Compare commits

...

180 Commits

Author SHA1 Message Date
admin de471d0fa0 refactor: move SidebarTrigger to header left, flatten system settings into sidebar
Docker Build / Build and Push Docker Image (push) Successful in 3m41s
2026-06-15 04:17:37 +08:00
admin ea5b152313 fix: TryUserAuth set role in context, fix docs page pagination response handling
Docker Build / Build and Push Docker Image (push) Successful in 3m48s
2026-06-15 04:07:15 +08:00
admin 75e608bdc8 refactor: sidebar full-height with logo+toggle at top, header beside sidebar
Docker Build / Build and Push Docker Image (push) Successful in 3m51s
2026-06-15 04:02:05 +08:00
admin c52ff411a7 fix: move SidebarTrigger to sidebar edge, move Dialog/AlertDialog outside SectionPageLayout
Docker Build / Build and Push Docker Image (push) Successful in 3m29s
2026-06-15 03:57:06 +08:00
admin ff407c4607 fix: stagger meteor delays for full-width appearance, move SidebarTrigger to AppHeader after logo
Docker Build / Build and Push Docker Image (push) Successful in 3m34s
2026-06-15 03:48:41 +08:00
admin ae380a5a81 fix: meteor trail full-width from start, fix dialog not rendering in doc management, move sidebar trigger after logo
Docker Build / Build and Push Docker Image (push) Successful in 3m25s
2026-06-15 03:15:21 +08:00
admin 71519301e4 fix: improve meteor effects and fix document management add functionality
Docker Build / Build and Push Docker Image (push) Successful in 3m45s
2026-06-15 03:01:39 +08:00
admin 497ed7c39a chore: empty commit to trigger CI
Docker Build / Build and Push Docker Image (push) Successful in 3m35s
2026-06-15 02:03:45 +08:00
admin 8be5131c29 style: update pricing page header to match docs page style with subtitle
Docker Build / Build and Push Docker Image (push) Successful in 3m30s
2026-06-15 01:56:39 +08:00
admin 6a9fd7611a fix: close overflow-hidden div properly in docs page JSX
Docker Build / Build and Push Docker Image (push) Successful in 4m26s
2026-06-15 01:31:24 +08:00
admin c42757f9af feat: add nebula clouds and meteor effects to pricing and docs pages, unify header style with homepage
Docker Build / Build and Push Docker Image (push) Failing after 1m37s
2026-06-15 01:26:15 +08:00
admin c450214d78 style: improve meteor effects with glowing head, gradient trail, and smoother animation
Docker Build / Build and Push Docker Image (push) Successful in 4m24s
2026-06-15 01:16:29 +08:00
admin 73e701e49a ci: trigger rebuild
Docker Build / Build and Push Docker Image (push) Successful in 4m33s
2026-06-15 01:02:37 +08:00
admin 537c292801 feat: add shooting star meteors and drifting nebula cloud effects to homepage
Docker Build / Build and Push Docker Image (push) Failing after 4m36s
2026-06-15 00:55:36 +08:00
admin b330253fe2 fix: add scrollbar-gutter:stable to prevent nav shift on page switch
Docker Build / Build and Push Docker Image (push) Successful in 4m36s
2026-06-15 00:51:23 +08:00
admin d703ccbb79 fix: unify PublicLayout usage with showMainContainer={false} across all pages to prevent nav shift
Docker Build / Build and Push Docker Image (push) Successful in 4m25s
2026-06-15 00:38:36 +08:00
admin 697ef3fcca fix: unify page container max-w and padding across all pages
Docker Build / Build and Push Docker Image (push) Successful in 4m18s
- Home CTA buttons: hover now fills with primary color for clear feedback
- Pricing/Docs/About: change px-4 to px-6 to match home page
- Wallet: change max-w-7xl to max-w-6xl to match nav
- Footer: change max-w-5xl to max-w-6xl for consistency
2026-06-15 00:05:12 +08:00
admin 1249eebf46 style: simplify pricing page header, update home CTA buttons to match filter chip style
Docker Build / Build and Push Docker Image (push) Successful in 4m33s
2026-06-14 23:07:13 +08:00
admin a34c284017 fix: adjust docs page layout to match other pages (max-w-6xl, pt-20, centered title)
Docker Build / Build and Push Docker Image (push) Successful in 4m43s
2026-06-14 21:50:22 +08:00
admin 3b2c0a0471 fix: resolve Gin route conflict between /docs/:slug and /docs/admin, move admin docs API to /admin/docs
Docker Build / Build and Push Docker Image (push) Successful in 4m33s
2026-06-14 21:40:48 +08:00
admin 207f98252d fix: docs nav always links to /docs, add admin GET routes for doc management, fix data parsing
Docker Build / Build and Push Docker Image (push) Failing after 4m1s
2026-06-14 21:32:55 +08:00
admin 113b9c8ecb fix: unify light/dark primary color, fix doc management API, remove compliance restriction
Docker Build / Build and Push Docker Image (push) Successful in 4m23s
2026-06-14 21:20:47 +08:00
admin a4c069f88d feat: redesign pricing page - move filters/search to top, remove sidebar, match homepage layout
Docker Build / Build and Push Docker Image (push) Successful in 4m21s
2026-06-14 20:51:14 +08:00
admin 7caf77db63 feat: remove Get Started arrow, redesign pricing table with Model/Group/Billing/PerRequest/Input/CachedInput/Output columns, default to table view
Docker Build / Build and Push Docker Image (push) Successful in 4m13s
2026-06-14 20:42:06 +08:00
admin 6faf989549 fix: remove header inner padding px-0
Docker Build / Build and Push Docker Image (push) Successful in 4m18s
2026-06-14 20:35:15 +08:00
admin 719d06ecd6 fix: consistent header height h-14, reduce side padding px-4
Docker Build / Build and Push Docker Image (push) Successful in 4m16s
2026-06-14 20:28:58 +08:00
admin 5464d14973 fix: header full-width, remove floating style, use border-bottom instead
Docker Build / Build and Push Docker Image (push) Successful in 4m17s
2026-06-14 20:21:50 +08:00
admin 446e1ce10b fix: move provider marquee after code demo section, center align
Docker Build / Build and Push Docker Image (push) Successful in 4m24s
2026-06-14 20:15:15 +08:00
admin cd8cdebdcb fix: move providers marquee between sections, restore footer, add notification to mobile
Docker Build / Build and Push Docker Image (push) Successful in 4m32s
2026-06-14 20:09:03 +08:00
admin 790d832756 fix: remove Codex from provider list, it's part of OpenAI/GPT
Docker Build / Build and Push Docker Image (push) Successful in 4m21s
2026-06-14 20:01:14 +08:00
admin 63c83d5abd fix: update docs_link description to mention built-in docs page
Docker Build / Build and Push Docker Image (push) Successful in 4m16s
2026-06-14 19:55:24 +08:00
admin 64326a86d5 refactor: homepage - remove demo section, adapt for token relay station, move providers to bottom
Docker Build / Build and Push Docker Image (push) Successful in 4m17s
2026-06-14 19:52:31 +08:00
admin 5c4ed6206e feat: redesign homepage, add docs page, admin doc/category management
Docker Build / Build and Push Docker Image (push) Successful in 4m15s
2026-06-14 19:34:55 +08:00
admin 43f9869246 fix: Termius light mode colors, remove social proof/features/CTA sections
Docker Build / Build and Push Docker Image (push) Successful in 4m27s
2026-06-14 18:53:29 +08:00
admin 8eb31a7c82 redesign: homepage layout referencing company project structure
Docker Build / Build and Push Docker Image (push) Successful in 4m19s
2026-06-14 18:44:58 +08:00
admin ca47fd18f5 fix: light mode primary to dark, cyan as accent via oklch classes
Docker Build / Build and Push Docker Image (push) Successful in 4m19s
2026-06-14 18:33:13 +08:00
admin 296120641a fix: light mode Termius theme + redesign homepage
Docker Build / Build and Push Docker Image (push) Successful in 4m10s
2026-06-14 18:24:49 +08:00
admin 5d18485e9a fix: resolve JSX quote mismatch and missing profile-dropdown import
Docker Build / Build and Push Docker Image (push) Successful in 4m25s
2026-06-14 18:14:36 +08:00
admin 60fc1eee31 fix: simplify home barrel export and use semantic tokens
Docker Build / Build and Push Docker Image (push) Failing after 1m39s
2026-06-14 18:08:41 +08:00
admin da8cf3eef0 fix: proper light/dark theme support for all public pages
Docker Build / Build and Push Docker Image (push) Failing after 1m31s
2026-06-14 17:42:31 +08:00
admin 1ddffd236c fix: force dark mode on auth/home, add grid texture and glow to homepage
Docker Build / Build and Push Docker Image (push) Successful in 4m3s
2026-06-14 17:27:06 +08:00
admin 5eeb3c9f18 fix: auth button hover, nav links, minimal homepage redesign
Docker Build / Build and Push Docker Image (push) Successful in 4m18s
2026-06-14 17:19:03 +08:00
admin 2d8bdc1b7c fix: auth buttons, dropdown ring, redesign homepage
Docker Build / Build and Push Docker Image (push) Successful in 4m33s
2026-06-14 16:49:59 +08:00
admin 2723bd66ad style: revert to cyan accent, fix dark hover button priority
Docker Build / Build and Push Docker Image (push) Successful in 4m15s
2026-06-14 16:13:53 +08:00
admin 2c4c2002c6 fix: light mode header button hover background
Docker Build / Build and Push Docker Image (push) Successful in 4m8s
2026-06-14 15:45:13 +08:00
admin 8cf49db0a8 fix: button hover visibility, card ring, homepage with code features
Docker Build / Build and Push Docker Image (push) Successful in 4m3s
2026-06-14 15:37:56 +08:00
admin 53a884a6fe design: Termius-dark theme - black bg, green accent, terminal style
Docker Build / Build and Push Docker Image (push) Successful in 4m6s
2026-06-14 15:23:40 +08:00
admin 60d63f87ff fix: light mode button colors, redesign homepage with features grid
Docker Build / Build and Push Docker Image (push) Successful in 4m13s
2026-06-14 15:05:00 +08:00
admin 603886d422 fix: correct primary colors, redesign hero, fix header text colors
Docker Build / Build and Push Docker Image (push) Successful in 4m7s
2026-06-14 14:40:08 +08:00
admin 3c056b00b1 design: apply Termius theme across all pages (light+dark)
Docker Build / Build and Push Docker Image (push) Successful in 4m8s
2026-06-14 13:53:31 +08:00
admin e9d408e718 design: redesign homepage strictly following Termius dark theme style
Docker Build / Build and Push Docker Image (push) Successful in 4m21s
2026-06-14 12:34:50 +08:00
admin c175a41a2b design: rewrite homepage with proper light/dark mode using Tailwind semantic tokens
Docker Build / Build and Push Docker Image (push) Successful in 4m23s
2026-06-14 12:20:39 +08:00
admin 4e335a9997 design: redesign default frontend homepage with Termius-dark style (black/blue/green)
Docker Build / Build and Push Docker Image (push) Successful in 4m23s
2026-06-14 11:51:56 +08:00
admin 6f1f8b60f1 design: redesign homepage with tech-minimalist style, remove blur-ball and shine-text
Docker Build / Build and Push Docker Image (push) Successful in 4m27s
2026-06-14 11:28:31 +08:00
admin a318279d5f feat: simplify SSH deploy step, update docker-compose to use own registry
Docker Build / Build and Push Docker Image (push) Failing after 3m56s
2026-06-14 11:22:50 +08:00
admin d1542a65ac feat: add SSH deploy step to CI/CD workflow
Docker Build / Build and Push Docker Image (push) Successful in 3m58s
2026-06-14 11:16:51 +08:00
admin 26ac8b5dc1 refactor: remove mt theme, clean up New API branding from classic frontend
Docker Build / Build and Push Docker Image (push) Successful in 3m54s
2026-06-14 11:06:47 +08:00
admin a8a96a7e60 redesign mt homepage with clean data-driven style, remove footer branding
Docker Build / Build and Push Docker Image (push) Successful in 4m16s
2026-06-14 00:10:06 +08:00
admin 45f676514a update favicon and logo for all themes
Docker Build / Build and Push Docker Image (push) Successful in 4m5s
2026-06-13 21:55:50 +08:00
admin 9e4ab6e7bb refactor: rename daisy frontend to mt, copy from classic
Docker Build / Build and Push Docker Image (push) Successful in 4m30s
2026-06-13 21:08:17 +08:00
admin 45aea22451 fix: resolve Gin route conflict between :slug and :id by moving doc admin routes to /docs/admin
Docker Build / Build and Push Docker Image (push) Successful in 4m36s
2026-06-13 12:29:21 +08:00
admin 5b9aa9e77a fix: specify varchar(255) for slug columns to fix MySQL unique index error
Docker Build / Build and Push Docker Image (push) Successful in 4m11s
2026-06-13 12:16:39 +08:00
admin 80bf73b41c fix: add missing logs pages and @hookform/resolvers dependency for Docker build
Docker Build / Build and Push Docker Image (push) Successful in 4m5s
2026-06-13 10:11:09 +08:00
admin e83ec743c8 feat: add DaisyUI frontend theme and document management system
Docker Build / Build and Push Docker Image (push) Failing after 1m35s
2026-06-13 01:36:06 +08:00
admin 824c6d9133 feat: add Tencent Cloud Coding Plan support and handle coding plan URLs in OpenAI adaptor
Docker Build / Build and Push Docker Image (push) Failing after 2m43s
2026-06-12 17:40:53 +08:00
admin 76f8112ee7 chore: remove payment compliance check system entirely
Docker Build / Build and Push Docker Image (push) Failing after 2m36s
2026-06-12 11:30:08 +08:00
admin 8afd1d1247 chore: bypass payment compliance check
Docker Build / Build and Push Docker Image (push) Successful in 3m35s
2026-06-12 10:12:33 +08:00
admin b4b58ee887 fix: re-apply brand rename with correct UTF-8 encoding (fixes Docker build)
Docker Build / Build and Push Docker Image (push) Successful in 3m50s
2026-06-12 04:57:18 +08:00
admin 97ddfeab59 chore: rebrand copyright from QuantumNous to modelstoken
Docker Build / Build and Push Docker Image (push) Failing after 1m2s
2026-06-12 01:05:46 +08:00
admin 6db071543c ci: add Docker cleanup step to prevent disk space exhaustion
Docker Build / Build and Push Docker Image (push) Successful in 4m41s
2026-06-12 00:12:00 +08:00
admin d83db15801 ci: fix docker-build workflow output and version handling
Docker Build / Build and Push Docker Image (push) Successful in 2m59s
2026-06-12 00:02:17 +08:00
admin 2780f71b8f ci: add main/master branch trigger to docker-build workflow
Docker Build / Build and Push Docker Image (push) Failing after 4m42s
2026-06-11 23:45:52 +08:00
admin 052da457eb ci: add Gitea Actions workflows 2026-06-11 23:22:08 +08:00
QuentinHsu 6f415428d3 perf(web): improve frontend table rendering and pinned columns/UI table (#5405)
* refactor(web): centralize data table implementation

- route all TanStack table setup through a shared data-table hook to remove repeated state and row model wiring.
- move table rendering, static table wrappers, empty states, and primitive exports behind the data-table module.
- update feature tables and configuration editors to share the same table UX while preserving their existing workflows.

