phsl / new-api

- Add IP field to Log model with database index and default empty value
- Implement conditional IP recording based on user setting in RecordConsumeLog and RecordErrorLog
- Add UserSettingRecordIpLog constant and update user settings API to handle record_ip_log field
- Create dedicated "IP记录" tab in personal settings under "其他设置" section
- Add IP column to logs table with help tooltip explaining recording conditions
- Make IP column visible to all users (not admin-only) with proper filtering for consume/error log types
- Restrict display of use_time and retry columns to consume and error log types only
- Update personal settings UI structure: rename "通知设置" to "其他设置" to accommodate new functionality
- Add proper translation support and maintain consistent styling across components

The IP logging feature is disabled by default and only records client IP addresses
for consume (type 2) and error (type 5) logs when explicitly enabled by users
in their personal settings.

feat: 分组特殊倍率

🔧 refactor(LogsTable, render): remove undefined parameters for improved clarity and consistency in function signatures

Changes
1. web/src/helpers/token.js
   • `fetchTokenKeys` now calls `/api/token/?p=1&size=10` (1-based paging).
   • Supports new response shape `{ items, total, page, page_size }`; falls back gracefully if array is returned.
   • Filters active tokens from `tokenItems`, not `data` directly.

`useTokenKeys` remains unchanged—its consumer code receives the same list of active keys.

Fix the search button loading state to be consistent with other table components.
The search button now properly shows loading animation when the table data is
being fetched.

Changes:
- Update search button loading prop from `loading={searching}` to
  `loading={loading || searching}` in TokensTable.js
- This ensures loading state is shown both when searching with keywords
  (searching=true) and when loading default data (loading=true)
- Aligns with the behavior of other table components like ChannelsTable,
  UsersTable, and RedemptionsTable

Before: Search button only showed loading when searching with keywords
After: Search button shows loading for all table data fetch operations

- Replace `showTotal` with `formatPageText` in Dashboard table components
- Unify pagination text format to match table components pattern
- Update SettingsAnnouncements.js, SettingsAPIInfo.js, and SettingsFAQ.js
- Change from "共 X 条记录，显示第 Y-Z 条" to "第 Y - Z 条，共 X 条" format
- Ensure consistent user experience across all table components

This change improves UI consistency by standardizing the pagination
text format across Dashboard and table components.

SUMMARY
• Migrated Token, Task, Midjourney, Channel, Redemption tables to true server-side pagination.
• Added total / page / page_size metadata in API responses; switched all affected React tables to consume new structure.
• Implemented counting helpers:
  – model/token.go CountUserTokens
  – model/task.go TaskCountAllTasks / TaskCountAllUserTask
  – model/midjourney.go CountAllTasks / CountAllUserTask
  – model/channel.go CountAllChannels / CountAllTags
• Refactored controllers (token, task, midjourney, channel) for 1-based paging & aggregated returns.
• Redesigned `ChannelsTable.js`:
  – `loadChannels`, `syncPageData`, `enrichChannels` for tag-mode grouping without recursion.
  – Fixed runtime white-screen (maximum call-stack) by removing child duplication.
  – Pagination, search, tag-mode, idSort all hot-reload correctly.
• Removed unused `log` import in controller/midjourney.go.

BREAKING CHANGES
Front-end consumers must now expect data.items / total / page / page_size from list endpoints (`/api/channel`, `/api/task`, `/api/mj`, `/api/token`, etc.).

🎨 style(dashboard): Optimize the layout of the Uptime card legend on the dashboard to resolve the issue where the last monitoring item is obscured

Previously, the uptime status endpoint returned HTTP 400 with
“未配置 Uptime Kuma URL/Slug” when either option was not set, resulting in
frontend error states.

Changes:
• Treat absence of `UptimeKumaUrl` or `UptimeKumaSlug` as a valid scenario.
• Immediately respond with HTTP 200, `success: true`, and an empty `data` array.
• Preserve existing behavior when both options are provided.

This prevents unnecessary error notifications on the dashboard when
Uptime Kuma integration is not configured and improves overall UX.

* Removed `size="middle"` and `centered` props from the column-selector
  `Modal` in `ChannelsTable.js` to match the visual style used in
  `LogsTable`.
* Re-added `size="middle"` to the main `Table` component to preserve the
  original table sizing.
* Ensures consistent UI/UX across both channel and log column settings
  modals.

feat: add column visibility settings for channels

fix playground-sse

Summary
• Centralized uptime status definition via `uptimeStatusMap`, containing color / label / text for each status.
• Generated `uptimeLegendData`, `getUptimeStatusColor`, `getUptimeStatusText` directly from the map, removing multiple switch-case blocks.

