phsl / new-api

Move the Turnstile verification component from the renderOAuthOptions method to the main render function in both LoginForm and RegisterForm components. This ensures the Turnstile verification is globally visible and accessible regardless of which authentication method the user chooses (email login/register or third-party OAuth options).

The change improves UI consistency and ensures the verification mechanism works properly across all authentication flows.

Replace the main card icon in ModelPricing component with IconLayers to better represent the model catalog functionality. Increase icon size from 'large' to 'extra-large' for improved visibility and visual hierarchy.

This change enhances the user interface by using more appropriate iconography and giving the primary navigation elements greater emphasis.

This commit improves the model display interface with several enhancements:

1. Add model count badges to each category tab and dropdown menu item
2. Highlight active category with red badge and use grey for inactive ones
3. Optimize performance by caching category counts with useMemo
4. Standardize model tag rendering across components:
   - Replace direct Tag component with centralized renderModelTag function
   - Update renderModelTag to use stringToColor for consistent coloring
   - Remove redundant color calculations in LogsTable

These changes improve the UI by providing users with visual cues about model distribution across categories while ensuring consistent styling throughout the application.

feat: support X-goog-api-key；remove [Done] in stream mode

Refactor model tag rendering to ensure consistency throughout the application:

- Replace direct Tag component in ModelPricing with centralized renderModelTag function
- Update renderModelTag in render.js to use stringToColor for consistent color generation
- Remove redundant stringToColor calls in LogsTable.js renderModelName function

This change improves UI consistency by ensuring all model tags have the same styling, iconography, and color generation logic. Model tags now automatically display appropriate vendor icons based on the model name pattern.

Fix JSON unmarshal error that occurred when copying channels after testing. The JavaScript timestamp (floating point number) in the test_time field was causing type conversion errors in Go backend which expected an int64.

Solution:
- Create deep copy of channel record instead of modifying original
- Remove test_time and response_time fields before sending to backend
- Allow backend to use default values for these fields

Error fixed: "json: cannot unmarshal number into Go struct field Channel.test_time of type int64"

Improve the logs table by implementing brand-specific model icons and better
redirection visualization:

- Replace standard tags with `renderModelTag` to display appropriate brand
  icons for each model (OpenAI, Claude, Gemini, etc.)
- Fix background colors by explicitly passing color parameters
- Restore descriptive text for model redirection in popover
- Replace refresh icon with forward icon for better representation of model
  redirection
- Clean up unused imports

This change provides a more intuitive visual representation of models and
their relationships, making the logs table easier to understand at a glance.

This commit removes all fixed width constraints from the model pricing table columns, allowing them to naturally expand and adjust based on content. This improves the table's responsiveness and ensures better space utilization across different screen sizes.

✨ feat(ui): add the attribute `mode="password"` to the password input on the Setup page to enable password visibility toggle

This commit refactors the login and registration forms to enhance user experience by conditionally displaying OAuth-related UI elements.

- In `LoginForm.js` and `RegisterForm.js`:
    - The "Other login/registration options" button and the "or" divider are now only displayed if at least one OAuth provider is enabled in the system settings.
    - This change ensures a cleaner interface when no OAuth options are configured, preventing user confusion.
- In `RegisterForm.js`:
    - Changed `div` class from `relative` to `min-h-screen relative` to ensure the registration form an take up the entire screen height.

- Refactored system initialization page using TailwindCSS 3 and SemiUI components
- Changed layout from step navigation to single page display for all configurations
- Modified top area from vertical to more compact horizontal layout
- Updated gradient color scheme from blue/purple to orange/pink
- Fixed form field name duplication issues and optimized Form implementation
- Changed usage mode selection from three-column grid to vertical layout
- Replaced usage mode card icons from settings to more appropriate Layers icon
- Added specific prompts for different database types (SQLite/MySQL/PostgreSQL)
- Removed configuration summary section while keeping the initialization button
- Fixed useSetupCheck issue by using SetupCheck as a direct component for proper redirection

- Integrated Docker Buildx setup step in docker-image-alpha.yml to enhance multi-platform build capabilities.