* refactor(web): trim data table public API

- remove unused data-table exports and dead static table helper types.
- keep internal table header, skeleton, empty state, and faceted filter helpers private to the data-table module.
- route feature imports through the data-table barrel to avoid subpath coupling.

* refactor(web): unify table rendering components

- centralize static table headers, bodies, empty states, and shared class names behind the data-table package.
- migrate settings, pricing, channel, key, subscription, and model tables to the shared table APIs.
- remove data-table exports for low-level table primitives so feature code uses one supported abstraction.

* perf(web): keep list tables fixed within page content

- make shared data table pages fill available height and scroll row data inside the table body.
- add a fixed content layout mode so selected list pages avoid page-level scrolling.
- apply the fixed table behavior to keys, logs, channels, models, users, redemptions, and subscriptions.

* perf(web): refine table pagination controls

- show total row counts instead of redundant page range text.
- tighten visible page buttons so pagination fits constrained table widths.
- align pagination controls and tune text hierarchy for clearer scanning.

* perf(web): stabilize model pricing table columns

- keep model pricing columns at fixed widths so headers do not collapse in narrow layouts.
- truncate long model names and pricing summaries within their cells instead of squeezing adjacent columns.

* refactor(web): simplify data table rendering internals

- split table body rendering into focused helpers for loading, empty, and row states.
- extract static table row and cell class resolution to reduce branching in the main component.
- reuse a single pagination page-size option list to avoid duplicated constants.

* perf(pricing): reduce dynamic pricing table render work

- reuse dynamic pricing field metadata instead of rebuilding it inside table columns.
- precompute formatted dynamic prices per tier and group to avoid repeated entry mapping for each cell.
- simplify select option construction in related dialogs while preserving the same choices.

* refactor(web): streamline pricing table rendering

- reuse translated endpoint select options between trigger data and menu items.
- precompute dynamic pricing maps per group so table cells only resolve formatted values.
- add local dynamic pricing type aliases to keep helper signatures readable.

* refactor(web): merge pricing table imports

* refactor(web): merge upstream ratio table imports

* refactor(web): merge channel selector table imports

* refactor(web): simplify tiered pricing select items

* refactor(web): reuse model ratio row state

* refactor(web): rely on table view row defaults

* refactor(web): reuse pagination state values

* refactor(web): hoist pagination size select items

* refactor(web): clarify static table body rows

* refactor(web): extract table page pagination rendering

* fix(web): remove direct hast type dependency

- rely on Shiki transformer contextual typing for line nodes.
- allow frontend typecheck to pass without an undeclared hast package.

* refactor(web): trim data table hook return API

- return only the TanStack table instance from useDataTable.
- keep internal state handling private because callers do not consume it directly.

* refactor(web): keep static table empty row private

- stop exporting the internal StaticDataTableEmptyRow helper.
- keep the public static table API focused on the table component and column type.

* refactor(web): hide data table view props from barrel

* refactor(web): remove stale long text lint override

* fix(web): keep pinned table columns opaque

- apply pinned column background classes after custom column classes.
- use an opaque hover background so scrolled content cannot show through fixed cells.

* refactor(data-table): organize shared table components

- group table primitives, page composition, toolbar controls, static tables, and hooks by responsibility.
- split shared view types, row rendering, header rendering, and pinned-column styling out of the main table view.
- keep the public data-table barrel stable while documenting the new ownership boundaries.

* fix(web): stabilize split table column sizing

- derive default colgroup widths from visible columns when split headers or header sizing are enabled.
- apply a fixed table layout with computed minimum width so header and body columns stay aligned.
- keep split-header containers from leaking horizontal overflow and avoid extra pinned-column borders.

* fix(web): set stable table utility column widths

- assign fixed widths to selection columns so shared colgroup sizing keeps checkbox cells compact.
- size id columns in redemption and user tables to keep split headers aligned with body rows.

* fix(web): align model metadata icon cells

- render compact provider avatars in the metadata icon column instead of wide wordmarks.
- position icons in a fixed-size wrapper so they line up with the existing icon header alignment.

* fix(status-badge): hide status dot by default

* fix(web): prevent user invite info overlap

- give the invite info and created-at columns explicit widths so table sizing reserves enough space.
- allow invite badges to wrap within the cell instead of spilling into adjacent columns.

* perf(data-table): cache pinned column class resolution

- reuse the pinned column lookup while table props stay stable to reduce repeated per-render work.
- share the resolved column class handler across unified and split-header table layouts.
- localize page-number screen reader labels so pagination remains accessible in every locale.

* refactor(data-table): tighten static table modes

- make StaticDataTable distinguish data-driven and children-only usage through explicit prop shapes.
- remove unsupported columns-without-data fallback after confirming no repository callers rely on it.
- default manual table modes away from unused local row models to reduce repeated table work.

* fix(data-table): make pinned edit column opaque

- use an opaque muted background for the active action column so sticky cells do not reveal scrolled content underneath.

* fix(data-table): prevent narrow column overlap

- apply stable header sizing to remaining desktop data table pages so constrained layouts scroll instead of compressing cells.
- add explicit widths for key, quota, badge, and timestamp columns that contain fixed-format content.
- constrain masked values and timestamp cells with truncation to keep content inside its assigned column.

* fix(table): align table cell content with headers

- remove extra inline padding from masked table text buttons so values start at the cell edge.
- tag status badges and offset leading badges inside table cells to match header text alignment.

* fix(table): prevent admin list column overflow

- widen redemption and subscription table columns so masked codes, timestamps, and localized headers fit.
- localize subscription ID headers and add Received amount translations across supported locales.

* fix(provider-badge): unify provider icon spacing

- add a shared provider badge component for icon and status label layout.
- reuse it in channel type and model vendor columns so OpenAI icons align consistently.
2026-06-11 02:36:41 +08:00
CaIon 59a93cf5c7 fix(openai): align image streaming relay governance
Route OpenAI image streaming through shared stream handling, split image/realtime/usage helpers for maintainability, and include the related image request and rate limit updates.
2026-06-10 17:47:37 +08:00
Benson Yan 867d8acfc3 fix: normalize kimi k2.6 temperature (#5390) 2026-06-10 17:19:57 +08:00
Q.A.zh 30d3a3a5f7 perf(web): add debounce channel search and skip during IME composition (#5393) 2026-06-10 17:18:51 +08:00
gaoren002 d2576ddcd3 fix(openai): support streaming image relay and image edit for images API (#4608)
* fix(openai): support streaming image relay

* fix(openai): keep image edit multipart body reusable

* test(openai): cover image stream usage details

* test(openai): cover image edit fallback stream field

* fix(openai): wrap image json fallback as stream

* fix(relay): support OpenAI image streaming

* fix(openai): record image stream upstream error events

* fix(openai): harden image stream relay

* fix(openai): return image JSON errors

* fix(relay): reset stream status per scanner run

* fix(relay): drop upstream credit passthrough

* fix(openai): keep image errors minimal

* fix(openai): keep image error status from response

---------

Co-authored-by: CaIon <i@caion.me>
2026-06-08 18:36:17 +08:00
同語 4ca47ee236 fix: support six-decimal steps in model pricing editor (#5332)
Merge pull request #5332 from yyhhyyyyyy/fix/model-pricing-six-decimal-step
2026-06-06 23:22:37 +08:00
同語 16dd7237c0 fix: align mobile usage log cost badge (#5161)
Merge pull request #5161 from yyhhyyyyyy/fix/mobile-usage-log-cost-alignment
2026-06-06 23:19:07 +08:00
同語 1915344838 fix: respect theme for multiselect combobox popover (#5328)
Merge pull request #5328 from yyhhyyyyyy/fix/multiselect-popover-theme
2026-06-06 23:18:04 +08:00
同語 15ff8e0268 chore(web): improve frontend dialog layout and sizing (#5346)
Merge pull request #5346 from QuantumNous/perf/ui-dialog
2026-06-06 23:16:53 +08:00
同語 a1c82841b5 chore(web): simplify public page hero copy (#5339)
Merge pull request #5339 from QuantumNous/perf/compact-display
2026-06-06 23:15:05 +08:00
同語 1e6f31b235 perf(model-pricing): improve model pricing editor UX (#5275)
Merge pull request #5275 from QuantumNous/fix/model-pricing-draft-save
2026-06-06 23:14:18 +08:00
QuentinHsu 2eaa943d9f perf(web): improve dialog sizing and footer layout
- migrate frontend dialogs to the shared footer API so actions stay separated from scrollable body content.
- tune dialog dimensions for model analytics, prefill groups, billing history, channel model sync, and related workflows.
- update channel terminology and dialog action translations across supported locales.
2026-06-06 21:49:33 +08:00
QuentinHsu 7a5348caa3 feat(web): add shared dialog wrapper
- introduce a reusable dialog component for consistent header, body, and footer layout.
- support per-dialog sizing, trigger rendering, initial focus, and close button controls.
- preserve base dialog open and close motion classes while allowing content-specific styling.
2026-06-06 18:47:10 +08:00
QuentinHsu f5753a2b31 perf(web): simplify public page hero copy 2026-06-06 15:49:38 +08:00
QuentinHsu 4dd68bad52 perf(model-pricing): move pricing tabs into page title
- place the model pricing tab switcher beside the page title instead of spanning the content area.
- keep the switcher width tied to its labels while preserving spacing around title status content.
2026-06-06 15:26:53 +08:00
QuentinHsu 0f043ae404 feat(json-editor): add reusable JSON code editor
- introduce a shared themed JSON editor with line numbers, formatting, status feedback, and keyboard editing helpers.
- use the shared editor in model pricing JSON mode so pricing maps get consistent editor behavior.
- localize structured JSON validation messages so parse errors avoid browser-specific English text.
2026-06-06 15:14:26 +08:00
QuentinHsu 75c05bb4b8 perf(model-pricing): improve JSON pricing editor layout
- render pricing JSON fields from shared configuration to reduce duplicated form markup.
- use fixed-height JSON textareas so long model maps scroll internally instead of stretching the page.
- arrange JSON editors in responsive columns to make wider settings pages easier to scan.
2026-06-06 14:36:21 +08:00
QuentinHsu 81d3dc08e5 perf(model-pricing): reduce duplicate model name display 2026-06-06 14:15:44 +08:00
同語 adc390c5fb feat(web): show user id on profile page (#5317)
Merge pull request #5317 from P2K0/feat/profile-show-user-id
2026-06-06 00:45:13 +08:00
yyhhyyyyyy e8c36762fd fix: support six-decimal steps in model pricing editor 2026-06-05 17:24:33 +08:00
yyhhyyyyyy e2dbd02cbb Merge remote-tracking branch 'upstream/main' into fix/mobile-usage-log-cost-alignment
# Conflicts:
#	web/default/src/features/usage-logs/components/usage-logs-mobile-card.tsx
2026-06-05 14:11:55 +08:00
yyhhyyyyyy c8d3768087 fix: respect theme for multiselect combobox popover 2026-06-05 14:02:26 +08:00
xujiantop-crypto 32805849d6 fix: reuse stream scanner buffer in channel handlers (#5225) 2026-06-05 12:18:57 +08:00
Don Ganesh 01c2128e23 fix: 收窄 OpenAI o 系列模型适配范围 (#5293)
* fix: 收窄 OpenAI o 系列模型适配范围

* fix(openai): 限制 gpt-5 适配仅作用于 OpenAI 模型

* fix(openai): narrow o-series reasoning model detection

---------

Co-authored-by: Seefs <i@seefs.me>
2026-06-05 12:12:45 +08:00
QuentinHsu 189913b7a0 fix(i18n): clarify thinking adapter copy (#5242)
- update the global thinking blacklist label to describe skipped suffix processing instead of disabled model thinking.
- rename Claude and Gemini adapter labels to thinking suffix adapter and sync all default locales.
- revise Claude helper text to clarify suffix request adaptation while keeping billing predictable.
2026-06-05 11:54:57 +08:00
Seefs d2f7f9ee3a fix: limit anonymous request body (#5244)
* fix: limit anonymous request body (env ANONYMOUS_REQUEST_BODY_LIMIT_KB = 512)

* fix: allow disabling anonymous request body limit
2026-06-05 11:39:29 +08:00
Chen011214 83068d115e fix(relay): fix Anthropic-compatible compatibility for GLM (avoid chunked encoding) (#5307) 2026-06-05 11:31:20 +08:00
XiaoDingSiRen 4a188deeaa feat: 支持配置渠道被禁用后是否清空渠道粘性 (#5306)
* fix: evict stale channel affinity

