测试性。所有测试均可本地运行,无需依赖 CI。请参阅[开发者指南](dev.md#test)中的Test部分。
未经测试的功能等同于不存在。为确保代码真正有效,应通过单元测试和功能测试覆盖相关流程。因此,在考虑贡献时,也请考虑可测试性。所有测试均可本地运行,无需依赖 CI。请参阅[开发者指南](./development.md#test)中的"Test"部分。
### 拉取请求的自动化测试
@@ -60,11 +60,11 @@ git commit --signoff -m "Your commit message"
### 获取代码审查/合并
维护者在此帮助您在合理时间内实现您的用例。他们会尽力在合理时间内审查您的代码并提供建设性反馈。但如果您在审查过程中受阻,或认为您的 Pull Request 未得到应有的关注,请通过 Issue 中的评论或者[社群](README.zh.md#-community)联系我们
维护者在此帮助您在合理时间内实现您的用例。他们会尽力在合理时间内审查您的代码并提供建设性反馈。但如果您在审查过程中受阻,或认为您的 Pull Request 未得到应有的关注,请通过 Issue 中的评论或者[社群](../README.md#-community)联系我们
### 参与测试计划
测试计划旨在为用户提供更稳定的应用体验和更快的迭代速度,详细情况请参阅[测试计划](testplan-zh.md)。
测试计划旨在为用户提供更稳定的应用体验和更快的迭代速度,详细情况请参阅[测试计划](./test-plan.md)。
### 其他建议

View File

@@ -0,0 +1,73 @@
# 🖥️ Develop
## IDE Setup
- Editor: [Cursor](https://www.cursor.com/), etc. Any VS Code compatible editor.
- Linter: [ESLint](https://marketplace.visualstudio.com/items?itemName=dbaeumer.vscode-eslint)
- Formatter: [Biome](https://marketplace.visualstudio.com/items?itemName=biomejs.biome)
## Project Setup
### Install
```bash
yarn
```
### Development
### Setup Node.js
Download and install [Node.js v22.x.x](https://nodejs.org/en/download)
### Setup Yarn
```bash
corepack enable
corepack prepare yarn@4.9.1 --activate
```
### Install Dependencies
```bash
yarn install
```
### ENV
```bash
copy .env.example .env
```
### Start
```bash
yarn dev
```
### Debug
```bash
yarn debug
```
Then input chrome://inspect in browser
### Test
```bash
yarn test
```
### Build
```bash
# For windows
$ yarn build:win
# For macOS
$ yarn build:mac
# For Linux
$ yarn build:linux
```

View File

@@ -15,11 +15,11 @@ i18n ally是一个强大的VSCode插件它能在开发阶段提供实时反
### 效果展示
![demo-1](./.assets.how-to-i18n/demo-1.png)
![demo-1](../../assets/images/i18n/demo-1.png)
![demo-2](./.assets.how-to-i18n/demo-2.png)
![demo-2](../../assets/images/i18n/demo-2.png)
![demo-3](./.assets.how-to-i18n/demo-3.png)
![demo-3](../../assets/images/i18n/demo-3.png)
## i18n 约定

View File

@@ -19,7 +19,7 @@
### 参与测试计划
开发者按照[贡献者指南](CONTRIBUTING.zh.md)要求正常提交`PR`并注意提交target为`main`)。仓库维护者会综合考虑(例如该功能对应用的影响程度,功能的重要性,是否需要更广泛的测试等),决定该`PR`是否应加入测试计划。
开发者按照[贡献者指南](./contributing.md)要求正常提交`PR`并注意提交target为`main`)。仓库维护者会综合考虑(例如该功能对应用的影响程度,功能的重要性,是否需要更广泛的测试等),决定该`PR`是否应加入测试计划。
若该`PR`加入测试计划,仓库维护者会做如下操作:

View File

@@ -85,7 +85,7 @@ graph TD
- **SvgPreview**: SVG 图像预览
- **GraphvizPreview**: Graphviz 图表预览
所有特殊视图组件共享通用架构,以确保一致的用户体验和功能。有关这些组件及其实现的详细信息,请参阅 [图像预览组件文档](./ImagePreview-zh.md)。
所有特殊视图组件共享通用架构,以确保一致的用户体验和功能。有关这些组件及其实现的详细信息,请参阅[图像预览组件文档](./image-preview.md)。
#### StatusBar 状态栏

View File

@@ -192,4 +192,4 @@ const { containerRef, error, isLoading, triggerRender, cancelRender, clearError,
- 共享状态管理
- 响应式布局适应
有关整体 CodeBlockView 架构的更多信息,请参阅 [CodeBlockView 文档](./CodeBlockView-zh.md)。
有关整体 CodeBlockView 架构的更多信息,请参阅 [CodeBlockView 文档](./code-block-view.md)。

View File

@@ -1,6 +1,24 @@
# `translate_languages` 表技术文档
# 数据库参考文档
## 📄 概述
本文档介绍 Cherry Studio 的数据库结构,包括设置字段和翻译语言表。
---
## 设置字段 (settings)
此部分包含设置相关字段的数据类型说明。
### 翻译相关字段
| 字段名 | 类型 | 说明 |
| ------------------------------ | ------------------------------ | ------------ |
| `translate:target:language` | `LanguageCode` | 翻译目标语言 |
| `translate:source:language` | `LanguageCode` | 翻译源语言 |
| `translate:bidirectional:pair` | `[LanguageCode, LanguageCode]` | 双向翻译对 |
---
## 翻译语言表 (translate_languages)
`translate_languages` 记录用户自定义的的语言类型(`Language`)。

View File

@@ -0,0 +1,404 @@
# 消息系统
本文档介绍 Cherry Studio 的消息系统架构,包括消息生命周期、状态管理和操作接口。
## 消息的生命周期
![消息生命周期](../../assets/images/message-lifecycle.png)
---
# messageBlock.ts 使用指南
该文件定义了用于管理应用程序中所有 `MessageBlock` 实体的 Redux Slice。它使用 Redux Toolkit 的 `createSlice``createEntityAdapter` 来高效地处理规范化的状态,并提供了一系列 actions 和 selectors 用于与消息块数据交互。
## 核心目标
- **状态管理**: 集中管理所有 `MessageBlock` 的状态。`MessageBlock` 代表消息中的不同内容单元(如文本、代码、图片、引用等)。
- **规范化**: 使用 `createEntityAdapter``MessageBlock` 数据存储在规范化的结构中(`{ ids: [], entities: {} }`),这有助于提高性能和简化更新逻辑。
- **可预测性**: 提供明确的 actions 来修改状态,并通过 selectors 安全地访问状态。
## 关键概念
- **Slice (`createSlice`)**: Redux Toolkit 的核心 API用于创建包含 reducer 逻辑、action creators 和初始状态的 Redux 模块。
- **Entity Adapter (`createEntityAdapter`)**: Redux Toolkit 提供的工具,用于简化对规范化数据的 CRUD创建、读取、更新、删除操作。它会自动生成 reducer 函数和 selectors。
- **Selectors**: 用于从 Redux store 中派生和计算数据的函数。Selectors 可以被记忆化memoized以提高性能。
## State 结构
`messageBlocks` slice 的状态结构由 `createEntityAdapter` 定义,大致如下:
```typescript
{
ids: string[]; // 存储所有 MessageBlock ID 的有序列表
entities: { [id: string]: MessageBlock }; // 按 ID 存储 MessageBlock 对象的字典
loadingState: 'idle' | 'loading' | 'succeeded' | 'failed'; // (可选) 其他状态,如加载状态
error: string | null; // (可选) 错误信息
}
```
## Actions
该 slice 导出以下 actions (由 `createSlice``createEntityAdapter` 自动生成或自定义)
- **`upsertOneBlock(payload: MessageBlock)`**:
- 添加一个新的 `MessageBlock` 或更新一个已存在的 `MessageBlock`。如果 payload 中的 `id` 已存在,则执行更新;否则执行插入。
- **`upsertManyBlocks(payload: MessageBlock[])`**:
- 添加或更新多个 `MessageBlock`。常用于批量加载数据(例如,加载一个 Topic 的所有消息块)。
- **`removeOneBlock(payload: string)`**:
- 根据提供的 `id` (payload) 移除单个 `MessageBlock`
- **`removeManyBlocks(payload: string[])`**:
- 根据提供的 `id` 数组 (payload) 移除多个 `MessageBlock`。常用于删除消息或清空 Topic 时清理相关的块。
- **`removeAllBlocks()`**:
- 移除 state 中的所有 `MessageBlock` 实体。
- **`updateOneBlock(payload: { id: string; changes: Partial<MessageBlock> })`**:
- 更新一个已存在的 `MessageBlock``payload` 需要包含块的 `id` 和一个包含要更改的字段的 `changes` 对象。
- **`setMessageBlocksLoading(payload: 'idle' | 'loading')`**:
- (自定义) 设置 `loadingState` 属性。
- **`setMessageBlocksError(payload: string)`**:
- (自定义) 设置 `loadingState``'failed'` 并记录错误信息。
**使用示例 (在 Thunk 或其他 Dispatch 的地方):**
```typescript
import { upsertOneBlock, removeManyBlocks, updateOneBlock } from './messageBlock'
import store from './store' // 假设这是你的 Redux store 实例
// 添加或更新一个块
const newBlock: MessageBlock = {
/* ... block data ... */
}
store.dispatch(upsertOneBlock(newBlock))
// 更新一个块的内容
store.dispatch(updateOneBlock({ id: blockId, changes: { content: 'New content' } }))
// 删除多个块
const blockIdsToRemove = ['id1', 'id2']
store.dispatch(removeManyBlocks(blockIdsToRemove))
```
## Selectors
该 slice 导出由 `createEntityAdapter` 生成的基础 selectors并通过 `messageBlocksSelectors` 对象访问:
- **`messageBlocksSelectors.selectIds(state: RootState): string[]`**: 返回包含所有块 ID 的数组。
- **`messageBlocksSelectors.selectEntities(state: RootState): { [id: string]: MessageBlock }`**: 返回块 ID 到块对象的映射字典。
- **`messageBlocksSelectors.selectAll(state: RootState): MessageBlock[]`**: 返回包含所有块对象的数组。
- **`messageBlocksSelectors.selectTotal(state: RootState): number`**: 返回块的总数。
- **`messageBlocksSelectors.selectById(state: RootState, id: string): MessageBlock | undefined`**: 根据 ID 返回单个块对象,如果找不到则返回 `undefined`
**此外,还提供了一个自定义的、记忆化的 selector**
- **`selectFormattedCitationsByBlockId(state: RootState, blockId: string | undefined): Citation[]`**:
- 接收一个 `blockId`
- 如果该 ID 对应的块是 `CITATION` 类型,则提取并格式化其包含的引用信息(来自网页搜索、知识库等),进行去重和重新编号,最后返回一个 `Citation[]` 数组,用于在 UI 中显示。
- 如果块不存在或类型不匹配,返回空数组 `[]`
- 这个 selector 封装了处理不同引用来源Gemini, OpenAI, OpenRouter, Zhipu 等)的复杂逻辑。
**使用示例 (在 React 组件或 `useSelector` 中):**
```typescript
import { useSelector } from 'react-redux'
import { messageBlocksSelectors, selectFormattedCitationsByBlockId } from './messageBlock'
import type { RootState } from './store'
// 获取所有块
const allBlocks = useSelector(messageBlocksSelectors.selectAll)
// 获取特定 ID 的块
const specificBlock = useSelector((state: RootState) => messageBlocksSelectors.selectById(state, someBlockId))
// 获取特定引用块格式化后的引用列表
const formattedCitations = useSelector((state: RootState) => selectFormattedCitationsByBlockId(state, citationBlockId))
// 在组件中使用引用数据
// {formattedCitations.map(citation => ...)}
```
## 集成
`messageBlock.ts` slice 通常与 `messageThunk.ts` 中的 Thunks 紧密协作。Thunks 负责处理异步逻辑(如 API 调用、数据库操作),并在需要时 dispatch `messageBlock` slice 的 actions 来更新状态。例如,当 `messageThunk` 接收到流式响应时,它会 dispatch `upsertOneBlock``updateOneBlock` 来实时更新对应的 `MessageBlock`。同样,删除消息的 Thunk 会 dispatch `removeManyBlocks`
理解 `messageBlock.ts` 的职责是管理**状态本身**,而 `messageThunk.ts` 负责**触发状态变更**的异步流程,这对于维护清晰的应用架构至关重要。
---
# messageThunk.ts 使用指南
该文件包含用于管理应用程序中消息流、处理助手交互以及同步 Redux 状态与 IndexedDB 数据库的核心 Thunk Action Creators。主要围绕 `Message``MessageBlock` 对象进行操作。
## 核心功能
1. **发送/接收消息**: 处理用户消息的发送,触发助手响应,并流式处理返回的数据,将其解析为不同的 `MessageBlock`
2. **状态管理**: 确保 Redux store 中的消息和消息块状态与 IndexedDB 中的持久化数据保持一致。
3. **消息操作**: 提供删除、重发、重新生成、编辑后重发、追加响应、克隆等消息生命周期管理功能。
4. **Block 处理**: 动态创建、更新和保存各种类型的 `MessageBlock`(文本、思考过程、工具调用、引用、图片、错误、翻译等)。
## 主要 Thunks
以下是一些关键的 Thunk 函数及其用途:
1. **`sendMessage(userMessage, userMessageBlocks, assistant, topicId)`**
- **用途**: 发送一条新的用户消息。
- **流程**:
- 保存用户消息 (`userMessage`) 及其块 (`userMessageBlocks`) 到 Redux 和 DB。
- 检查 `@mentions` 以确定是单模型响应还是多模型响应。
- 创建助手消息(们)的存根 (Stub)。
- 将存根添加到 Redux 和 DB。
- 将核心处理逻辑 `fetchAndProcessAssistantResponseImpl` 添加到该 `topicId` 的队列中以获取实际响应。
- **Block 相关**: 主要处理用户消息的初始 `MessageBlock` 保存。
2. **`fetchAndProcessAssistantResponseImpl(dispatch, getState, topicId, assistant, assistantMessage)`**
- **用途**: (内部函数) 获取并处理单个助手响应的核心逻辑,被 `sendMessage`, `resend...`, `regenerate...`, `append...` 等调用。
- **流程**:
- 设置 Topic 加载状态。
- 准备上下文消息。
- 调用 `fetchChatCompletion` API 服务。
- 使用 `createStreamProcessor` 处理流式响应。
- 通过各种回调 (`onTextChunk`, `onThinkingChunk`, `onToolCallComplete`, `onImageGenerated`, `onError`, `onComplete` 等) 处理不同类型的事件。
- **Block 相关**:
- 根据流事件创建初始 `UNKNOWN` 块。
- 实时创建和更新 `MAIN_TEXT``THINKING` 块,使用 `throttledBlockUpdate``throttledBlockDbUpdate` 进行节流更新。
- 创建 `TOOL`, `CITATION`, `IMAGE`, `ERROR` 等类型的块。
- 在事件完成时(如 `onTextComplete`, `onToolCallComplete`)将块状态标记为 `SUCCESS``ERROR`,并使用 `saveUpdatedBlockToDB` 保存最终状态。
- 使用 `handleBlockTransition` 管理非流式块(如 `TOOL`, `CITATION`)的添加和状态更新。
3. **`loadTopicMessagesThunk(topicId, forceReload)`**
- **用途**: 从数据库加载指定主题的所有消息及其关联的 `MessageBlock`
- **流程**:
- 从 DB 获取 `Topic` 及其 `messages` 列表。
- 根据消息 ID 列表从 DB 获取所有相关的 `MessageBlock`
- 使用 `upsertManyBlocks` 将块更新到 Redux。
- 将消息更新到 Redux。
- **Block 相关**: 负责将持久化的 `MessageBlock` 加载到 Redux 状态。
4. **删除 Thunks**
- `deleteSingleMessageThunk(topicId, messageId)`: 删除单个消息及其所有 `MessageBlock`
- `deleteMessageGroupThunk(topicId, askId)`: 删除一个用户消息及其所有相关的助手响应消息和它们的所有 `MessageBlock`
- `clearTopicMessagesThunk(topicId)`: 清空主题下的所有消息及其所有 `MessageBlock`
- **Block 相关**: 从 Redux 和 DB 中移除指定的 `MessageBlock`
5. **重发/重新生成 Thunks**
- `resendMessageThunk(topicId, userMessageToResend, assistant)`: 重发用户消息。会重置(清空 Block 并标记为 PENDING所有与该用户消息关联的助手响应然后重新请求生成。
- `resendUserMessageWithEditThunk(topicId, originalMessage, mainTextBlockId, editedContent, assistant)`: 用户编辑消息内容后重发。先更新用户消息的 `MAIN_TEXT` 块内容,然后调用 `resendMessageThunk`
- `regenerateAssistantResponseThunk(topicId, assistantMessageToRegenerate, assistant)`: 重新生成单个助手响应。重置该助手消息(清空 Block 并标记为 PENDING然后重新请求生成。
- **Block 相关**: 删除旧的 `MessageBlock`,并在重新生成过程中创建新的 `MessageBlock`
6. **`appendAssistantResponseThunk(topicId, existingAssistantMessageId, newModel, assistant)`**
- **用途**: 在已有的对话上下文中,针对同一个用户问题,使用新选择的模型追加一个新的助手响应。
- **流程**:
- 找到现有助手消息以获取原始 `askId`
- 创建使用 `newModel` 的新助手消息存根(使用相同的 `askId`)。
- 添加新存根到 Redux 和 DB。
-`fetchAndProcessAssistantResponseImpl` 添加到队列以生成新响应。
- **Block 相关**: 为新的助手响应创建全新的 `MessageBlock`
7. **`cloneMessagesToNewTopicThunk(sourceTopicId, branchPointIndex, newTopic)`**
- **用途**: 将源主题的部分消息(及其 Block克隆到一个**已存在**的新主题中。
- **流程**:
- 复制指定索引前的消息。
- 为所有克隆的消息和 Block 生成新的 UUID。
- 正确映射克隆消息之间的 `askId` 关系。
- 复制 `MessageBlock` 内容,更新其 `messageId` 指向新的消息 ID。
- 更新文件引用计数(如果 Block 是文件或图片)。
- 将克隆的消息和 Block 保存到新主题的 Redux 状态和 DB 中。
- **Block 相关**: 创建 `MessageBlock` 的副本,并更新其 ID 和 `messageId`
8. **`initiateTranslationThunk(messageId, topicId, targetLanguage, sourceBlockId?, sourceLanguage?)`**
- **用途**: 为指定消息启动翻译流程,创建一个初始的 `TRANSLATION` 类型的 `MessageBlock`
- **流程**:
- 创建一个状态为 `STREAMING``TranslationMessageBlock`
- 将其添加到 Redux 和 DB。
- 更新原消息的 `blocks` 列表以包含新的翻译块 ID。
- **Block 相关**: 创建并保存一个占位的 `TranslationMessageBlock`。实际翻译内容的获取和填充需要后续步骤。
## 内部机制和注意事项
- **数据库交互**: 通过 `saveMessageAndBlocksToDB`, `updateExistingMessageAndBlocksInDB`, `saveUpdatesToDB`, `saveUpdatedBlockToDB`, `throttledBlockDbUpdate` 等辅助函数与 IndexedDB (`db`) 交互,确保数据持久化。
- **状态同步**: Thunks 负责协调 Redux Store 和 IndexedDB 之间的数据一致性。
- **队列 (`getTopicQueue`)**: 使用 `AsyncQueue` 确保对同一主题的操作(尤其是 API 请求)按顺序执行,避免竞态条件。
- **节流 (`throttle`)**: 对流式响应中频繁的 Block 更新(文本、思考)使用 `lodash.throttle` 优化性能,减少 Redux dispatch 和 DB 写入次数。
- **错误处理**: `fetchAndProcessAssistantResponseImpl` 内的回调函数(特别是 `onError`)处理流处理和 API 调用中可能出现的错误,并创建 `ERROR` 类型的 `MessageBlock`
开发者在使用这些 Thunks 时,通常需要提供 `dispatch`, `getState` (由 Redux Thunk 中间件注入),以及如 `topicId`, `assistant` 配置对象, 相关的 `Message``MessageBlock` 对象/ID 等参数。理解每个 Thunk 的职责和它如何影响消息及块的状态至关重要。
---
# useMessageOperations.ts 使用指南
该文件定义了一个名为 `useMessageOperations` 的自定义 React Hook。这个 Hook 的主要目的是为 React 组件提供一个便捷的接口用于执行与特定主题Topic相关的各种消息操作。它封装了调用 Redux Thunks (`messageThunk.ts`) 和 Actions (`newMessage.ts`, `messageBlock.ts`) 的逻辑,简化了组件与消息数据交互的代码。
## 核心目标
- **封装**: 将复杂的消息操作逻辑(如删除、重发、重新生成、编辑、翻译等)封装在易于使用的函数中。
- **简化**: 让组件可以直接调用这些操作函数,而无需直接与 Redux `dispatch` 或 Thunks 交互。
- **上下文关联**: 所有操作都与传入的 `topic` 对象相关联,确保操作作用于正确的主题。
## 如何使用
在你的 React 函数组件中,导入并调用 `useMessageOperations` Hook并传入当前活动的 `Topic` 对象。
```typescript
import React from 'react';
import { useMessageOperations } from '@renderer/hooks/useMessageOperations';
import type { Topic, Message, Assistant, Model } from '@renderer/types';
interface MyComponentProps {
currentTopic: Topic;
currentAssistant: Assistant;
}
function MyComponent({ currentTopic, currentAssistant }: MyComponentProps) {
const {
deleteMessage,
resendMessage,
regenerateAssistantMessage,
appendAssistantResponse,
getTranslationUpdater,
createTopicBranch,
// ... 其他操作函数
} = useMessageOperations(currentTopic);
const handleDelete = (messageId: string) => {
deleteMessage(messageId);
};
const handleResend = (message: Message) => {
resendMessage(message, currentAssistant);
};
const handleAppend = (existingMsg: Message, newModel: Model) => {
appendAssistantResponse(existingMsg, newModel, currentAssistant);
}
// ... 在组件中使用其他操作函数
return (
<div>
{/* Component UI */}
<button onClick={() => handleDelete('some-message-id')}>Delete Message</button>
{/* ... */}
</div>
);
}
```
## 返回值
`useMessageOperations(topic)` Hook 返回一个包含以下函数和值的对象:
- **`deleteMessage(id: string)`**:
- 删除指定 `id` 的单个消息。
- 内部调用 `deleteSingleMessageThunk`
- **`deleteGroupMessages(askId: string)`**:
- 删除与指定 `askId` 相关联的一组消息(通常是用户提问及其所有助手回答)。
- 内部调用 `deleteMessageGroupThunk`
- **`editMessage(messageId: string, updates: Partial<Message>)`**:
- 更新指定 `messageId` 的消息的部分属性。
- **注意**: 目前主要用于更新 Redux 状态
- 内部调用 `newMessagesActions.updateMessage`
- **`resendMessage(message: Message, assistant: Assistant)`**:
- 重新发送指定的用户消息 (`message`),这将触发其所有关联助手响应的重新生成。
- 内部调用 `resendMessageThunk`
- **`resendUserMessageWithEdit(message: Message, editedContent: string, assistant: Assistant)`**:
- 在用户消息的主要文本块被编辑后,重新发送该消息。
- 会先查找消息的 `MAIN_TEXT` 块 ID然后调用 `resendUserMessageWithEditThunk`
- **`clearTopicMessages(_topicId?: string)`**:
- 清除当前主题(或可选的指定 `_topicId`)下的所有消息。
- 内部调用 `clearTopicMessagesThunk`
- **`createNewContext()`**:
- 发出一个全局事件 (`EVENT_NAMES.NEW_CONTEXT`),通常用于通知 UI 清空显示,准备新的上下文。不直接修改 Redux 状态。
- **`displayCount`**:
- (非操作函数) 从 Redux store 中获取当前的 `displayCount` 值。
- **`pauseMessages()`**:
- 尝试中止当前主题中正在进行的消息生成(状态为 `processing``pending`)。
- 通过查找相关的 `askId` 并调用 `abortCompletion` 来实现。
- 同时会 dispatch `setTopicLoading` action 将加载状态设为 `false`
- **`resumeMessage(message: Message, assistant: Assistant)`**:
- 恢复/重新发送一个用户消息。目前实现为直接调用 `resendMessage`
- **`regenerateAssistantMessage(message: Message, assistant: Assistant)`**:
- 重新生成指定的**助手**消息 (`message`) 的响应。
- 内部调用 `regenerateAssistantResponseThunk`
- **`appendAssistantResponse(existingAssistantMessage: Message, newModel: Model, assistant: Assistant)`**:
- 针对 `existingAssistantMessage` 所回复的**同一用户提问**,使用 `newModel` 追加一个新的助手响应。
- 内部调用 `appendAssistantResponseThunk`
- **`getTranslationUpdater(messageId: string, targetLanguage: string, sourceBlockId?: string, sourceLanguage?: string)`**:
- **用途**: 获取一个用于逐步更新翻译块内容的函数。
- **流程**:
1. 内部调用 `initiateTranslationThunk` 来创建或获取一个 `TRANSLATION` 类型的 `MessageBlock`,并获取其 `blockId`
2. 返回一个**异步更新函数**。
- **返回的更新函数 `(accumulatedText: string, isComplete?: boolean) => void`**:
- 接收累积的翻译文本和完成状态。
- 调用 `updateOneBlock` 更新 Redux 中的翻译块内容和状态 (`STREAMING``SUCCESS`)。
- 调用 `throttledBlockDbUpdate` 将更新(节流地)保存到数据库。
- 如果初始化失败Thunk 返回 `undefined`),则此函数返回 `null`
- **`createTopicBranch(sourceTopicId: string, branchPointIndex: number, newTopic: Topic)`**:
- 创建一个主题分支,将 `sourceTopicId` 主题中 `branchPointIndex` 索引之前的消息克隆到 `newTopic` 中。
- **注意**: `newTopic` 对象必须是调用此函数**之前**已经创建并添加到 Redux 和数据库中的。
- 内部调用 `cloneMessagesToNewTopicThunk`
## 依赖
- **`topic: Topic`**: 必须传入当前操作上下文的主题对象。Hook 返回的操作函数将始终作用于这个主题的 `topic.id`
- **Redux `dispatch`**: Hook 内部使用 `useAppDispatch` 获取 `dispatch` 函数来调用 actions 和 thunks。
## 相关 Hooks
在同一文件中还定义了两个辅助 Hook
- **`useTopicMessages(topic: Topic)`**:
- 使用 `selectMessagesForTopic` selector 来获取并返回指定主题的消息列表。
- **`useTopicLoading(topic: Topic)`**:
- 使用 `selectNewTopicLoading` selector 来获取并返回指定主题的加载状态。
这些 Hook 可以与 `useMessageOperations` 结合使用,方便地在组件中获取消息数据、加载状态,并执行相关操作。