UI Improvements
1. Added statuses 2 (High Latency) & 3 (Maintenance) with dedicated colors.
2. Relocated status legend to a styled footer wrapped in a borderless sub-Card; header now only shows title + refresh button.
3. Footer (and its negative margin) renders only when `uptimeData` is present, preventing empty legend display.
4. Applied rounded, blurred badge style and always-on shadow to legend container for clearer separation.

Maintenance
• Simplified code paths, reduced duplication, and improved readability without breaking existing functionality.

Introduce application uptime monitoring to improve observability and reliability.

• Add UptimeService to track process start time and expose uptime in seconds
• Create /health/uptime endpoint returning the current uptime in JSON format
• Integrate uptime metric into existing health-check middleware
• Update README with instructions for consuming the new endpoint
• Add unit tests covering UptimeService and new health route

This change enables operations teams and dashboards to programmatically
determine how long the service has been running, facilitating automated
alerts and trend analysis.

- Modified returnUrl configuration in RequestEpay function
- Changed payment success redirect path to match updated frontend routing
- Updated controller/topup.go line 116 to use correct callback path

- Add SettingsAnnouncements component with full CRUD operations for system announcements
  * Support multiple announcement types (default, ongoing, success, warning, error)
  * Include publish date, content, type classification and additional notes
  * Implement batch operations and pagination for better data management
  * Add real-time preview with relative time display and date formatting

- Add SettingsFAQ component for comprehensive FAQ management
  * Support question-answer pairs with rich text content
  * Include full editing, deletion and creation capabilities
  * Implement batch delete operations and paginated display
  * Add validation for complete Q&A information

- Integrate announcement and FAQ modules into DashboardSetting
  * Add unified configuration interface in admin console
  * Implement auto-refresh functionality for real-time updates
  * Add loading states and error handling for better UX

- Enhance backend API support in controller and setting modules
  * Add validation functions for console settings
  * Include time and sorting utilities for announcement management
  * Extend API endpoints for announcement and FAQ data persistence

- Improve frontend infrastructure
  * Add new translation keys for internationalization support
  * Update utility functions for date/time formatting
  * Enhance CSS styles for better component presentation
  * Add icons and visual improvements for announcements and FAQ sections

This implementation provides administrators with comprehensive tools to manage
system-wide announcements and user FAQ content through an intuitive console interface.

- Replace IconSearch with Server icon for API info card title to better represent server/API related content
- Add Server imports from lucide-react

This change improves the semantic meaning of icons and provides better visual representation of their respective functionalities.

- Refactor api_info.go to console.go for broader console settings support
- Update URL regex pattern to accept both domain names and IP addresses
- Add support for IPv4 addresses with optional port numbers
- Improve validation to handle formats like http://192.168.1.1:8080
- Add ValidateConsoleSettings function for extensible settings validation
- Maintain backward compatibility with existing ValidateApiInfo function
- Add comprehensive comments explaining supported URL formats

Fixes issue where IP-based URLs were incorrectly rejected as invalid format.
Prepares infrastructure for additional console settings validation.

- Update URL regex pattern to accept both domain names and IP addresses
- Add support for IPv4 addresses with optional port numbers
- Improve validation to handle formats like http://192.168.1.1:8080
- Add comprehensive comments explaining supported URL formats
- Maintain backward compatibility with existing domain-based URLs

Fixes issue where IP-based URLs were incorrectly rejected as invalid format.

Add request deduplication mechanism to prevent duplicate GET requests
to the same endpoint within the same timeframe, significantly reducing
unnecessary network overhead.

**Changes:**
- Add `patchAPIInstance()` function to intercept and deduplicate GET requests
- Implement in-flight request tracking using Map with URL+params as unique keys
- Apply deduplication patch to both initial API instance and `updateAPI()` recreated instances
- Add `disableDuplicate: true` config option to bypass deduplication when needed

**Benefits:**
- Eliminates redundant API calls caused by component re-renders or rapid user interactions
- Reduces server load and improves application performance
- Provides automatic protection against accidental duplicate requests
- Maintains backward compatibility with existing code

**Technical Details:**
- Uses Promise sharing for identical concurrent requests
- Automatically cleans up completed requests from tracking map
- Preserves original axios functionality with minimal overhead
- Zero breaking changes to existing API usage

Addresses the issue observed in EditChannel.js where multiple calls
were made to the same endpoints during component lifecycle.