* feat: configure disabled channel affinity retention

---------

Co-authored-by: Seefs <i@seefs.me>
2026-06-05 11:30:29 +08:00
Seefs 933ea0cddc fix: add relay idle connection timeout config (#5309) 2026-06-05 11:30:08 +08:00
P2K0 b53319361f feat(web): show user id on profile page 2026-06-05 07:37:02 +08:00
QuentinHsu 5681c92b3f perf(model-pricing): refine visual editor actions
- keep the global reset action in the top toolbar while moving visual-mode saves into the model editor footer.
- pin the actions header with the rest of the model table headers so horizontal scrolling keeps context visible.
- add action icons to make save and reset controls easier to scan.
2026-06-05 01:04:47 +08:00
QuentinHsu 6e5a359110 refactor(model-pricing): split visual pricing editor modules
- extract pricing form primitives, snapshot helpers, and table column setup to keep the editor components smaller.
- remove draft comparison UI now that switching models discards unsaved edits.
- refine the model list with a fixed actions column and tighter mode and price summary display.
2026-06-05 00:06:41 +08:00
Rain 87cc22d7ec fix(distributor): resolve model for GET /v1/video/generations/:task_id (#5133) 2026-06-04 18:48:30 +08:00
Rain 3aa113b5a3 fix(dify): initialize file pointer before remote-image field assignment (#5134) 2026-06-04 18:21:35 +08:00
QuentinHsu 77d3157592 fix(model-pricing): commit visual pricing drafts on save
- Commit the open visual editor draft before saving model pricing settings
- Show unsaved draft differences against persisted model pricing values
- Move model pricing actions into the editor toolbar and refine the visual editor layout
2026-06-04 17:22:50 +08:00
同語 00d23abf64 fix: 修复余额显示时只切换了单位未切换数值 #5296
Merge pull request #5296 from feitianbubu/pr/27fe9a3a82f51bac2b7645213e3b1480cb7f14f2
2026-06-04 02:55:23 +08:00
feitianbubu 580ad97c02 fix: convert usd amount by exchange rate in classic quota display 2026-06-03 22:23:12 +08:00
QuentinHsu 39e05118ff fix(model-pricing): align pricing mode editor spacing
- add consistent tab and field spacing so each pricing mode keeps the same visual rhythm.
- wrap per-request and tiered sections in shared field groups to match the per-token form structure.
- keep fixed-price descriptions and validation messages aligned with the updated field layout.
2026-06-03 18:27:40 +08:00
QuentinHsu 9e59ffc3d8 fix(model-pricing): align pricing mode editor spacing
- add consistent tab and field spacing so each pricing mode keeps the same visual rhythm.
- wrap per-request and tiered sections in shared field groups to match the per-token form structure.
- keep fixed-price descriptions and validation messages aligned with the updated field layout.
2026-06-03 18:27:07 +08:00
QuentinHsu abad0d3cc0 fix(model-pricing): detect visual pricing draft changes on save
- expose a draft commit handle from the model pricing editor panel before saving.
- commit the open visual editor into the parent form before page-level save runs.
- support both desktop side editor and mobile sheet save paths.
2026-06-03 14:49:08 +08:00
t0ng7u b0ac0429cf fix(web): resolve TypeScript errors in usage logs mobile card
Cast row.original to Record<string, unknown> before accessing created_at and type in CommonLogsCard, matching the pattern used elsewhere in the same component.

Close: #5243
2026-06-03 12:37:36 +08:00
Seefs d17b566bcc docs: refine issue templates (#5271) 2026-06-03 12:04:40 +08:00
同語 7aaa533265 fix(channels): reveal advanced validation errors #5239
Merge pull request #5239 from QuantumNous/fix/channel-advanced-errors
2026-06-02 14:30:20 +08:00
t0ng7u 7791b78429 chore(fd): delete the test file 2026-06-02 14:28:35 +08:00
QuentinHsu cb5c0453f5 fix(channels): avoid expanding advanced settings for model mapping
- remove model mapping from advanced settings error detection so visible model configuration errors do not expand the advanced panel.
- update the edit-time advanced settings auto-expand check to only depend on fields actually rendered in the advanced section.
- add regression coverage to verify model_mapping errors are not classified as advanced settings errors.
2026-06-02 12:31:32 +08:00
QuentinHsu 4d20e053cb fix(channels): reveal advanced validation errors
- add channel form error detection for JSON validation errors inside advanced settings.
- expand advanced settings on invalid channel drawer submit and prompt users to fix highlighted fields.
- add regression tests for advanced error detection, non-advanced exclusions, and schema error classification.
2026-06-02 12:09:47 +08:00
同語 0ff9c35e62 feat(web): support classic Rsbuild dev and build
Merge pull request #5232 from QuantumNous/feat/classic-rsbuild-dev-workflow
2026-06-02 11:33:33 +08:00
QuentinHsu 0bbcaa8999 fix(classic): inject Semi React 19 adapter 2026-06-02 00:50:29 +08:00
QuentinHsu 1e9ff8a0de feat(web): support classic Rsbuild dev and build
- migrate the classic frontend from Vite to Rsbuild with JSX, Semi UI, proxy, and production build config.
- update make dev-web to run both default and classic frontends for local theme switching.
- fix classic public page height, footer, CORS proxy, error handling, and constant export warnings.
- update Dockerfile and release workflow to install from the web workspace root with the shared lockfile.
2026-06-02 00:32:16 +08:00
同語 9a2e60dff2 chore(web): centralize shared frontend dependency versions #5227
Merge pull request #5227 from QuantumNous/chore/web-shared-dependency-catalog
2026-06-01 19:19:13 +08:00
QuentinHsu b596de739d chore(web): centralize shared frontend dependency versions
- add a web workspace catalog to manage dependency versions shared by default and classic frontends.
- switch shared dependencies including @lobehub/icons to catalog references and align @lobehub/icons on 5.10.0.
- replace separate frontend Bun lockfiles with a unified web/bun.lock to reduce duplicate maintenance.
2026-06-01 19:12:39 +08:00
同語 45d54c1613 fix(pricing): sync custom model icons #5224
Merge pull request #5224 from QuantumNous/fix/pricing-model-icons
2026-06-01 18:17:58 +08:00
QuentinHsu 086044650d fix(pricing): sync custom model icons
- add the icon field to the pricing model type to consume model-level icons returned by the backend.
- prefer model icons in cards, table model cells, and detail headers while falling back to vendor icons.
2026-06-01 17:58:02 +08:00
GGXH 0c7aceb831 feat: add claude opus 4.8 support (#5177) 2026-05-31 13:50:52 +08:00
dependabot[bot] b2e25b7df2 chore(deps): bump axios from 1.15.2 to 1.16.0 in /web/classic (#5185)
Bumps [axios](https://github.com/axios/axios) from 1.15.2 to 1.16.0.
- [Release notes](https://github.com/axios/axios/releases)
- [Changelog](https://github.com/axios/axios/blob/v1.x/CHANGELOG.md)
- [Commits](https://github.com/axios/axios/compare/v1.15.2...v1.16.0)

---
updated-dependencies:
- dependency-name: axios
  dependency-version: 1.16.0
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-05-31 13:49:50 +08:00
skynono 230a3592f8 perf: order admin logs by created_at to use composite index (#5116) 2026-05-30 20:00:02 +08:00
YuPeng Wu afb470e405 fix(model): correct idx_created_at_id index column order to (created_at, id) (#5191)
The idx_created_at_id composite index on the logs table was defined as
(id, created_at) because the GORM `priority` values on Id and CreatedAt were
swapped. Since `id` is the auto-increment primary key, a secondary composite
index leading with `id` is redundant with the PK and cannot accelerate
`created_at` range scans (a range column must sit at the index prefix).

This defeats the common log-listing queries
(`WHERE created_at BETWEEN ? AND ? ORDER BY id DESC LIMIT n` in
GetAllLogs/GetUserLogs) that the index name implies it should serve — the
optimizer falls back to scanning the primary key, degrading to near full-table
scans on large logs tables.

Swap the priorities so the column order becomes (created_at, id), matching the
index name and its intended purpose. idx_user_id_id and idx_created_at_type are
unaffected.

Note: GORM AutoMigrate does not change the column order of an already-existing
index with the same name, so existing deployments must rebuild the index
manually (see PR description for per-database DDL).

Co-authored-by: wuyupeng <wuyupeng@floatmiracle.com>
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-05-30 19:54:02 +08:00
CaIon 1588027084 feat: add subscription balance redemption toggle (#3071) 2026-05-29 12:54:00 +08:00
zengwei 38bf2d8daa feat(keys/cc-switch-dialog): 修复自定义cc-switch名称失焦后重置问题 (#5170) 2026-05-29 12:18:52 +08:00
同語 e8c836d705 fix(web): improve form validation error focus #5163
Merge pull request #5163 from QuantumNous/fix/form-validation-focus
2026-05-28 23:34:02 +08:00
yyhhyyyyyy 979aeceb5c fix: align mobile usage log cost badge 2026-05-28 19:17:47 +08:00
QuentinHsu e79cee1e9e perf(form): focus first validation error on submit
- scope validation queries with a form root id so feedback stays inside the submitted form.
- scroll to the earliest invalid control or message and move focus without fighting scroll position.
- avoid handling the same failed submit twice to keep retries from jumping unexpectedly.
2026-05-28 15:10:17 +08:00
QuentinHsu 63ead2bf7f chore(repo): ignore playwright mcp artifacts 2026-05-28 15:02:00 +08:00
CaIon 5b86ce0d70 fix: optimize batch update process 2026-05-27 13:23:05 +08:00
CaIon 74985fa877 fix: keep token log filters exact 2026-05-26 21:17:25 +08:00
yyhhyyyyyy 1d32037364 fix: keep usage log filters exact unless wildcard is explicit (#5097) 2026-05-26 21:00:32 +08:00
CaIon dc245ae764 fix(web): improve channel and usage log UI
Fixes #5121
2026-05-26 20:28:28 +08:00
CaIon f8add4ca49 feat(theme): add simple-large preset, xl scale and clean up channel badge dots
Implement the Simple Large-font theme preset and xl font scale options to enhance interface accessibility. Remove status indicator dots from channel badges in logs to keep the table layout visual and clean.
2026-05-26 18:35:51 +08:00
t0ng7u 65f8afe922 🐛 fix(system-settings): resolve save detection and number input NaN issues
System settings forms that used flat dotted API keys (e.g.
`performance_setting.monitor_cpu_threshold`) with React Hook Form were
broken: RHF stores dotted paths as nested objects on update, while dirty
checks and submit comparisons still read flat keys from defaults. Users
could edit values but always saw "No changes to save".

Refactor affected sections to use nested Zod schemas and default values
for RHF, with explicit helpers to convert between nested form state and
flat API keys. Track a normalized baseline in refs for accurate change
detection and post-save resets.

Add `safeNumberFieldProps` to prevent native `<input type="number">`
from writing NaN into form state when cleared. NaN caused Zod validation
to fail silently and made the save button appear unresponsive. The
helper ignores non-finite updates so controlled inputs snap back to the
last valid value, matching legacy Semi InputNumber behavior.

Sections refactored for dotted-key handling:
- maintenance/performance-section
- models/grok-settings-card
- auth/passkey-section
- auth/oauth-section
- auth/section-registry (pass attachment_preference raw; normalize in section)

Sections migrated to safeNumberFieldProps:
- maintenance/performance-section
- models/grok-settings-card
- integrations/monitoring-settings-section
- integrations/payment-settings-section
- integrations/creem-product-dialog
- general/pricing-section (USD exchange rate)
- general/system-behavior-section
- content/dashboard-section

Optional numeric fields (e.g. custom currency exchange rate) keep their
existing empty-to-undefined semantics and are intentionally unchanged.
2026-05-26 15:43:56 +08:00
CaIon 5bc4c74813 🎨 fix(logs): tune usage table typography
Set usage log tables to a balanced 13px default and keep log badges aligned with the active theme font.
2026-05-26 12:41:00 +08:00
Seefs 30025aeba3 fix: use actual user id for channel tests (#5109) 2026-05-26 12:32:20 +08:00
Seefs c91ba0c4eb fix: consolidate Waffo payment settings save flow (#5110) 2026-05-26 12:32:05 +08:00
CaIon f223db9330 🎨 fix(charts): improve dark mode chart readability
Ensure VChart labels and grid lines use theme-aware colors, and remove oversized rounded corners from ranking bar charts.
2026-05-26 12:30:13 +08:00
CaIon 9e283ab10b 🎨 fix(logs): remove hardcoded font-mono to support global theme font inheritance
- Remove explicit 'font-mono' and custom size classes from model and token
  badges in usage logs.
- Allow model name and token badges to naturally inherit the active theme's
  font family (Sans or Serif) and text size from the parent container.
- Restore visual consistency and proportion across all table badge components.
2026-05-26 12:23:52 +08:00
CaIon a8b7c92e5f 🎨 fix(logs): restore timing background badges and optimize model/token spacing
- Re-introduce the custom translucent background color and thin border scheme
  for timing and duration badges in common, drawing, and task logs.
- Remove strict max-width constraints on model badges to ensure complete
  names (with version suffixes) are always visible and wrap gracefully.
- Adjust spacing on model and token badges (h-6 height, larger gaps, and
  proper padding) to prevent crowded elements and restore a balanced,
  high-quality look in the log tables.
2026-05-26 12:03:43 +08:00
CaIon 6b6c9904ac feat(subscription): support balance purchases
Refs #3071.
2026-05-26 12:03:02 +08:00
CaIon 1011934987 🎨 fix(theme): default theme font preset falls back to Sans instead of Serif
Adjust PRESET_DEFAULT_FONT so that the shipped 'default' preset falls back to the humanist 'sans' (Public Sans) out-of-the-box instead of forcing the editorial 'serif' (Lora). Keeps the 'anthropic' preset on 'serif' as intended.
2026-05-26 11:29:38 +08:00
CaIon bc8110ce36 🎨 refactor(badge): restore status-badge sizes and classic color scheme
- Restore StatusBadge sizes to h-5/text-xs (sm, md) and h-6/text-xs (lg).
- Restore classic textColorMap coloring and status indicator dot.
- Embed channel type icon directly inside StatusBadge as custom children.
- Re-align status badge colors: danger for manual disabled, warning for auto disabled.
2026-05-26 11:23:18 +08:00
yyhhyyyyyy ad224ecf5b fix: prevent duplicate channel action toasts (#5015)
* fix: prevent duplicate channel action toasts

* fix: localize api error fallbacks
2026-05-26 10:20:54 +08:00
t0ng7u a64f26d1d2 🎨 feat(web/default): add Anthropic theme preset and configurable serif typography
Introduce a switchable Anthropic-inspired color preset and a new Font customization axis so users can adopt the editorial serif look across the entire UI, including sidebar navigation, tabs, form controls, buttons, and table headers.