View File

@@ -134,58 +134,56 @@ artifactBuildCompleted: scripts/artifact-build-completed.js
releaseInfo:
releaseNotes: |
<!--LANG:en-->
What's New in v1.7.0-rc.1
🎉 MAJOR NEW FEATURE: AI Agents
- Create and manage custom AI agents with specialized tools and permissions
- Dedicated agent sessions with persistent SQLite storage, separate from regular chats
- Real-time tool approval system - review and approve agent actions dynamically
- MCP (Model Context Protocol) integration for connecting external tools
- Slash commands support for quick agent interactions
- OpenAI-compatible REST API for agent access
What's New in v1.7.0-rc.3
✨ New Features:
- AI Providers: Added support for Hugging Face, Mistral, Perplexity, and SophNet
- Knowledge Base: OpenMinerU document preprocessor, full-text search in notes, enhanced tool selection
- Image & OCR: Intel OVMS painting provider and Intel OpenVINO (NPU) OCR support
- MCP Management: Redesigned interface with dual-column layout for easier management
- Languages: Added German language support
- Provider: Added Silicon provider support for Anthropic API compatibility
- Provider: AIHubMix support for nano banana
🐛 Bug Fixes:
- i18n: Clean up translation tags and untranslated strings
- Provider: Fixed Silicon provider code list
- Provider: Fixed Poe API reasoning parameters for GPT-5 and reasoning models
- Provider: Fixed duplicate /v1 in Anthropic API endpoints
- Provider: Fixed Azure provider handling in AI SDK integration
- Models: Added Claude Opus 4.5 pattern to THINKING_TOKEN_MAP
- Models: Improved Gemini reasoning and message handling
- Models: Fixed custom parameters for Gemini models
- Models: Fixed qwen-mt-flash text delta support
- Models: Fixed Groq verbosity setting
- UI: Fixed quota display and quota tips
- UI: Fixed web search button condition
- Settings: Fixed updateAssistantPreset reducer to properly update preset
- Settings: Respect enableMaxTokens setting when maxTokens is not configured
- SDK: Fixed header merging logic in AI SDK
⚡ Improvements:
- Upgraded to Electron 38.7.0
- Enhanced system shutdown handling and automatic update checks
- Improved proxy bypass rules
🐛 Important Bug Fixes:
- Fixed streaming response issues across multiple AI providers
- Fixed session list scrolling problems
- Fixed knowledge base deletion errors
- SDK: Upgraded @anthropic-ai/claude-agent-sdk to 0.1.53
<!--LANG:zh-CN-->
v1.7.0-rc.1 新特性
🎉 重大更新AI Agent 智能体系统
- 创建和管理专属 AI Agent配置专用工具和权限
- 独立的 Agent 会话,使用 SQLite 持久化存储,与普通聊天分离
- 实时工具审批系统 - 动态审查和批准 Agent 操作
- MCP模型上下文协议集成连接外部工具
- 支持斜杠命令快速交互
- 兼容 OpenAI 的 REST API 访问
v1.7.0-rc.3 更新内容
✨ 新功能:
- AI 提供商:新增 Hugging Face、Mistral、Perplexity 和 SophNet 支持
- 知识库OpenMinerU 文档预处理器、笔记全文搜索、增强的工具选择
- 图像与 OCRIntel OVMS 绘图提供商和 Intel OpenVINO (NPU) OCR 支持
- MCP 管理:重构管理界面,采用双列布局,更加方便管理
- 语言:新增德语支持
- 提供商:新增 Silicon 提供商对 Anthropic API 的兼容性支持
- 提供商AIHubMix 支持 nano banana
🐛 问题修复:
- 国际化:清理翻译标签和未翻译字符串
- 提供商:修复 Silicon 提供商代码列表
- 提供商:修复 Poe API 对 GPT-5 和推理模型的推理参数
- 提供商:修复 Anthropic API 端点重复 /v1 问题
- 提供商:修复 Azure 提供商在 AI SDK 集成中的处理
- 模型Claude Opus 4.5 添加到 THINKING_TOKEN_MAP
- 模型:改进 Gemini 推理和消息处理
- 模型:修复 Gemini 模型自定义参数
- 模型:修复 qwen-mt-flash text delta 支持
- 模型:修复 Groq verbosity 设置
- 界面:修复配额显示和配额提示
- 界面:修复 Web 搜索按钮条件
- 设置:修复 updateAssistantPreset reducer 正确更新 preset
- 设置:尊重 enableMaxTokens 设置
- SDK修复 AI SDK 中 header 合并逻辑
⚡ 改进:
- 升级到 Electron 38.7.0
- 增强的系统关机处理和自动更新检查
- 改进的代理绕过规则
🐛 重要修复:
- 修复多个 AI 提供商的流式响应问题
- 修复会话列表滚动问题
- 修复知识库删除错误
- SDK升级 @anthropic-ai/claude-agent-sdk 到 0.1.53
<!--LANG:END-->

View File

@@ -1,6 +1,6 @@
{
"name": "CherryStudio",
"version": "1.7.0-rc.1",
"version": "1.7.0-rc.3",
"private": true,
"description": "A powerful AI assistant for producer.",
"main": "./out/main/index.js",
@@ -80,12 +80,13 @@
"release:ai-sdk-provider": "yarn workspace @cherrystudio/ai-sdk-provider version patch --immediate && yarn workspace @cherrystudio/ai-sdk-provider build && yarn workspace @cherrystudio/ai-sdk-provider npm publish --access public"
},
"dependencies": {
"@anthropic-ai/claude-agent-sdk": "patch:@anthropic-ai/claude-agent-sdk@npm%3A0.1.30#~/.yarn/patches/@anthropic-ai-claude-agent-sdk-npm-0.1.30-b50a299674.patch",
"@anthropic-ai/claude-agent-sdk": "patch:@anthropic-ai/claude-agent-sdk@npm%3A0.1.53#~/.yarn/patches/@anthropic-ai-claude-agent-sdk-npm-0.1.53-4b77f4cf29.patch",
"@libsql/client": "0.14.0",
"@libsql/win32-x64-msvc": "^0.4.7",
"@napi-rs/system-ocr": "patch:@napi-rs/system-ocr@npm%3A1.0.2#~/.yarn/patches/@napi-rs-system-ocr-npm-1.0.2-59e7a78e8b.patch",
"@paymoapp/electron-shutdown-handler": "^1.1.2",
"@strongtz/win32-arm64-msvc": "^0.4.7",
"emoji-picker-element-data": "^1",
"express": "^5.1.0",
"font-list": "^2.0.0",
"graceful-fs": "^4.2.11",
@@ -108,16 +109,17 @@
"@agentic/exa": "^7.3.3",
"@agentic/searxng": "^7.3.3",
"@agentic/tavily": "^7.3.3",
"@ai-sdk/amazon-bedrock": "^3.0.53",
"@ai-sdk/anthropic": "^2.0.44",
"@ai-sdk/amazon-bedrock": "^3.0.61",
"@ai-sdk/anthropic": "^2.0.49",
"@ai-sdk/cerebras": "^1.0.31",
"@ai-sdk/gateway": "^2.0.9",
"@ai-sdk/google": "patch:@ai-sdk/google@npm%3A2.0.36#~/.yarn/patches/@ai-sdk-google-npm-2.0.36-6f3cc06026.patch",
"@ai-sdk/google-vertex": "^3.0.68",
"@ai-sdk/huggingface": "patch:@ai-sdk/huggingface@npm%3A0.0.8#~/.yarn/patches/@ai-sdk-huggingface-npm-0.0.8-d4d0aaac93.patch",
"@ai-sdk/mistral": "^2.0.23",
"@ai-sdk/openai": "patch:@ai-sdk/openai@npm%3A2.0.64#~/.yarn/patches/@ai-sdk-openai-npm-2.0.64-48f99f5bf3.patch",
"@ai-sdk/perplexity": "^2.0.17",
"@ai-sdk/gateway": "^2.0.15",
"@ai-sdk/google": "patch:@ai-sdk/google@npm%3A2.0.43#~/.yarn/patches/@ai-sdk-google-npm-2.0.43-689ed559b3.patch",
"@ai-sdk/google-vertex": "^3.0.79",
"@ai-sdk/huggingface": "^0.0.10",
"@ai-sdk/mistral": "^2.0.24",
"@ai-sdk/openai": "patch:@ai-sdk/openai@npm%3A2.0.72#~/.yarn/patches/@ai-sdk-openai-npm-2.0.72-234e68da87.patch",
"@ai-sdk/perplexity": "^2.0.20",
"@ai-sdk/test-server": "^0.0.1",
"@ant-design/v5-patch-for-react-19": "^1.0.3",
"@anthropic-ai/sdk": "^0.41.0",
"@anthropic-ai/vertex-sdk": "patch:@anthropic-ai/vertex-sdk@npm%3A0.11.4#~/.yarn/patches/@anthropic-ai-vertex-sdk-npm-0.11.4-c19cb41edb.patch",
@@ -162,14 +164,14 @@
"@modelcontextprotocol/sdk": "^1.17.5",
"@mozilla/readability": "^0.6.0",
"@notionhq/client": "^2.2.15",
"@openrouter/ai-sdk-provider": "^1.2.0",
"@openrouter/ai-sdk-provider": "^1.2.5",
"@opentelemetry/api": "^1.9.0",
"@opentelemetry/core": "2.0.0",
"@opentelemetry/exporter-trace-otlp-http": "^0.200.0",
"@opentelemetry/sdk-trace-base": "^2.0.0",
"@opentelemetry/sdk-trace-node": "^2.0.0",
"@opentelemetry/sdk-trace-web": "^2.0.0",
"@opeoginni/github-copilot-openai-compatible": "0.1.21",
"@opeoginni/github-copilot-openai-compatible": "^0.1.21",
"@playwright/test": "^1.52.0",
"@radix-ui/react-context-menu": "^2.2.16",
"@reduxjs/toolkit": "^2.2.5",
@@ -215,8 +217,8 @@
"@types/mime-types": "^3",
"@types/node": "^22.17.1",
"@types/pako": "^1.0.2",
"@types/react": "^19.0.12",
"@types/react-dom": "^19.0.4",
"@types/react": "^19.2.7",
"@types/react-dom": "^19.2.3",
"@types/react-infinite-scroll-component": "^5.0.0",
"@types/react-transition-group": "^4.4.12",
"@types/react-window": "^1",
@@ -238,7 +240,7 @@
"@viz-js/lang-dot": "^1.0.5",
"@viz-js/viz": "^3.14.0",
"@xyflow/react": "^12.4.4",
"ai": "^5.0.90",
"ai": "^5.0.98",
"antd": "patch:antd@npm%3A5.27.0#~/.yarn/patches/antd-npm-5.27.0-aa91c36546.patch",
"archiver": "^7.0.1",
"async-mutex": "^0.5.0",
@@ -410,9 +412,9 @@
"@langchain/openai@npm:>=0.1.0 <0.6.0": "patch:@langchain/openai@npm%3A1.0.0#~/.yarn/patches/@langchain-openai-npm-1.0.0-474d0ad9d4.patch",
"@langchain/openai@npm:^0.3.16": "patch:@langchain/openai@npm%3A1.0.0#~/.yarn/patches/@langchain-openai-npm-1.0.0-474d0ad9d4.patch",
"@langchain/openai@npm:>=0.2.0 <0.7.0": "patch:@langchain/openai@npm%3A1.0.0#~/.yarn/patches/@langchain-openai-npm-1.0.0-474d0ad9d4.patch",
"@ai-sdk/openai@npm:2.0.64": "patch:@ai-sdk/openai@npm%3A2.0.64#~/.yarn/patches/@ai-sdk-openai-npm-2.0.64-48f99f5bf3.patch",
"@ai-sdk/openai@npm:^2.0.42": "patch:@ai-sdk/openai@npm%3A2.0.64#~/.yarn/patches/@ai-sdk-openai-npm-2.0.64-48f99f5bf3.patch",
"@ai-sdk/google@npm:2.0.36": "patch:@ai-sdk/google@npm%3A2.0.36#~/.yarn/patches/@ai-sdk-google-npm-2.0.36-6f3cc06026.patch"
"@ai-sdk/openai@npm:^2.0.42": "patch:@ai-sdk/openai@npm%3A2.0.72#~/.yarn/patches/@ai-sdk-openai-npm-2.0.72-234e68da87.patch",
"@ai-sdk/google@npm:^2.0.40": "patch:@ai-sdk/google@npm%3A2.0.40#~/.yarn/patches/@ai-sdk-google-npm-2.0.40-47e0eeee83.patch",
"@ai-sdk/openai-compatible@npm:^1.0.27": "patch:@ai-sdk/openai-compatible@npm%3A1.0.27#~/.yarn/patches/@ai-sdk-openai-compatible-npm-1.0.27-06f74278cf.patch"
},
"packageManager": "yarn@4.9.1",
"lint-staged": {

View File

@@ -1,6 +1,6 @@
{
"name": "@cherrystudio/ai-sdk-provider",
"version": "0.1.2",
"version": "0.1.3",
"description": "Cherry Studio AI SDK provider bundle with CherryIN routing.",
"keywords": [
"ai-sdk",
@@ -42,7 +42,7 @@
},
"dependencies": {
"@ai-sdk/provider": "^2.0.0",
"@ai-sdk/provider-utils": "^3.0.12"
"@ai-sdk/provider-utils": "^3.0.17"
},
"devDependencies": {
"tsdown": "^0.13.3",

View File

@@ -67,6 +67,10 @@ export interface CherryInProviderSettings {
* Optional static headers applied to every request.
*/
headers?: HeadersInput
/**
* Optional endpoint type to distinguish different endpoint behaviors.
*/
endpointType?: 'openai' | 'openai-response' | 'anthropic' | 'gemini' | 'image-generation' | 'jina-rerank'
}
export interface CherryInProvider extends ProviderV2 {
@@ -151,7 +155,8 @@ export const createCherryIn = (options: CherryInProviderSettings = {}): CherryIn
baseURL = DEFAULT_CHERRYIN_BASE_URL,
anthropicBaseURL = DEFAULT_CHERRYIN_ANTHROPIC_BASE_URL,
geminiBaseURL = DEFAULT_CHERRYIN_GEMINI_BASE_URL,
fetch
fetch,
endpointType
} = options
const getJsonHeaders = createJsonHeadersGetter(options)
@@ -205,7 +210,7 @@ export const createCherryIn = (options: CherryInProviderSettings = {}): CherryIn
fetch
})
const createChatModel = (modelId: string, settings: OpenAIProviderSettings = {}) => {
const createChatModelByModelId = (modelId: string, settings: OpenAIProviderSettings = {}) => {
if (isAnthropicModel(modelId)) {
return createAnthropicModel(modelId)
}
@@ -223,6 +228,29 @@ export const createCherryIn = (options: CherryInProviderSettings = {}): CherryIn
})
}
const createChatModel = (modelId: string, settings: OpenAIProviderSettings = {}) => {
if (!endpointType) return createChatModelByModelId(modelId, settings)
switch (endpointType) {
case 'anthropic':
return createAnthropicModel(modelId)
case 'gemini':
return createGeminiModel(modelId)
case 'openai':
return createOpenAIChatModel(modelId)
case 'openai-response':
default:
return new OpenAIResponsesLanguageModel(modelId, {
provider: `${CHERRYIN_PROVIDER_NAME}.openai`,
url,
headers: () => ({
...getJsonHeaders(),
...settings.headers
}),
fetch
})
}
}
const createCompletionModel = (modelId: string, settings: OpenAIProviderSettings = {}) =>
new OpenAICompletionLanguageModel(modelId, {
provider: `${CHERRYIN_PROVIDER_NAME}.completion`,

View File

@@ -35,17 +35,17 @@
"peerDependencies": {
"@ai-sdk/google": "^2.0.36",
"@ai-sdk/openai": "^2.0.64",
"@cherrystudio/ai-sdk-provider": "^0.1.2",
"@cherrystudio/ai-sdk-provider": "^0.1.3",
"ai": "^5.0.26"
},
"dependencies": {
"@ai-sdk/anthropic": "^2.0.43",
"@ai-sdk/azure": "^2.0.66",
"@ai-sdk/deepseek": "^1.0.27",
"@ai-sdk/openai-compatible": "^1.0.26",
"@ai-sdk/anthropic": "^2.0.49",
"@ai-sdk/azure": "^2.0.74",
"@ai-sdk/deepseek": "^1.0.29",
"@ai-sdk/openai-compatible": "patch:@ai-sdk/openai-compatible@npm%3A1.0.27#~/.yarn/patches/@ai-sdk-openai-compatible-npm-1.0.27-06f74278cf.patch",
"@ai-sdk/provider": "^2.0.0",
"@ai-sdk/provider-utils": "^3.0.16",
"@ai-sdk/xai": "^2.0.31",
"@ai-sdk/provider-utils": "^3.0.17",
"@ai-sdk/xai": "^2.0.36",
"zod": "^4.1.5"
},
"devDependencies": {

View File

@@ -0,0 +1,180 @@
/**
* Mock Provider Instances
* Provides mock implementations for all supported AI providers
*/
import type { ImageModelV2, LanguageModelV2 } from '@ai-sdk/provider'
import { vi } from 'vitest'
/**
* Creates a mock language model with customizable behavior
*/
export function createMockLanguageModel(overrides?: Partial<LanguageModelV2>): LanguageModelV2 {
return {
specificationVersion: 'v1',
provider: 'mock-provider',
modelId: 'mock-model',
defaultObjectGenerationMode: 'tool',
doGenerate: vi.fn().mockResolvedValue({
text: 'Mock response text',
finishReason: 'stop',
usage: {
promptTokens: 10,
completionTokens: 20,
totalTokens: 30
},
rawCall: { rawPrompt: null, rawSettings: {} },
rawResponse: { headers: {} },
warnings: []
}),
doStream: vi.fn().mockReturnValue({
stream: (async function* () {
yield {
type: 'text-delta',
textDelta: 'Mock '
}
yield {
type: 'text-delta',
textDelta: 'streaming '
}
yield {
type: 'text-delta',
textDelta: 'response'
}
yield {
type: 'finish',
finishReason: 'stop',
usage: {
promptTokens: 10,
completionTokens: 15,
totalTokens: 25
}
}
})(),
rawCall: { rawPrompt: null, rawSettings: {} },
rawResponse: { headers: {} },
warnings: []
}),
...overrides
} as LanguageModelV2
}
/**
* Creates a mock image model with customizable behavior
*/
export function createMockImageModel(overrides?: Partial<ImageModelV2>): ImageModelV2 {
return {
specificationVersion: 'v2',
provider: 'mock-provider',
modelId: 'mock-image-model',
doGenerate: vi.fn().mockResolvedValue({
images: [
{
base64: 'mock-base64-image-data',
uint8Array: new Uint8Array([1, 2, 3, 4, 5]),
mimeType: 'image/png'
}
],
warnings: []
}),
...overrides
} as ImageModelV2
}
/**
* Mock provider configurations for testing
*/
export const mockProviderConfigs = {
openai: {
apiKey: 'sk-test-openai-key-123456789',
baseURL: 'https://api.openai.com/v1',
organization: 'test-org'
},
anthropic: {
apiKey: 'sk-ant-test-key-123456789',
baseURL: 'https://api.anthropic.com'
},
google: {
apiKey: 'test-google-api-key-123456789',
baseURL: 'https://generativelanguage.googleapis.com/v1'
},
xai: {
apiKey: 'xai-test-key-123456789',
baseURL: 'https://api.x.ai/v1'
},
azure: {
apiKey: 'test-azure-key-123456789',
resourceName: 'test-resource',
deployment: 'test-deployment'
},
deepseek: {
apiKey: 'sk-test-deepseek-key-123456789',
baseURL: 'https://api.deepseek.com/v1'
},
openrouter: {
apiKey: 'sk-or-test-key-123456789',
baseURL: 'https://openrouter.ai/api/v1'
},
huggingface: {
apiKey: 'hf_test_key_123456789',
baseURL: 'https://api-inference.huggingface.co'
},
'openai-compatible': {
apiKey: 'test-compatible-key-123456789',
baseURL: 'https://api.example.com/v1',
name: 'test-provider'
},
'openai-chat': {
apiKey: 'sk-test-chat-key-123456789',
baseURL: 'https://api.openai.com/v1'
}
} as const
/**
* Mock provider instances for testing
*/
export const mockProviderInstances = {
openai: {
name: 'openai-mock',
languageModel: createMockLanguageModel({ provider: 'openai', modelId: 'gpt-4' }),
imageModel: createMockImageModel({ provider: 'openai', modelId: 'dall-e-3' })
},
anthropic: {
name: 'anthropic-mock',
languageModel: createMockLanguageModel({ provider: 'anthropic', modelId: 'claude-3-5-sonnet-20241022' })
},
google: {
name: 'google-mock',
languageModel: createMockLanguageModel({ provider: 'google', modelId: 'gemini-2.0-flash-exp' }),
imageModel: createMockImageModel({ provider: 'google', modelId: 'imagen-3.0-generate-001' })
},
xai: {
name: 'xai-mock',
languageModel: createMockLanguageModel({ provider: 'xai', modelId: 'grok-2-latest' }),
imageModel: createMockImageModel({ provider: 'xai', modelId: 'grok-2-image-latest' })
},
deepseek: {
name: 'deepseek-mock',
languageModel: createMockLanguageModel({ provider: 'deepseek', modelId: 'deepseek-chat' })
}
}
export type ProviderId = keyof typeof mockProviderConfigs

View File

@@ -0,0 +1,331 @@
/**
* Mock Responses
* Provides realistic mock responses for all provider types
*/
import { jsonSchema, type ModelMessage, type Tool } from 'ai'
/**
* Standard test messages for all scenarios
*/
export const testMessages = {
simple: [{ role: 'user' as const, content: 'Hello, how are you?' }],
conversation: [
{ role: 'user' as const, content: 'What is the capital of France?' },
{ role: 'assistant' as const, content: 'The capital of France is Paris.' },
{ role: 'user' as const, content: 'What is its population?' }
],
withSystem: [
{ role: 'system' as const, content: 'You are a helpful assistant that provides concise answers.' },
{ role: 'user' as const, content: 'Explain quantum computing in one sentence.' }
],
withImages: [
{
role: 'user' as const,
content: [
{ type: 'text' as const, text: 'What is in this image?' },
{
type: 'image' as const,
image:
'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg=='
}
]
}
],
toolUse: [{ role: 'user' as const, content: 'What is the weather in San Francisco?' }],
multiTurn: [
{ role: 'user' as const, content: 'Can you help me with a math problem?' },
{ role: 'assistant' as const, content: 'Of course! What math problem would you like help with?' },
{ role: 'user' as const, content: 'What is 15 * 23?' },
{ role: 'assistant' as const, content: '15 * 23 = 345' },
{ role: 'user' as const, content: 'Now divide that by 5' }
]
} satisfies Record<string, ModelMessage[]>
/**
* Standard test tools for tool calling scenarios
*/
export const testTools: Record<string, Tool> = {
getWeather: {
description: 'Get the current weather in a given location',
inputSchema: jsonSchema({
type: 'object',
properties: {
location: {
type: 'string',
description: 'The city and state, e.g. San Francisco, CA'
},
unit: {
type: 'string',
enum: ['celsius', 'fahrenheit'],
description: 'The temperature unit to use'
}
},
required: ['location']
}),
execute: async ({ location, unit = 'fahrenheit' }) => {
return {
location,
temperature: unit === 'celsius' ? 22 : 72,
unit,
condition: 'sunny'
}
}
},
calculate: {
description: 'Perform a mathematical calculation',
inputSchema: jsonSchema({
type: 'object',
properties: {
operation: {
type: 'string',
enum: ['add', 'subtract', 'multiply', 'divide'],
description: 'The operation to perform'
},
a: {
type: 'number',
description: 'The first number'
},
b: {
type: 'number',
description: 'The second number'
}
},
required: ['operation', 'a', 'b']
}),
execute: async ({ operation, a, b }) => {
const operations = {
add: (x: number, y: number) => x + y,
subtract: (x: number, y: number) => x - y,
multiply: (x: number, y: number) => x * y,
divide: (x: number, y: number) => x / y
}
return { result: operations[operation as keyof typeof operations](a, b) }
}
},
searchDatabase: {
description: 'Search for information in a database',
inputSchema: jsonSchema({
type: 'object',
properties: {
query: {
type: 'string',
description: 'The search query'
},
limit: {
type: 'number',
description: 'Maximum number of results to return',
default: 10
}
},
required: ['query']
}),
execute: async ({ query, limit = 10 }) => {
return {
results: [
{ id: 1, title: `Result 1 for ${query}`, relevance: 0.95 },
{ id: 2, title: `Result 2 for ${query}`, relevance: 0.87 }
].slice(0, limit)
}
}
}
}
/**
* Mock streaming chunks for different providers
*/
export const mockStreamingChunks = {
text: [
{ type: 'text-delta' as const, textDelta: 'Hello' },
{ type: 'text-delta' as const, textDelta: ', ' },
{ type: 'text-delta' as const, textDelta: 'this ' },
{ type: 'text-delta' as const, textDelta: 'is ' },
{ type: 'text-delta' as const, textDelta: 'a ' },
{ type: 'text-delta' as const, textDelta: 'test.' }
],
withToolCall: [
{ type: 'text-delta' as const, textDelta: 'Let me check the weather for you.' },
{
type: 'tool-call-delta' as const,
toolCallType: 'function' as const,
toolCallId: 'call_123',
toolName: 'getWeather',
argsTextDelta: '{"location":'
},
{
type: 'tool-call-delta' as const,
toolCallType: 'function' as const,
toolCallId: 'call_123',
toolName: 'getWeather',
argsTextDelta: ' "San Francisco, CA"}'
},
{
type: 'tool-call' as const,
toolCallType: 'function' as const,
toolCallId: 'call_123',
toolName: 'getWeather',
args: { location: 'San Francisco, CA' }
}
],
withFinish: [
{ type: 'text-delta' as const, textDelta: 'Complete response.' },
{
type: 'finish' as const,
finishReason: 'stop' as const,
usage: {
promptTokens: 10,
completionTokens: 5,
totalTokens: 15
}
}
]
}
/**
* Mock complete responses for non-streaming scenarios
*/
export const mockCompleteResponses = {
simple: {
text: 'This is a simple response.',
finishReason: 'stop' as const,
usage: {
promptTokens: 15,
completionTokens: 8,
totalTokens: 23
}
},
withToolCalls: {
text: 'I will check the weather for you.',
toolCalls: [
{
toolCallId: 'call_456',
toolName: 'getWeather',
args: { location: 'New York, NY', unit: 'celsius' }
}
],
finishReason: 'tool-calls' as const,
usage: {
promptTokens: 25,
completionTokens: 12,
totalTokens: 37
}
},
withWarnings: {
text: 'Response with warnings.',
finishReason: 'stop' as const,
usage: {
promptTokens: 10,
completionTokens: 5,
totalTokens: 15
},
warnings: [
{
type: 'unsupported-setting' as const,
message: 'Temperature parameter not supported for this model'
}
]
}
}
/**
* Mock image generation responses
*/
export const mockImageResponses = {
single: {
image: {
base64: 'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==',
uint8Array: new Uint8Array([137, 80, 78, 71, 13, 10, 26, 10, 0, 0, 0, 13, 73, 72, 68, 82]),
mimeType: 'image/png' as const
},
warnings: []
},
multiple: {
images: [
{
base64: 'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==',
uint8Array: new Uint8Array([137, 80, 78, 71]),
mimeType: 'image/png' as const
},
{
base64: 'iVBORw0KGgoAAAANSUhEUgAAAAIAAAACCAYAAABytg0kAAAAEklEQVR42mNk+M9QzwAEjDAGACCKAgdZ9zImAAAAAElFTkSuQmCC',
uint8Array: new Uint8Array([137, 80, 78, 71]),
mimeType: 'image/png' as const
}
],
warnings: []
},
withProviderMetadata: {
image: {
base64: 'iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==',
uint8Array: new Uint8Array([137, 80, 78, 71]),
mimeType: 'image/png' as const
},
providerMetadata: {
openai: {
images: [
{
revisedPrompt: 'A detailed and enhanced version of the original prompt'
}
]
}
},
warnings: []
}
}
/**
* Mock error responses
*/
export const mockErrors = {
invalidApiKey: {
name: 'APIError',
message: 'Invalid API key provided',
statusCode: 401
},
rateLimitExceeded: {
name: 'RateLimitError',
message: 'Rate limit exceeded. Please try again later.',
statusCode: 429,
headers: {
'retry-after': '60'
}
},
modelNotFound: {
name: 'ModelNotFoundError',
message: 'The requested model was not found',
statusCode: 404
},
contextLengthExceeded: {
name: 'ContextLengthError',
message: "This model's maximum context length is 4096 tokens",
statusCode: 400
},
timeout: {
name: 'TimeoutError',
message: 'Request timed out after 30000ms',
code: 'ETIMEDOUT'
},
networkError: {
name: 'NetworkError',
message: 'Network connection failed',
code: 'ECONNREFUSED'
}
}

