BREAKING CHANGE: Refactor message stream processing and think tag handling logic
- Improve automatic collapse behavior of reasoning panel
- Collapse panel immediately after reasoning content completion
- Add detection for complete think tag pairs
- Remove dependency on full content completion
- Enhance think tag processing
- Unify tag handling logic across stream and stop states
- Add robust handling for unclosed think tags
- Prevent markdown rendering errors from malformed tags
- Simplify state management
- Reduce complexity in collapse condition checks
- Remove redundant state transitions
- Improve code readability by removing unnecessary comments
This commit ensures a more stable and predictable behavior in the chat interface, particularly in handling streaming responses and reasoning content display.
- Modified the `renderCustomChatContent` function to strip `<think>` and `</think>` tags from `finalExtractedThinkingContent` before passing it to the `MarkdownRender` component. This resolves potential parsing errors caused by unclosed or malformed tags.
- Added a conditional check to ensure that the thinking process block is only rendered if `finalExtractedThinkingContent` has actual content and the section is expanded, improving rendering stability.
This commit significantly refactors the handling of <think> tags within the Playground component to improve robustness, prevent Markdown rendering errors, and ensure accurate separation of reasoning and displayable content.
Key changes include:
- Modified `handleNonStreamRequest`, `onStopGenerator`, and `renderCustomChatContent` for more resilient <think> tag parsing.
- Implemented logic to correctly extract content from fully paired `<think>...</think>` tags.
- Added handling for unclosed `<think>` tags, particularly relevant during streaming responses, to capture ongoing thought processes.
- Ensured `reasoningContent` is accurately populated from all sources (API, paired tags, and unclosed streaming tags).
- Thoroughly sanitized the final `currentDisplayableFinalContent` to remove all residual `<think>` and `</think>` string instances, preventing issues with the Markdown renderer.
- Corrected the usage of `thinkTagRegex.lastIndex = 0;` to ensure proper regex state resetting before each use in loops.
Additionally, numerous explanatory comments detailing the "how" of the code (rather than the "why") have been removed to improve code readability and reduce noise.
This commit improves the handling of message generation interruption in the
playground chat interface, ensuring both content and thinking process are
properly preserved.
Key changes:
- Preserve existing content when stopping message generation
- Handle both direct reasoningContent and <think> tag formats
- Extract and merge thinking process from unclosed <think> tags
- Maintain consistent thinking chain format with separators
- Auto-collapse reasoning panel after stopping for cleaner UI
Technical details:
- Modified onStopGenerator to properly handle SSE connection closure
- Added regex pattern to extract thinking content from <think> tags
- Implemented proper state management for incomplete messages
- Ensured all content types are preserved in their respective fields
This fix resolves the issue where thinking chain content would be lost
when stopping message generation mid-stream.
- Add click-to-navigate functionality on balance cards in both Detail and TokensTable components
- Implement navigation to '/console/topup' when clicking on current balance
- Add useNavigate hook and necessary imports
- Keep consistent navigation behavior across components
Components affected:
- web/src/pages/Detail/index.js
- web/src/components/TokensTable.js
This commit introduces the following changes:
- Add i18n support to the Loading component
- Import useTranslation hook from react-i18next
- Replace hardcoded Chinese text with translation keys
- Support dynamic content interpolation for loading prompts
- Use {{name}} variable in translation template
Technical details:
- Added: import { useTranslation } from 'react-i18next'
- Modified: Loading text from static Chinese to i18n keys
- Translation keys added:
- "加载中..."
- "加载{{name}}中..."
File changed: web/src/components/Loading.js
This commit introduces the following changes:
- Create a reusable NoticeModal component to handle system announcements
- Extract notice functionality from Home and HeaderBar components
- Add loading and empty states using Semi UI illustrations
- Implement "close for today" feature with localStorage
- Support both light and dark mode for empty state illustrations
- Add proper error handling and loading states
- Improve code reusability and maintainability
Breaking changes: None
Related components: HeaderBar.js, Home/index.js, NoticeModal.js
- Remove unnecessary undefined status check in readystatechange event
- Only show disconnection message for actual non-200 status codes
- Remove redundant else branch for normal status handling
- Prevent false "Connection lost" messages on successful completion
- Add comprehensive error handling for SSE events
- Implement proper error state UI with Semi Typography
- Prevent white screen issues on non-200 responses
- Add error logging for better debugging
- Enhance message state management for error conditions
- Improve user feedback with i18n error messages
- Ensure UI stability during error states
- Add try-catch blocks for JSON parsing and stream initialization
- Handle connection termination gracefully
- Preserve error states in message stream updates
- Replace emoji icons with Semi Design Avatar components
- Standardize statistics cards layout across TokensTable and LogsTable
- Remove informational text banners for cleaner interface
- Implement consistent grid layout for metric cards
- Add elevated shadow effect to white Avatar for better contrast
Changes include:
- Convert emoji indicators to Semi Icons (MoneyExchangeStroked, Pulse, Typograph)
- Unify card styling with rounded corners and hover shadows
- Implement responsive grid layout (1 column on mobile, 3 columns on desktop)
- Standardize typography and spacing across all stat cards
- Apply Semi Design's elevated shadow to improve white Avatar visibility