- Added support for multiple platforms (linux/amd64, linux/arm64) in docker-image-alpha.yml.
- Removed outdated docker-image-amd64.yml and docker-image-arm64.yml workflows.
- Deleted linux-release.yml, macos-release.yml, and windows-release.yml as part of workflow cleanup.

Fix ReferenceError caused by undefined `isValidMessage` and `MESSAGE_ROLES`
in the buildApiPayload function within api.js.

Changes:
- Add missing `isValidMessage` import from utils.js
- Add missing `MESSAGE_ROLES` import from playground constants
- Consolidate duplicate `formatMessageForAPI` import
- Clean up import statements organization

Resolves: ReferenceError: isValidMessage is not defined at buildApiPayload (api.js:39:13)

- Rename files to follow camelCase naming convention:
  • auth-header.js → authUtils.js
  • other.js → logUtils.js
  • rehypeSplitWordsIntoSpans.js → textAnimationUtils.js

- Update import paths in affected components:
  • Update exports in helpers/index.js
  • Fix import in LogsTable.js for logUtils
  • Fix import in MarkdownRenderer.js for textAnimationUtils

- Remove old files after successful migration

- Improve file naming clarity:
  • authUtils.js better describes authentication utilities
  • logUtils.js clearly indicates log processing functions
  • textAnimationUtils.js concisely describes text animation functionality

This refactoring enhances code maintainability and follows consistent naming patterns throughout the helpers directory.

Gemini 格式

- Remove automatic insertion of "你好" placeholder message in preview payload
- Keep messages array empty when no user messages exist in conversation
- Only process image handling logic when user messages are present
- Ensure preview request body accurately reflects current conversation state

Previously, the preview panel would automatically inject a default "你好"
user message when the conversation was empty, which could be misleading.
This change ensures the preview payload shows exactly what would be sent
based on the current conversation state, improving accuracy and user
understanding of the actual API request structure.

- Add clear context button positioned on the left side of input area
- Create symmetric layout with clear button (left) and send button (right)
- Standardize both buttons to 32x32px size for consistent appearance
- Apply distinct styling: gray background for clear (red on hover), purple for send
- Add smooth transition animations for better user experience
- Align buttons vertically centered for improved visual balance

The clear conversation button provides quick access to context clearing
functionality directly from the input area, matching the design patterns
shown in Semi Design documentation and improving overall user workflow.

- Add onMessageReset reference comparison to OptimizedMessageActions memo
- Force component re-render when model selection changes
- Prevent stale closure issue in retry functionality
- Ensure first retry attempt uses newly selected model

Previously, when changing the model selection, the retry button would
still use the previous model due to React memo optimization preventing
re-renders. By comparing the onMessageReset callback reference, the
component now properly updates when the model changes, ensuring the
retry functionality immediately uses the currently selected model.

- Store editing message object reference using useRef to avoid ID conflicts
- Use object reference comparison first before falling back to ID matching
- Apply fix to both save and cancel operations in message editing
- Clear reference after edit completion to prevent stale object issues

Previously, when editing imported messages that contained duplicate IDs,
the findIndex operation would match the first occurrence rather than the
intended message, causing conversation history truncation when saving
edits. This change stores and uses object references for accurate message
identification during the editing process.

- Use object reference comparison first before falling back to ID matching
- Prevent incorrect message index lookup when duplicate IDs exist
- Apply fix to both handleMessageReset and handleMessageDelete functions
- Maintain backward compatibility with ID-based message identification

Previously, when importing messages that contained duplicate IDs, the
findIndex operation would match the first occurrence rather than the
intended message, causing history truncation on retry. This change uses
object reference comparison as the primary method, ensuring accurate
message identification and preserving conversation history.

- Extend handleMessageReset condition to include 'system' role messages
- Allow system messages to trigger regeneration like assistant messages
- Fix disabled retry button issue when message role is switched to system
- Maintain consistent user experience across different message roles

Previously, when an assistant message was switched to system role,
the retry button became non-functional. This change ensures that
system messages can be regenerated by finding the previous user
message and resending it, maintaining feature parity with assistant
messages.