View File

@@ -0,0 +1,329 @@
/**
* Provider-Specific Test Utilities
* Helper functions for testing individual providers with all their parameters
*/
import type { Tool } from 'ai'
import { expect } from 'vitest'
/**
* Provider parameter configurations for comprehensive testing
*/
export const providerParameterMatrix = {
openai: {
models: ['gpt-4', 'gpt-4-turbo', 'gpt-3.5-turbo', 'gpt-4o'],
parameters: {
temperature: [0, 0.5, 0.7, 1.0, 1.5, 2.0],
maxTokens: [100, 500, 1000, 2000, 4000],
topP: [0.1, 0.5, 0.9, 1.0],
frequencyPenalty: [-2.0, -1.0, 0, 1.0, 2.0],
presencePenalty: [-2.0, -1.0, 0, 1.0, 2.0],
stop: [undefined, ['stop'], ['STOP', 'END']],
seed: [undefined, 12345, 67890],
responseFormat: [undefined, { type: 'json_object' as const }],
user: [undefined, 'test-user-123']
},
toolChoice: ['auto', 'required', 'none', { type: 'function' as const, name: 'getWeather' }],
parallelToolCalls: [true, false]
},
anthropic: {
models: ['claude-3-5-sonnet-20241022', 'claude-3-opus-20240229', 'claude-3-haiku-20240307'],
parameters: {
temperature: [0, 0.5, 1.0],
maxTokens: [100, 1000, 4000, 8000],
topP: [0.1, 0.5, 0.9, 1.0],
topK: [undefined, 1, 5, 10, 40],
stop: [undefined, ['Human:', 'Assistant:']],
metadata: [undefined, { userId: 'test-123' }]
},
toolChoice: ['auto', 'any', { type: 'tool' as const, name: 'getWeather' }]
},
google: {
models: ['gemini-2.0-flash-exp', 'gemini-1.5-pro', 'gemini-1.5-flash'],
parameters: {
temperature: [0, 0.5, 0.9, 1.0],
maxTokens: [100, 1000, 2000, 8000],
topP: [0.1, 0.5, 0.95, 1.0],
topK: [undefined, 1, 16, 40],
stopSequences: [undefined, ['END'], ['STOP', 'TERMINATE']]
},
safetySettings: [
undefined,
[
{ category: 'HARM_CATEGORY_HARASSMENT', threshold: 'BLOCK_MEDIUM_AND_ABOVE' },
{ category: 'HARM_CATEGORY_HATE_SPEECH', threshold: 'BLOCK_ONLY_HIGH' }
]
]
},
xai: {
models: ['grok-2-latest', 'grok-2-1212'],
parameters: {
temperature: [0, 0.5, 1.0, 1.5],
maxTokens: [100, 500, 2000, 4000],
topP: [0.1, 0.5, 0.9, 1.0],
stop: [undefined, ['STOP'], ['END', 'TERMINATE']],
seed: [undefined, 12345]
}
},
deepseek: {
models: ['deepseek-chat', 'deepseek-coder'],
parameters: {
temperature: [0, 0.5, 1.0],
maxTokens: [100, 1000, 4000],
topP: [0.1, 0.5, 0.95],
frequencyPenalty: [0, 0.5, 1.0],
presencePenalty: [0, 0.5, 1.0],
stop: [undefined, ['```'], ['END']]
}
},
azure: {
deployments: ['gpt-4-deployment', 'gpt-35-turbo-deployment'],
parameters: {
temperature: [0, 0.7, 1.0],
maxTokens: [100, 1000, 2000],
topP: [0.1, 0.5, 0.95],
frequencyPenalty: [0, 1.0],
presencePenalty: [0, 1.0],
stop: [undefined, ['STOP']]
}
}
} as const
/**
* Creates test cases for all parameter combinations
*/
export function generateParameterTestCases<T extends Record<string, any[]>>(
params: T,
maxCombinations = 50
): Array<Partial<{ [K in keyof T]: T[K][number] }>> {
const keys = Object.keys(params) as Array<keyof T>
const testCases: Array<Partial<{ [K in keyof T]: T[K][number] }>> = []
// Generate combinations using sampling strategy for large parameter spaces
const totalCombinations = keys.reduce((acc, key) => acc * params[key].length, 1)
if (totalCombinations <= maxCombinations) {
// Generate all combinations if total is small
generateAllCombinations(params, keys, 0, {}, testCases)
} else {
// Sample diverse combinations if total is large
generateSampledCombinations(params, keys, maxCombinations, testCases)
}
return testCases
}
function generateAllCombinations<T extends Record<string, any[]>>(
params: T,
keys: Array<keyof T>,
index: number,
current: Partial<{ [K in keyof T]: T[K][number] }>,
results: Array<Partial<{ [K in keyof T]: T[K][number] }>>
) {
if (index === keys.length) {
results.push({ ...current })
return
}
const key = keys[index]
for (const value of params[key]) {
generateAllCombinations(params, keys, index + 1, { ...current, [key]: value }, results)
}
}
function generateSampledCombinations<T extends Record<string, any[]>>(
params: T,
keys: Array<keyof T>,
count: number,
results: Array<Partial<{ [K in keyof T]: T[K][number] }>>
) {
// Generate edge cases first (min/max values)
const edgeCase1: any = {}
const edgeCase2: any = {}
for (const key of keys) {
edgeCase1[key] = params[key][0]
edgeCase2[key] = params[key][params[key].length - 1]
}
results.push(edgeCase1, edgeCase2)
// Generate random combinations for the rest
for (let i = results.length; i < count; i++) {
const combination: any = {}
for (const key of keys) {
const values = params[key]
combination[key] = values[Math.floor(Math.random() * values.length)]
}
results.push(combination)
}
}
/**
* Validates that all provider-specific parameters are correctly passed through
*/
export function validateProviderParams(providerId: string, actualParams: any, expectedParams: any): void {
const requiredFields: Record<string, string[]> = {
openai: ['model', 'messages'],
anthropic: ['model', 'messages'],
google: ['model', 'contents'],
xai: ['model', 'messages'],
deepseek: ['model', 'messages'],
azure: ['messages']
}
const fields = requiredFields[providerId] || ['model', 'messages']
for (const field of fields) {
expect(actualParams).toHaveProperty(field)
}
// Validate optional parameters if they were provided
const optionalParams = ['temperature', 'max_tokens', 'top_p', 'stop', 'tools']
for (const param of optionalParams) {
if (expectedParams[param] !== undefined) {
expect(actualParams[param]).toEqual(expectedParams[param])
}
}
}
/**
* Creates a comprehensive test suite for a provider
*/
// oxlint-disable-next-line no-unused-vars
export function createProviderTestSuite(_providerId: string) {
return {
testBasicCompletion: async (executor: any, model: string) => {
const result = await executor.generateText({
model,
messages: [{ role: 'user' as const, content: 'Hello' }]
})
expect(result).toBeDefined()
expect(result.text).toBeDefined()
expect(typeof result.text).toBe('string')
},
testStreaming: async (executor: any, model: string) => {
const chunks: any[] = []
const result = await executor.streamText({
model,
messages: [{ role: 'user' as const, content: 'Hello' }]
})
for await (const chunk of result.textStream) {
chunks.push(chunk)
}
expect(chunks.length).toBeGreaterThan(0)
},
testTemperature: async (executor: any, model: string, temperatures: number[]) => {
for (const temperature of temperatures) {
const result = await executor.generateText({
model,
messages: [{ role: 'user' as const, content: 'Hello' }],
temperature
})
expect(result).toBeDefined()
}
},
testMaxTokens: async (executor: any, model: string, maxTokensValues: number[]) => {
for (const maxTokens of maxTokensValues) {
const result = await executor.generateText({
model,
messages: [{ role: 'user' as const, content: 'Hello' }],
maxTokens
})
expect(result).toBeDefined()
if (result.usage?.completionTokens) {
expect(result.usage.completionTokens).toBeLessThanOrEqual(maxTokens)
}
}
},
testToolCalling: async (executor: any, model: string, tools: Record<string, Tool>) => {
const result = await executor.generateText({
model,
messages: [{ role: 'user' as const, content: 'What is the weather in SF?' }],
tools
})
expect(result).toBeDefined()
},
testStopSequences: async (executor: any, model: string, stopSequences: string[][]) => {
for (const stop of stopSequences) {
const result = await executor.generateText({
model,
messages: [{ role: 'user' as const, content: 'Count to 10' }],
stop
})
expect(result).toBeDefined()
}
}
}
}
/**
* Generates test data for vision/multimodal testing
*/
export function createVisionTestData() {
return {
imageUrl: 'https://example.com/test-image.jpg',
base64Image:
'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==',
messages: [
{
role: 'user' as const,
content: [
{ type: 'text' as const, text: 'What is in this image?' },
{
type: 'image' as const,
image:
'data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg=='
}
]
}
]
}
}
/**
* Creates mock responses for different finish reasons
*/
export function createFinishReasonMocks() {
return {
stop: {
text: 'Complete response.',
finishReason: 'stop' as const,
usage: { promptTokens: 10, completionTokens: 5, totalTokens: 15 }
},
length: {
text: 'Incomplete response due to',
finishReason: 'length' as const,
usage: { promptTokens: 10, completionTokens: 100, totalTokens: 110 }
},
'tool-calls': {
text: 'Calling tools',
finishReason: 'tool-calls' as const,
toolCalls: [{ toolCallId: 'call_1', toolName: 'getWeather', args: { location: 'SF' } }],
usage: { promptTokens: 10, completionTokens: 8, totalTokens: 18 }
},
'content-filter': {
text: '',
finishReason: 'content-filter' as const,
usage: { promptTokens: 10, completionTokens: 0, totalTokens: 10 }
}
}
}

View File

@@ -0,0 +1,291 @@
/**
* Test Utilities
* Helper functions for testing AI Core functionality
*/
import { expect, vi } from 'vitest'
import type { ProviderId } from '../fixtures/mock-providers'
import { createMockImageModel, createMockLanguageModel, mockProviderConfigs } from '../fixtures/mock-providers'
/**
* Creates a test provider with streaming support
*/
export function createTestStreamingProvider(chunks: any[]) {
return createMockLanguageModel({
doStream: vi.fn().mockReturnValue({
stream: (async function* () {
for (const chunk of chunks) {
yield chunk
}
})(),
rawCall: { rawPrompt: null, rawSettings: {} },
rawResponse: { headers: {} },
warnings: []
})
})
}
/**
* Creates a test provider that throws errors
*/
export function createErrorProvider(error: Error) {
return createMockLanguageModel({
doGenerate: vi.fn().mockRejectedValue(error),
doStream: vi.fn().mockImplementation(() => {
throw error
})
})
}
/**
* Collects all chunks from a stream
*/
export async function collectStreamChunks<T>(stream: AsyncIterable<T>): Promise<T[]> {
const chunks: T[] = []
for await (const chunk of stream) {
chunks.push(chunk)
}
return chunks
}
/**
* Waits for a specific number of milliseconds
*/
export function wait(ms: number): Promise<void> {
return new Promise((resolve) => setTimeout(resolve, ms))
}
/**
* Creates a mock abort controller that aborts after a delay
*/
export function createDelayedAbortController(delayMs: number): AbortController {
const controller = new AbortController()
setTimeout(() => controller.abort(), delayMs)
return controller
}
/**
* Asserts that a function throws an error with a specific message
*/
export async function expectError(fn: () => Promise<any>, expectedMessage?: string | RegExp): Promise<Error> {
try {
await fn()
throw new Error('Expected function to throw an error, but it did not')
} catch (error) {
if (expectedMessage) {
const message = (error as Error).message
if (typeof expectedMessage === 'string') {
if (!message.includes(expectedMessage)) {
throw new Error(`Expected error message to include "${expectedMessage}", but got "${message}"`)
}
} else {
if (!expectedMessage.test(message)) {
throw new Error(`Expected error message to match ${expectedMessage}, but got "${message}"`)
}
}
}
return error as Error
}
}
/**
* Creates a spy function that tracks calls and arguments
*/
export function createSpy<T extends (...args: any[]) => any>() {
const calls: Array<{ args: Parameters<T>; result?: ReturnType<T>; error?: Error }> = []
const spy = vi.fn((...args: Parameters<T>) => {
try {
const result = undefined as ReturnType<T>
calls.push({ args, result })
return result
} catch (error) {
calls.push({ args, error: error as Error })
throw error
}
})
return {
fn: spy,
calls,
getCalls: () => calls,
getCallCount: () => calls.length,
getLastCall: () => calls[calls.length - 1],
reset: () => {
calls.length = 0
spy.mockClear()
}
}
}
/**
* Validates provider configuration
*/
export function validateProviderConfig(providerId: ProviderId) {
const config = mockProviderConfigs[providerId]
if (!config) {
throw new Error(`No mock configuration found for provider: ${providerId}`)
}
if (!config.apiKey) {
throw new Error(`Provider ${providerId} is missing apiKey in mock config`)
}
return config
}
/**
* Creates a test context with common setup
*/
export function createTestContext() {
const mocks = {
languageModel: createMockLanguageModel(),
imageModel: createMockImageModel(),
providers: new Map<string, any>()
}
const cleanup = () => {
mocks.providers.clear()
vi.clearAllMocks()
}
return {
mocks,
cleanup
}
}
/**
* Measures execution time of an async function
*/
export async function measureTime<T>(fn: () => Promise<T>): Promise<{ result: T; duration: number }> {
const start = Date.now()
const result = await fn()
const duration = Date.now() - start
return { result, duration }
}
/**
* Retries a function until it succeeds or max attempts reached
*/
export async function retryUntilSuccess<T>(fn: () => Promise<T>, maxAttempts = 3, delayMs = 100): Promise<T> {
let lastError: Error | undefined
for (let attempt = 1; attempt <= maxAttempts; attempt++) {
try {
return await fn()
} catch (error) {
lastError = error as Error
if (attempt < maxAttempts) {
await wait(delayMs)
}
}
}
throw lastError || new Error('All retry attempts failed')
}
/**
* Creates a mock streaming response that emits chunks at intervals
*/
export function createTimedStream<T>(chunks: T[], intervalMs = 10) {
return {
async *[Symbol.asyncIterator]() {
for (const chunk of chunks) {
await wait(intervalMs)
yield chunk
}
}
}
}
/**
* Asserts that two objects are deeply equal, ignoring specified keys
*/
export function assertDeepEqualIgnoring<T extends Record<string, any>>(
actual: T,
expected: T,
ignoreKeys: string[] = []
): void {
const filterKeys = (obj: T): Partial<T> => {
const filtered = { ...obj }
for (const key of ignoreKeys) {
delete filtered[key]
}
return filtered
}
const filteredActual = filterKeys(actual)
const filteredExpected = filterKeys(expected)
expect(filteredActual).toEqual(filteredExpected)
}
/**
* Creates a provider mock that simulates rate limiting
*/
export function createRateLimitedProvider(limitPerSecond: number) {
const calls: number[] = []
return createMockLanguageModel({
doGenerate: vi.fn().mockImplementation(async () => {
const now = Date.now()
calls.push(now)
// Remove calls older than 1 second
const recentCalls = calls.filter((time) => now - time < 1000)
if (recentCalls.length > limitPerSecond) {
throw new Error('Rate limit exceeded')
}
return {
text: 'Rate limited response',
finishReason: 'stop' as const,
usage: { promptTokens: 10, completionTokens: 5, totalTokens: 15 },
rawCall: { rawPrompt: null, rawSettings: {} },
rawResponse: { headers: {} },
warnings: []
}
})
})
}
/**
* Validates streaming response structure
*/
export function validateStreamChunk(chunk: any): void {
expect(chunk).toBeDefined()
expect(chunk).toHaveProperty('type')
if (chunk.type === 'text-delta') {
expect(chunk).toHaveProperty('textDelta')
expect(typeof chunk.textDelta).toBe('string')
} else if (chunk.type === 'finish') {
expect(chunk).toHaveProperty('finishReason')
expect(chunk).toHaveProperty('usage')
} else if (chunk.type === 'tool-call') {
expect(chunk).toHaveProperty('toolCallId')
expect(chunk).toHaveProperty('toolName')
expect(chunk).toHaveProperty('args')
}
}
/**
* Creates a test logger that captures log messages
*/
export function createTestLogger() {
const logs: Array<{ level: string; message: string; meta?: any }> = []
return {
info: (message: string, meta?: any) => logs.push({ level: 'info', message, meta }),
warn: (message: string, meta?: any) => logs.push({ level: 'warn', message, meta }),
error: (message: string, meta?: any) => logs.push({ level: 'error', message, meta }),
debug: (message: string, meta?: any) => logs.push({ level: 'debug', message, meta }),
getLogs: () => logs,
clear: () => {
logs.length = 0
}
}
}

View File

@@ -0,0 +1,12 @@
/**
* Test Infrastructure Exports
* Central export point for all test utilities, fixtures, and helpers
*/
// Fixtures
export * from './fixtures/mock-providers'
export * from './fixtures/mock-responses'
// Helpers
export * from './helpers/provider-test-utils'
export * from './helpers/test-utils'