Theme preset

Add anthropic to the theme preset registry with warm cream canvas, slate foreground, and clay/coral accent tokens for light and dark modes
Define explicit surface colors for the Anthropic preset instead of relying on the semantic surface bridge
Exclude anthropic from the primary-color surface bridge so bespoke warm neutrals are not overridden by accent-tinted mixes
Typography system

Add @fontsource-variable/lora and a global --font-serif token with CJK serif fallbacks (Noto Serif SC, Source Han Serif, Songti SC, etc.)
Introduce a --font-body token and drive <body> font-family from it
Add a Font axis (default | sans | serif) parallel to radius/scale
Resolve font: 'default' against preset defaults (anthropic → serif)
Persist font preference via cookie and apply data-theme-font on <body>
Apply serif OpenType features (kern, liga, calt, tnum) and heading display tuning when serif is active
Remove per-component sans opt-outs so serif inherits through sidebar, tabs, inputs, buttons, badges, and table headers via natural CSS cascade
Keep monospace contexts unchanged via Tailwind preflight and .font-mono
UI and i18n

Add Font selector to the theme config drawer (Auto / Sans / Serif)
Add "Font" and "Select body font" translations for en, zh, fr, ja, ru, vi
Misc

Tighten group and status badge sizing for better balance with serif text
2026-05-26 04:31:13 +08:00
t0ng7u 3360882642 ♻️ refactor(channels): rebuild channel editor UX with modular sections and Base UI multi-select
Restructure the default-theme channel create/edit experience to match classic
frontend behavior, improve form UX, and align with the project's Base UI design
system.

Channel editor architecture:
- Split the monolithic channel mutate drawer into focused section components
  (basic, API access, auth, models, advanced) with shared drawer layout
  primitives
- Extract submission, toast handling, and react-query cache invalidation into
  `useChannelMutateForm`
- Add a dedicated loading skeleton for channel detail fetch during edit mode
- Remove the top-level configuration summary block

Form validation and data handling:
- Strengthen `channel-form` Zod schema with JSON, model mapping, status code
  mapping, Codex credential, and Vertex AI key refinements
- Move type-specific conditional validation into `superRefine`
- Normalize base URL formatting and tighten model mapping value validation

Model mapping editor:
- Add Visual/JSON tabbed editing with inline JSON and duplicate-key feedback
- Improve accessibility for icon-only actions and add model suggestion datalists

MultiSelect component:
- Replace the custom cmdk-based implementation with Base UI Combobox chips
- Align focus, border, ring, disabled, and invalid states with standard Input
  styling via `ComboboxChips`
- Preserve existing API for all current callers (`options`, `selected`,
  `onChange`, `allowCreate`, `createLabel`)
- Support inline custom value creation and comma/newline batch input
- Limit visible chips with a compact "+N more" overflow summary via
  `maxVisibleChips` (8 in the channel editor)
- Anchor the dropdown to the full chips container via `useComboboxAnchor` so
  the popup matches input width and long model names are no longer truncated

Models & groups UX:
- Integrate manual custom model entry directly into the model MultiSelect
- Remove the separate manual model input/button block
- Keep selected-model count badge and existing model-mapping guardrail behavior

i18n:
- Add and sync translation keys for section descriptions, validation messages,
  model mapping UI, and MultiSelect labels across en, zh, fr, ja, ru, and vi
- Fix missing translations for "Name, provider type, and availability.",
  "Endpoint, provider-specific settings, and credentials.", and "Published
  models, groups, and model remapping rules."
- Remove obsolete keys tied to the deprecated summary and manual model entry UI
2026-05-26 01:55:27 +08:00
t0ng7u b37b6d80b3 Merge remote-tracking branch 'origin/main' 2026-05-26 01:22:56 +08:00
t0ng7u 3d850d38b6 ♻️ refactor(channels): rebuild channel create/edit drawer with modular sections and improved form UX
Restructure the default-theme channel create/edit experience to align with
classic frontend behavior, modern form UX patterns, and the project's Base UI
design system.

Channel editor architecture:
- Split the monolithic channel mutate drawer into focused section components
  (basic, API access, auth, models, advanced) with shared drawer layout
  primitives
- Extract submission, toast handling, and react-query cache invalidation into
  `useChannelMutateForm`
- Add a dedicated loading skeleton for channel detail fetch during edit mode
- Remove the top-level configuration summary block per UX feedback

Form validation and data handling:
- Strengthen `channel-form` Zod schema with JSON, model mapping, status code
  mapping, Codex credential, and Vertex AI key refinements
- Move type-specific conditional validation into `superRefine`
- Normalize base URL formatting and tighten model mapping value validation

Model mapping editor:
- Add Visual/JSON tabbed editing with inline JSON and duplicate-key feedback
- Improve accessibility for icon-only actions and add model suggestion datalists

MultiSelect component:
- Replace the custom cmdk-based implementation with Base UI Combobox chips
- Align focus, border, ring, disabled, and invalid states with standard Input
  styling via `ComboboxChips`
- Preserve existing API (`options`, `selected`, `onChange`, `allowCreate`,
  `createLabel`) for all current callers
- Support inline custom value creation, comma/newline batch input, searchable
  options, portal-based dropdown positioning, and chip removal

Models & groups UX:
- Integrate manual custom model entry directly into the model MultiSelect
- Remove the separate manual model input/button block
- Keep selected-model count and existing model-mapping guardrail behavior

i18n:
- Add and sync translation keys for new editor sections, validation messages,
  model mapping UI, and MultiSelect empty/create labels across en, zh, fr, ja,
  ru, and vi
- Remove obsolete keys tied to the deprecated summary and manual model entry UI

Affected areas:
- `web/default/src/features/channels/components/drawers/`
- `web/default/src/features/channels/hooks/use-channel-mutate-form.ts`
- `web/default/src/features/channels/lib/channel-form.ts`
- `web/default/src/features/channels/lib/model-mapping-validation.ts`
- `web/default/src/features/channels/components/model-mapping-editor.tsx`
- `web/default/src/components/multi-select.tsx`
- `web/default/src/i18n/locales/*.json`
2026-05-26 01:22:49 +08:00
yyhhyyyyyy 349d5429ca fix: handle paginated API key search response (#5014)
* fix: handle paginated API key search response

* fix: add accessible label to API key filter
2026-05-25 23:15:59 +08:00
花月喵梦 465c5edab9 fix:gemini to claude tool_use err (#5041) 2026-05-25 23:14:01 +08:00
learner-i ff06067a18 fix: 移除 fcIdx -1 偏移,修复并发工具调用撞键问题 (#5095)
当 Claude 直接以多个 tool_use 块起始(无文本前导,index=0)时,
-1 偏移导致 index=0 和 index=1 同时映射到 fcIdx=0:
- index=0 的工具 args 先流完,发出一次合法调用 ✓
- index=1 的 args 追加到同一 map 槽位,污染后为非法 JSON,该工具丢失 ✗
- index=2 以后的工具各自独占唯一 fcIdx,正常发出 ✓

结果:每轮并发调用中第 2 个工具必然丢失,
模型收不到对应的工具结果后重试剩余工具,
产生雪球效应(10个→9个→8个...逐轮收缩)。

修复:直接使用 Claude 的 block index 作为 fcIdx,不做偏移。
fcIdx 仅作为本地 map 的 key,只需保证唯一性,无需从 0 开始。
2026-05-25 23:13:06 +08:00
CaIon 51ca897cf4 refactor(home): redesign hero section to dual-column layout with compliant copywriting
Redesigns the hero section into a balanced horizontal dual-column layout:
- Left Column: Features title, clean legal-compliant descriptions, CTA buttons with BookOpen Docs link, and enlarged supported apps buttons (Cherry Studio and CC Switch with lobe icons)
- Right Column: Smoothly integrates the terminal API demo with top horizontal alignment
- i18n: Configures compliance translations for en, zh, fr, ja, ru, and vi locales
2026-05-25 23:11:05 +08:00
Seefs 1288028181 fix: truncate oversized upstream error logs (#5083) 2026-05-25 23:10:30 +08:00
真的非她不可 2a528d46cb fix(relay): correct image quality parameter handling (#5103) 2026-05-25 22:57:02 +08:00
t0ng7u 583da45296 refactor(ui): Improve usage log filter responsiveness and mobile UX
Refactor the usage log filter toolbar into a shared reusable component for common, drawing, and task logs. Optimize desktop filters with a responsive grid, move secondary filters into a mobile drawer, standardize filter typography, remove redundant filter icons, and add the missing i18n translations for the new drawer description.
2026-05-25 05:35:44 +08:00
t0ng7u b302be30e3 🛠️ fix: v1 interface feedback regressions
Resolve verified V1 frontend feedback by improving channel workflows, auth behavior, API key interactions, user filtering, layout persistence, subscription quota handling, i18n text, pricing metadata, and stale frontend cache recovery.

- Add a global frontend cache version cleanup to prevent old frontend localStorage from causing page errors after upgrades.
- Fix channel copy refresh, model mapping input focus loss, create-channel fetch-model title state, upstream model update confirmation, and batch test toast behavior.
- Respect password login settings and improve Turnstile, forgot-password, registration, and invite-link flows.
- Make user role/status filtering server-side and preserve table page size in URLs.
- Improve API key edit validation feedback and prefetch real keys for reliable copy actions.
- Fix rankings access fail-open behavior, double scrollbars, subscription received amount conversion/display, token i18n wording, model deletion confirmation grammar, and Claude pricing context inference.
- Add clearer Playground model/group loading errors.

Validation:
- bun run typecheck
- bun run i18n:sync
- gofmt on modified Go files
- go test ./controller ./model -run '^$'
2026-05-25 02:42:22 +08:00
t0ng7u 88437a1869 ⬆️ chore(deps): Upgrade default frontend dependencies
Upgrade all web/default dependencies to their latest versions and refresh the Bun lockfile. Add dependency overrides for vulnerable transitive packages so bun audit reports no known vulnerabilities.

Update TypeScript configuration for TypeScript 6 by removing deprecated baseUrl usage and explicitly enabling Node types where needed. Adapt the calendar component to react-day-picker v10 by replacing the removed table class key with month_grid.

Validation:
- bun outdated: no outdated dependencies
- bun audit: no vulnerabilities found
- bun run typecheck: passed
- bun run build: passed
2026-05-25 01:06:42 +08:00
t0ng7u b08febaa3c refactor: system settings UI for consistent, compact layouts
Redesign the system settings interface to align with the rest of the console experience by using fixed header actions, removing redundant subtitles, respecting global content width, and standardizing responsive form layouts.

Introduce reusable settings layout primitives for forms, switch rows, grouped controls, nested control sections, title status indicators, and page action portals. Replace duplicated card-style switch markup with explicit compact components, improve nested switch readability, and reduce visual noise across authentication, billing, content, integrations, maintenance, models, and request-limit settings.

Also complete missing i18n translations, remove obsolete subtitle translation keys, refine i18n sync reporting, fix sidebar truncation for long labels, and verify the frontend with type checking and lint diagnostics.
2026-05-25 00:34:26 +08:00
t0ng7u 92a0959448 refactor(web/default): adopt drill-in sidebar pattern for System Settings
Replace the ad-hoc "workspace" abstraction with a focused, URL-driven
"sidebar view" registry that implements the modern Vercel / Cloudflare
drill-in pattern: clicking a top-level entry (e.g. System Settings)
swaps the sidebar to a contextual workspace, with a `← Back to
Dashboard` affordance, instead of stacking sub-navigation in the root.

Architecture
------------
- types.ts
    + SidebarView           — declarative nested view config
                              (id, pathPattern, parent, getNavGroups)
    + SidebarViewParent     — back-navigation descriptor
    + ResolvedSidebarView   — { key, view, navGroups } returned by hook
    + SidebarData           — slimmed to { navGroups } only
    - Workspace             — removed (logo/plan never rendered)

- lib/sidebar-view-registry.ts (new, replaces workspace-registry.ts)
    + SIDEBAR_VIEWS array — single source of truth for nested views
    + resolveSidebarView(pathname)
    + getNavGroupsForPath(pathname, t) — back-compat helper for the
      command palette

- config/system-settings.config.ts
    Refactored to export a single SYSTEM_SETTINGS_VIEW (SidebarView)
    with parent `/dashboard/overview` + label `Back to Dashboard`.

- components/sidebar-view-header.tsx (new)
    Renders only the back affordance (chevron + label). Uses the
    default SidebarMenuButton size so its typography matches the
    nav items below; collapses gracefully into icon mode via the
    existing tooltip behavior. The redundant "title + icon" row was
    removed — workspace context is already carried by the nav groups.

- hooks/use-sidebar-view.ts (new)
    Encapsulates view resolution and root-nav filtering:
      · matched view  → returns its nav groups verbatim (route-level
                        beforeLoad guards already enforce access);
      · no match      → returns root nav groups, narrowed by user
                        role (admin gate) and useSidebarConfig
                        (admin × user sidebar_modules overlay).

- components/app-sidebar.tsx
    Now a thin presentation layer: reads { key, view, navGroups }
    from useSidebarView() and orchestrates the view transition via
    AnimatePresence + MOTION_VARIANTS.sidebarSlide (respects
    prefers-reduced-motion). No logic, no role checks, no path
    matching — those live in the hook.

- components/command-menu.tsx
    Switched to the new getNavGroupsForPath() API; behavior preserved.

Cleanup
-------
- Deleted layout/context/workspace-context.tsx (zero consumers).
- Deleted layout/lib/workspace-registry.ts and its
  workspace-registry.example.ts companion (over-abstracted: name/id
  metadata, isInWorkspace / getAllWorkspaces / WORKSPACE_IDS were
  registered but never read).
- Removed `workspaces` field from useSidebarData (never consumed
  after the top-switcher was dropped).
- Dropped WorkspaceProvider from authenticated-layout.tsx.
- Trimmed dead `Manage and configure` translation key from all six
  locale files and from static-keys.ts.

i18n
----
Added the `Back to Dashboard` key to en, zh, fr, ja, ru, vi, and
registered it in static-keys.ts under "Sidebar views".