View File

@@ -0,0 +1,499 @@
/**
* RuntimeExecutor.generateText Comprehensive Tests
* Tests non-streaming text generation across all providers with various parameters
*/
import { generateText } from 'ai'
import { beforeEach, describe, expect, it, vi } from 'vitest'
import {
createMockLanguageModel,
mockCompleteResponses,
mockProviderConfigs,
testMessages,
testTools
} from '../../../__tests__'
import type { AiPlugin } from '../../plugins'
import { globalRegistryManagement } from '../../providers/RegistryManagement'
import { RuntimeExecutor } from '../executor'
// Mock AI SDK
vi.mock('ai', () => ({
generateText: vi.fn()
}))
vi.mock('../../providers/RegistryManagement', () => ({
globalRegistryManagement: {
languageModel: vi.fn()
},
DEFAULT_SEPARATOR: '|'
}))
describe('RuntimeExecutor.generateText', () => {
let executor: RuntimeExecutor<'openai'>
let mockLanguageModel: any
beforeEach(() => {
vi.clearAllMocks()
executor = RuntimeExecutor.create('openai', mockProviderConfigs.openai)
mockLanguageModel = createMockLanguageModel({
provider: 'openai',
modelId: 'gpt-4'
})
vi.mocked(globalRegistryManagement.languageModel).mockReturnValue(mockLanguageModel)
vi.mocked(generateText).mockResolvedValue(mockCompleteResponses.simple as any)
})
describe('Basic Functionality', () => {
it('should generate text with minimal parameters', async () => {
const result = await executor.generateText({
model: 'gpt-4',
messages: testMessages.simple
})
expect(generateText).toHaveBeenCalledWith({
model: mockLanguageModel,
messages: testMessages.simple
})
expect(result.text).toBe('This is a simple response.')
expect(result.finishReason).toBe('stop')
expect(result.usage).toBeDefined()
})
it('should generate with system messages', async () => {
await executor.generateText({
model: 'gpt-4',
messages: testMessages.withSystem
})
expect(generateText).toHaveBeenCalledWith({
model: mockLanguageModel,
messages: testMessages.withSystem
})
})
it('should generate with conversation history', async () => {
await executor.generateText({
model: 'gpt-4',
messages: testMessages.conversation
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
messages: testMessages.conversation
})
)
})
})
describe('All Parameter Combinations', () => {
it('should support all parameters together', async () => {
await executor.generateText({
model: 'gpt-4',
messages: testMessages.simple,
temperature: 0.7,
maxOutputTokens: 500,
topP: 0.9,
frequencyPenalty: 0.5,
presencePenalty: 0.3,
stopSequences: ['STOP'],
seed: 12345
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
temperature: 0.7,
maxOutputTokens: 500,
topP: 0.9,
frequencyPenalty: 0.5,
presencePenalty: 0.3,
stopSequences: ['STOP'],
seed: 12345
})
)
})
it('should support partial parameters', async () => {
await executor.generateText({
model: 'gpt-4',
messages: testMessages.simple,
temperature: 0.5,
maxOutputTokens: 100
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
temperature: 0.5,
maxOutputTokens: 100
})
)
})
})
describe('Tool Calling', () => {
beforeEach(() => {
vi.mocked(generateText).mockResolvedValue(mockCompleteResponses.withToolCalls as any)
})
it('should support tool calling', async () => {
const result = await executor.generateText({
model: 'gpt-4',
messages: testMessages.toolUse,
tools: testTools
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
tools: testTools
})
)
expect(result.toolCalls).toBeDefined()
expect(result.toolCalls).toHaveLength(1)
})
it('should support toolChoice auto', async () => {
await executor.generateText({
model: 'gpt-4',
messages: testMessages.toolUse,
tools: testTools,
toolChoice: 'auto'
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
toolChoice: 'auto'
})
)
})
it('should support toolChoice required', async () => {
await executor.generateText({
model: 'gpt-4',
messages: testMessages.toolUse,
tools: testTools,
toolChoice: 'required'
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
toolChoice: 'required'
})
)
})
it('should support toolChoice none', async () => {
vi.mocked(generateText).mockResolvedValue(mockCompleteResponses.simple as any)
await executor.generateText({
model: 'gpt-4',
messages: testMessages.simple,
tools: testTools,
toolChoice: 'none'
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
toolChoice: 'none'
})
)
})
it('should support specific tool selection', async () => {
await executor.generateText({
model: 'gpt-4',
messages: testMessages.toolUse,
tools: testTools,
toolChoice: {
type: 'tool',
toolName: 'getWeather'
}
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
toolChoice: {
type: 'tool',
toolName: 'getWeather'
}
})
)
})
})
describe('Multiple Providers', () => {
it('should work with Anthropic provider', async () => {
const anthropicExecutor = RuntimeExecutor.create('anthropic', mockProviderConfigs.anthropic)
const anthropicModel = createMockLanguageModel({
provider: 'anthropic',
modelId: 'claude-3-5-sonnet-20241022'
})
vi.mocked(globalRegistryManagement.languageModel).mockReturnValue(anthropicModel)
await anthropicExecutor.generateText({
model: 'claude-3-5-sonnet-20241022',
messages: testMessages.simple
})
expect(globalRegistryManagement.languageModel).toHaveBeenCalledWith('anthropic|claude-3-5-sonnet-20241022')
})
it('should work with Google provider', async () => {
const googleExecutor = RuntimeExecutor.create('google', mockProviderConfigs.google)
const googleModel = createMockLanguageModel({
provider: 'google',
modelId: 'gemini-2.0-flash-exp'
})
vi.mocked(globalRegistryManagement.languageModel).mockReturnValue(googleModel)
await googleExecutor.generateText({
model: 'gemini-2.0-flash-exp',
messages: testMessages.simple
})
expect(globalRegistryManagement.languageModel).toHaveBeenCalledWith('google|gemini-2.0-flash-exp')
})
it('should work with xAI provider', async () => {
const xaiExecutor = RuntimeExecutor.create('xai', mockProviderConfigs.xai)
const xaiModel = createMockLanguageModel({
provider: 'xai',
modelId: 'grok-2-latest'
})
vi.mocked(globalRegistryManagement.languageModel).mockReturnValue(xaiModel)
await xaiExecutor.generateText({
model: 'grok-2-latest',
messages: testMessages.simple
})
expect(globalRegistryManagement.languageModel).toHaveBeenCalledWith('xai|grok-2-latest')
})
it('should work with DeepSeek provider', async () => {
const deepseekExecutor = RuntimeExecutor.create('deepseek', mockProviderConfigs.deepseek)
const deepseekModel = createMockLanguageModel({
provider: 'deepseek',
modelId: 'deepseek-chat'
})
vi.mocked(globalRegistryManagement.languageModel).mockReturnValue(deepseekModel)
await deepseekExecutor.generateText({
model: 'deepseek-chat',
messages: testMessages.simple
})
expect(globalRegistryManagement.languageModel).toHaveBeenCalledWith('deepseek|deepseek-chat')
})
})
describe('Plugin Integration', () => {
it('should execute all plugin hooks', async () => {
const pluginCalls: string[] = []
const testPlugin: AiPlugin = {
name: 'test-plugin',
onRequestStart: vi.fn(async () => {
pluginCalls.push('onRequestStart')
}),
transformParams: vi.fn(async (params) => {
pluginCalls.push('transformParams')
return { ...params, temperature: 0.8 }
}),
transformResult: vi.fn(async (result) => {
pluginCalls.push('transformResult')
return { ...result, text: result.text + ' [modified]' }
}),
onRequestEnd: vi.fn(async () => {
pluginCalls.push('onRequestEnd')
})
}
const executorWithPlugin = RuntimeExecutor.create('openai', mockProviderConfigs.openai, [testPlugin])
const result = await executorWithPlugin.generateText({
model: 'gpt-4',
messages: testMessages.simple
})
expect(pluginCalls).toEqual(['onRequestStart', 'transformParams', 'transformResult', 'onRequestEnd'])
// Verify transformed parameters
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
temperature: 0.8
})
)
// Verify transformed result
expect(result.text).toContain('[modified]')
})
it('should handle multiple plugins in order', async () => {
const pluginOrder: string[] = []
const plugin1: AiPlugin = {
name: 'plugin-1',
transformParams: vi.fn(async (params) => {
pluginOrder.push('plugin-1')
return { ...params, temperature: 0.5 }
})
}
const plugin2: AiPlugin = {
name: 'plugin-2',
transformParams: vi.fn(async (params) => {
pluginOrder.push('plugin-2')
return { ...params, maxTokens: 200 }
})
}
const executorWithPlugins = RuntimeExecutor.create('openai', mockProviderConfigs.openai, [plugin1, plugin2])
await executorWithPlugins.generateText({
model: 'gpt-4',
messages: testMessages.simple
})
expect(pluginOrder).toEqual(['plugin-1', 'plugin-2'])
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
temperature: 0.5,
maxTokens: 200
})
)
})
})
describe('Error Handling', () => {
it('should handle API errors', async () => {
const error = new Error('API request failed')
vi.mocked(generateText).mockRejectedValue(error)
await expect(
executor.generateText({
model: 'gpt-4',
messages: testMessages.simple
})
).rejects.toThrow('API request failed')
})
it('should execute onError plugin hook', async () => {
const error = new Error('Generation failed')
vi.mocked(generateText).mockRejectedValue(error)
const errorPlugin: AiPlugin = {
name: 'error-handler',
onError: vi.fn()
}
const executorWithPlugin = RuntimeExecutor.create('openai', mockProviderConfigs.openai, [errorPlugin])
await expect(
executorWithPlugin.generateText({
model: 'gpt-4',
messages: testMessages.simple
})
).rejects.toThrow('Generation failed')
expect(errorPlugin.onError).toHaveBeenCalledWith(
error,
expect.objectContaining({
providerId: 'openai',
modelId: 'gpt-4'
})
)
})
it('should handle model not found error', async () => {
const error = new Error('Model not found: invalid-model')
vi.mocked(globalRegistryManagement.languageModel).mockImplementation(() => {
throw error
})
await expect(
executor.generateText({
model: 'invalid-model',
messages: testMessages.simple
})
).rejects.toThrow('Model not found')
})
})
describe('Usage and Metadata', () => {
it('should return usage information', async () => {
const result = await executor.generateText({
model: 'gpt-4',
messages: testMessages.simple
})
expect(result.usage).toBeDefined()
expect(result.usage.inputTokens).toBe(15)
expect(result.usage.outputTokens).toBe(8)
expect(result.usage.totalTokens).toBe(23)
})
it('should handle warnings', async () => {
vi.mocked(generateText).mockResolvedValue(mockCompleteResponses.withWarnings as any)
const result = await executor.generateText({
model: 'gpt-4',
messages: testMessages.simple,
temperature: 2.5 // Unsupported value
})
expect(result.warnings).toBeDefined()
expect(result.warnings).toHaveLength(1)
expect(result.warnings![0].type).toBe('unsupported-setting')
})
})
describe('Abort Signal', () => {
it('should support abort signal', async () => {
const abortController = new AbortController()
await executor.generateText({
model: 'gpt-4',
messages: testMessages.simple,
abortSignal: abortController.signal
})
expect(generateText).toHaveBeenCalledWith(
expect.objectContaining({
abortSignal: abortController.signal
})
)
})
it('should handle aborted request', async () => {
const abortError = new Error('Request aborted')
abortError.name = 'AbortError'
vi.mocked(generateText).mockRejectedValue(abortError)
const abortController = new AbortController()
abortController.abort()
await expect(
executor.generateText({
model: 'gpt-4',
messages: testMessages.simple,
abortSignal: abortController.signal
})
).rejects.toThrow('Request aborted')
})
})
})

View File

@@ -0,0 +1,525 @@
/**
* RuntimeExecutor.streamText Comprehensive Tests
* Tests streaming text generation across all providers with various parameters
*/
import { streamText } from 'ai'
import { beforeEach, describe, expect, it, vi } from 'vitest'
import { collectStreamChunks, createMockLanguageModel, mockProviderConfigs, testMessages } from '../../../__tests__'
import type { AiPlugin } from '../../plugins'
import { globalRegistryManagement } from '../../providers/RegistryManagement'
import { RuntimeExecutor } from '../executor'
// Mock AI SDK
vi.mock('ai', () => ({
streamText: vi.fn()
}))
vi.mock('../../providers/RegistryManagement', () => ({
globalRegistryManagement: {
languageModel: vi.fn()
},
DEFAULT_SEPARATOR: '|'
}))
describe('RuntimeExecutor.streamText', () => {
let executor: RuntimeExecutor<'openai'>
let mockLanguageModel: any
beforeEach(() => {
vi.clearAllMocks()
executor = RuntimeExecutor.create('openai', mockProviderConfigs.openai)
mockLanguageModel = createMockLanguageModel({
provider: 'openai',
modelId: 'gpt-4'
})
vi.mocked(globalRegistryManagement.languageModel).mockReturnValue(mockLanguageModel)
})
describe('Basic Functionality', () => {
it('should stream text with minimal parameters', async () => {
const mockStream = {
textStream: (async function* () {
yield 'Hello'
yield ' '
yield 'World'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Hello' }
yield { type: 'text-delta', textDelta: ' ' }
yield { type: 'text-delta', textDelta: 'World' }
})(),
usage: Promise.resolve({ promptTokens: 5, completionTokens: 3, totalTokens: 8 })
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
const result = await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple
})
expect(streamText).toHaveBeenCalledWith({
model: mockLanguageModel,
messages: testMessages.simple
})
const chunks = await collectStreamChunks(result.textStream)
expect(chunks).toEqual(['Hello', ' ', 'World'])
})
it('should stream with system messages', async () => {
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.withSystem
})
expect(streamText).toHaveBeenCalledWith({
model: mockLanguageModel,
messages: testMessages.withSystem
})
})
it('should stream multi-turn conversations', async () => {
const mockStream = {
textStream: (async function* () {
yield 'Multi-turn response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Multi-turn response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.multiTurn
})
expect(streamText).toHaveBeenCalled()
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
messages: testMessages.multiTurn
})
)
})
})
describe('Temperature Parameter', () => {
const temperatures = [0, 0.3, 0.5, 0.7, 0.9, 1.0, 1.5, 2.0]
it.each(temperatures)('should support temperature=%s', async (temperature) => {
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
temperature
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
temperature
})
)
})
})
describe('Max Tokens Parameter', () => {
const maxTokensValues = [10, 50, 100, 500, 1000, 2000, 4000]
it.each(maxTokensValues)('should support maxTokens=%s', async (maxTokens) => {
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
maxOutputTokens: maxTokens
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
maxTokens
})
)
})
})
describe('Top P Parameter', () => {
const topPValues = [0.1, 0.3, 0.5, 0.7, 0.9, 0.95, 1.0]
it.each(topPValues)('should support topP=%s', async (topP) => {
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
topP
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
topP
})
)
})
})
describe('Frequency and Presence Penalty', () => {
it('should support frequency penalty', async () => {
const penalties = [-2.0, -1.0, 0, 0.5, 1.0, 1.5, 2.0]
for (const frequencyPenalty of penalties) {
vi.clearAllMocks()
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
frequencyPenalty
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
frequencyPenalty
})
)
}
})
it('should support presence penalty', async () => {
const penalties = [-2.0, -1.0, 0, 0.5, 1.0, 1.5, 2.0]
for (const presencePenalty of penalties) {
vi.clearAllMocks()
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
presencePenalty
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
presencePenalty
})
)
}
})
it('should support both penalties together', async () => {
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
frequencyPenalty: 0.5,
presencePenalty: 0.5
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
frequencyPenalty: 0.5,
presencePenalty: 0.5
})
)
})
})
describe('Seed Parameter', () => {
it('should support seed for deterministic output', async () => {
const seeds = [0, 12345, 67890, 999999]
for (const seed of seeds) {
vi.clearAllMocks()
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
seed
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
seed
})
)
}
})
})
describe('Abort Signal', () => {
it('should support abort signal', async () => {
const abortController = new AbortController()
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
abortSignal: abortController.signal
})
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
abortSignal: abortController.signal
})
)
})
it('should handle abort during streaming', async () => {
const abortController = new AbortController()
const mockStream = {
textStream: (async function* () {
yield 'Start'
// Simulate abort
abortController.abort()
throw new Error('Aborted')
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Start' }
throw new Error('Aborted')
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
const result = await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple,
abortSignal: abortController.signal
})
await expect(async () => {
// oxlint-disable-next-line no-unused-vars
for await (const _chunk of result.textStream) {
// Stream should be interrupted
}
}).rejects.toThrow('Aborted')
})
})
describe('Plugin Integration', () => {
it('should execute plugins during streaming', async () => {
const pluginCalls: string[] = []
const testPlugin: AiPlugin = {
name: 'test-plugin',
onRequestStart: vi.fn(async () => {
pluginCalls.push('onRequestStart')
}),
transformParams: vi.fn(async (params) => {
pluginCalls.push('transformParams')
return { ...params, temperature: 0.5 }
}),
onRequestEnd: vi.fn(async () => {
pluginCalls.push('onRequestEnd')
})
}
const executorWithPlugin = RuntimeExecutor.create('openai', mockProviderConfigs.openai, [testPlugin])
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
const result = await executorWithPlugin.streamText({
model: 'gpt-4',
messages: testMessages.simple
})
// Consume stream
// oxlint-disable-next-line no-unused-vars
for await (const _chunk of result.textStream) {
// Stream chunks
}
expect(pluginCalls).toContain('onRequestStart')
expect(pluginCalls).toContain('transformParams')
// Verify transformed parameters were used
expect(streamText).toHaveBeenCalledWith(
expect.objectContaining({
temperature: 0.5
})
)
})
})
describe('Full Stream with Finish Reason', () => {
it('should provide finish reason in full stream', async () => {
const mockStream = {
textStream: (async function* () {
yield 'Response'
})(),
fullStream: (async function* () {
yield { type: 'text-delta', textDelta: 'Response' }
yield {
type: 'finish',
finishReason: 'stop',
usage: { promptTokens: 5, completionTokens: 3, totalTokens: 8 }
}
})()
}
vi.mocked(streamText).mockResolvedValue(mockStream as any)
const result = await executor.streamText({
model: 'gpt-4',
messages: testMessages.simple
})
const fullChunks = await collectStreamChunks(result.fullStream)
expect(fullChunks).toHaveLength(2)
expect(fullChunks[0]).toEqual({ type: 'text-delta', textDelta: 'Response' })
expect(fullChunks[1]).toEqual({
type: 'finish',
finishReason: 'stop',
usage: { promptTokens: 5, completionTokens: 3, totalTokens: 8 }
})
})
})
describe('Error Handling', () => {
it('should handle streaming errors', async () => {
const error = new Error('Streaming failed')
vi.mocked(streamText).mockRejectedValue(error)
await expect(
executor.streamText({
model: 'gpt-4',
messages: testMessages.simple
})
).rejects.toThrow('Streaming failed')
})
it('should execute onError plugin hook on failure', async () => {
const error = new Error('Stream error')
vi.mocked(streamText).mockRejectedValue(error)
const errorPlugin: AiPlugin = {
name: 'error-handler',
onError: vi.fn()
}
const executorWithPlugin = RuntimeExecutor.create('openai', mockProviderConfigs.openai, [errorPlugin])
await expect(
executorWithPlugin.streamText({
model: 'gpt-4',
messages: testMessages.simple
})
).rejects.toThrow('Stream error')
expect(errorPlugin.onError).toHaveBeenCalledWith(
error,
expect.objectContaining({
providerId: 'openai',
modelId: 'gpt-4'
})
)
})
})
})

View File

@@ -235,6 +235,7 @@ export enum IpcChannel {
System_GetDeviceType = 'system:getDeviceType',
System_GetHostname = 'system:getHostname',
System_GetCpuName = 'system:getCpuName',
System_CheckGitBash = 'system:checkGitBash',
// DevTools
System_ToggleDevTools = 'system:toggleDevTools',
@@ -373,5 +374,13 @@ export enum IpcChannel {
WebSocket_Stop = 'webSocket:stop',
WebSocket_Status = 'webSocket:status',
WebSocket_SendFile = 'webSocket:send-file',
WebSocket_GetAllCandidates = 'webSocket:get-all-candidates'
WebSocket_GetAllCandidates = 'webSocket:get-all-candidates',
// Volcengine
Volcengine_SaveCredentials = 'volcengine:save-credentials',
Volcengine_HasCredentials = 'volcengine:has-credentials',
Volcengine_ClearCredentials = 'volcengine:clear-credentials',
Volcengine_ListModels = 'volcengine:list-models',
Volcengine_GetAuthHeaders = 'volcengine:get-auth-headers',
Volcengine_MakeRequest = 'volcengine:make-request'
}

View File

@@ -88,11 +88,16 @@ export function getSdkClient(
}
})
}
const baseURL =
let baseURL =
provider.type === 'anthropic'
? provider.apiHost
: (provider.anthropicApiHost && provider.anthropicApiHost.trim()) || provider.apiHost
// Anthropic SDK automatically appends /v1 to all endpoints (like /v1/messages, /v1/models)
// We need to strip api version from baseURL to avoid duplication (e.g., /v3/v1/models)
// formatProviderApiHost adds /v1 for AI SDK compatibility, but Anthropic SDK needs it removed
baseURL = baseURL.replace(/\/v\d+(?:alpha|beta)?(?=\/|$)/i, '')
logger.debug('Anthropic API baseURL', { baseURL, providerId: provider.id })
if (provider.id === 'aihubmix') {

View File

@@ -0,0 +1,48 @@
/**
* @fileoverview Shared provider configuration for Claude Code and Anthropic API compatibility
*
* This module defines which models from specific providers support the Anthropic API endpoint.
* Used by both the Code Tools page and the Anthropic SDK client.
*/
/**
* Silicon provider models that support Anthropic API endpoint.
* These models can be used with Claude Code via the Anthropic-compatible API.
*
* @see https://docs.siliconflow.cn/cn/api-reference/chat-completions/messages
*/
export const SILICON_ANTHROPIC_COMPATIBLE_MODELS: readonly string[] = [
// DeepSeek V3.1 series
'Pro/deepseek-ai/DeepSeek-V3.1-Terminus',
'deepseek-ai/DeepSeek-V3.1',
'Pro/deepseek-ai/DeepSeek-V3.1',
// DeepSeek V3 series
'deepseek-ai/DeepSeek-V3',
'Pro/deepseek-ai/DeepSeek-V3',
// Moonshot/Kimi series
'moonshotai/Kimi-K2-Instruct-0905',
'Pro/moonshotai/Kimi-K2-Instruct-0905',
'moonshotai/Kimi-Dev-72B',
// Baidu ERNIE
'baidu/ERNIE-4.5-300B-A47B'
]
/**
* Creates a Set for efficient lookup of silicon Anthropic-compatible model IDs.
*/
const SILICON_ANTHROPIC_COMPATIBLE_MODEL_SET = new Set(SILICON_ANTHROPIC_COMPATIBLE_MODELS)
/**
* Checks if a model ID is compatible with Anthropic API on Silicon provider.
*
* @param modelId - The model ID to check
* @returns true if the model supports Anthropic API endpoint
*/
export function isSiliconAnthropicCompatibleModel(modelId: string): boolean {
return SILICON_ANTHROPIC_COMPATIBLE_MODEL_SET.has(modelId)
}
/**
* Silicon provider's Anthropic API host URL.
*/
export const SILICON_ANTHROPIC_API_HOST = 'https://api.siliconflow.cn'

View File

@@ -4,3 +4,34 @@ export const defaultAppHeaders = () => {
'X-Title': 'Cherry Studio'
}
}
// Following two function are not being used for now.
// I may use them in the future, so just keep them commented. - by eurfelux
/**
* Converts an `undefined` value to `null`, otherwise returns the value as-is.
* @param value - The value to check
* @returns `null` if the input is `undefined`; otherwise the input value
*/
// export function toNullIfUndefined<T>(value: T | undefined): T | null {
// if (value === undefined) {
// return null
// } else {
// return value
// }
// }
/**
* Converts a `null` value to `undefined`, otherwise returns the value as-is.
* @param value - The value to check
* @returns `undefined` if the input is `null`; otherwise the input value
*/
// export function toUndefinedIfNull<T>(value: T | null): T | undefined {
// if (value === null) {
// return undefined
// } else {
// return value
// }
// }

View File

@@ -104,12 +104,6 @@ const router = express
logger.warn('No models available from providers', { filter })
}
logger.info('Models response ready', {
filter,
total: response.total,
modelIds: response.data.map((m) => m.id)
})
return res.json(response satisfies ApiModelsResponse)
} catch (error: any) {
logger.error('Error fetching models', { error })

View File

@@ -3,7 +3,6 @@ import { createServer } from 'node:http'
import { loggerService } from '@logger'
import { IpcChannel } from '@shared/IpcChannel'
import { agentService } from '../services/agents'
import { windowService } from '../services/WindowService'
import { app } from './app'
import { config } from './config'
@@ -32,11 +31,6 @@ export class ApiServer {
// Load config
const { port, host } = await config.load()
// Initialize AgentService
logger.info('Initializing AgentService')
await agentService.initialize()
logger.info('AgentService initialized')
// Create server with Express app
this.server = createServer(app)
this.applyServerTimeouts(this.server)

View File

@@ -32,7 +32,7 @@ export class ModelsService {
for (const model of models) {
const provider = providers.find((p) => p.id === model.provider)
logger.debug(`Processing model ${model.id}`)
// logger.debug(`Processing model ${model.id}`)
if (!provider) {
logger.debug(`Skipping model ${model.id} . Reason: Provider not found.`)
continue

View File

@@ -1,6 +1,7 @@
import { CacheService } from '@main/services/CacheService'
import { loggerService } from '@main/services/LoggerService'
import { reduxService } from '@main/services/ReduxService'
import { isSiliconAnthropicCompatibleModel } from '@shared/config/providers'
import type { ApiModel, Model, Provider } from '@types'
const logger = loggerService.withContext('ApiServerUtils')
@@ -287,6 +288,8 @@ export const getProviderAnthropicModelChecker = (providerId: string): ((m: Model
return (m: Model) => m.endpoint_type === 'anthropic'
case 'aihubmix':
return (m: Model) => m.id.includes('claude')
case 'silicon':
return (m: Model) => isSiliconAnthropicCompatibleModel(m.id)
default:
// allow all models when checker not configured
return () => true

View File

@@ -34,6 +34,7 @@ import { TrayService } from './services/TrayService'
import { versionService } from './services/VersionService'
import { windowService } from './services/WindowService'
import { initWebviewHotkeys } from './services/WebviewService'
import { runAsyncFunction } from './utils'
const logger = loggerService.withContext('MainEntry')
@@ -170,39 +171,33 @@ if (!app.requestSingleInstanceLock()) {
//start selection assistant service
initSelectionService()
// Initialize Agent Service
try {
await agentService.initialize()
logger.info('Agent service initialized successfully')
} catch (error: any) {
logger.error('Failed to initialize Agent service:', error)
}
runAsyncFunction(async () => {
// Start API server if enabled or if agents exist
try {
const config = await apiServerService.getCurrentConfig()
logger.info('API server config:', config)
// Start API server if enabled or if agents exist
try {
const config = await apiServerService.getCurrentConfig()
logger.info('API server config:', config)
// Check if there are any agents
let shouldStart = config.enabled
if (!shouldStart) {
try {
const { total } = await agentService.listAgents({ limit: 1 })
if (total > 0) {
shouldStart = true
logger.info(`Detected ${total} agent(s), auto-starting API server`)
// Check if there are any agents
let shouldStart = config.enabled
if (!shouldStart) {
try {
const { total } = await agentService.listAgents({ limit: 1 })
if (total > 0) {
shouldStart = true
logger.info(`Detected ${total} agent(s), auto-starting API server`)
}
} catch (error: any) {
logger.warn('Failed to check agent count:', error)
}
} catch (error: any) {
logger.warn('Failed to check agent count:', error)
}
}
if (shouldStart) {
await apiServerService.start()
if (shouldStart) {
await apiServerService.start()
}
} catch (error: any) {
logger.error('Failed to check/start API server:', error)
}
} catch (error: any) {
logger.error('Failed to check/start API server:', error)
}
})
})
registerProtocolClient(app)

View File

@@ -73,6 +73,7 @@ import {
import storeSyncService from './services/StoreSyncService'
import { themeService } from './services/ThemeService'
import VertexAIService from './services/VertexAIService'
import VolcengineService from './services/VolcengineService'
import WebSocketService from './services/WebSocketService'
import { setOpenLinkExternal } from './services/WebviewService'
import { windowService } from './services/WindowService'
@@ -493,6 +494,44 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) {
ipcMain.handle(IpcChannel.System_GetDeviceType, () => (isMac ? 'mac' : isWin ? 'windows' : 'linux'))
ipcMain.handle(IpcChannel.System_GetHostname, () => require('os').hostname())
ipcMain.handle(IpcChannel.System_GetCpuName, () => require('os').cpus()[0].model)
ipcMain.handle(IpcChannel.System_CheckGitBash, () => {
if (!isWin) {
return true // Non-Windows systems don't need Git Bash
}
try {
// Check common Git Bash installation paths
const commonPaths = [
path.join(process.env.ProgramFiles || 'C:\\Program Files', 'Git', 'bin', 'bash.exe'),
path.join(process.env['ProgramFiles(x86)'] || 'C:\\Program Files (x86)', 'Git', 'bin', 'bash.exe'),
path.join(process.env.LOCALAPPDATA || '', 'Programs', 'Git', 'bin', 'bash.exe')
]
// Check if any of the common paths exist
for (const bashPath of commonPaths) {
if (fs.existsSync(bashPath)) {
logger.debug('Git Bash found', { path: bashPath })
return true
}
}
// Check if git is in PATH
const { execSync } = require('child_process')
try {
execSync('git --version', { stdio: 'ignore' })
logger.debug('Git found in PATH')
return true
} catch {
// Git not in PATH
}
logger.debug('Git Bash not found on Windows system')
return false
} catch (error) {
logger.error('Error checking Git Bash', error as Error)
return false
}
})
ipcMain.handle(IpcChannel.System_ToggleDevTools, (e) => {
const win = BrowserWindow.fromWebContents(e.sender)
win && win.webContents.toggleDevTools()
@@ -1039,6 +1078,14 @@ export function registerIpc(mainWindow: BrowserWindow, app: Electron.App) {
ipcMain.handle(IpcChannel.WebSocket_SendFile, WebSocketService.sendFile)
ipcMain.handle(IpcChannel.WebSocket_GetAllCandidates, WebSocketService.getAllCandidates)
// Volcengine
ipcMain.handle(IpcChannel.Volcengine_SaveCredentials, VolcengineService.saveCredentials)
ipcMain.handle(IpcChannel.Volcengine_HasCredentials, VolcengineService.hasCredentials)
ipcMain.handle(IpcChannel.Volcengine_ClearCredentials, VolcengineService.clearCredentials)
ipcMain.handle(IpcChannel.Volcengine_ListModels, VolcengineService.listModels)
ipcMain.handle(IpcChannel.Volcengine_GetAuthHeaders, VolcengineService.getAuthHeaders)
ipcMain.handle(IpcChannel.Volcengine_MakeRequest, VolcengineService.makeRequest)
ipcMain.handle(IpcChannel.APP_CrashRenderProcess, () => {
mainWindow.webContents.forcefullyCrashRenderer()
})

View File

@@ -0,0 +1,732 @@
import { loggerService } from '@logger'
import crypto from 'crypto'
import { app, net, safeStorage } from 'electron'
import fs from 'fs'
import path from 'path'
import * as z from 'zod'
import { getConfigDir } from '../utils/file'
const logger = loggerService.withContext('VolcengineService')
// Configuration constants
const CONFIG = {
ALGORITHM: 'HMAC-SHA256',
REQUEST_TYPE: 'request',
DEFAULT_REGION: 'cn-beijing',
SERVICE_NAME: 'ark',
DEFAULT_HEADERS: {
'content-type': 'application/json',
accept: 'application/json'
},
API_URLS: {
ARK_HOST: 'open.volcengineapi.com'
},
CREDENTIALS_FILE_NAME: '.volcengine_credentials',
API_VERSION: '2024-01-01',
DEFAULT_PAGE_SIZE: 100
} as const
// Request schemas
const ListFoundationModelsRequestSchema = z.object({
PageNumber: z.optional(z.number()),
PageSize: z.optional(z.number())
})
const ListEndpointsRequestSchema = z.object({
ProjectName: z.optional(z.string()),
PageNumber: z.optional(z.number()),
PageSize: z.optional(z.number())
})
// Response schemas - only keep fields needed for model list
const FoundationModelItemSchema = z.object({
Name: z.string(),
DisplayName: z.optional(z.string()),
Description: z.optional(z.string())
})
const EndpointItemSchema = z.object({
Id: z.string(),
Name: z.optional(z.string()),
Description: z.optional(z.string()),
ModelReference: z.optional(
z.object({
FoundationModel: z.optional(
z.object({
Name: z.optional(z.string()),
ModelVersion: z.optional(z.string())
})
),
CustomModelId: z.optional(z.string())
})
)
})
const ListFoundationModelsResponseSchema = z.object({
Result: z.object({
TotalCount: z.number(),
Items: z.array(FoundationModelItemSchema)
})
})
const ListEndpointsResponseSchema = z.object({
Result: z.object({
TotalCount: z.number(),
Items: z.array(EndpointItemSchema)
})
})
// Infer types from schemas
type ListFoundationModelsRequest = z.infer<typeof ListFoundationModelsRequestSchema>
type ListEndpointsRequest = z.infer<typeof ListEndpointsRequestSchema>
type ListFoundationModelsResponse = z.infer<typeof ListFoundationModelsResponseSchema>
type ListEndpointsResponse = z.infer<typeof ListEndpointsResponseSchema>
// ============= Internal Type Definitions =============
interface VolcengineCredentials {
accessKeyId: string
secretAccessKey: string
}
interface SignedRequestParams {
method: 'GET' | 'POST'
host: string
path: string
query: Record<string, string>
headers: Record<string, string>
body?: string
service: string
region: string
}
interface SignedHeaders {
Authorization: string
'X-Date': string
'X-Content-Sha256': string
Host: string
}
interface ModelInfo {
id: string
name: string
description?: string
created?: number
}
interface ListModelsResult {
models: ModelInfo[]
total?: number
warnings?: string[]
}
// Custom error class
class VolcengineServiceError extends Error {
constructor(
message: string,
public readonly cause?: unknown
) {
super(message)
this.name = 'VolcengineServiceError'
}
}
/**
* Volcengine API Signing Service
*
* Implements HMAC-SHA256 signing algorithm for Volcengine API authentication.
* Securely stores credentials using Electron's safeStorage.
*/
class VolcengineService {
private readonly credentialsFilePath: string
constructor() {
this.credentialsFilePath = this.getCredentialsFilePath()
}
/**
* Get the path for storing encrypted credentials
*/
private getCredentialsFilePath(): string {
const oldPath = path.join(app.getPath('userData'), CONFIG.CREDENTIALS_FILE_NAME)
if (fs.existsSync(oldPath)) {
return oldPath
}
return path.join(getConfigDir(), CONFIG.CREDENTIALS_FILE_NAME)
}
// ============= Cryptographic Helper Methods =============
/**
* Calculate SHA256 hash of data and return hex encoded string
*/
private sha256Hash(data: string | Buffer): string {
return crypto.createHash('sha256').update(data).digest('hex')
}
/**
* Calculate HMAC-SHA256 and return buffer
*/
private hmacSha256(key: Buffer | string, data: string): Buffer {
return crypto.createHmac('sha256', key).update(data, 'utf8').digest()
}
/**
* Calculate HMAC-SHA256 and return hex encoded string
*/
private hmacSha256Hex(key: Buffer | string, data: string): string {
return crypto.createHmac('sha256', key).update(data, 'utf8').digest('hex')
}
/**
* URL encode according to RFC3986
*/
private uriEncode(str: string, encodeSlash: boolean = true): string {
if (!str) return ''
return str
.split('')
.map((char) => {
if (
(char >= 'A' && char <= 'Z') ||
(char >= 'a' && char <= 'z') ||
(char >= '0' && char <= '9') ||
char === '_' ||
char === '-' ||
char === '~' ||
char === '.'
) {
return char
}
if (char === '/' && !encodeSlash) {
return char
}
return encodeURIComponent(char)
})
.join('')
}
// ============= Signing Implementation =============
/**
* Get current UTC time in ISO8601 format (YYYYMMDD'T'HHMMSS'Z')
*/
private getIso8601DateTime(): string {
const now = new Date()
return now
.toISOString()
.replace(/[-:]/g, '')
.replace(/\.\d{3}/, '')
}
/**
* Get date portion from datetime (YYYYMMDD)
*/
private getDateFromDateTime(dateTime: string): string {
return dateTime.substring(0, 8)
}
/**
* Build canonical query string from query parameters
*/
private buildCanonicalQueryString(query: Record<string, string>): string {
if (!query || Object.keys(query).length === 0) {
return ''
}
return Object.keys(query)
.sort()
.map((key) => `${this.uriEncode(key)}=${this.uriEncode(query[key])}`)
.join('&')
}
/**
* Build canonical headers string
*/
private buildCanonicalHeaders(headers: Record<string, string>): {
canonicalHeaders: string
signedHeaders: string
} {
const sortedKeys = Object.keys(headers)
.map((k) => k.toLowerCase())
.sort()
const canonicalHeaders = sortedKeys.map((key) => `${key}:${headers[key]?.trim() || ''}`).join('\n') + '\n'
const signedHeaders = sortedKeys.join(';')
return { canonicalHeaders, signedHeaders }
}
/**
* Create the signing key through a series of HMAC operations
*
* kSecret = SecretAccessKey
* kDate = HMAC(kSecret, Date)
* kRegion = HMAC(kDate, Region)
* kService = HMAC(kRegion, Service)
* kSigning = HMAC(kService, "request")
*/
private deriveSigningKey(secretKey: string, date: string, region: string, service: string): Buffer {
const kDate = this.hmacSha256(secretKey, date)
const kRegion = this.hmacSha256(kDate, region)
const kService = this.hmacSha256(kRegion, service)
const kSigning = this.hmacSha256(kService, CONFIG.REQUEST_TYPE)
return kSigning
}
/**
* Create canonical request string
*
* CanonicalRequest =
* HTTPRequestMethod + '\n' +
* CanonicalURI + '\n' +
* CanonicalQueryString + '\n' +
* CanonicalHeaders + '\n' +
* SignedHeaders + '\n' +
* HexEncode(Hash(RequestPayload))
*/
private createCanonicalRequest(
method: string,
canonicalUri: string,
canonicalQueryString: string,
canonicalHeaders: string,
signedHeaders: string,
payloadHash: string
): string {
return [method, canonicalUri, canonicalQueryString, canonicalHeaders, signedHeaders, payloadHash].join('\n')
}
/**
* Create string to sign
*
* StringToSign =
* Algorithm + '\n' +
* RequestDateTime + '\n' +
* CredentialScope + '\n' +
* HexEncode(Hash(CanonicalRequest))
*/
private createStringToSign(dateTime: string, credentialScope: string, canonicalRequest: string): string {
const hashedCanonicalRequest = this.sha256Hash(canonicalRequest)
return [CONFIG.ALGORITHM, dateTime, credentialScope, hashedCanonicalRequest].join('\n')
}
/**
* Generate signature for the request
*/
private generateSignature(params: SignedRequestParams, credentials: VolcengineCredentials): SignedHeaders {
const { method, host, path: requestPath, query, body, service, region } = params
// Step 1: Prepare datetime
const dateTime = this.getIso8601DateTime()
const date = this.getDateFromDateTime(dateTime)
// Step 2: Calculate payload hash
const payloadHash = this.sha256Hash(body || '')
// Step 3: Prepare headers for signing
const headersToSign: Record<string, string> = {
host: host,
'x-date': dateTime,
'x-content-sha256': payloadHash,
'content-type': 'application/json'
}
// Step 4: Build canonical components
const canonicalUri = this.uriEncode(requestPath, false) || '/'
const canonicalQueryString = this.buildCanonicalQueryString(query)
const { canonicalHeaders, signedHeaders } = this.buildCanonicalHeaders(headersToSign)
// Step 5: Create canonical request
const canonicalRequest = this.createCanonicalRequest(
method.toUpperCase(),
canonicalUri,
canonicalQueryString,
canonicalHeaders,
signedHeaders,
payloadHash
)
// Step 6: Create credential scope and string to sign
const credentialScope = `${date}/${region}/${service}/${CONFIG.REQUEST_TYPE}`
const stringToSign = this.createStringToSign(dateTime, credentialScope, canonicalRequest)
// Step 7: Calculate signature
const signingKey = this.deriveSigningKey(credentials.secretAccessKey, date, region, service)
const signature = this.hmacSha256Hex(signingKey, stringToSign)
// Step 8: Build authorization header
const authorization = `${CONFIG.ALGORITHM} Credential=${credentials.accessKeyId}/${credentialScope}, SignedHeaders=${signedHeaders}, Signature=${signature}`
return {
Authorization: authorization,
'X-Date': dateTime,
'X-Content-Sha256': payloadHash,
Host: host
}
}
// ============= Credential Management =============
/**
* Save credentials securely using Electron's safeStorage
*/
public saveCredentials = async (
_: Electron.IpcMainInvokeEvent,
accessKeyId: string,
secretAccessKey: string
): Promise<void> => {
try {
if (!accessKeyId || !secretAccessKey) {
throw new VolcengineServiceError('Access Key ID and Secret Access Key are required')
}
const credentials: VolcengineCredentials = { accessKeyId, secretAccessKey }
const credentialsJson = JSON.stringify(credentials)
const encryptedData = safeStorage.encryptString(credentialsJson)
// Ensure directory exists
const dir = path.dirname(this.credentialsFilePath)
if (!fs.existsSync(dir)) {
await fs.promises.mkdir(dir, { recursive: true })
}
await fs.promises.writeFile(this.credentialsFilePath, encryptedData)
logger.info('Volcengine credentials saved successfully')
} catch (error) {
logger.error('Failed to save Volcengine credentials:', error as Error)
throw new VolcengineServiceError('Failed to save credentials', error)
}
}
/**
* Load credentials from encrypted storage
* @throws VolcengineServiceError if credentials file exists but is corrupted
*/
private async loadCredentials(): Promise<VolcengineCredentials | null> {
if (!fs.existsSync(this.credentialsFilePath)) {
return null
}
try {
const encryptedData = await fs.promises.readFile(this.credentialsFilePath)
const decryptedJson = safeStorage.decryptString(Buffer.from(encryptedData))
return JSON.parse(decryptedJson) as VolcengineCredentials
} catch (error) {
logger.error('Failed to load Volcengine credentials:', error as Error)
throw new VolcengineServiceError(
'Credentials file exists but could not be loaded. Please re-enter your credentials.',
error
)
}
}
/**
* Check if credentials exist
*/
public hasCredentials = async (): Promise<boolean> => {
return fs.existsSync(this.credentialsFilePath)
}
/**
* Clear stored credentials
*/
public clearCredentials = async (): Promise<void> => {
try {
if (fs.existsSync(this.credentialsFilePath)) {
await fs.promises.unlink(this.credentialsFilePath)
logger.info('Volcengine credentials cleared')
}
} catch (error) {
logger.error('Failed to clear Volcengine credentials:', error as Error)
throw new VolcengineServiceError('Failed to clear credentials', error)
}
}
// ============= API Methods =============
/**
* Make a signed request to Volcengine API
*/
private async makeSignedRequest<T>(
method: 'GET' | 'POST',
host: string,
path: string,
action: string,
version: string,
query?: Record<string, string>,
body?: Record<string, unknown>,
service: string = CONFIG.SERVICE_NAME,
region: string = CONFIG.DEFAULT_REGION
): Promise<T> {
const credentials = await this.loadCredentials()
if (!credentials) {
throw new VolcengineServiceError('No credentials found. Please save credentials first.')
}
const fullQuery: Record<string, string> = {
Action: action,
Version: version,
...query
}
const bodyString = body ? JSON.stringify(body) : ''
const signedHeaders = this.generateSignature(
{
method,
host,
path,
query: fullQuery,
headers: {},
body: bodyString,
service,
region
},
credentials
)
// Build URL with query string (use simple encoding for URL, canonical encoding is only for signature)
const urlParams = new URLSearchParams(fullQuery)
const url = `https://${host}${path}?${urlParams.toString()}`
const requestHeaders: Record<string, string> = {
...CONFIG.DEFAULT_HEADERS,
Authorization: signedHeaders.Authorization,
'X-Date': signedHeaders['X-Date'],
'X-Content-Sha256': signedHeaders['X-Content-Sha256']
}
logger.debug('Making Volcengine API request', { url, method, action })
try {
const response = await net.fetch(url, {
method,
headers: requestHeaders,
body: method === 'POST' && bodyString ? bodyString : undefined
})
if (!response.ok) {
const errorText = await response.text()
logger.error(`Volcengine API error: ${response.status}`, { errorText })
throw new VolcengineServiceError(`API request failed: ${response.status} - ${errorText}`)
}
return (await response.json()) as T
} catch (error) {
if (error instanceof VolcengineServiceError) {
throw error
}
logger.error('Volcengine API request failed:', error as Error)
throw new VolcengineServiceError('API request failed', error)
}
}
/**
* List foundation models from Volcengine ARK
*/
private async listFoundationModels(region: string = CONFIG.DEFAULT_REGION): Promise<ListFoundationModelsResponse> {
const requestBody: ListFoundationModelsRequest = {
PageNumber: 1,
PageSize: CONFIG.DEFAULT_PAGE_SIZE
}
const response = await this.makeSignedRequest<unknown>(
'POST',
CONFIG.API_URLS.ARK_HOST,
'/',
'ListFoundationModels',
CONFIG.API_VERSION,
{},
requestBody,
CONFIG.SERVICE_NAME,
region
)
return ListFoundationModelsResponseSchema.parse(response)
}
/**
* List user-created endpoints from Volcengine ARK
*/
private async listEndpoints(
projectName?: string,
region: string = CONFIG.DEFAULT_REGION
): Promise<ListEndpointsResponse> {
const requestBody: ListEndpointsRequest = {
ProjectName: projectName || 'default',
PageNumber: 1,
PageSize: CONFIG.DEFAULT_PAGE_SIZE
}
const response = await this.makeSignedRequest<unknown>(
'POST',
CONFIG.API_URLS.ARK_HOST,
'/',
'ListEndpoints',
CONFIG.API_VERSION,
{},
requestBody,
CONFIG.SERVICE_NAME,
region
)
return ListEndpointsResponseSchema.parse(response)
}
/**
* List all available models from Volcengine ARK
* Combines foundation models and user-created endpoints
*/
public listModels = async (
_?: Electron.IpcMainInvokeEvent,
projectName?: string,
region?: string
): Promise<ListModelsResult> => {
try {
const effectiveRegion = region || CONFIG.DEFAULT_REGION
const [foundationModelsResult, endpointsResult] = await Promise.allSettled([
this.listFoundationModels(effectiveRegion),
this.listEndpoints(projectName, effectiveRegion)
])
const models: ModelInfo[] = []
const warnings: string[] = []
if (foundationModelsResult.status === 'fulfilled') {
const foundationModels = foundationModelsResult.value
for (const item of foundationModels.Result.Items) {
models.push({
id: item.Name,
name: item.DisplayName || item.Name,
description: item.Description
})
}
logger.info(`Found ${foundationModels.Result.Items.length} foundation models`)
} else {
const errorMsg = `Failed to fetch foundation models: ${foundationModelsResult.reason}`
logger.warn(errorMsg)
warnings.push(errorMsg)
}
// Process endpoints
if (endpointsResult.status === 'fulfilled') {
const endpoints = endpointsResult.value
for (const item of endpoints.Result.Items) {
const modelRef = item.ModelReference
const foundationModelName = modelRef?.FoundationModel?.Name
const modelVersion = modelRef?.FoundationModel?.ModelVersion
const customModelId = modelRef?.CustomModelId
let displayName = item.Name || item.Id
if (foundationModelName) {
displayName = modelVersion ? `${foundationModelName} (${modelVersion})` : foundationModelName
} else if (customModelId) {
displayName = customModelId
}
models.push({
id: item.Id,
name: displayName,
description: item.Description
})
}
logger.info(`Found ${endpoints.Result.Items.length} endpoints`)
} else {
const errorMsg = `Failed to fetch endpoints: ${endpointsResult.reason}`
logger.warn(errorMsg)
warnings.push(errorMsg)
}
// If both failed, throw error
if (foundationModelsResult.status === 'rejected' && endpointsResult.status === 'rejected') {
throw new VolcengineServiceError('Failed to fetch both foundation models and endpoints')
}
const total =
(foundationModelsResult.status === 'fulfilled' ? foundationModelsResult.value.Result.TotalCount : 0) +
(endpointsResult.status === 'fulfilled' ? endpointsResult.value.Result.TotalCount : 0)
logger.info(`Total models found: ${models.length}`)
return {
models,
total,
warnings: warnings.length > 0 ? warnings : undefined
}
} catch (error) {
logger.error('Failed to list Volcengine models:', error as Error)
throw new VolcengineServiceError('Failed to list models', error)
}
}
/**
* Get authorization headers for external use
* This allows the renderer process to make direct API calls with proper authentication
*/
public getAuthHeaders = async (
_: Electron.IpcMainInvokeEvent,
params: {
method: 'GET' | 'POST'
host: string
path: string
query?: Record<string, string>
body?: string
service?: string
region?: string
}
): Promise<SignedHeaders> => {
const credentials = await this.loadCredentials()
if (!credentials) {
throw new VolcengineServiceError('No credentials found. Please save credentials first.')
}
return this.generateSignature(
{
method: params.method,
host: params.host,
path: params.path,
query: params.query || {},
headers: {},
body: params.body,
service: params.service || CONFIG.SERVICE_NAME,
region: params.region || CONFIG.DEFAULT_REGION
},
credentials
)
}
/**
* Make a generic signed API request
* This is a more flexible method that allows custom API calls
*/
public makeRequest = async (
_: Electron.IpcMainInvokeEvent,
params: {
method: 'GET' | 'POST'
host: string
path: string
action: string
version: string
query?: Record<string, string>
body?: Record<string, unknown>
service?: string
region?: string
}
): Promise<unknown> => {
return this.makeSignedRequest(
params.method,
params.host,
params.path,
params.action,
params.version,
params.query,
params.body,
params.service || CONFIG.SERVICE_NAME,
params.region || CONFIG.DEFAULT_REGION
)
}
}
export default new VolcengineService()

View File

@@ -1,17 +1,13 @@
import { type Client, createClient } from '@libsql/client'
import { loggerService } from '@logger'
import { mcpApiService } from '@main/apiServer/services/mcp'
import type { ModelValidationError } from '@main/apiServer/utils'
import { validateModelId } from '@main/apiServer/utils'
import type { AgentType, MCPTool, SlashCommand, Tool } from '@types'
import { objectKeys } from '@types'
import { drizzle, type LibSQLDatabase } from 'drizzle-orm/libsql'
import fs from 'fs'
import path from 'path'
import { MigrationService } from './database/MigrationService'
import * as schema from './database/schema'
import { dbPath } from './drizzle.config'
import { DatabaseManager } from './database/DatabaseManager'
import type { AgentModelField } from './errors'
import { AgentModelValidationError } from './errors'
import { builtinSlashCommands } from './services/claudecode/commands'
@@ -20,22 +16,16 @@ import { builtinTools } from './services/claudecode/tools'
const logger = loggerService.withContext('BaseService')
/**
* Base service class providing shared database connection and utilities
* for all agent-related services.
* Base service class providing shared utilities for all agent-related services.
*
* Features:
* - Programmatic schema management (no CLI dependencies)
* - Automatic table creation and migration
* - Schema version tracking and compatibility checks
* - Transaction-based operations for safety
* - Development vs production mode handling
* - Connection retry logic with exponential backoff
* - Database access through DatabaseManager singleton
* - JSON field serialization/deserialization
* - Path validation and creation
* - Model validation
* - MCP tools and slash commands listing
*/
export abstract class BaseService {
protected static client: Client | null = null
protected static db: LibSQLDatabase<typeof schema> | null = null
protected static isInitialized = false
protected static initializationPromise: Promise<void> | null = null
protected jsonFields: string[] = [
'tools',
'mcps',
@@ -45,23 +35,6 @@ export abstract class BaseService {
'slash_commands'
]
/**
* Initialize database with retry logic and proper error handling
*/
protected static async initialize(): Promise<void> {
// Return existing initialization if in progress
if (BaseService.initializationPromise) {
return BaseService.initializationPromise
}
if (BaseService.isInitialized) {
return
}
BaseService.initializationPromise = BaseService.performInitialization()
return BaseService.initializationPromise
}
public async listMcpTools(agentType: AgentType, ids?: string[]): Promise<Tool[]> {
const tools: Tool[] = []
if (agentType === 'claude-code') {
@@ -101,78 +74,13 @@ export abstract class BaseService {
return []
}
private static async performInitialization(): Promise<void> {
const maxRetries = 3
let lastError: Error
for (let attempt = 1; attempt <= maxRetries; attempt++) {
try {
logger.info(`Initializing Agent database at: ${dbPath} (attempt ${attempt}/${maxRetries})`)
// Ensure the database directory exists
const dbDir = path.dirname(dbPath)
if (!fs.existsSync(dbDir)) {
logger.info(`Creating database directory: ${dbDir}`)
fs.mkdirSync(dbDir, { recursive: true })
}
BaseService.client = createClient({
url: `file:${dbPath}`
})
BaseService.db = drizzle(BaseService.client, { schema })
// Run database migrations
const migrationService = new MigrationService(BaseService.db, BaseService.client)
await migrationService.runMigrations()
BaseService.isInitialized = true
logger.info('Agent database initialized successfully')
return
} catch (error) {
lastError = error as Error
logger.warn(`Database initialization attempt ${attempt} failed:`, lastError)
// Clean up on failure
if (BaseService.client) {
try {
BaseService.client.close()
} catch (closeError) {
logger.warn('Failed to close client during cleanup:', closeError as Error)
}
}
BaseService.client = null
BaseService.db = null
// Wait before retrying (exponential backoff)
if (attempt < maxRetries) {
const delay = Math.pow(2, attempt) * 1000 // 2s, 4s, 8s
logger.info(`Retrying in ${delay}ms...`)
await new Promise((resolve) => setTimeout(resolve, delay))
}
}
}
// All retries failed
BaseService.initializationPromise = null
logger.error('Failed to initialize Agent database after all retries:', lastError!)
throw lastError!
}
protected ensureInitialized(): void {
if (!BaseService.isInitialized || !BaseService.db || !BaseService.client) {
throw new Error('Database not initialized. Call initialize() first.')
}
}
protected get database(): LibSQLDatabase<typeof schema> {
this.ensureInitialized()
return BaseService.db!
}
protected get rawClient(): Client {
this.ensureInitialized()
return BaseService.client!
/**
* Get database instance
* Automatically waits for initialization to complete
*/
protected async getDatabase() {
const dbManager = await DatabaseManager.getInstance()
return dbManager.getDatabase()
}
protected serializeJsonFields(data: any): any {
@@ -284,7 +192,7 @@ export abstract class BaseService {
}
/**
* Force re-initialization (for development/testing)
* Validate agent model configuration
*/
protected async validateAgentModels(
agentType: AgentType,
@@ -325,22 +233,4 @@ export abstract class BaseService {
}
}
}
static async reinitialize(): Promise<void> {
BaseService.isInitialized = false
BaseService.initializationPromise = null
if (BaseService.client) {
try {
BaseService.client.close()
} catch (error) {
logger.warn('Failed to close client during reinitialize:', error as Error)
}
}
BaseService.client = null
BaseService.db = null
await BaseService.initialize()
}
}