Verification
------------
- bun run typecheck: passes
- Lint: no new warnings/errors on the touched files
- Adding a new drill-in workspace now only requires registering a
  SidebarView in SIDEBAR_VIEWS — no changes to AppSidebar required.
2026-05-24 22:09:05 +08:00
Seefs 49bc3a1175 fix(payment): hide classic Waffo Pancake settings (#5085) 2026-05-24 16:37:43 +08:00
Hill-waffo 0354c38bef [BugFix] fix webhook process (#5047) 2026-05-24 16:19:27 +08:00
同語 ebbe315533 🐛 fix(channel): evict auto-disabled multi-key channels from cache (#4983)
* 🐛 fix(channel): evict auto-disabled multi-key channels from cache

Ensure multi-key channels are removed from the in-memory routing cache when all keys become auto-disabled, preventing subsequent requests from repeatedly selecting channels with no available keys.

Also make multi-key status updates more robust by handling missing key matches, checking actual enabled key availability, and restoring the channel status when a key is re-enabled. Add regression coverage for disabled cached channels and multi-key cache eviction.
2026-05-23 13:24:56 +08:00
CaIon fddf54ccc5 perf: reduce heap residency for large base64 relay requests
Three layered optimizations targeting Gemini-style 5MB base64 payloads where
RSS could balloon to tens of GB under concurrent load:

1. Byte-based param override (relay/common/override.go)
   - Switch legacy/operations hot paths from common.Marshal round-trips and
     map[string]any conversions to gjson/sjson on []byte directly.
   - Avoids cloning 5MB strings during each Set/Delete operation.

2. strings.Builder for Gemini response markdown (relay/channel/gemini/relay-gemini.go)
   - Replace string concatenation + strings.Join when assembling
     "![image](data:...;base64,DATA)" content for inline image responses.
   - Pre-allocates capacity from inline_data byte sizes.

3. Outbound BodyStorage + streaming Decoder (this commit's core)
   - New relay/common/outbound_body.go helper wraps marshaled upstream bodies
     in common.BodyStorage, allowing disk-cache mode to offload jsonData to
     a temp file while waiting for upstream TTFB. The original []byte can
     then be GC'd, removing ~5MB/req of heap residency during the longest
     window of a request.
   - All 7 relay handlers (gemini/claude/responses/embedding/image/compatible/
     rerank) plus chat_completions_via_responses adopt the helper with
     defer closer.Close() and explicit jsonData = nil.
   - relay/common/relay_info.go: new UpstreamRequestBodySize so
     relay/channel/api_request.go can populate req.ContentLength (lost when
     body becomes a type-erased io.Reader).
   - common/gin.go UnmarshalBodyReusable: when storage is disk-backed and
     content-type is JSON, decode via DecodeJson(storage) instead of
     storage.Bytes()+Unmarshal, removing one transient 5MB copy per request.
     memory mode and form/multipart paths unchanged.
2026-05-22 19:08:38 +08:00
CaIon b9bc6f0e21 Revert "fix: correct usage logs filtering (#4883)"
This reverts commit 554defe4f4.
2026-05-22 16:19:54 +08:00
Seefs f2c7647ecf fix: enforce Waffo subscription compliance and product ID update (#5038)
* fix: enforce Waffo subscription compliance and product ID updates

* fix: hide Waffo Pancake settings in classic UI
2026-05-22 11:48:32 +08:00
Hill-waffo 19f1821fc8 [Feature Request] Waffo Pancake gateway — full integration with subscription support + admin catalog binding flow (#4935) 2026-05-22 11:00:58 +08:00
JunXiaoRuo 8e5e89bb5b 修复 切换新版前端Turnstile 开启后注册页未显示验证的问题 (#5011)
Co-authored-by: Codex <codex@users.noreply.github.com>
2026-05-22 10:39:24 +08:00
yyhhyyyyyy e13d673454 fix: update default frontend hardcoded route links (#5016) 2026-05-22 10:36:50 +08:00
Seefs ae6a03364d perf: optimize request metadata extraction and disabled field filtering (#5009)
* perf: optimize request metadata extraction and disabled field filtering

* perf: optimize stream usage estimation path
2026-05-22 10:32:11 +08:00
yyhhyyyyyy 006e801652 fix: resolve model owned_by from active channels (#4416)
* fix: resolve model owned_by from active channels

* fix: respect token group when resolving model owners
2026-05-21 11:16:17 +08:00
yyhhyyyyyy 6f11d19877 fix: normalize model pricing display drift (#4985) 2026-05-21 11:10:22 +08:00
yyhhyyyyyy 58ba867dd6 fix: improve channel test failure details UX (#4988)
* fix: improve channel test failure details UX