View File

@@ -0,0 +1,156 @@
import { type Client, createClient } from '@libsql/client'
import { loggerService } from '@logger'
import type { LibSQLDatabase } from 'drizzle-orm/libsql'
import { drizzle } from 'drizzle-orm/libsql'
import fs from 'fs'
import path from 'path'
import { dbPath } from '../drizzle.config'
import { MigrationService } from './MigrationService'
import * as schema from './schema'
const logger = loggerService.withContext('DatabaseManager')
/**
* Database initialization state
*/
enum InitState {
INITIALIZING = 'initializing',
INITIALIZED = 'initialized',
FAILED = 'failed'
}
/**
* DatabaseManager - Singleton class for managing libsql database connections
*
* Responsibilities:
* - Single source of truth for database connection
* - Thread-safe initialization with state management
* - Automatic migration handling
* - Safe connection cleanup
* - Error recovery and retry logic
* - Windows platform compatibility fixes
*/
export class DatabaseManager {
private static instance: DatabaseManager | null = null
private client: Client | null = null
private db: LibSQLDatabase<typeof schema> | null = null
private state: InitState = InitState.INITIALIZING
/**
* Get the singleton instance (database initialization starts automatically)
*/
public static async getInstance(): Promise<DatabaseManager> {
if (DatabaseManager.instance) {
return DatabaseManager.instance
}
const instance = new DatabaseManager()
await instance.initialize()
DatabaseManager.instance = instance
return instance
}
/**
* Perform the actual initialization
*/
public async initialize(): Promise<void> {
if (this.state === InitState.INITIALIZED) {
return
}
try {
logger.info(`Initializing database at: ${dbPath}`)
// Ensure database directory exists
const dbDir = path.dirname(dbPath)
if (!fs.existsSync(dbDir)) {
logger.info(`Creating database directory: ${dbDir}`)
fs.mkdirSync(dbDir, { recursive: true })
}
// Check if database file is corrupted (Windows specific check)
if (fs.existsSync(dbPath)) {
const stats = fs.statSync(dbPath)
if (stats.size === 0) {
logger.warn('Database file is empty, removing corrupted file')
fs.unlinkSync(dbPath)
}
}
// Create client with platform-specific options
this.client = createClient({
url: `file:${dbPath}`,
// intMode: 'number' helps avoid some Windows compatibility issues
intMode: 'number'
})
// Create drizzle instance
this.db = drizzle(this.client, { schema })
// Run migrations
const migrationService = new MigrationService(this.db, this.client)
await migrationService.runMigrations()
this.state = InitState.INITIALIZED
logger.info('Database initialized successfully')
} catch (error) {
const err = error as Error
logger.error('Database initialization failed:', {
error: err.message,
stack: err.stack
})
// Clean up failed initialization
this.cleanupFailedInit()
// Set failed state
this.state = InitState.FAILED
throw new Error(`Database initialization failed: ${err.message || 'Unknown error'}`)
}
}
/**
* Clean up after failed initialization
*/
private cleanupFailedInit(): void {
if (this.client) {
try {
// On Windows, closing a partially initialized client can crash
// Wrap in try-catch and ignore errors during cleanup
this.client.close()
} catch (error) {
logger.warn('Failed to close client during cleanup:', error as Error)
}
}
this.client = null
this.db = null
}
/**
* Get the database instance
* Automatically waits for initialization to complete
* @throws Error if database initialization failed
*/
public getDatabase(): LibSQLDatabase<typeof schema> {
return this.db!
}
/**
* Get the raw client (for advanced operations)
* Automatically waits for initialization to complete
* @throws Error if database initialization failed
*/
public async getClient(): Promise<Client> {
return this.client!
}
/**
* Check if database is initialized
*/
public isInitialized(): boolean {
return this.state === InitState.INITIALIZED
}
}

View File

@@ -7,8 +7,14 @@
* Schema evolution is handled by Drizzle Kit migrations.
*/
// Database Manager (Singleton)
export * from './DatabaseManager'
// Drizzle ORM schemas
export * from './schema'
// Repository helpers
export * from './sessionMessageRepository'
// Migration Service
export * from './MigrationService'

View File

@@ -15,26 +15,16 @@ import { sessionMessagesTable } from './schema'
const logger = loggerService.withContext('AgentMessageRepository')
type TxClient = any
export type PersistUserMessageParams = AgentMessageUserPersistPayload & {
sessionId: string
agentSessionId?: string
tx?: TxClient
}
export type PersistAssistantMessageParams = AgentMessageAssistantPersistPayload & {
sessionId: string
agentSessionId: string
tx?: TxClient
}
type PersistExchangeParams = AgentMessagePersistExchangePayload & {
tx?: TxClient
}
type PersistExchangeResult = AgentMessagePersistExchangeResult
class AgentMessageRepository extends BaseService {
private static instance: AgentMessageRepository | null = null
@@ -87,17 +77,13 @@ class AgentMessageRepository extends BaseService {
return deserialized
}
private getWriter(tx?: TxClient): TxClient {
return tx ?? this.database
}
private async findExistingMessageRow(
writer: TxClient,
sessionId: string,
role: string,
messageId: string
): Promise<SessionMessageRow | null> {
const candidateRows: SessionMessageRow[] = await writer
const database = await this.getDatabase()
const candidateRows: SessionMessageRow[] = await database
.select()
.from(sessionMessagesTable)
.where(and(eq(sessionMessagesTable.session_id, sessionId), eq(sessionMessagesTable.role, role)))
@@ -122,10 +108,7 @@ class AgentMessageRepository extends BaseService {
private async upsertMessage(
params: PersistUserMessageParams | PersistAssistantMessageParams
): Promise<AgentSessionMessageEntity> {
await AgentMessageRepository.initialize()
this.ensureInitialized()
const { sessionId, agentSessionId = '', payload, metadata, createdAt, tx } = params
const { sessionId, agentSessionId = '', payload, metadata, createdAt } = params
if (!payload?.message?.role) {
throw new Error('Message payload missing role')
@@ -135,18 +118,18 @@ class AgentMessageRepository extends BaseService {
throw new Error('Message payload missing id')
}
const writer = this.getWriter(tx)
const database = await this.getDatabase()
const now = createdAt ?? payload.message.createdAt ?? new Date().toISOString()
const serializedPayload = this.serializeMessage(payload)
const serializedMetadata = this.serializeMetadata(metadata)
const existingRow = await this.findExistingMessageRow(writer, sessionId, payload.message.role, payload.message.id)
const existingRow = await this.findExistingMessageRow(sessionId, payload.message.role, payload.message.id)
if (existingRow) {
const metadataToPersist = serializedMetadata ?? existingRow.metadata ?? undefined
const agentSessionToPersist = agentSessionId || existingRow.agent_session_id || ''
await writer
await database
.update(sessionMessagesTable)
.set({
content: serializedPayload,
@@ -175,7 +158,7 @@ class AgentMessageRepository extends BaseService {
updated_at: now
}
const [saved] = await writer.insert(sessionMessagesTable).values(insertData).returning()
const [saved] = await database.insert(sessionMessagesTable).values(insertData).returning()
return this.deserialize(saved)
}
@@ -188,49 +171,38 @@ class AgentMessageRepository extends BaseService {
return this.upsertMessage(params)
}
async persistExchange(params: PersistExchangeParams): Promise<PersistExchangeResult> {
await AgentMessageRepository.initialize()
this.ensureInitialized()
async persistExchange(params: AgentMessagePersistExchangePayload): Promise<AgentMessagePersistExchangeResult> {
const { sessionId, agentSessionId, user, assistant } = params
const result = await this.database.transaction(async (tx) => {
const exchangeResult: PersistExchangeResult = {}
const exchangeResult: AgentMessagePersistExchangeResult = {}
if (user?.payload) {
exchangeResult.userMessage = await this.persistUserMessage({
sessionId,
agentSessionId,
payload: user.payload,
metadata: user.metadata,
createdAt: user.createdAt,
tx
})
}
if (user?.payload) {
exchangeResult.userMessage = await this.persistUserMessage({
sessionId,
agentSessionId,
payload: user.payload,
metadata: user.metadata,
createdAt: user.createdAt
})
}
if (assistant?.payload) {
exchangeResult.assistantMessage = await this.persistAssistantMessage({
sessionId,
agentSessionId,
payload: assistant.payload,
metadata: assistant.metadata,
createdAt: assistant.createdAt,
tx
})
}
if (assistant?.payload) {
exchangeResult.assistantMessage = await this.persistAssistantMessage({
sessionId,
agentSessionId,
payload: assistant.payload,
metadata: assistant.metadata,
createdAt: assistant.createdAt
})
}
return exchangeResult
})
return result
return exchangeResult
}
async getSessionHistory(sessionId: string): Promise<AgentPersistedMessage[]> {
await AgentMessageRepository.initialize()
this.ensureInitialized()
try {
const rows = await this.database
const database = await this.getDatabase()
const rows = await database
.select()
.from(sessionMessagesTable)
.where(eq(sessionMessagesTable.session_id, sessionId))

View File

@@ -32,14 +32,8 @@ export class AgentService extends BaseService {
return AgentService.instance
}
async initialize(): Promise<void> {
await BaseService.initialize()
}
// Agent Methods
async createAgent(req: CreateAgentRequest): Promise<CreateAgentResponse> {
this.ensureInitialized()
const id = `agent_${Date.now()}_${Math.random().toString(36).substring(2, 11)}`
const now = new Date().toISOString()
@@ -75,8 +69,9 @@ export class AgentService extends BaseService {
updated_at: now
}
await this.database.insert(agentsTable).values(insertData)
const result = await this.database.select().from(agentsTable).where(eq(agentsTable.id, id)).limit(1)
const database = await this.getDatabase()
await database.insert(agentsTable).values(insertData)
const result = await database.select().from(agentsTable).where(eq(agentsTable.id, id)).limit(1)
if (!result[0]) {
throw new Error('Failed to create agent')
}
@@ -86,9 +81,8 @@ export class AgentService extends BaseService {
}
async getAgent(id: string): Promise<GetAgentResponse | null> {
this.ensureInitialized()
const result = await this.database.select().from(agentsTable).where(eq(agentsTable.id, id)).limit(1)
const database = await this.getDatabase()
const result = await database.select().from(agentsTable).where(eq(agentsTable.id, id)).limit(1)
if (!result[0]) {
return null
@@ -118,9 +112,9 @@ export class AgentService extends BaseService {
}
async listAgents(options: ListOptions = {}): Promise<{ agents: AgentEntity[]; total: number }> {
this.ensureInitialized() // Build query with pagination
const totalResult = await this.database.select({ count: count() }).from(agentsTable)
// Build query with pagination
const database = await this.getDatabase()
const totalResult = await database.select({ count: count() }).from(agentsTable)
const sortBy = options.sortBy || 'created_at'
const orderBy = options.orderBy || 'desc'
@@ -128,7 +122,7 @@ export class AgentService extends BaseService {
const sortField = agentsTable[sortBy]
const orderFn = orderBy === 'asc' ? asc : desc
const baseQuery = this.database.select().from(agentsTable).orderBy(orderFn(sortField))
const baseQuery = database.select().from(agentsTable).orderBy(orderFn(sortField))
const result =
options.limit !== undefined
@@ -151,8 +145,6 @@ export class AgentService extends BaseService {
updates: UpdateAgentRequest,
options: { replace?: boolean } = {}
): Promise<UpdateAgentResponse | null> {
this.ensureInitialized()
// Check if agent exists
const existing = await this.getAgent(id)
if (!existing) {
@@ -195,22 +187,21 @@ export class AgentService extends BaseService {
}
}
await this.database.update(agentsTable).set(updateData).where(eq(agentsTable.id, id))
const database = await this.getDatabase()
await database.update(agentsTable).set(updateData).where(eq(agentsTable.id, id))
return await this.getAgent(id)
}
async deleteAgent(id: string): Promise<boolean> {
this.ensureInitialized()
const result = await this.database.delete(agentsTable).where(eq(agentsTable.id, id))
const database = await this.getDatabase()
const result = await database.delete(agentsTable).where(eq(agentsTable.id, id))
return result.rowsAffected > 0
}
async agentExists(id: string): Promise<boolean> {
this.ensureInitialized()
const result = await this.database
const database = await this.getDatabase()
const result = await database
.select({ id: agentsTable.id })
.from(agentsTable)
.where(eq(agentsTable.id, id))

View File

@@ -104,14 +104,9 @@ export class SessionMessageService extends BaseService {
return SessionMessageService.instance
}
async initialize(): Promise<void> {
await BaseService.initialize()
}
async sessionMessageExists(id: number): Promise<boolean> {
this.ensureInitialized()
const result = await this.database
const database = await this.getDatabase()
const result = await database
.select({ id: sessionMessagesTable.id })
.from(sessionMessagesTable)
.where(eq(sessionMessagesTable.id, id))
@@ -124,10 +119,9 @@ export class SessionMessageService extends BaseService {
sessionId: string,
options: ListOptions = {}
): Promise<{ messages: AgentSessionMessageEntity[] }> {
this.ensureInitialized()
// Get messages with pagination
const baseQuery = this.database
const database = await this.getDatabase()
const baseQuery = database
.select()
.from(sessionMessagesTable)
.where(eq(sessionMessagesTable.session_id, sessionId))
@@ -146,9 +140,8 @@ export class SessionMessageService extends BaseService {
}
async deleteSessionMessage(sessionId: string, messageId: number): Promise<boolean> {
this.ensureInitialized()
const result = await this.database
const database = await this.getDatabase()
const result = await database
.delete(sessionMessagesTable)
.where(and(eq(sessionMessagesTable.id, messageId), eq(sessionMessagesTable.session_id, sessionId)))
@@ -160,8 +153,6 @@ export class SessionMessageService extends BaseService {
messageData: CreateSessionMessageRequest,
abortController: AbortController
): Promise<SessionStreamResult> {
this.ensureInitialized()
return await this.startSessionMessageStream(session, messageData, abortController)
}
@@ -270,10 +261,9 @@ export class SessionMessageService extends BaseService {
}
private async getLastAgentSessionId(sessionId: string): Promise<string> {
this.ensureInitialized()
try {
const result = await this.database
const database = await this.getDatabase()
const result = await database
.select({ agent_session_id: sessionMessagesTable.agent_session_id })
.from(sessionMessagesTable)
.where(and(eq(sessionMessagesTable.session_id, sessionId), not(eq(sessionMessagesTable.agent_session_id, ''))))

View File

@@ -30,10 +30,6 @@ export class SessionService extends BaseService {
return SessionService.instance
}
async initialize(): Promise<void> {
await BaseService.initialize()
}
/**
* Override BaseService.listSlashCommands to merge builtin and plugin commands
*/
@@ -84,13 +80,12 @@ export class SessionService extends BaseService {
agentId: string,
req: Partial<CreateSessionRequest> = {}
): Promise<GetAgentSessionResponse | null> {
this.ensureInitialized()
// Validate agent exists - we'll need to import AgentService for this check
// For now, we'll skip this validation to avoid circular dependencies
// The database foreign key constraint will handle this
const agents = await this.database.select().from(agentsTable).where(eq(agentsTable.id, agentId)).limit(1)
const database = await this.getDatabase()
const agents = await database.select().from(agentsTable).where(eq(agentsTable.id, agentId)).limit(1)
if (!agents[0]) {
throw new Error('Agent not found')
}
@@ -135,9 +130,10 @@ export class SessionService extends BaseService {
updated_at: now
}
await this.database.insert(sessionsTable).values(insertData)
const db = await this.getDatabase()
await db.insert(sessionsTable).values(insertData)
const result = await this.database.select().from(sessionsTable).where(eq(sessionsTable.id, id)).limit(1)
const result = await db.select().from(sessionsTable).where(eq(sessionsTable.id, id)).limit(1)
if (!result[0]) {
throw new Error('Failed to create session')
@@ -148,9 +144,8 @@ export class SessionService extends BaseService {
}
async getSession(agentId: string, id: string): Promise<GetAgentSessionResponse | null> {
this.ensureInitialized()
const result = await this.database
const database = await this.getDatabase()
const result = await database
.select()
.from(sessionsTable)
.where(and(eq(sessionsTable.id, id), eq(sessionsTable.agent_id, agentId)))
@@ -176,8 +171,6 @@ export class SessionService extends BaseService {
agentId?: string,
options: ListOptions = {}
): Promise<{ sessions: AgentSessionEntity[]; total: number }> {
this.ensureInitialized()
// Build where conditions
const whereConditions: SQL[] = []
if (agentId) {
@@ -192,16 +185,13 @@ export class SessionService extends BaseService {
: undefined
// Get total count
const totalResult = await this.database.select({ count: count() }).from(sessionsTable).where(whereClause)
const database = await this.getDatabase()
const totalResult = await database.select({ count: count() }).from(sessionsTable).where(whereClause)
const total = totalResult[0].count
// Build list query with pagination - sort by updated_at descending (latest first)
const baseQuery = this.database
.select()
.from(sessionsTable)
.where(whereClause)
.orderBy(desc(sessionsTable.updated_at))
const baseQuery = database.select().from(sessionsTable).where(whereClause).orderBy(desc(sessionsTable.updated_at))
const result =
options.limit !== undefined
@@ -220,8 +210,6 @@ export class SessionService extends BaseService {
id: string,
updates: UpdateSessionRequest
): Promise<UpdateSessionResponse | null> {
this.ensureInitialized()
// Check if session exists
const existing = await this.getSession(agentId, id)
if (!existing) {
@@ -262,15 +250,15 @@ export class SessionService extends BaseService {
}
}
await this.database.update(sessionsTable).set(updateData).where(eq(sessionsTable.id, id))
const database = await this.getDatabase()
await database.update(sessionsTable).set(updateData).where(eq(sessionsTable.id, id))
return await this.getSession(agentId, id)
}
async deleteSession(agentId: string, id: string): Promise<boolean> {
this.ensureInitialized()
const result = await this.database
const database = await this.getDatabase()
const result = await database
.delete(sessionsTable)
.where(and(eq(sessionsTable.id, id), eq(sessionsTable.agent_id, agentId)))
@@ -278,9 +266,8 @@ export class SessionService extends BaseService {
}
async sessionExists(agentId: string, id: string): Promise<boolean> {
this.ensureInitialized()
const result = await this.database
const database = await this.getDatabase()
const result = await database
.select({ id: sessionsTable.id })
.from(sessionsTable)
.where(and(eq(sessionsTable.id, id), eq(sessionsTable.agent_id, agentId)))

View File

@@ -21,6 +21,11 @@ describe('stripLocalCommandTags', () => {
'<local-command-stdout>line1</local-command-stdout>\nkeep\n<local-command-stderr>Error</local-command-stderr>'
expect(stripLocalCommandTags(input)).toBe('line1\nkeep\nError')
})
it('if no tags present, returns original string', () => {
const input = 'just some normal text'
expect(stripLocalCommandTags(input)).toBe(input)
})
})
describe('Claude → AiSDK transform', () => {
@@ -188,6 +193,111 @@ describe('Claude → AiSDK transform', () => {
expect(toolResult.output).toBe('ok')
})
it('handles tool calls without streaming events (no content_block_start/stop)', () => {
const state = new ClaudeStreamState({ agentSessionId: '12344' })
const parts: ReturnType<typeof transformSDKMessageToStreamParts>[number][] = []
const messages: SDKMessage[] = [
{
...baseStreamMetadata,
type: 'assistant',
uuid: uuid(20),
message: {
id: 'msg-tool-no-stream',
type: 'message',
role: 'assistant',
model: 'claude-test',
content: [
{
type: 'tool_use',
id: 'tool-read',
name: 'Read',
input: { file_path: '/test.txt' }
},
{
type: 'tool_use',
id: 'tool-bash',
name: 'Bash',
input: { command: 'ls -la' }
}
],
stop_reason: 'tool_use',
stop_sequence: null,
usage: {
input_tokens: 10,
output_tokens: 20
}
}
} as unknown as SDKMessage,
{
...baseStreamMetadata,
type: 'user',
uuid: uuid(21),
message: {
role: 'user',
content: [
{
type: 'tool_result',
tool_use_id: 'tool-read',
content: 'file contents',
is_error: false
}
]
}
} as SDKMessage,
{
...baseStreamMetadata,
type: 'user',
uuid: uuid(22),
message: {
role: 'user',
content: [
{
type: 'tool_result',
tool_use_id: 'tool-bash',
content: 'total 42\n...',
is_error: false
}
]
}
} as SDKMessage
]
for (const message of messages) {
const transformed = transformSDKMessageToStreamParts(message, state)
parts.push(...transformed)
}
const types = parts.map((part) => part.type)
expect(types).toEqual(['tool-call', 'tool-call', 'tool-result', 'tool-result'])
const toolCalls = parts.filter((part) => part.type === 'tool-call') as Extract<
(typeof parts)[number],
{ type: 'tool-call' }
>[]
expect(toolCalls).toHaveLength(2)
expect(toolCalls[0].toolName).toBe('Read')
expect(toolCalls[0].toolCallId).toBe('12344:tool-read')
expect(toolCalls[1].toolName).toBe('Bash')
expect(toolCalls[1].toolCallId).toBe('12344:tool-bash')
const toolResults = parts.filter((part) => part.type === 'tool-result') as Extract<
(typeof parts)[number],
{ type: 'tool-result' }
>[]
expect(toolResults).toHaveLength(2)
// This is the key assertion - toolName should NOT be 'unknown'
expect(toolResults[0].toolName).toBe('Read')
expect(toolResults[0].toolCallId).toBe('12344:tool-read')
expect(toolResults[0].input).toEqual({ file_path: '/test.txt' })
expect(toolResults[0].output).toBe('file contents')
expect(toolResults[1].toolName).toBe('Bash')
expect(toolResults[1].toolCallId).toBe('12344:tool-bash')
expect(toolResults[1].input).toEqual({ command: 'ls -la' })
expect(toolResults[1].output).toBe('total 42\n...')
})
it('handles streaming text completion', () => {
const state = new ClaudeStreamState({ agentSessionId: baseStreamMetadata.session_id })
const parts: ReturnType<typeof transformSDKMessageToStreamParts>[number][] = []
@@ -300,4 +410,87 @@ describe('Claude → AiSDK transform', () => {
expect(finishStep.finishReason).toBe('stop')
expect(finishStep.usage).toEqual({ inputTokens: 2, outputTokens: 4, totalTokens: 6 })
})
it('emits fallback text when Claude sends a snapshot instead of deltas', () => {
const state = new ClaudeStreamState({ agentSessionId: '12344' })
const parts: ReturnType<typeof transformSDKMessageToStreamParts>[number][] = []
const messages: SDKMessage[] = [
{
...baseStreamMetadata,
type: 'stream_event',
uuid: uuid(30),
event: {
type: 'message_start',
message: {
id: 'msg-fallback',
type: 'message',
role: 'assistant',
model: 'claude-test',
content: [],
stop_reason: null,
stop_sequence: null,
usage: {}
}
}
} as unknown as SDKMessage,
{
...baseStreamMetadata,
type: 'stream_event',
uuid: uuid(31),
event: {
type: 'content_block_start',
index: 0,
content_block: {
type: 'text',
text: ''
}
}
} as unknown as SDKMessage,
{
...baseStreamMetadata,
type: 'assistant',
uuid: uuid(32),
message: {
id: 'msg-fallback-content',
type: 'message',
role: 'assistant',
model: 'claude-test',
content: [
{
type: 'text',
text: 'Final answer without streaming deltas.'
}
],
stop_reason: 'end_turn',
stop_sequence: null,
usage: {
input_tokens: 3,
output_tokens: 7
}
}
} as unknown as SDKMessage
]
for (const message of messages) {
const transformed = transformSDKMessageToStreamParts(message, state)
parts.push(...transformed)
}
const types = parts.map((part) => part.type)
expect(types).toEqual(['start-step', 'text-start', 'text-delta', 'text-end', 'finish-step'])
const delta = parts.find((part) => part.type === 'text-delta') as Extract<
(typeof parts)[number],
{ type: 'text-delta' }
>
expect(delta.text).toBe('Final answer without streaming deltas.')
const finish = parts.find((part) => part.type === 'finish-step') as Extract<
(typeof parts)[number],
{ type: 'finish-step' }
>
expect(finish.usage).toEqual({ inputTokens: 3, outputTokens: 7, totalTokens: 10 })
expect(finish.finishReason).toBe('stop')
})
})

View File

@@ -153,6 +153,20 @@ export class ClaudeStreamState {
return this.blocksByIndex.get(index)
}
getFirstOpenTextBlock(): TextBlockState | undefined {
const candidates: TextBlockState[] = []
for (const block of this.blocksByIndex.values()) {
if (block.kind === 'text') {
candidates.push(block)
}
}
if (candidates.length === 0) {
return undefined
}
candidates.sort((a, b) => a.index - b.index)
return candidates[0]
}
getToolBlockById(toolCallId: string): ToolBlockState | undefined {
const index = this.toolIndexByNamespacedId.get(toolCallId)
if (index === undefined) return undefined
@@ -217,10 +231,10 @@ export class ClaudeStreamState {
* Persists the final input payload for a tool block once the provider signals
* completion so that downstream tool results can reference the original call.
*/
completeToolBlock(toolCallId: string, input: unknown, providerMetadata?: ProviderMetadata): void {
completeToolBlock(toolCallId: string, toolName: string, input: unknown, providerMetadata?: ProviderMetadata): void {
const block = this.getToolBlockByRawId(toolCallId)
this.registerToolCall(toolCallId, {
toolName: block?.toolName ?? 'unknown',
toolName,
input,
providerMetadata
})

View File

@@ -2,7 +2,14 @@
import { EventEmitter } from 'node:events'
import { createRequire } from 'node:module'
import type { CanUseTool, McpHttpServerConfig, Options, SDKMessage } from '@anthropic-ai/claude-agent-sdk'
import type {
CanUseTool,
HookCallback,
McpHttpServerConfig,
Options,
PreToolUseHookInput,
SDKMessage
} from '@anthropic-ai/claude-agent-sdk'
import { query } from '@anthropic-ai/claude-agent-sdk'
import { loggerService } from '@logger'
import { config as apiConfigService } from '@main/apiServer/config'
@@ -157,6 +164,63 @@ class ClaudeCodeService implements AgentServiceInterface {
})
}
const preToolUseHook: HookCallback = async (input, toolUseID, options) => {
// Type guard to ensure we're handling PreToolUse event
if (input.hook_event_name !== 'PreToolUse') {
return {}
}
const hookInput = input as PreToolUseHookInput
const toolName = hookInput.tool_name
logger.debug('PreToolUse hook triggered', {
session_id: hookInput.session_id,
tool_name: hookInput.tool_name,
tool_use_id: toolUseID,
tool_input: hookInput.tool_input,
cwd: hookInput.cwd,
permission_mode: hookInput.permission_mode,
autoAllowTools: autoAllowTools
})
if (options?.signal?.aborted) {
logger.debug('PreToolUse hook signal already aborted; skipping tool use', {
tool_name: hookInput.tool_name
})
return {}
}
// handle auto approved tools since it never triggers canUseTool
const normalizedToolName = normalizeToolName(toolName)
if (toolUseID) {
const bypassAll = input.permission_mode === 'bypassPermissions'
const autoAllowed = autoAllowTools.has(toolName) || autoAllowTools.has(normalizedToolName)
if (bypassAll || autoAllowed) {
const namespacedToolCallId = buildNamespacedToolCallId(session.id, toolUseID)
logger.debug('handling auto approved tools', {
toolName,
normalizedToolName,
namespacedToolCallId,
permission_mode: input.permission_mode,
autoAllowTools
})
const isRecord = (v: unknown): v is Record<string, unknown> => {
return !!v && typeof v === 'object' && !Array.isArray(v)
}
const toolInput = isRecord(input.tool_input) ? input.tool_input : {}
await promptForToolApproval(toolName, toolInput, {
...options,
toolCallId: namespacedToolCallId,
autoApprove: true
})
}
}
// Return to proceed without modification
return {}
}
// Build SDK options from parameters
const options: Options = {
abortController,
@@ -180,7 +244,14 @@ class ClaudeCodeService implements AgentServiceInterface {
permissionMode: session.configuration?.permission_mode,
maxTurns: session.configuration?.max_turns,
allowedTools: session.allowed_tools,
canUseTool
canUseTool,
hooks: {
PreToolUse: [
{
hooks: [preToolUseHook]
}
]
}
}
if (session.accessible_paths.length > 1) {
@@ -414,23 +485,6 @@ class ClaudeCodeService implements AgentServiceInterface {
}
}
if (message.type === 'assistant' || message.type === 'user') {
logger.silly('claude response', {
message,
content: JSON.stringify(message.message.content)
})
} else if (message.type === 'stream_event') {
// logger.silly('Claude stream event', {
// message,
// event: JSON.stringify(message.event)
// })
} else {
logger.silly('Claude response', {
message,
event: JSON.stringify(message)
})
}
const chunks = transformSDKMessageToStreamParts(message, streamState)
for (const chunk of chunks) {
stream.emit('data', {

View File

@@ -31,6 +31,7 @@ type PendingPermissionRequest = {
abortListener?: () => void
originalInput: Record<string, unknown>
toolName: string
toolCallId?: string
}
type RendererPermissionRequestPayload = {
@@ -45,6 +46,7 @@ type RendererPermissionRequestPayload = {
createdAt: number
expiresAt: number
suggestions: PermissionUpdate[]
autoApprove?: boolean
}
type RendererPermissionResultPayload = {
@@ -52,6 +54,7 @@ type RendererPermissionResultPayload = {
behavior: ToolPermissionBehavior
message?: string
reason: 'response' | 'timeout' | 'aborted' | 'no-window'
toolCallId?: string
}
const pendingRequests = new Map<string, PendingPermissionRequest>()
@@ -145,7 +148,8 @@ const finalizeRequest = (
requestId,
behavior: update.behavior,
message: update.behavior === 'deny' ? update.message : undefined,
reason
reason,
toolCallId: pending.toolCallId
}
const dispatched = broadcastToRenderer(IpcChannel.AgentToolPermission_Result, resultPayload)
@@ -210,6 +214,7 @@ const ensureIpcHandlersRegistered = () => {
type PromptForToolApprovalOptions = {
signal: AbortSignal
suggestions?: PermissionUpdate[]
autoApprove?: boolean
// NOTICE: This ID is namespaced with session ID, not the raw SDK tool call ID.
// Format: `${sessionId}:${rawToolCallId}`, e.g., `session_123:WebFetch_0`
@@ -270,7 +275,8 @@ export async function promptForToolApproval(
inputPreview,
createdAt,
expiresAt,
suggestions: sanitizedSuggestions
suggestions: sanitizedSuggestions,
autoApprove: options.autoApprove
}
const defaultDenyUpdate: PermissionResult = { behavior: 'deny', message: 'Tool request aborted before user decision' }
@@ -299,7 +305,8 @@ export async function promptForToolApproval(
timeout,
originalInput: sanitizedInput,
toolName,
signal: options?.signal
signal: options?.signal,
toolCallId: options.toolCallId
}
if (options?.signal) {

View File

@@ -110,7 +110,7 @@ const sdkMessageToProviderMetadata = (message: SDKMessage): ProviderMetadata =>
* blocks across calls so that incremental deltas can be correlated correctly.
*/
export function transformSDKMessageToStreamParts(sdkMessage: SDKMessage, state: ClaudeStreamState): AgentStreamPart[] {
logger.silly('Transforming SDKMessage', { message: sdkMessage })
logger.silly('Transforming SDKMessage', { message: JSON.stringify(sdkMessage) })
switch (sdkMessage.type) {
case 'assistant':
return handleAssistantMessage(sdkMessage, state)
@@ -186,14 +186,13 @@ function handleAssistantMessage(
for (const block of content) {
switch (block.type) {
case 'text':
if (!isStreamingActive) {
const sanitizedText = stripLocalCommandTags(block.text)
if (sanitizedText) {
textBlocks.push(sanitizedText)
}
case 'text': {
const sanitizedText = stripLocalCommandTags(block.text)
if (sanitizedText) {
textBlocks.push(sanitizedText)
}
break
}
case 'tool_use':
handleAssistantToolUse(block as ToolUseContent, providerMetadata, state, chunks)
break
@@ -203,7 +202,16 @@ function handleAssistantMessage(
}
}
if (!isStreamingActive && textBlocks.length > 0) {
if (textBlocks.length === 0) {
return chunks
}
const combinedText = textBlocks.join('')
if (!combinedText) {
return chunks
}
if (!isStreamingActive) {
const id = message.uuid?.toString() || generateMessageId()
state.beginStep()
chunks.push({
@@ -219,7 +227,7 @@ function handleAssistantMessage(
chunks.push({
type: 'text-delta',
id,
text: textBlocks.join(''),
text: combinedText,
providerMetadata
})
chunks.push({
@@ -230,7 +238,27 @@ function handleAssistantMessage(
return finalizeNonStreamingStep(message, state, chunks)
}
return chunks
const existingTextBlock = state.getFirstOpenTextBlock()
const fallbackId = existingTextBlock?.id || message.uuid?.toString() || generateMessageId()
if (!existingTextBlock) {
chunks.push({
type: 'text-start',
id: fallbackId,
providerMetadata
})
}
chunks.push({
type: 'text-delta',
id: fallbackId,
text: combinedText,
providerMetadata
})
chunks.push({
type: 'text-end',
id: fallbackId,
providerMetadata
})
return finalizeNonStreamingStep(message, state, chunks)
}
/**
@@ -252,7 +280,7 @@ function handleAssistantToolUse(
providerExecuted: true,
providerMetadata
})
state.completeToolBlock(block.id, block.input, providerMetadata)
state.completeToolBlock(block.id, block.name, block.input, providerMetadata)
}
/**
@@ -459,6 +487,9 @@ function handleStreamEvent(
}
case 'message_stop': {
if (!state.hasActiveStep()) {
break
}
const pending = state.getPendingUsage()
chunks.push({
type: 'finish-step',

View File

@@ -122,7 +122,8 @@ const api = {
system: {
getDeviceType: () => ipcRenderer.invoke(IpcChannel.System_GetDeviceType),
getHostname: () => ipcRenderer.invoke(IpcChannel.System_GetHostname),
getCpuName: () => ipcRenderer.invoke(IpcChannel.System_GetCpuName)
getCpuName: () => ipcRenderer.invoke(IpcChannel.System_GetCpuName),
checkGitBash: (): Promise<boolean> => ipcRenderer.invoke(IpcChannel.System_CheckGitBash)
},
devTools: {
toggle: () => ipcRenderer.invoke(IpcChannel.System_ToggleDevTools)
@@ -571,6 +572,41 @@ const api = {
status: () => ipcRenderer.invoke(IpcChannel.WebSocket_Status),
sendFile: (filePath: string) => ipcRenderer.invoke(IpcChannel.WebSocket_SendFile, filePath),
getAllCandidates: () => ipcRenderer.invoke(IpcChannel.WebSocket_GetAllCandidates)
},
volcengine: {
saveCredentials: (accessKeyId: string, secretAccessKey: string): Promise<void> =>
ipcRenderer.invoke(IpcChannel.Volcengine_SaveCredentials, accessKeyId, secretAccessKey),
hasCredentials: (): Promise<boolean> => ipcRenderer.invoke(IpcChannel.Volcengine_HasCredentials),
clearCredentials: (): Promise<void> => ipcRenderer.invoke(IpcChannel.Volcengine_ClearCredentials),
listModels: (
projectName?: string,
region?: string
): Promise<{
models: Array<{ id: string; name: string; description?: string; created?: number }>
total?: number
warnings?: string[]
}> => ipcRenderer.invoke(IpcChannel.Volcengine_ListModels, projectName, region),
getAuthHeaders: (params: {
method: 'GET' | 'POST'
host: string
path: string
query?: Record<string, string>
body?: string
service?: string
region?: string
}): Promise<{ Authorization: string; 'X-Date': string; 'X-Content-Sha256': string; Host: string }> =>
ipcRenderer.invoke(IpcChannel.Volcengine_GetAuthHeaders, params),
makeRequest: (params: {
method: 'GET' | 'POST'
host: string
path: string
action: string
version: string
query?: Record<string, string>
body?: Record<string, unknown>
service?: string
region?: string
}): Promise<unknown> => ipcRenderer.invoke(IpcChannel.Volcengine_MakeRequest, params)
}
}

View File

@@ -386,14 +386,13 @@ export class AiSdkToChunkAdapter {
case 'error':
this.onChunk({
type: ChunkType.ERROR,
error:
chunk.error instanceof AISDKError
? chunk.error
: new ProviderSpecificError({
message: formatErrorMessage(chunk.error),
provider: 'unknown',
cause: chunk.error
})
error: AISDKError.isInstance(chunk.error)
? chunk.error
: new ProviderSpecificError({
message: formatErrorMessage(chunk.error),
provider: 'unknown',
cause: chunk.error
})
})
break

View File

@@ -32,6 +32,7 @@ import {
prepareSpecialProviderConfig,
providerToAiSdkConfig
} from './provider/providerConfig'
import type { AiSdkConfig } from './types'
const logger = loggerService.withContext('ModernAiProvider')
@@ -44,12 +45,45 @@ export type ModernAiProviderConfig = AiSdkMiddlewareConfig & {
export default class ModernAiProvider {
private legacyProvider: LegacyAiProvider
private config?: ReturnType<typeof providerToAiSdkConfig>
private config?: AiSdkConfig
private actualProvider: Provider
private model?: Model
private localProvider: Awaited<AiSdkProvider> | null = null
// 构造函数重载签名
/**
* Constructor for ModernAiProvider
*
* @param modelOrProvider - Model or Provider object
* @param provider - Optional Provider object (only used when first param is Model)
*
* @remarks
* **Important behavior notes**:
*
* 1. When called with `(model)`:
* - Calls `getActualProvider(model)` to retrieve and format the provider
* - URL will be automatically formatted via `formatProviderApiHost`, adding version suffixes like `/v1`
*
* 2. When called with `(model, provider)`:
* - **Directly uses the provided provider WITHOUT going through `getActualProvider`**
* - **URL will NOT be automatically formatted, `/v1` suffix will NOT be added**
* - This is legacy behavior kept for backward compatibility
*
* 3. When called with `(provider)`:
* - Directly uses the provider without requiring a model
* - Used for operations that don't need a model (e.g., fetchModels)
*
* @example
* ```typescript
* // Recommended: Auto-format URL
* const ai = new ModernAiProvider(model)
*
* // Not recommended: Skip URL formatting (only for special cases)
* const ai = new ModernAiProvider(model, customProvider)
*
* // For operations that don't need a model
* const ai = new ModernAiProvider(provider)
* ```
*/
constructor(model: Model, provider?: Provider)
constructor(provider: Provider)
constructor(modelOrProvider: Model | Provider, provider?: Provider)
@@ -89,6 +123,11 @@ export default class ModernAiProvider {
// 每次请求时重新生成配置以确保API key轮换生效
this.config = providerToAiSdkConfig(this.actualProvider, this.model)
logger.debug('Generated provider config for completions', this.config)
// 检查 config 是否存在
if (!this.config) {
throw new Error('Provider config is undefined; cannot proceed with completions')
}
if (SUPPORTED_IMAGE_ENDPOINT_LIST.includes(this.config.options.endpoint)) {
providerConfig.isImageGenerationEndpoint = true
}
@@ -149,7 +188,8 @@ export default class ModernAiProvider {
params: StreamTextParams,
config: ModernAiProviderConfig
): Promise<CompletionsResult> {
if (config.isImageGenerationEndpoint) {
// ai-gateway不是image/generation 端点所以就先不走legacy了
if (config.isImageGenerationEndpoint && this.getActualProvider().id !== SystemProviderIds['ai-gateway']) {
// 使用 legacy 实现处理图像生成(支持图片编辑等高级功能)
if (!config.uiMessages) {
throw new Error('uiMessages is required for image generation endpoint')
@@ -315,10 +355,10 @@ export default class ModernAiProvider {
}
}
/**
* 使用现代化 AI SDK 的图像生成实现,支持流式输出
* @deprecated 已改为使用 legacy 实现以支持图片编辑等高级功能
*/
// /**
// * 使用现代化 AI SDK 的图像生成实现,支持流式输出
// * @deprecated 已改为使用 legacy 实现以支持图片编辑等高级功能
// */
/*
private async modernImageGeneration(
model: ImageModel,
@@ -463,8 +503,13 @@ export default class ModernAiProvider {
// 如果支持新的 AI SDK使用现代化实现
if (isModernSdkSupported(this.actualProvider)) {
try {
// 确保 config 已定义
if (!this.config) {
throw new Error('Provider config is undefined; cannot proceed with generateImage')
}
// 确保本地provider已创建
if (!this.localProvider) {
if (!this.localProvider && this.config) {
this.localProvider = await createAiSdkProvider(this.config)
if (!this.localProvider) {
throw new Error('Local provider not created')

View File

@@ -1,6 +1,6 @@
import { loggerService } from '@logger'
import { isNewApiProvider } from '@renderer/config/providers'
import type { Provider } from '@renderer/types'
import { isNewApiProvider } from '@renderer/utils/provider'
import { AihubmixAPIClient } from './aihubmix/AihubmixAPIClient'
import { AnthropicAPIClient } from './anthropic/AnthropicAPIClient'
@@ -14,6 +14,7 @@ import { OpenAIAPIClient } from './openai/OpenAIApiClient'
import { OpenAIResponseAPIClient } from './openai/OpenAIResponseAPIClient'
import { OVMSClient } from './ovms/OVMSClient'
import { PPIOAPIClient } from './ppio/PPIOAPIClient'
import { VolcengineAPIClient } from './volcengine/VolcengineAPIClient'
import { ZhipuAPIClient } from './zhipu/ZhipuAPIClient'
const logger = loggerService.withContext('ApiClientFactory')
@@ -64,6 +65,12 @@ export class ApiClientFactory {
return instance
}
if (provider.id === 'doubao') {
logger.debug(`Creating VolcengineAPIClient for provider: ${provider.id}`)
instance = new VolcengineAPIClient(provider) as BaseApiClient
return instance
}
if (provider.id === 'ovms') {
logger.debug(`Creating OVMSClient for provider: ${provider.id}`)
instance = new OVMSClient(provider) as BaseApiClient

View File

@@ -7,7 +7,6 @@ import {
isSupportFlexServiceTierModel
} from '@renderer/config/models'
import { REFERENCE_PROMPT } from '@renderer/config/prompts'
import { isSupportServiceTierProvider } from '@renderer/config/providers'
import { getLMStudioKeepAliveTime } from '@renderer/hooks/useLMStudio'
import { getAssistantSettings } from '@renderer/services/AssistantService'
import type {
@@ -19,7 +18,6 @@ import type {
MCPToolResponse,
MemoryItem,
Model,
OpenAIVerbosity,
Provider,
ToolCallResponse,
WebSearchProviderResponse,
@@ -33,6 +31,7 @@ import {
OpenAIServiceTiers,
SystemProviderIds
} from '@renderer/types'
import type { OpenAIVerbosity } from '@renderer/types/aiCoreTypes'
import type { Message } from '@renderer/types/newMessage'
import type {
RequestOptions,
@@ -48,6 +47,7 @@ import type {
import { isJSON, parseJSON } from '@renderer/utils'
import { addAbortController, removeAbortController } from '@renderer/utils/abortController'
import { findFileBlocks, getMainTextContent } from '@renderer/utils/messageUtils/find'
import { isSupportServiceTierProvider } from '@renderer/utils/provider'
import { defaultTimeout } from '@shared/config/constant'
import { defaultAppHeaders } from '@shared/utils'
import { isEmpty } from 'lodash'

View File

@@ -58,10 +58,27 @@ vi.mock('../aws/AwsBedrockAPIClient', () => ({
AwsBedrockAPIClient: vi.fn().mockImplementation(() => ({}))
}))
vi.mock('@renderer/services/AssistantService.ts', () => ({
getDefaultAssistant: () => {
return {
id: 'default',
name: 'default',
emoji: '😀',
prompt: '',
topics: [],
messages: [],
type: 'assistant',
regularPhrases: [],
settings: {}
}
}
}))
// Mock the models config to prevent circular dependency issues
vi.mock('@renderer/config/models', () => ({
findTokenLimit: vi.fn(),
isReasoningModel: vi.fn(),
isOpenAILLMModel: vi.fn(),
SYSTEM_MODELS: {
silicon: [],
defaultModel: []

View File

@@ -1,7 +1,8 @@
import { GoogleGenAI } from '@google/genai'
import { loggerService } from '@logger'
import { createVertexProvider, isVertexAIConfigured, isVertexProvider } from '@renderer/hooks/useVertexAI'
import { createVertexProvider, isVertexAIConfigured } from '@renderer/hooks/useVertexAI'
import type { Model, Provider, VertexProvider } from '@renderer/types'
import { isVertexProvider } from '@renderer/utils/provider'
import { isEmpty } from 'lodash'
import { AnthropicVertexClient } from '../anthropic/AnthropicVertexClient'

View File

@@ -10,12 +10,9 @@ import { DEFAULT_MAX_TOKENS } from '@renderer/config/constant'
import {
findTokenLimit,
GEMINI_FLASH_MODEL_REGEX,
getOpenAIWebSearchParams,
getThinkModelType,
isClaudeReasoningModel,
isDeepSeekHybridInferenceModel,
isDoubaoThinkingAutoModel,
isGeminiReasoningModel,
isGPT5SeriesModel,
isGrokReasoningModel,
isNotSupportSystemMessageModel,
@@ -40,12 +37,6 @@ import {
MODEL_SUPPORTED_REASONING_EFFORT,
ZHIPU_RESULT_TOKENS
} from '@renderer/config/models'
import {
isSupportArrayContentProvider,
isSupportDeveloperRoleProvider,
isSupportEnableThinkingProvider,
isSupportStreamOptionsProvider
} from '@renderer/config/providers'
import { mapLanguageToQwenMTModel } from '@renderer/config/translate'
import { processPostsuffixQwen3Model, processReqMessages } from '@renderer/services/ModelMessageService'
import { estimateTextTokens } from '@renderer/services/TokenService'
@@ -89,6 +80,12 @@ import {
openAIToolsToMcpTool
} from '@renderer/utils/mcp-tools'
import { findFileBlocks, findImageBlocks } from '@renderer/utils/messageUtils/find'
import {
isSupportArrayContentProvider,
isSupportDeveloperRoleProvider,
isSupportEnableThinkingProvider,
isSupportStreamOptionsProvider
} from '@renderer/utils/provider'
import { t } from 'i18next'
import type { GenericChunk } from '../../middleware/schemas'
@@ -652,7 +649,6 @@ export class OpenAIAPIClient extends OpenAIBaseClient<
logger.warn('No user message. Some providers may not support.')
}
// poe 需要通过用户消息传递 reasoningEffort
const reasoningEffort = this.getReasoningEffort(assistant, model)
const lastUserMsg = userMessages.findLast((m) => m.role === 'user')
@@ -663,22 +659,6 @@ export class OpenAIAPIClient extends OpenAIBaseClient<
lastUserMsg.content = processPostsuffixQwen3Model(currentContent, qwenThinkModeEnabled)
}
if (this.provider.id === SystemProviderIds.poe) {
// 如果以后 poe 支持 reasoning_effort 参数了,可以删掉这部分
let suffix = ''
if (isGPT5SeriesModel(model) && reasoningEffort.reasoning_effort) {
suffix = ` --reasoning_effort ${reasoningEffort.reasoning_effort}`
} else if (isClaudeReasoningModel(model) && reasoningEffort.thinking?.budget_tokens) {
suffix = ` --thinking_budget ${reasoningEffort.thinking.budget_tokens}`
} else if (isGeminiReasoningModel(model) && reasoningEffort.extra_body?.google?.thinking_config) {
suffix = ` --thinking_budget ${reasoningEffort.extra_body.google.thinking_config.thinking_budget}`
}
// FIXME: poe 不支持多个text part上传文本文件的时候用的不是file part而是text part因此会出问题
// 临时解决方案是强制poe用string content但是其实poe部分支持array
if (typeof lastUserMsg.content === 'string') {
lastUserMsg.content += suffix
}
}
}
// 4. 最终请求消息
@@ -743,7 +723,7 @@ export class OpenAIAPIClient extends OpenAIBaseClient<
: {}),
...this.getProviderSpecificParameters(assistant, model),
...reasoningEffort,
...getOpenAIWebSearchParams(model, enableWebSearch),
// ...getOpenAIWebSearchParams(model, enableWebSearch),
// OpenRouter usage tracking
...(this.provider.id === 'openrouter' ? { usage: { include: true } } : {}),
...extra_body,

View File

@@ -12,7 +12,6 @@ import {
isSupportVerbosityModel,
isVisionModel
} from '@renderer/config/models'
import { isSupportDeveloperRoleProvider } from '@renderer/config/providers'
import { estimateTextTokens } from '@renderer/services/TokenService'
import type {
FileMetadata,
@@ -43,6 +42,7 @@ import {
openAIToolsToMcpTool
} from '@renderer/utils/mcp-tools'
import { findFileBlocks, findImageBlocks } from '@renderer/utils/messageUtils/find'
import { isSupportDeveloperRoleProvider } from '@renderer/utils/provider'
import { MB } from '@shared/config/constant'
import { t } from 'i18next'
import { isEmpty } from 'lodash'

View File

@@ -0,0 +1,74 @@
import type OpenAI from '@cherrystudio/openai'
import { loggerService } from '@logger'
import { getVolcengineProjectName, getVolcengineRegion } from '@renderer/hooks/useVolcengine'
import type { Provider } from '@renderer/types'
import { OpenAIAPIClient } from '../openai/OpenAIApiClient'
const logger = loggerService.withContext('VolcengineAPIClient')
/**
* Volcengine (Doubao) API Client
*
* Extends OpenAIAPIClient for standard chat completions (OpenAI-compatible),
* but overrides listModels to use Volcengine's signed API via IPC.
*/
export class VolcengineAPIClient extends OpenAIAPIClient {
constructor(provider: Provider) {
super(provider)
}
/**
* List models using Volcengine's signed API
* This calls the main process VolcengineService which handles HMAC-SHA256 signing
*/
override async listModels(): Promise<OpenAI.Models.Model[]> {
try {
const hasCredentials = await window.api.volcengine.hasCredentials()
if (!hasCredentials) {
logger.info('Volcengine credentials not configured, falling back to OpenAI-compatible list')
// Fall back to standard OpenAI-compatible API if no Volcengine credentials
return super.listModels()
}
logger.info('Fetching models from Volcengine API using signed request')
const projectName = getVolcengineProjectName()
const region = getVolcengineRegion()
const response = await window.api.volcengine.listModels(projectName, region)
if (!response || !response.models) {
logger.warn('Empty response from Volcengine listModels')
return []
}
// Notify user of any partial failures
if (response.warnings && response.warnings.length > 0) {
for (const warning of response.warnings) {
logger.warn(warning)
}
window.toast?.warning('Some Volcengine models could not be fetched. Check logs for details.')
}
const models: OpenAI.Models.Model[] = response.models.map((model) => ({
id: model.id,
object: 'model' as const,
created: model.created || Math.floor(Date.now() / 1000),
owned_by: 'volcengine',
// @ts-ignore - description is used by UI to display model name
name: model.name || model.id
}))
logger.info(`Found ${models.length} models from Volcengine API`)
return models
} catch (error) {
logger.error('Failed to list Volcengine models:', error as Error)
// Notify user before falling back
window.toast?.warning('Failed to fetch Volcengine models. Check credentials if this persists.')
// Fall back to standard OpenAI-compatible API on error
logger.info('Falling back to OpenAI-compatible model list')
return super.listModels()
}
}
}

View File

@@ -1,6 +1,7 @@
import { loggerService } from '@logger'
import { isZhipuModel } from '@renderer/config/models'
import { getStoreProviders } from '@renderer/hooks/useStore'
import { getDefaultModel } from '@renderer/services/AssistantService'
import type { Chunk } from '@renderer/types/chunk'
import type { CompletionsParams, CompletionsResult } from '../schemas'
@@ -66,7 +67,7 @@ export const ErrorHandlerMiddleware =
}
function handleError(error: any, params: CompletionsParams): any {
if (isZhipuModel(params.assistant.model) && error.status && !params.enableGenerateImage) {
if (isZhipuModel(params.assistant.model || getDefaultModel()) && error.status && !params.enableGenerateImage) {
return handleZhipuError(error)
}

View File

@@ -1,18 +1,21 @@
import type { WebSearchPluginConfig } from '@cherrystudio/ai-core/built-in/plugins'
import { loggerService } from '@logger'
import { isSupportedThinkingTokenQwenModel } from '@renderer/config/models'
import { isSupportEnableThinkingProvider } from '@renderer/config/providers'
import { isGemini3Model, isSupportedThinkingTokenQwenModel } from '@renderer/config/models'
import type { MCPTool } from '@renderer/types'
import { type Assistant, type Message, type Model, type Provider } from '@renderer/types'
import { type Assistant, type Message, type Model, type Provider, SystemProviderIds } from '@renderer/types'
import type { Chunk } from '@renderer/types/chunk'
import { isSupportEnableThinkingProvider } from '@renderer/utils/provider'
import type { LanguageModelMiddleware } from 'ai'
import { extractReasoningMiddleware, simulateStreamingMiddleware } from 'ai'
import { isEmpty } from 'lodash'
import { getAiSdkProviderId } from '../provider/factory'
import { isOpenRouterGeminiGenerateImageModel } from '../utils/image'
import { noThinkMiddleware } from './noThinkMiddleware'
import { openrouterGenerateImageMiddleware } from './openrouterGenerateImageMiddleware'
import { openrouterReasoningMiddleware } from './openrouterReasoningMiddleware'
import { qwenThinkingMiddleware } from './qwenThinkingMiddleware'
import { skipGeminiThoughtSignatureMiddleware } from './skipGeminiThoughtSignatureMiddleware'
import { toolChoiceMiddleware } from './toolChoiceMiddleware'
const logger = loggerService.withContext('AiSdkMiddlewareBuilder')
@@ -217,6 +220,14 @@ function addProviderSpecificMiddlewares(builder: AiSdkMiddlewareBuilder, config:
middleware: noThinkMiddleware()
})
}
if (config.provider.id === SystemProviderIds.openrouter && config.enableReasoning) {
builder.add({
name: 'openrouter-reasoning-redaction',
middleware: openrouterReasoningMiddleware()
})
logger.debug('Added OpenRouter reasoning redaction middleware')
}
}
/**
@@ -248,6 +259,15 @@ function addModelSpecificMiddlewares(builder: AiSdkMiddlewareBuilder, config: Ai
middleware: openrouterGenerateImageMiddleware()
})
}
if (isGemini3Model(config.model)) {
const aiSdkId = getAiSdkProviderId(config.provider)
builder.add({
name: 'skip-gemini3-thought-signature',
middleware: skipGeminiThoughtSignatureMiddleware(aiSdkId)
})
logger.debug('Added skip Gemini3 thought signature middleware')
}
}
/**