* fix: add accessible label to channel models region
2026-05-21 11:09:51 +08:00
1484 changed files with 40962 additions and 31285 deletions
+2
View File
@@ -56,6 +56,8 @@
# 对话超时设置
# 所有请求超时时间,单位秒,默认为0,表示不限制
# RELAY_TIMEOUT=0
# Relay HTTP 客户端空闲连接超时时间,单位秒,默认跟随 Go 标准库,设置为0表示不限制
# RELAY_IDLE_CONN_TIMEOUT=90
# 流模式无响应超时时间,单位秒,如果出现空补全可以尝试改为更大值
# STREAMING_TIMEOUT=300
+92
View File
@@ -0,0 +1,92 @@
name: Docker Build
on:
push:
branches:
- main
- master
tags:
- 'v*'
workflow_dispatch:
jobs:
build-and-push:
name: Build and Push Docker Image
runs-on: act-runner-4c6g
env:
RUNNER_TOOL_CACHE: /toolcache
steps:
- name: Install Docker CLI
run: |
if ! command -v docker &> /dev/null; then
if command -v apk &> /dev/null; then
apk add --no-cache docker-cli
elif command -v apt-get &> /dev/null; then
apt-get update && apt-get install -y docker.io
else
curl -fsSL https://download.docker.com/linux/static/stable/x86_64/docker-24.0.7.tgz | tar xz -C /tmp
mv /tmp/docker/docker /usr/local/bin/
chmod +x /usr/local/bin/docker
fi
fi
docker --version
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Resolve tag & write VERSION
id: version
run: |
if echo "${{ github.ref }}" | grep -q "^refs/tags/"; then
TAG=${GITHUB_REF#refs/tags/}
else
SHORT_SHA=$(git rev-parse --short HEAD)
TAG="dev-${SHORT_SHA}"
fi
echo "TAG=${TAG}" >> $GITHUB_ENV
echo "${TAG}" > VERSION
echo "Building tag: ${TAG}"
cat VERSION
- name: Login to Gitea Container Registry
run: |
echo "${{ secrets.PACKAGES_TOKEN }}" | docker login git.viaeon.com -u "${{ github.actor }}" --password-stdin
- name: Build Docker image
run: |
echo "Building image with tag: ${{ env.TAG }}"
docker build \
--label "org.opencontainers.image.source=https://git.viaeon.com/admin/new-api" \
--label "org.opencontainers.image.revision=${{ github.sha }}" \
-t git.viaeon.com/admin/new-api:${{ env.TAG }} \
-t git.viaeon.com/admin/new-api:latest .
- name: Push Docker image
run: |
echo "Pushing ${{ env.TAG }}..."
docker push git.viaeon.com/admin/new-api:${{ env.TAG }}
echo "Pushing latest..."
docker push git.viaeon.com/admin/new-api:latest
- name: Cleanup Docker
if: always()
run: |
echo "Removing local images..."
docker rmi git.viaeon.com/admin/new-api:${{ env.TAG }} git.viaeon.com/admin/new-api:latest 2>/dev/null || true
echo "Pruning unused Docker resources..."
docker system prune -af --volumes 2>/dev/null || true
echo "Docker disk usage:"
docker system df
- name: Deploy via SSH
if: success()
run: |
if [ -z "${{ secrets.DEPLOY_SSH_HOST }}" ]; then
echo "DEPLOY_SSH_HOST not set, skip deploy"
exit 0
fi
apk add --no-cache sshpass 2>/dev/null || apt-get update && apt-get install -y sshpass 2>/dev/null || true
sshpass -p "${{ secrets.DEPLOY_SSH_PASS }}" ssh -o StrictHostKeyChecking=no -p ${{ secrets.DEPLOY_SSH_PORT || 22 }} ${{ secrets.DEPLOY_SSH_USER }}@${{ secrets.DEPLOY_SSH_HOST }} "cd ${{ secrets.DEPLOY_DIR || '/opt/new-api' }} && docker compose pull && docker compose up -d"
+73
View File
@@ -0,0 +1,73 @@
name: Docker Build (alpha)
on:
push:
branches:
- alpha
workflow_dispatch:
jobs:
build-and-push:
name: Build and Push Alpha Docker Image
runs-on: act-runner-4c6g
env:
RUNNER_TOOL_CACHE: /toolcache
steps:
- name: Install Docker CLI
run: |
if ! command -v docker &> /dev/null; then
if command -v apk &> /dev/null; then
apk add --no-cache docker-cli
elif command -v apt-get &> /dev/null; then
apt-get update && apt-get install -y docker.io
else
curl -fsSL https://download.docker.com/linux/static/stable/x86_64/docker-24.0.7.tgz | tar xz -C /tmp
mv /tmp/docker/docker /usr/local/bin/
chmod +x /usr/local/bin/docker
fi
fi
docker --version
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 1
- name: Determine alpha version
id: version
run: |
VERSION="alpha-$(date +'%Y%m%d')-$(git rev-parse --short HEAD)"
echo "$VERSION" > VERSION
echo "VERSION=$VERSION" >> $GITHUB_ENV
echo "Publishing version: $VERSION"
- name: Login to Gitea Container Registry
run: |
echo "${{ secrets.PACKAGES_TOKEN }}" | docker login git.viaeon.com -u "${{ github.actor }}" --password-stdin 2>&1
- name: Build Docker image
run: |
echo "Building alpha image..."
docker build \
--label "org.opencontainers.image.source=https://git.viaeon.com/admin/new-api" \
--label "org.opencontainers.image.revision=${{ github.sha }}" \
-t git.viaeon.com/admin/new-api:${{ env.VERSION }} \
-t git.viaeon.com/admin/new-api:alpha . 2>&1
- name: Push Docker image
run: |
echo "Pushing ${{ env.VERSION }}..."
docker push git.viaeon.com/admin/new-api:${{ env.VERSION }}
echo "Pushing alpha..."
docker push git.viaeon.com/admin/new-api:alpha
- name: Cleanup Docker
if: always()
run: |
echo "Removing local images..."
docker rmi git.viaeon.com/admin/new-api:${{ env.VERSION }} git.viaeon.com/admin/new-api:alpha 2>/dev/null || true
echo "Pruning unused Docker resources..."
docker system prune -af --volumes 2>/dev/null || true
echo "Docker disk usage:"
docker system df
@@ -0,0 +1,73 @@
name: Docker Build (nightly)
on:
push:
branches:
- nightly
workflow_dispatch:
jobs:
build-and-push:
name: Build and Push Nightly Docker Image
runs-on: act-runner-4c6g
env:
RUNNER_TOOL_CACHE: /toolcache
steps:
- name: Install Docker CLI
run: |
if ! command -v docker &> /dev/null; then
if command -v apk &> /dev/null; then
apk add --no-cache docker-cli
elif command -v apt-get &> /dev/null; then
apt-get update && apt-get install -y docker.io
else
curl -fsSL https://download.docker.com/linux/static/stable/x86_64/docker-24.0.7.tgz | tar xz -C /tmp
mv /tmp/docker/docker /usr/local/bin/
chmod +x /usr/local/bin/docker
fi
fi
docker --version
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 1
- name: Determine nightly version
id: version
run: |
VERSION="nightly-$(date +'%Y%m%d')-$(git rev-parse --short HEAD)"
echo "$VERSION" > VERSION
echo "VERSION=$VERSION" >> $GITHUB_ENV
echo "Publishing version: $VERSION"
- name: Login to Gitea Container Registry
run: |
echo "${{ secrets.PACKAGES_TOKEN }}" | docker login git.viaeon.com -u "${{ github.actor }}" --password-stdin 2>&1
- name: Build Docker image
run: |
echo "Building nightly image..."
docker build \
--label "org.opencontainers.image.source=https://git.viaeon.com/admin/new-api" \
--label "org.opencontainers.image.revision=${{ github.sha }}" \
-t git.viaeon.com/admin/new-api:${{ env.VERSION }} \
-t git.viaeon.com/admin/new-api:nightly . 2>&1
- name: Push Docker image
run: |
echo "Pushing ${{ env.VERSION }}..."
docker push git.viaeon.com/admin/new-api:${{ env.VERSION }}
echo "Pushing nightly..."
docker push git.viaeon.com/admin/new-api:nightly
- name: Cleanup Docker
if: always()
run: |
echo "Removing local images..."
docker rmi git.viaeon.com/admin/new-api:${{ env.VERSION }} git.viaeon.com/admin/new-api:nightly 2>/dev/null || true
echo "Pruning unused Docker resources..."
docker system prune -af --volumes 2>/dev/null || true
echo "Docker disk usage:"
docker system df
+82
View File
@@ -0,0 +1,82 @@
name: PR Check
on:
pull_request:
types: [opened, reopened]
jobs:
pr-quality:
name: PR Quality Check
runs-on: act-runner-4c6g
env:
RUNNER_TOOL_CACHE: /toolcache
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Check PR description
env:
GITEA_TOKEN: ${{ secrets.PACKAGES_TOKEN }}
PR_NUMBER: ${{ github.event.pull_request.number }}
REPO: admin/new-api
GITEA_URL: https://git.viaeon.com
run: |
# 获取 PR 信息
PR_INFO=$(curl -s -H "Authorization: token ${GITEA_TOKEN}" \
"${GITEA_URL}/api/v1/repos/${REPO}/pulls/${PR_NUMBER}")
PR_BODY=$(echo "$PR_INFO" | jq -r '.body // empty')
PR_TITLE=$(echo "$PR_INFO" | jq -r '.title // empty')
PR_USER=$(echo "$PR_INFO" | jq -r '.user.login // empty')
FAILED=0
REASONS=""
# 检查 PR 描述是否为空
if [ -z "$PR_BODY" ] || [ "$PR_BODY" = "null" ]; then
FAILED=1
REASONS="${REASONS}- PR description is empty\n"
fi
# 检查 PR 标题是否为空
if [ -z "$PR_TITLE" ] || [ "$PR_TITLE" = "null" ]; then
FAILED=1
REASONS="${REASONS}- PR title is empty\n"
fi
# 检查是否包含纯 AI 生成标记
if echo "$PR_BODY" | grep -qi "Generated with Claude Code"; then
FAILED=1
REASONS="${REASONS}- PR appears to be purely AI-generated without meaningful human involvement\n"
fi
if [ "$FAILED" -eq 1 ]; then
echo "PR check failed:"
echo -e "$REASONS"
# 添加标签
curl -s -X POST \
"${GITEA_URL}/api/v1/repos/${REPO}/issues/${PR_NUMBER}/labels" \
-H "Authorization: token ${GITEA_TOKEN}" \
-H "Content-Type: application/json" \
-d '{"labels": ["pr-check-failed"]}'
# 添加评论
curl -s -X POST \
"${GITEA_URL}/api/v1/repos/${REPO}/issues/${PR_NUMBER}/comments" \
-H "Authorization: token ${GITEA_TOKEN}" \
-H "Content-Type: application/json" \
-d '{"body": "感谢您的提交。由于该 PR 未遵循我们的贡献模板,且被识别为缺乏人工参与的纯 AI 生成内容,我们将先予以关闭。我们更欢迎经过人工审核、验证并带有个人思考的贡献。如果您认为这其中存在误解,请回复告知。"}'
# 关闭 PR
curl -s -X PATCH \
"${GITEA_URL}/api/v1/repos/${REPO}/pulls/${PR_NUMBER}" \
-H "Authorization: token ${GITEA_TOKEN}" \
-H "Content-Type: application/json" \
-d '{"state": "closed"}'
exit 1
fi
echo "PR check passed!"
+161
View File
@@ -0,0 +1,161 @@
name: Release (Linux)
on:
push:
tags:
- 'v*'
- '!*-alpha*'
workflow_dispatch:
inputs:
tag:
description: 'Tag name to build (e.g., v0.10.8)'
required: true
type: string
jobs:
build-linux:
name: Linux Release
runs-on: act-runner-4c6g
env:
RUNNER_TOOL_CACHE: /toolcache
steps:
- name: Install dependencies
run: |
export PATH="/toolcache/bin:$PATH"
# Install Go
if ! command -v go &> /dev/null; then
curl -fsSL https://go.dev/dl/go1.25.1.linux-amd64.tar.gz | tar -C /usr/local -xzf -
echo "export PATH=\$PATH:/usr/local/go/bin" >> ~/.bashrc
export PATH=$PATH:/usr/local/go/bin
fi
go version
# Install Bun
if ! command -v bun &> /dev/null; then
curl -fsSL https://bun.sh/install | bash
export PATH="$HOME/.bun/bin:$PATH"
fi
bun --version
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Determine Version
run: |
if [ -n "${{ github.event.inputs.tag }}" ]; then
TAG="${{ github.event.inputs.tag }}"
else
TAG=${GITHUB_REF#refs/tags/}
fi
VERSION=$(git describe --tags 2>/dev/null || echo "$TAG")
echo "VERSION=$VERSION" >> $GITHUB_ENV
echo "Building version: $VERSION"
- name: Build Frontend (default)
env:
CI: ""
run: |
export PATH="$HOME/.bun/bin:/usr/local/go/bin:$PATH"
cd web
bun install --frozen-lockfile
cd default
DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Build Frontend (classic)
env:
CI: ""
run: |
export PATH="$HOME/.bun/bin:/usr/local/go/bin:$PATH"
cd web
bun install --frozen-lockfile
cd classic
VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Build Backend (amd64)
run: |
export PATH="/usr/local/go/bin:$PATH"
go mod download
go build -ldflags "-s -w -X 'new-api/common.Version=$VERSION' -extldflags '-static'" -o new-api-$VERSION
- name: Build Backend (arm64)
run: |
export PATH="/usr/local/go/bin:$PATH"
sudo apt-get update
DEBIAN_FRONTEND=noninteractive sudo apt-get install -y gcc-aarch64-linux-gnu
CC=aarch64-linux-gnu-gcc CGO_ENABLED=1 GOOS=linux GOARCH=arm64 go build -ldflags "-s -w -X 'new-api/common.Version=$VERSION' -extldflags '-static'" -o new-api-arm64-$VERSION
- name: Generate checksums
run: sha256sum new-api-* > checksums-linux.txt
- name: Upload artifacts
uses: actions/upload-artifact@v4
with:
name: linux-build
path: |
new-api-*
checksums-linux.txt
release:
name: Create Gitea Release
needs: [build-linux]
runs-on: act-runner-4c6g
env:
RUNNER_TOOL_CACHE: /toolcache
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Determine Version
run: |
if [ -n "${{ github.event.inputs.tag }}" ]; then
TAG="${{ github.event.inputs.tag }}"
else
TAG=${GITHUB_REF#refs/tags/}
fi
echo "TAG=$TAG" >> $GITHUB_ENV
- name: Download artifacts
uses: actions/download-artifact@v4
with:
path: artifacts
- name: Create Gitea Release
env:
GITEA_TOKEN: ${{ secrets.PACKAGES_TOKEN }}
run: |
# 使用 Gitea API 创建 Release
TAG="${{ env.TAG }}"
REPO="admin/new-api"
GITEA_URL="https://git.viaeon.com"
# 创建 Release
RELEASE_ID=$(curl -s -X POST \
"${GITEA_URL}/api/v1/repos/${REPO}/releases" \
-H "Authorization: token ${GITEA_TOKEN}" \
-H "Content-Type: application/json" \
-d "{
\"tag_name\": \"${TAG}\",
\"name\": \"${TAG}\",
\"body\": \"Release ${TAG}\",
\"draft\": false,
\"prerelease\": false
}" | jq -r '.id')
echo "Created release ID: ${RELEASE_ID}"
# 上传附件
find artifacts -type f | while read file; do
echo "Uploading: ${file}"
curl -s -X POST \
"${GITEA_URL}/api/v1/repos/${REPO}/releases/${RELEASE_ID}/assets" \
-H "Authorization: token ${GITEA_TOKEN}" \
-F "attachment=@${file}" \
-F "name=$(basename ${file})"
done
echo "Release ${TAG} created successfully!"
+11 -4
View File
@@ -11,6 +11,8 @@ assignees: ''
- 文档:https://docs.newapi.ai/
- 使用问题先看或先问:https://deepwiki.com/QuantumNous/new-api
- 开启透传后的转发相关反馈不接受 issue;透传模式会直接转发请求,请自行确认上游行为。
- 不接受 coding plan、逆向渠道等技术支持类 issue。
- 警告:删除本模板、删除小节标题或随意清空内容的 issue,可能会被直接关闭;重复恶意提交者可能会被 block。
**您当前的 newapi 版本**
@@ -20,13 +22,18 @@ assignees: ''
**提交确认**
[//]: # (方框内删除已有的空格,填 x 号)
+ [ ] 我已确认目前没有类似 issue
+ [ ] 我已完整查看文档 https://docs.newapi.ai/项目 README,尤其是常见问题部分
+ [ ] 我未删除此模板中的任何引导内容或小节标题,并会按要求完整填写