View File

@@ -0,0 +1,50 @@
import type { LanguageModelV2StreamPart } from '@ai-sdk/provider'
import type { LanguageModelMiddleware } from 'ai'
/**
* https://openrouter.ai/docs/docs/best-practices/reasoning-tokens#example-preserving-reasoning-blocks-with-openrouter-and-claude
*
* @returns LanguageModelMiddleware - a middleware filter redacted block
*/
export function openrouterReasoningMiddleware(): LanguageModelMiddleware {
const REDACTED_BLOCK = '[REDACTED]'
return {
middlewareVersion: 'v2',
wrapGenerate: async ({ doGenerate }) => {
const { content, ...rest } = await doGenerate()
const modifiedContent = content.map((part) => {
if (part.type === 'reasoning' && part.text.includes(REDACTED_BLOCK)) {
return {
...part,
text: part.text.replace(REDACTED_BLOCK, '')
}
}
return part
})
return { content: modifiedContent, ...rest }
},
wrapStream: async ({ doStream }) => {
const { stream, ...rest } = await doStream()
return {
stream: stream.pipeThrough(
new TransformStream<LanguageModelV2StreamPart, LanguageModelV2StreamPart>({
transform(
chunk: LanguageModelV2StreamPart,
controller: TransformStreamDefaultController<LanguageModelV2StreamPart>
) {
if (chunk.type === 'reasoning-delta' && chunk.delta.includes(REDACTED_BLOCK)) {
controller.enqueue({
...chunk,
delta: chunk.delta.replace(REDACTED_BLOCK, '')
})
} else {
controller.enqueue(chunk)
}
}
})
),
...rest
}
}
}
}

View File

@@ -0,0 +1,36 @@
import type { LanguageModelMiddleware } from 'ai'
/**
* skip Gemini Thought Signature Middleware
* 由于多模型客户端请求的复杂性(可以中途切换其他模型),这里选择通过中间件方式添加跳过所有 Gemini3 思考签名
* Due to the complexity of multi-model client requests (which can switch to other models mid-process),
* it was decided to add a skip for all Gemini3 thinking signatures via middleware.
* @param aiSdkId AI SDK Provider ID
* @returns LanguageModelMiddleware
*/
export function skipGeminiThoughtSignatureMiddleware(aiSdkId: string): LanguageModelMiddleware {
const MAGIC_STRING = 'skip_thought_signature_validator'
return {
middlewareVersion: 'v2',
transformParams: async ({ params }) => {
const transformedParams = { ...params }
// Process messages in prompt
if (transformedParams.prompt && Array.isArray(transformedParams.prompt)) {
transformedParams.prompt = transformedParams.prompt.map((message) => {
if (typeof message.content !== 'string') {
for (const part of message.content) {
const googleOptions = part?.providerOptions?.[aiSdkId]
if (googleOptions?.thoughtSignature) {
googleOptions.thoughtSignature = MAGIC_STRING
}
}
}
return message
})
}
return transformedParams
}
}
}

View File

@@ -0,0 +1,239 @@
import type { Message, Model } from '@renderer/types'
import type { FileMetadata } from '@renderer/types/file'
import { FileTypes } from '@renderer/types/file'
import {
AssistantMessageStatus,
type FileMessageBlock,
type ImageMessageBlock,
MessageBlockStatus,
MessageBlockType,
type ThinkingMessageBlock,
UserMessageStatus
} from '@renderer/types/newMessage'
import { beforeEach, describe, expect, it, vi } from 'vitest'
const { convertFileBlockToFilePartMock, convertFileBlockToTextPartMock } = vi.hoisted(() => ({
convertFileBlockToFilePartMock: vi.fn(),
convertFileBlockToTextPartMock: vi.fn()
}))
vi.mock('../fileProcessor', () => ({
convertFileBlockToFilePart: convertFileBlockToFilePartMock,
convertFileBlockToTextPart: convertFileBlockToTextPartMock
}))
const visionModelIds = new Set(['gpt-4o-mini', 'qwen-image-edit'])
const imageEnhancementModelIds = new Set(['qwen-image-edit'])
vi.mock('@renderer/config/models', () => ({
isVisionModel: (model: Model) => visionModelIds.has(model.id),
isImageEnhancementModel: (model: Model) => imageEnhancementModelIds.has(model.id)
}))
type MockableMessage = Message & {
__mockContent?: string
__mockFileBlocks?: FileMessageBlock[]
__mockImageBlocks?: ImageMessageBlock[]
__mockThinkingBlocks?: ThinkingMessageBlock[]
}
vi.mock('@renderer/utils/messageUtils/find', () => ({
getMainTextContent: (message: Message) => (message as MockableMessage).__mockContent ?? '',
findFileBlocks: (message: Message) => (message as MockableMessage).__mockFileBlocks ?? [],
findImageBlocks: (message: Message) => (message as MockableMessage).__mockImageBlocks ?? [],
findThinkingBlocks: (message: Message) => (message as MockableMessage).__mockThinkingBlocks ?? []
}))
import { convertMessagesToSdkMessages, convertMessageToSdkParam } from '../messageConverter'
let messageCounter = 0
let blockCounter = 0
const createModel = (overrides: Partial<Model> = {}): Model => ({
id: 'gpt-4o-mini',
name: 'GPT-4o mini',
provider: 'openai',
group: 'openai',
...overrides
})
const createMessage = (role: Message['role']): MockableMessage =>
({
id: `message-${++messageCounter}`,
role,
assistantId: 'assistant-1',
topicId: 'topic-1',
createdAt: new Date(2024, 0, 1, 0, 0, messageCounter).toISOString(),
status: role === 'assistant' ? AssistantMessageStatus.SUCCESS : UserMessageStatus.SUCCESS,
blocks: []
}) as MockableMessage
const createFileBlock = (
messageId: string,
overrides: Partial<Omit<FileMessageBlock, 'file' | 'messageId' | 'type'>> & { file?: Partial<FileMetadata> } = {}
): FileMessageBlock => {
const { file, ...blockOverrides } = overrides
const timestamp = new Date(2024, 0, 1, 0, 0, ++blockCounter).toISOString()
return {
id: blockOverrides.id ?? `file-block-${blockCounter}`,
messageId,
type: MessageBlockType.FILE,
createdAt: blockOverrides.createdAt ?? timestamp,
status: blockOverrides.status ?? MessageBlockStatus.SUCCESS,
file: {
id: file?.id ?? `file-${blockCounter}`,
name: file?.name ?? 'document.txt',
origin_name: file?.origin_name ?? 'document.txt',
path: file?.path ?? '/tmp/document.txt',
size: file?.size ?? 1024,
ext: file?.ext ?? '.txt',
type: file?.type ?? FileTypes.TEXT,
created_at: file?.created_at ?? timestamp,
count: file?.count ?? 1,
...file
},
...blockOverrides
}
}
const createImageBlock = (
messageId: string,
overrides: Partial<Omit<ImageMessageBlock, 'type' | 'messageId'>> = {}
): ImageMessageBlock => ({
id: overrides.id ?? `image-block-${++blockCounter}`,
messageId,
type: MessageBlockType.IMAGE,
createdAt: overrides.createdAt ?? new Date(2024, 0, 1, 0, 0, blockCounter).toISOString(),
status: overrides.status ?? MessageBlockStatus.SUCCESS,
url: overrides.url ?? 'https://example.com/image.png',
...overrides
})
describe('messageConverter', () => {
beforeEach(() => {
convertFileBlockToFilePartMock.mockReset()
convertFileBlockToTextPartMock.mockReset()
convertFileBlockToFilePartMock.mockResolvedValue(null)
convertFileBlockToTextPartMock.mockResolvedValue(null)
messageCounter = 0
blockCounter = 0
})
describe('convertMessageToSdkParam', () => {
it('includes text and image parts for user messages on vision models', async () => {
const model = createModel()
const message = createMessage('user')
message.__mockContent = 'Describe this picture'
message.__mockImageBlocks = [createImageBlock(message.id, { url: 'https://example.com/cat.png' })]
const result = await convertMessageToSdkParam(message, true, model)
expect(result).toEqual({
role: 'user',
content: [
{ type: 'text', text: 'Describe this picture' },
{ type: 'image', image: 'https://example.com/cat.png' }
]
})
})
it('returns file instructions as a system message when native uploads succeed', async () => {
const model = createModel()
const message = createMessage('user')
message.__mockContent = 'Summarize the PDF'
message.__mockFileBlocks = [createFileBlock(message.id)]
convertFileBlockToFilePartMock.mockResolvedValueOnce({
type: 'file',
filename: 'document.pdf',
mediaType: 'application/pdf',
data: 'fileid://remote-file'
})
const result = await convertMessageToSdkParam(message, false, model)
expect(result).toEqual([
{
role: 'system',
content: 'fileid://remote-file'
},
{
role: 'user',
content: [{ type: 'text', text: 'Summarize the PDF' }]
}
])
})
})
describe('convertMessagesToSdkMessages', () => {
it('appends assistant images to the final user message for image enhancement models', async () => {
const model = createModel({ id: 'qwen-image-edit', name: 'Qwen Image Edit', provider: 'qwen', group: 'qwen' })
const initialUser = createMessage('user')
initialUser.__mockContent = 'Start editing'
const assistant = createMessage('assistant')
assistant.__mockContent = 'Here is the current preview'
assistant.__mockImageBlocks = [createImageBlock(assistant.id, { url: 'https://example.com/preview.png' })]
const finalUser = createMessage('user')
finalUser.__mockContent = 'Increase the brightness'
const result = await convertMessagesToSdkMessages([initialUser, assistant, finalUser], model)
expect(result).toEqual([
{
role: 'user',
content: [{ type: 'text', text: 'Start editing' }]
},
{
role: 'assistant',
content: [{ type: 'text', text: 'Here is the current preview' }]
},
{
role: 'user',
content: [
{ type: 'text', text: 'Increase the brightness' },
{ type: 'image', image: 'https://example.com/preview.png' }
]
}
])
})
it('preserves preceding system instructions when building enhancement payloads', async () => {
const model = createModel({ id: 'qwen-image-edit', name: 'Qwen Image Edit', provider: 'qwen', group: 'qwen' })
const fileUser = createMessage('user')
fileUser.__mockContent = 'Use this document as inspiration'
fileUser.__mockFileBlocks = [createFileBlock(fileUser.id, { file: { ext: '.pdf', type: FileTypes.DOCUMENT } })]
convertFileBlockToFilePartMock.mockResolvedValueOnce({
type: 'file',
filename: 'reference.pdf',
mediaType: 'application/pdf',
data: 'fileid://reference'
})
const assistant = createMessage('assistant')
assistant.__mockContent = 'Generated previews ready'
assistant.__mockImageBlocks = [createImageBlock(assistant.id, { url: 'https://example.com/reference.png' })]
const finalUser = createMessage('user')
finalUser.__mockContent = 'Apply the edits'
const result = await convertMessagesToSdkMessages([fileUser, assistant, finalUser], model)
expect(result).toEqual([
{ role: 'system', content: 'fileid://reference' },
{ role: 'user', content: [{ type: 'text', text: 'Use this document as inspiration' }] },
{
role: 'assistant',
content: [{ type: 'text', text: 'Generated previews ready' }]
},
{
role: 'user',
content: [
{ type: 'text', text: 'Apply the edits' },
{ type: 'image', image: 'https://example.com/reference.png' }
]
}
])
})
})
})

View File

@@ -0,0 +1,218 @@
import type { Assistant, AssistantSettings, Model, Topic } from '@renderer/types'
import { TopicType } from '@renderer/types'
import { defaultTimeout } from '@shared/config/constant'
import { describe, expect, it, vi } from 'vitest'
import { getTemperature, getTimeout, getTopP } from '../modelParameters'
vi.mock('@renderer/services/AssistantService', () => ({
getAssistantSettings: (assistant: Assistant): AssistantSettings => ({
contextCount: assistant.settings?.contextCount ?? 4096,
temperature: assistant.settings?.temperature ?? 0.7,
enableTemperature: assistant.settings?.enableTemperature ?? true,
topP: assistant.settings?.topP ?? 1,
enableTopP: assistant.settings?.enableTopP ?? false,
enableMaxTokens: assistant.settings?.enableMaxTokens ?? false,
maxTokens: assistant.settings?.maxTokens,
streamOutput: assistant.settings?.streamOutput ?? true,
toolUseMode: assistant.settings?.toolUseMode ?? 'prompt',
defaultModel: assistant.defaultModel,
customParameters: assistant.settings?.customParameters ?? [],
reasoning_effort: assistant.settings?.reasoning_effort,
reasoning_effort_cache: assistant.settings?.reasoning_effort_cache,
qwenThinkMode: assistant.settings?.qwenThinkMode
})
}))
vi.mock('@renderer/hooks/useSettings', () => ({
getStoreSetting: vi.fn(),
useSettings: vi.fn(() => ({})),
useNavbarPosition: vi.fn(() => ({ navbarPosition: 'left', isLeftNavbar: true, isTopNavbar: false }))
}))
vi.mock('@renderer/hooks/useStore', () => ({
getStoreProviders: vi.fn(() => [])
}))
vi.mock('@renderer/store/settings', () => ({
default: (state = { settings: {} }) => state
}))
vi.mock('@renderer/store/assistants', () => ({
default: (state = { assistants: [] }) => state
}))
const createTopic = (assistantId: string): Topic => ({
id: `topic-${assistantId}`,
assistantId,
name: 'topic',
createdAt: new Date().toISOString(),
updatedAt: new Date().toISOString(),
messages: [],
type: TopicType.Chat
})
const createAssistant = (settings: Assistant['settings'] = {}): Assistant => {
const assistantId = 'assistant-1'
return {
id: assistantId,
name: 'Test Assistant',
prompt: 'prompt',
topics: [createTopic(assistantId)],
type: 'assistant',
settings
}
}
const createModel = (overrides: Partial<Model> = {}): Model => ({
id: 'gpt-4o',
provider: 'openai',
name: 'GPT-4o',
group: 'openai',
...overrides
})
describe('modelParameters', () => {
describe('getTemperature', () => {
it('returns undefined when reasoning effort is enabled for Claude models', () => {
const assistant = createAssistant({ reasoning_effort: 'medium' })
const model = createModel({ id: 'claude-opus-4', name: 'Claude Opus 4', provider: 'anthropic', group: 'claude' })
expect(getTemperature(assistant, model)).toBeUndefined()
})
it('returns undefined for models without temperature/topP support', () => {
const assistant = createAssistant({ enableTemperature: true })
const model = createModel({ id: 'qwen-mt-large', name: 'Qwen MT', provider: 'qwen', group: 'qwen' })
expect(getTemperature(assistant, model)).toBeUndefined()
})
it('returns undefined for Claude 4.5 reasoning models when only TopP is enabled', () => {
const assistant = createAssistant({ enableTopP: true, enableTemperature: false })
const model = createModel({
id: 'claude-sonnet-4.5',
name: 'Claude Sonnet 4.5',
provider: 'anthropic',
group: 'claude'
})
expect(getTemperature(assistant, model)).toBeUndefined()
})
it('returns configured temperature when enabled', () => {
const assistant = createAssistant({ enableTemperature: true, temperature: 0.42 })
const model = createModel({ id: 'gpt-4o', provider: 'openai', group: 'openai' })
expect(getTemperature(assistant, model)).toBe(0.42)
})
it('returns undefined when temperature is disabled', () => {
const assistant = createAssistant({ enableTemperature: false, temperature: 0.9 })
const model = createModel({ id: 'gpt-4o', provider: 'openai', group: 'openai' })
expect(getTemperature(assistant, model)).toBeUndefined()
})
it('clamps temperature to max 1.0 for Zhipu models', () => {
const assistant = createAssistant({ enableTemperature: true, temperature: 2.0 })
const model = createModel({ id: 'glm-4-plus', name: 'GLM-4 Plus', provider: 'zhipu', group: 'zhipu' })
expect(getTemperature(assistant, model)).toBe(1.0)
})
it('clamps temperature to max 1.0 for Anthropic models', () => {
const assistant = createAssistant({ enableTemperature: true, temperature: 1.5 })
const model = createModel({
id: 'claude-sonnet-3.5',
name: 'Claude 3.5 Sonnet',
provider: 'anthropic',
group: 'claude'
})
expect(getTemperature(assistant, model)).toBe(1.0)
})
it('clamps temperature to max 1.0 for Moonshot models', () => {
const assistant = createAssistant({ enableTemperature: true, temperature: 2.0 })
const model = createModel({
id: 'moonshot-v1-8k',
name: 'Moonshot v1 8k',
provider: 'moonshot',
group: 'moonshot'
})
expect(getTemperature(assistant, model)).toBe(1.0)
})
it('does not clamp temperature for OpenAI models', () => {
const assistant = createAssistant({ enableTemperature: true, temperature: 2.0 })
const model = createModel({ id: 'gpt-4o', provider: 'openai', group: 'openai' })
expect(getTemperature(assistant, model)).toBe(2.0)
})
it('does not clamp temperature when it is already within limits', () => {
const assistant = createAssistant({ enableTemperature: true, temperature: 0.8 })
const model = createModel({ id: 'glm-4-plus', name: 'GLM-4 Plus', provider: 'zhipu', group: 'zhipu' })
expect(getTemperature(assistant, model)).toBe(0.8)
})
})
describe('getTopP', () => {
it('returns undefined when reasoning effort is enabled for Claude models', () => {
const assistant = createAssistant({ reasoning_effort: 'high' })
const model = createModel({ id: 'claude-opus-4', provider: 'anthropic', group: 'claude' })
expect(getTopP(assistant, model)).toBeUndefined()
})
it('returns undefined for models without TopP support', () => {
const assistant = createAssistant({ enableTopP: true })
const model = createModel({ id: 'qwen-mt-small', name: 'Qwen MT', provider: 'qwen', group: 'qwen' })
expect(getTopP(assistant, model)).toBeUndefined()
})
it('returns undefined for Claude 4.5 reasoning models when temperature is enabled', () => {
const assistant = createAssistant({ enableTemperature: true })
const model = createModel({
id: 'claude-opus-4.5',
name: 'Claude Opus 4.5',
provider: 'anthropic',
group: 'claude'
})
expect(getTopP(assistant, model)).toBeUndefined()
})
it('returns configured TopP when enabled', () => {
const assistant = createAssistant({ enableTopP: true, topP: 0.73 })
const model = createModel({ id: 'gpt-4o', provider: 'openai', group: 'openai' })
expect(getTopP(assistant, model)).toBe(0.73)
})
it('returns undefined when TopP is disabled', () => {
const assistant = createAssistant({ enableTopP: false, topP: 0.5 })
const model = createModel({ id: 'gpt-4o', provider: 'openai', group: 'openai' })
expect(getTopP(assistant, model)).toBeUndefined()
})
})
describe('getTimeout', () => {
it('uses an extended timeout for flex service tier models', () => {
const model = createModel({ id: 'o3-pro', provider: 'openai', group: 'openai' })
expect(getTimeout(model)).toBe(15 * 1000 * 60)
})
it('falls back to the default timeout otherwise', () => {
const model = createModel({ id: 'gpt-4o', provider: 'openai', group: 'openai' })
expect(getTimeout(model)).toBe(defaultTimeout)
})
})
})

View File

@@ -1,13 +1,31 @@
import { isClaude45ReasoningModel } from '@renderer/config/models'
import { isClaude4SeriesModel, isClaude45ReasoningModel } from '@renderer/config/models'
import { getProviderByModel } from '@renderer/services/AssistantService'
import type { Assistant, Model } from '@renderer/types'
import { isToolUseModeFunction } from '@renderer/utils/assistant'
import { isAwsBedrockProvider, isVertexProvider } from '@renderer/utils/provider'
// https://docs.claude.com/en/docs/build-with-claude/extended-thinking#interleaved-thinking
const INTERLEAVED_THINKING_HEADER = 'interleaved-thinking-2025-05-14'
// https://docs.claude.com/en/docs/build-with-claude/context-windows#1m-token-context-window
const CONTEXT_100M_HEADER = 'context-1m-2025-08-07'
// https://docs.cloud.google.com/vertex-ai/generative-ai/docs/partner-models/claude/web-search
const WEBSEARCH_HEADER = 'web-search-2025-03-05'
export function addAnthropicHeaders(assistant: Assistant, model: Model): string[] {
const anthropicHeaders: string[] = []
if (isClaude45ReasoningModel(model) && isToolUseModeFunction(assistant)) {
const provider = getProviderByModel(model)
if (
isClaude45ReasoningModel(model) &&
isToolUseModeFunction(assistant) &&
!(isVertexProvider(provider) && isAwsBedrockProvider(provider))
) {
anthropicHeaders.push(INTERLEAVED_THINKING_HEADER)
}
if (isClaude4SeriesModel(model)) {
if (isVertexProvider(provider) && assistant.enableWebSearch) {
anthropicHeaders.push(WEBSEARCH_HEADER)
}
anthropicHeaders.push(CONTEXT_100M_HEADER)
}
return anthropicHeaders
}

View File

@@ -194,20 +194,20 @@ async function convertMessageToAssistantModelMessage(
* This function processes messages and transforms them into the format required by the SDK.
* It handles special cases for vision models and image enhancement models.
*
* @param messages - Array of messages to convert. Must contain at least 2 messages when using image enhancement models.
* @param messages - Array of messages to convert. Must contain at least 3 messages when using image enhancement models for special handling.
* @param model - The model configuration that determines conversion behavior
*
* @returns A promise that resolves to an array of SDK-compatible model messages
*
* @remarks
* For image enhancement models with 2+ messages:
* - Expects the second-to-last message (index length-2) to be an assistant message containing image blocks
* - Expects the last message (index length-1) to be a user message
* - Extracts images from the assistant message and appends them to the user message content
* - Returns only the last two processed messages [assistantSdkMessage, userSdkMessage]
* For image enhancement models with 3+ messages:
* - Examines the last 2 messages to find an assistant message containing image blocks
* - If found, extracts images from the assistant message and appends them to the last user message content
* - Returns all converted messages (not just the last two) with the images merged into the user message
* - Typical pattern: [system?, assistant(image), user] -> [system?, assistant, user(image)]
*
* For other models:
* - Returns all converted messages in order
* - Returns all converted messages in order without special image handling
*
* The function automatically detects vision model capabilities and adjusts conversion accordingly.
*/
@@ -220,29 +220,25 @@ export async function convertMessagesToSdkMessages(messages: Message[], model: M
sdkMessages.push(...(Array.isArray(sdkMessage) ? sdkMessage : [sdkMessage]))
}
// Special handling for image enhancement models
// Only keep the last two messages and merge images into the user message
// [system?, user, assistant, user]
// Only merge images into the user message
// [system?, assistant(image), user] -> [system?, assistant, user(image)]
if (isImageEnhancementModel(model) && messages.length >= 3) {
const needUpdatedMessages = messages.slice(-2)
const needUpdatedSdkMessages = sdkMessages.slice(-2)
const assistantMessage = needUpdatedMessages.filter((m) => m.role === 'assistant')[0]
const assistantSdkMessage = needUpdatedSdkMessages.filter((m) => m.role === 'assistant')[0]
const userSdkMessage = needUpdatedSdkMessages.filter((m) => m.role === 'user')[0]
const systemSdkMessages = sdkMessages.filter((m) => m.role === 'system')
const imageBlocks = findImageBlocks(assistantMessage)
const imageParts = await convertImageBlockToImagePart(imageBlocks)
const parts: Array<TextPart | ImagePart | FilePart> = []
if (typeof userSdkMessage.content === 'string') {
parts.push({ type: 'text', text: userSdkMessage.content })
parts.push(...imageParts)
userSdkMessage.content = parts
} else {
userSdkMessage.content.push(...imageParts)
const assistantMessage = needUpdatedMessages.find((m) => m.role === 'assistant')
const userSdkMessage = sdkMessages[sdkMessages.length - 1]
if (assistantMessage && userSdkMessage?.role === 'user') {
const imageBlocks = findImageBlocks(assistantMessage)
const imageParts = await convertImageBlockToImagePart(imageBlocks)
if (imageParts.length > 0) {
if (typeof userSdkMessage.content === 'string') {
userSdkMessage.content = [{ type: 'text', text: userSdkMessage.content }, ...imageParts]
} else if (Array.isArray(userSdkMessage.content)) {
userSdkMessage.content.push(...imageParts)
}
}
}
if (systemSdkMessages.length > 0) {
return [systemSdkMessages[0], assistantSdkMessage, userSdkMessage]
}
return [assistantSdkMessage, userSdkMessage]
}
return sdkMessages

View File

@@ -85,19 +85,6 @@ export function supportsLargeFileUpload(model: Model): boolean {
})
}
/**
* 检查模型是否支持TopP
*/
export function supportsTopP(model: Model): boolean {
const provider = getProviderByModel(model)
if (provider?.type === 'anthropic' || model?.endpoint_type === 'anthropic') {
return false
}
return true
}
/**
* 获取提供商特定的文件大小限制
*/

Some files were not shown because too many files have changed in this diff Show More