+ [ ] 我理解项目维护者精力有限,不遵循模板要求的 issue 可能会被无视或直接关闭
- [ ] **非重复 issue:** 我已搜索现有 [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue)确认目前没有类似 issue
- [ ] **提交前必读:** 我已完整阅读上方“提交前必读”,并已查看文档 https://docs.newapi.ai/项目 README 且向 AI 提问,确认这不是使用、配置或接入类问题。
- [ ] **模板完整:** 我未删除此模板中的任何引导内容或小节标题,并会按要求完整填写
- [ ] **维护成本:** 我理解项目维护者精力有限,不遵循模板要求的 issue 可能会被无视或直接关闭
**问题描述**
请尽可能说明问题现象、影响范围,以及你判断它是程序问题而不是上游行为或使用问题的依据。
- 转发问题请尽可能说明渠道类型、转换格式、上游原生支持依据和服务端日志。
- 计费问题请尽可能附请求返回的 `usage` 示例。
**复现步骤**
**预期结果**
+11 -4
View File
@@ -11,6 +11,8 @@ assignees: ''
- Docs: https://docs.newapi.ai/
- Usage questions first: https://deepwiki.com/QuantumNous/new-api
- Issues about forwarding behavior after enabling pass-through mode are not accepted; pass-through mode forwards requests directly, so please verify upstream behavior yourself.
- Technical support requests such as coding plans or reverse-engineering channels are not accepted as issues.
- Warning: issues with this template removed, section headings deleted, or content cleared may be closed directly. Repeated abusive submissions may result in a block.
**Your current newapi version**
@@ -20,13 +22,18 @@ Please fill this in, for example: `v1.0.0`
**Submission Checks**
[//]: # (Remove the space in the box and fill with an x)
+ [ ] I have confirmed there are no similar issues
+ [ ] I have thoroughly read the docs at https://docs.newapi.ai/ and the project README, especially the FAQ section
+ [ ] I have not removed any guidance or section headings from this template and will complete it as requested
+ [ ] I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly
- [ ] **Non-duplicate issue:** I have searched existing [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue) and confirmed there are no similar issues.
- [ ] **Read this first:** I have fully read the section above, reviewed the docs at https://docs.newapi.ai/ and the project README, and asked AI first, confirming this is not a usage, configuration, or integration question.
- [ ] **Template intact:** I have not removed any guidance or section headings from this template and will complete it as requested.
- [ ] **Maintainer time:** I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly.
**Issue Description**
Describe the symptom, impact scope, and why you believe this is an application issue rather than upstream behavior or a usage question with as much detail as possible.
- For forwarding issues, include the channel type, conversion format, upstream native-support evidence, and server logs when possible.
- For billing issues, include an example of the returned `usage` when possible.
**Steps to Reproduce**
**Expected Result**
+6 -4
View File
@@ -11,6 +11,8 @@ assignees: ''
- 文档:https://docs.newapi.ai/
- 使用问题先看或先问:https://deepwiki.com/QuantumNous/new-api
- 开启透传后的转发相关反馈不接受 issue;透传模式会直接转发请求,请自行确认上游行为。
- 不接受 coding plan、逆向渠道等技术支持类 issue。
- 警告:删除本模板、删除小节标题或随意清空内容的 issue,可能会被直接关闭;重复恶意提交者可能会被 block。
**您当前的 newapi 版本**
@@ -20,10 +22,10 @@ assignees: ''
**提交确认**
[//]: # (方框内删除已有的空格,填 x 号)
+ [ ] 我已确认目前没有类似 issue
+ [ ] 我已完整查看文档 https://docs.newapi.ai/项目 README,已确定现有版本无法满足需求
+ [ ] 我未删除此模板中的任何引导内容或小节标题,并会按要求完整填写
+ [ ] 我理解项目维护者精力有限,不遵循模板要求的 issue 可能会被无视或直接关闭
- [ ] **非重复 issue:** 我已搜索现有 [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue)确认目前没有类似 issue
- [ ] **提交前必读:** 我已完整阅读上方“提交前必读”,并已查看文档 https://docs.newapi.ai/项目 README 且向 AI 提问,确认这不是使用、配置或接入类问题,且现有版本无法满足需求
- [ ] **模板完整:** 我未删除此模板中的任何引导内容或小节标题,并会按要求完整填写
- [ ] **维护成本:** 我理解项目维护者精力有限,不遵循模板要求的 issue 可能会被无视或直接关闭
**功能描述**
+6 -4
View File
@@ -11,6 +11,8 @@ assignees: ''
- Docs: https://docs.newapi.ai/
- Usage questions first: https://deepwiki.com/QuantumNous/new-api
- Issues about forwarding behavior after enabling pass-through mode are not accepted; pass-through mode forwards requests directly, so please verify upstream behavior yourself.
- Technical support requests such as coding plans or reverse-engineering channels are not accepted as issues.
- Warning: issues with this template removed, section headings deleted, or content cleared may be closed directly. Repeated abusive submissions may result in a block.
**Your current newapi version**
@@ -20,10 +22,10 @@ Please fill this in, for example: `v1.0.0`
**Submission Checks**
[//]: # (Remove the space in the box and fill with an x)
+ [ ] I have confirmed there are no similar issues
+ [ ] I have thoroughly read the docs at https://docs.newapi.ai/ and the project README, and confirmed the current version cannot meet my needs
+ [ ] I have not removed any guidance or section headings from this template and will complete it as requested
+ [ ] I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly
- [ ] **Non-duplicate issue:** I have searched existing [Issues](https://github.com/QuantumNous/new-api/issues?q=is%3Aissue) and confirmed there are no similar issues.
- [ ] **Read this first:** I have fully read the section above, reviewed the docs at https://docs.newapi.ai/ and the project README, and asked AI first, confirming this is not a usage, configuration, or integration question, and that the current version cannot meet my needs.
- [ ] **Template intact:** I have not removed any guidance or section headings from this template and will complete it as requested.
- [ ] **Maintainer time:** I understand that maintainers have limited time and issues that do not follow this template may be ignored or closed directly.
**Feature Description**
+18 -12
View File
@@ -33,16 +33,18 @@ jobs:
env:
CI: ""
run: |
cd web/default
bun install
cd web
bun install --frozen-lockfile
cd default
DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Build Frontend (classic)
env:
CI: ""
run: |
cd web/classic
bun install
cd web
bun install --frozen-lockfile
cd classic
VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Set up Go
@@ -91,16 +93,18 @@ jobs:
CI: ""
NODE_OPTIONS: "--max-old-space-size=4096"
run: |
cd web/default
bun install
cd web
bun install --frozen-lockfile
cd default
DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Build Frontend (classic)
env:
CI: ""
run: |
cd web/classic
bun install
cd web
bun install --frozen-lockfile
cd classic
VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Set up Go
@@ -146,16 +150,18 @@ jobs:
env:
CI: ""
run: |
cd web/default
bun install
cd web
bun install --frozen-lockfile
cd default
DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Build Frontend (classic)
env:
CI: ""
run: |
cd web/classic
bun install
cd web
bun install --frozen-lockfile
cd classic
VITE_REACT_APP_VERSION=$VERSION bun run build
cd ../..
- name: Set up Go
+3 -1
View File
@@ -7,9 +7,10 @@ upload
*.db
build
*.db-journal
logs
/logs
web/default/dist
web/classic/dist
web/daisy/dist
web/node_modules
web/dist
.env
@@ -35,3 +36,4 @@ data/
.test
token_estimator_test.go
skills-lock.json
.playwright-mcp
+170
View File
@@ -0,0 +1,170 @@
# ModelsToken 管理平台 - 产品需求文档 (PRD)
## 1. 产品概述
ModelsToken 是一个 AI API 管理与分发平台,为开发者和企业提供统一的 AI 模型接入、密钥管理、用量计费、渠道代理等一站式服务。新前端将采用 React + DaisyUI 5 + TypeScript 构建,替换现有的 Default/Classic 双前端,同时新增本地文档管理功能。
- 目标用户:AI 应用开发者、企业运维人员、API 服务管理者
- 核心价值:简化 AI API 的管理复杂度,提供直观的操作界面和完整的文档支持
## 2. 核心功能
### 2.1 用户角色
| 角色 | 注册方式 | 核心权限 |
|------|----------|----------|
| 普通用户 | 用户名/邮箱/OAuth | 密钥管理、充值、订阅、日志查看、文档访问 |
| 管理员 | 由超级管理员指定 | 渠道管理、用户管理、兑换码、模型管理、订阅管理 |
| 超级管理员 | 系统初始化 | 全部权限 + 系统设置 |
### 2.2 功能模块
#### 公共页面(无需登录)
1. **首页**:Hero 区域、特性展示、快速入门指引
2. **登录页**:用户名/密码、OAuth 登录(GitHub/Discord/OIDC/LinuxDO/微信/Telegram/自定义)
3. **注册页**:注册表单 + Turnstile 人机验证
4. **忘记密码**:邮箱重置链接
5. **模型定价**:模型价格列表、搜索筛选
6. **关于页面**:项目信息、版本、许可证
7. **用户协议/隐私政策**
8. **初始化向导**:首次部署配置
#### 用户功能(需登录)
1. **仪表盘**:额度概览、使用趋势图、API 信息面板、公告、FAQ
2. **API 密钥管理**:创建/编辑/删除/批量操作、额度限制、模型限制、IP 限制
3. **钱包/充值**:余额查看、兑换码充值、在线支付(易支付/Stripe/Creem/Waffo)、签到
4. **订阅管理**:查看计划、购买订阅、当前订阅状态
5. **使用日志**:请求日志搜索/筛选、MJ 日志、任务日志、统计图表
6. **个人设置**:资料编辑、2FA 设置、Passkey 管理、OAuth 绑定、语言切换
7. **Playground**API 在线调试、Chat Completions 测试
8. **文档中心**(新增):本地文档管理、分类浏览、搜索、Markdown 渲染
#### 管理员功能
1. **渠道管理**:CRUD、测试、余额更新、标签管理、批量操作、多密钥、Codex OAuth、Ollama 管理
2. **用户管理**:列表/搜索/创建/编辑/升降级/启禁/额度调整
3. **兑换码管理**CRUD、批量删除无效码
4. **模型管理**:模型元数据 CRUD、上游同步、缺失模型检测
5. **供应商管理**CRUD
6. **订阅管理**:计划 CRUD、用户订阅管理
7. **部署管理**io.net 部署 CRUD、容器管理、日志
#### 超级管理员 - 系统设置
1. **站点设置**:名称/Logo/页脚/公告/首页内容/服务器地址
2. **认证设置**:注册/登录开关、OAuth 配置、Turnstile、Passkey、自定义 OAuth
3. **计费设置**:额度/倍率/支付配置/签到/分组倍率
4. **内容设置**:公告/FAQ/Uptime Kuma/聊天/绘图/Midjourney
5. **模型设置**:透传/思维模型/Gemini/Claude 配置
6. **运维设置**:重试/自动禁用/SMTP/性能监控/日志
7. **安全设置**:速率限制/敏感词/SSRF 防护/IP 过滤
### 2.3 新增功能 - 本地文档管理
| 功能 | 说明 |
|------|------|
| 文档分类 | 支持多级分类树,管理员可创建/编辑/删除分类 |
| 文档 CRUD | 管理员创建/编辑/删除文档,支持 Markdown 编辑器 |
| 文档浏览 | 用户按分类浏览文档,支持搜索 |
| 文档搜索 | 全文搜索文档标题和内容 |
| 文档版本 | 文档更新历史记录 |
| 权限控制 | 可设置文档为公开/登录可见/管理员可见 |
## 3. 核心流程
### 3.1 用户认证流程
```mermaid
flowchart TD
"访问平台" --> "已登录?"
"已登录?" -->|"是"| "仪表盘"
"已登录?" -->|"否"| "登录页"
"登录页" --> "输入凭证"
"输入凭证" --> "需要2FA?"
"需要2FA?" -->|"是"| "输入2FA码"
"需要2FA?" -->|"否"| "验证成功"
"输入2FA码" --> "验证成功"
"验证成功" --> "仪表盘"
"登录页" --> "OAuth登录"
"OAuth登录" --> "OAuth回调"
"OAuth回调" --> "已绑定账号?"
"已绑定账号?" -->|"是"| "仪表盘"
"已绑定账号?" -->|"否"| "绑定/注册"
```
### 3.2 API 调用流程
```mermaid
flowchart TD
"创建API密钥" --> "配置密钥参数"
"配置密钥参数" --> "使用密钥调用API"
"使用密钥调用API" --> "平台路由到渠道"
"平台路由到渠道" --> "返回结果"
"返回结果" --> "记录日志"
"记录日志" --> "扣除额度"
```
### 3.3 文档管理流程(新增)
```mermaid
flowchart TD
"管理员创建分类" --> "创建文档"
"创建文档" --> "Markdown编辑"
"Markdown编辑" --> "设置可见性"
"设置可见性" --> "发布文档"
"发布文档" --> "用户浏览/搜索"
```
## 4. 用户界面设计
### 4.1 设计风格
- **主色调**:深蓝 (#1e293b) + 亮蓝 (#3b82f6) 渐变,搭配 DaisyUI 的 `business` 主题
- **辅助色**:翡翠绿 (#10b981) 用于成功/在线状态,琥珀色 (#f59e0b) 用于警告
- **按钮风格**:DaisyUI 默认圆角按钮,主要操作用 `btn-primary`,危险操作用 `btn-error`
- **字体**JetBrains Mono(代码/密钥)+ Noto Sans SC(中文正文)
- **布局风格**:左侧固定导航栏 + 顶部状态栏 + 主内容区,响应式折叠
- **图标**Lucide React 图标库
- **动效**DaisyUI 内置动画 + 页面切换淡入
### 4.2 页面设计概览
| 页面 | 模块 | UI 元素 |
|------|------|---------|
| 首页 | Hero | 渐变背景、特性卡片、快速开始按钮 |
| 登录 | 表单 | 居中卡片、OAuth 按钮组、Turnstile |
| 仪表盘 | 统计卡片 | 4 列额度卡片、折线图、公告栏、API 信息 |
| 密钥管理 | 数据表 | 搜索栏、筛选器、表格、批量操作栏 |
| 渠道管理 | 数据表+表单 | 标签筛选、测试按钮、多密钥管理抽屉 |
| 系统设置 | 标签页 | 7 大分类侧边导航、表单分组、开关/输入框 |
| 文档中心 | 侧边树+内容 | 分类树导航、Markdown 渲染、搜索框、面包屑 |
| Playground | 分栏 | 左侧参数面板、右侧响应面板、模型选择器 |
### 4.3 响应式设计
- 桌面优先(1280px+
- 平板适配(768px-1279px):侧边栏折叠为抽屉
- 移动端适配(<768px):单列布局,表格改为卡片列表
### 4.4 布局结构
```
┌──────────────────────────────────────────────┐
│ 顶部导航栏 (Navbar) │
│ Logo | 搜索 | 通知 | 用户菜单 | 主题切换 │
├──────┬───────────────────────────────────────┤
│ │ │
│ 侧边 │ 主内容区 │
│ 导航 │ │
│ 栏 │ ┌─────────────────────────────────┐ │
│ │ │ 面包屑 + 页面标题 + 操作按钮 │ │
│ 仪表盘│ ├─────────────────────────────────┤ │
│ 密钥 │ │ │ │
│ 渠道 │ │ 页面内容 │ │
│ 用户 │ │ │ │
│ 日志 │ │ │ │
│ 钱包 │ └─────────────────────────────────┘ │
│ 订阅 │ │
│ 文档 │ │
│ 设置 │ │
│ │ │
└──────┴───────────────────────────────────────┘
```
+459
View File
@@ -0,0 +1,459 @@
# ModelsToken 管理平台 - 技术架构文档
## 1. 架构设计
```mermaid
flowchart TB
subgraph "前端 (React + DaisyUI 5)"
A["React 18"] --> B["React Router v6"]
B --> C["页面组件"]
C --> D["DaisyUI 5 组件"]
D --> E["Tailwind CSS 4"]
A --> F["Zustand 状态管理"]
A --> G["React Query 数据请求"]
A --> H["i18next 国际化"]
A --> I["React Markdown 渲染"]
end
subgraph "后端 (Go + Gin)"
J["Gin HTTP Server"]
J --> K["API 路由"]
J --> L["Relay 代理"]
K --> M["控制器"]
M --> N["模型层"]
N --> O["数据库 (SQLite/MySQL/PostgreSQL)"]
end
C -->|"Axios HTTP"| K
```
## 2. 技术说明
- **前端框架**React 18 + TypeScript
- **UI 库**DaisyUI 5 + Tailwind CSS 4
- **构建工具**Vite 6
- **路由**React Router v6(懒加载)
- **状态管理**Zustand(轻量级,替代 Redux
- **数据请求**TanStack React Query v5 + Axios
- **国际化**i18next + react-i18next
- **图表**Recharts
- **Markdown**react-markdown + remark-gfm + rehype-highlight
- **图标**Lucide React
- **代码高亮**highlight.js
- **表单验证**React Hook Form + Zod
- **通知**react-hot-toast
- **项目目录**`web/daisy/`
## 3. 路由定义
### 3.1 公共路由
| 路由 | 用途 |
|------|------|
| `/` | 首页 |
| `/login` | 登录 |
| `/register` | 注册 |
| `/forgot-password` | 忘记密码 |
| `/reset-password` | 密码重置确认 |
| `/setup` | 初始化向导 |
| `/pricing` | 模型定价 |
| `/about` | 关于 |
| `/user-agreement` | 用户协议 |
| `/privacy-policy` | 隐私政策 |
| `/oauth/callback/:provider` | OAuth 回调 |
### 3.2 认证后路由
| 路由 | 用途 |
|------|------|
| `/dashboard` | 仪表盘 |
| `/tokens` | API 密钥管理 |
| `/wallet` | 钱包/充值 |
| `/subscriptions` | 订阅管理 |
| `/logs` | 使用日志 |
| `/logs/midjourney` | MJ 日志 |
| `/logs/tasks` | 任务日志 |
| `/profile` | 个人设置 |
| `/playground` | Playground |
| `/docs` | 文档中心(新增) |
| `/docs/:slug` | 文档详情(新增) |
### 3.3 管理员路由
| 路由 | 用途 |
|------|------|
| `/admin/channels` | 渠道管理 |
| `/admin/users` | 用户管理 |
| `/admin/redemptions` | 兑换码管理 |
| `/admin/models` | 模型管理 |
| `/admin/vendors` | 供应商管理 |
| `/admin/deployments` | 部署管理 |
| `/admin/subscriptions` | 订阅计划管理 |
### 3.4 超级管理员路由
| 路由 | 用途 |
|------|------|
| `/settings/site` | 站点设置 |
| `/settings/auth` | 认证设置 |
| `/settings/billing` | 计费设置 |
| `/settings/content` | 内容设置 |
| `/settings/models` | 模型设置 |
| `/settings/operations` | 运维设置 |
| `/settings/security` | 安全设置 |
| `/settings/docs` | 文档管理(新增) |
## 4. API 定义
### 4.1 核心类型
```typescript
// 用户
interface User {
id: number;
username: string;
display_name: string;
email: string;
role: number; // 1=user, 10=admin, 100=root
status: number;
quota: number;
used_quota: number;
request_count: number;
group: string;
aff_code: string;
inviter_id: number;
language: string;
access_token: string;
created_time: number;
}
// 渠道
interface Channel {
id: number;
type: number;
key: string;
openai_organization?: string;
base_url: string;
models: string;
model_mapping?: string;
group: string;
groups: string[];
name: string;
priority: number;
weight: number;
status: number;
tag?: string;
setting?: string;
test_time: number;
response_time: number;
balance: number;
balance_updated_time: number;
created_time: number;
}
// 令牌
interface Token {
id: number;
user_id: number;
key: string;
status: number;
name: string;
created_time: number;
accessed_time: number;
expired_time: number;
remain_quota: number;
unlimited_quota: boolean;
used_quota: number;
models: string;
subnet: string;
group: string;
}
// 日志
interface Log {
id: number;
user_id: number;
created_at: number;
type: number;
content: string;
username: string;
token_name: string;
model_name: string;
quota: number;
prompt_tokens: number;
completion_tokens: number;
channel_id: number;
token_id: number;
group: string;
request_id: string;
ip: string;
detail: string;
}
// 订阅计划
interface SubscriptionPlan {
id: number;
name: string;
description: string;
price: number;
currency: string;
duration_days: number;
quota: number;
models: string;
enabled: boolean;
sort_order: number;
created_time: number;
}
// 文档(新增)
interface Document {
id: number;
title: string;
slug: string;
content: string; // Markdown
category_id: number;
category?: DocumentCategory;
visibility: 'public' | 'auth' | 'admin';
sort_order: number;
created_at: string;
updated_at: string;
author_id: number;
author?: User;
versions?: DocumentVersion[];
}
interface DocumentCategory {
id: number;
name: string;
slug: string;
parent_id: number | null;
children?: DocumentCategory[];
sort_order: number;
}
interface DocumentVersion {
id: number;
document_id: number;
content: string;
created_at: string;
author_id: number;
}
```
### 4.2 新增文档管理 API
| 端点 | 方法 | 权限 | 说明 |
|------|------|------|------|
| `/api/docs/categories` | GET | 公开 | 获取分类树 |
| `/api/docs/categories` | POST | Admin | 创建分类 |
| `/api/docs/categories/:id` | PUT | Admin | 更新分类 |
| `/api/docs/categories/:id` | DELETE | Admin | 删除分类 |
| `/api/docs/` | GET | 按可见性 | 文档列表(支持搜索) |
| `/api/docs/:slug` | GET | 按可见性 | 获取文档详情 |
| `/api/docs/` | POST | Admin | 创建文档 |
| `/api/docs/:id` | PUT | Admin | 更新文档 |
| `/api/docs/:id` | DELETE | Admin | 删除文档 |
| `/api/docs/:id/versions` | GET | Admin | 文档版本历史 |
## 5. 项目目录结构
```
web/daisy/
├── index.html
├── package.json
├── tsconfig.json
├── vite.config.ts
├── tailwind.config.ts
├── public/
│ └── manifest.json
└── src/
├── main.tsx # 入口
├── App.tsx # 根组件 + 路由
├── vite-env.d.ts
├── api/ # API 请求层
│ ├── client.ts # Axios 实例 + 拦截器
│ ├── auth.ts # 认证 API
│ ├── channel.ts # 渠道 API
│ ├── token.ts # 令牌 API
│ ├── user.ts # 用户 API
│ ├── log.ts # 日志 API
│ ├── subscription.ts # 订阅 API
│ ├── redemption.ts # 兑换码 API
│ ├── model.ts # 模型 API
│ ├── vendor.ts # 供应商 API
│ ├── deployment.ts # 部署 API
│ ├── option.ts # 系统设置 API
│ ├── payment.ts # 支付 API
│ └── doc.ts # 文档 API(新增)
├── stores/ # Zustand 状态
│ ├── auth.ts # 认证状态
│ └── ui.ts # UI 状态(侧边栏/主题)
├── hooks/ # 自定义 Hooks
│ ├── useAuth.ts
│ ├── usePermission.ts
│ └── useQuota.ts
├── components/ # 通用组件
│ ├── layout/
│ │ ├── AppLayout.tsx # 主布局
│ │ ├── Sidebar.tsx # 侧边导航
│ │ ├── Navbar.tsx # 顶部导航
│ │ └── Breadcrumb.tsx # 面包屑
│ ├── common/
│ │ ├── QuotaDisplay.tsx # 额度显示
│ │ ├── ModelBadge.tsx # 模型标签
│ │ ├── StatusBadge.tsx # 状态标签
│ │ ├── SearchInput.tsx # 搜索框
│ │ ├── DataTable.tsx # 数据表格
│ │ ├── ConfirmDialog.tsx # 确认对话框
│ │ └── LoadingSpinner.tsx # 加载动画
│ └── charts/
│ ├── QuotaChart.tsx # 额度趋势图
│ └── StatsChart.tsx # 统计图表
├── pages/ # 页面组件
│ ├── public/
│ │ ├── Home.tsx
│ │ ├── Login.tsx
│ │ ├── Register.tsx
│ │ ├── ForgotPassword.tsx
│ │ ├── Pricing.tsx
│ │ ├── About.tsx
│ │ └── Setup.tsx
│ ├── dashboard/
│ │ └── Dashboard.tsx
│ ├── tokens/
│ │ ├── TokenList.tsx
│ │ └── TokenForm.tsx
│ ├── channels/
│ │ ├── ChannelList.tsx
│ │ └── ChannelForm.tsx
│ ├── users/
│ │ ├── UserList.tsx
│ │ └── UserForm.tsx
│ ├── logs/
│ │ ├── LogList.tsx
│ │ ├── MidjourneyLog.tsx
│ │ └── TaskLog.tsx
│ ├── wallet/
│ │ └── Wallet.tsx
│ ├── subscriptions/
│ │ ├── PlanList.tsx
│ │ └── MySubscription.tsx
│ ├── redemptions/
│ │ └── RedemptionList.tsx
│ ├── models/
│ │ └── ModelList.tsx
│ ├── vendors/
│ │ └── VendorList.tsx
│ ├── deployments/
│ │ └── DeploymentList.tsx
│ ├── playground/
│ │ └── Playground.tsx
│ ├── profile/
│ │ └── Profile.tsx
│ ├── docs/ # 文档中心(新增)
│ │ ├── DocCenter.tsx # 文档浏览主页
│ │ ├── DocViewer.tsx # 文档阅读页
│ │ ├── DocEditor.tsx # 文档编辑页(管理员)
│ │ └── DocCategoryManager.tsx # 分类管理(管理员)
│ └── settings/
│ ├── SiteSettings.tsx
│ ├── AuthSettings.tsx
│ ├── BillingSettings.tsx
│ ├── ContentSettings.tsx
│ ├── ModelSettings.tsx
│ ├── OperationsSettings.tsx
│ ├── SecuritySettings.tsx
│ └── DocSettings.tsx # 文档设置(新增)
├── i18n/ # 国际化
│ ├── index.ts
│ └── locales/
│ ├── en.json
│ └── zh.json
├── lib/ # 工具函数
│ ├── constants.ts
│ ├── utils.ts
│ ├── quota.ts
│ └── channel-types.ts
└── types/ # TypeScript 类型
├── api.ts
├── channel.ts
├── token.ts
├── user.ts
├── log.ts
├── subscription.ts
├── doc.ts
└── option.ts
```
## 6. 数据模型(新增文档管理)
```mermaid
erDiagram
"document_categories" {
int id PK
string name
string slug UK
int parent_id FK
int sort_order
timestamp created_at
}
"documents" {
int id PK
string title
string slug UK
text content
int category_id FK
string visibility
int sort_order
int author_id FK
timestamp created_at
timestamp updated_at
}
"document_versions" {
int id PK
int document_id FK
text content
int author_id FK
timestamp created_at
}
"document_categories" ||--o{ "document_categories" : "parent"
"document_categories" ||--o{ "documents" : "has"
"documents" ||--o{ "document_versions" : "has"
```
### DDL
```sql
CREATE TABLE document_categories (
id INTEGER PRIMARY KEY AUTOINCREMENT,
name VARCHAR(100) NOT NULL,
slug VARCHAR(100) NOT NULL UNIQUE,
parent_id INTEGER REFERENCES document_categories(id) ON DELETE SET NULL,
sort_order INTEGER DEFAULT 0,
created_at DATETIME DEFAULT CURRENT_TIMESTAMP
);
CREATE TABLE documents (
id INTEGER PRIMARY KEY AUTOINCREMENT,
title VARCHAR(200) NOT NULL,
slug VARCHAR(200) NOT NULL UNIQUE,
content TEXT NOT NULL,
category_id INTEGER REFERENCES document_categories(id) ON DELETE SET NULL,
visibility VARCHAR(20) DEFAULT 'public' CHECK (visibility IN ('public', 'auth', 'admin')),
sort_order INTEGER DEFAULT 0,
author_id INTEGER NOT NULL,
created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
updated_at DATETIME DEFAULT CURRENT_TIMESTAMP
);
CREATE TABLE document_versions (
id INTEGER PRIMARY KEY AUTOINCREMENT,
document_id INTEGER NOT NULL REFERENCES documents(id) ON DELETE CASCADE,
content TEXT NOT NULL,
author_id INTEGER NOT NULL,
created_at DATETIME DEFAULT CURRENT_TIMESTAMP
);
CREATE INDEX idx_documents_slug ON documents(slug);
CREATE INDEX idx_documents_category ON documents(category_id);
CREATE INDEX idx_documents_visibility ON documents(visibility);
CREATE INDEX idx_document_versions_doc ON document_versions(document_id);
```
+18 -16
View File
@@ -1,22 +1,24 @@
FROM oven/bun:1@sha256:0733e50325078969732ebe3b15ce4c4be5082f18c4ac1a0f0ca4839c2e4e42a7 AS builder
WORKDIR /build
COPY web/default/package.json .
COPY web/default/bun.lock .
RUN bun install
COPY ./web/default .
COPY ./VERSION .
RUN DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat VERSION) bun run build
WORKDIR /build/web
COPY web/package.json web/bun.lock ./
COPY web/default/package.json ./default/package.json
COPY web/classic/package.json ./classic/package.json
RUN bun install --frozen-lockfile
COPY ./web/default ./default
COPY ./VERSION /build/VERSION
RUN cd default && DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat /build/VERSION) bun run build
FROM oven/bun:1@sha256:0733e50325078969732ebe3b15ce4c4be5082f18c4ac1a0f0ca4839c2e4e42a7 AS builder-classic
WORKDIR /build
COPY web/classic/package.json .
COPY web/classic/bun.lock .
RUN bun install
COPY ./web/classic .
COPY ./VERSION .
RUN VITE_REACT_APP_VERSION=$(cat VERSION) bun run build
WORKDIR /build/web
COPY web/package.json web/bun.lock ./
COPY web/default/package.json ./default/package.json
COPY web/classic/package.json ./classic/package.json
RUN bun install --frozen-lockfile
COPY ./web/classic ./classic
COPY ./VERSION /build/VERSION
RUN cd classic && VITE_REACT_APP_VERSION=$(cat /build/VERSION) bun run build
FROM golang:1.26.1-alpine@sha256:2389ebfa5b7f43eeafbd6be0c3700cc46690ef842ad962f6c5bd6be49ed82039 AS builder2
ENV GO111MODULE=on CGO_ENABLED=0
@@ -32,8 +34,8 @@ ADD go.mod go.sum ./
RUN go mod download
COPY . .
COPY --from=builder /build/dist ./web/default/dist
COPY --from=builder-classic /build/dist ./web/classic/dist
COPY --from=builder /build/web/default/dist ./web/default/dist
COPY --from=builder-classic /build/web/classic/dist ./web/classic/dist
RUN go build -ldflags "-s -w -X 'github.com/QuantumNous/new-api/common.Version=$(cat VERSION)'" -o new-api
FROM debian:bookworm-slim@sha256:f06537653ac770703bc45b4b113475bd402f451e85223f0f2837acbf89ab020a
+2 -2
View File
@@ -1,7 +1,7 @@
new-api Notices
new-api
Copyright (c) QuantumNous and contributors.
Copyright (c) modelstoken and contributors.
This project is licensed under the GNU Affero General Public License v3.0.
See LICENSE for the full project license terms.
@@ -19,7 +19,7 @@ Modified versions that present a user interface must also preserve a visible
link to the original project in a prominent about, legal, footer, or
attribution location:
https://github.com/QuantumNous/new-api
https://git.viaeon.com/admin/new-api
Modified versions must not misrepresent the origin of the software and must
mark their changes in accordance with AGPLv3 Section 7(c).
+1
View File
@@ -316,6 +316,7 @@ docker run --name new-api -d --restart always \
| `CRYPTO_SECRET` | Encryption secret (required for Redis) | - |
| `SQL_DSN` | Database connection string | - |
| `REDIS_CONN_STRING` | Redis connection string | - |
| `RELAY_IDLE_CONN_TIMEOUT` | Idle keep-alive timeout for relay HTTP clients, seconds. Defaults to Go standard library behavior; set `0` to disable | `90` |
| `STREAMING_TIMEOUT` | Streaming timeout (seconds) | `300` |
| `STREAM_SCANNER_MAX_BUFFER_MB` | Max per-line buffer (MB) for the stream scanner; increase when upstream sends huge image/base64 payloads | `64` |
| `MAX_REQUEST_BODY_MB` | Max request body size (MB, counted **after decompression**; prevents huge requests/zip bombs from exhausting memory). Exceeding it returns `413` | `32` |
+2 -1
View File
@@ -14,7 +14,7 @@ import (
var StartTime = time.Now().Unix() // unit: second
var Version = "v0.0.0" // this hard coding will be replaced automatically when building, no need to manually change
var SystemName = "New API"
var SystemName = "ModelsToken"
var Footer = ""
var Logo = ""
var TopUpLink = ""
@@ -170,6 +170,7 @@ var BatchUpdateInterval int
var RelayTimeout int // unit is second
var RelayIdleConnTimeout int // unit is second
var RelayMaxIdleConns int
var RelayMaxIdleConnsPerHost int
+2 -2
View File
@@ -37,7 +37,7 @@ func checkWriter(writer io.Writer) stringWriter {
// W3C Working Draft 29 October 2009
// http://www.w3.org/TR/2009/WD-eventsource-20091029/
var contentType = []string{"text/event-stream"}
var writeContentType = []string{"text/event-stream"}
var noCache = []string{"no-cache"}
var fieldReplacer = strings.NewReplacer(
@@ -79,7 +79,7 @@ func (r CustomEvent) WriteContentType(w http.ResponseWriter) {
r.Mutex.Lock()
defer r.Mutex.Unlock()
header := w.Header()
header["Content-Type"] = contentType
header["Content-Type"] = writeContentType
if _, exist := header["Cache-Control"]; !exist {
header["Cache-Control"] = noCache
+8 -4
View File
@@ -51,17 +51,21 @@ type themeAwareFileSystem struct {
}
func (t *themeAwareFileSystem) Exists(prefix string, path string) bool {
if GetTheme() == "classic" {
switch GetTheme() {
case "classic":
return t.classicFS.Exists(prefix, path)
default:
return t.defaultFS.Exists(prefix, path)
}
return t.defaultFS.Exists(prefix, path)
}
func (t *themeAwareFileSystem) Open(name string) (http.File, error) {
if GetTheme() == "classic" {
switch GetTheme() {
case "classic":
return t.classicFS.Open(name)
default:
return t.defaultFS.Open(name)
}
return t.defaultFS.Open(name)
}
func NewThemeAwareFS(defaultFS, classicFS static.ServeFileSystem) static.ServeFileSystem {
+19 -1
View File
@@ -110,11 +110,29 @@ func UnmarshalBodyReusable(c *gin.Context, v any) error {
if err != nil {
return err
}
contentType := c.Request.Header.Get("Content-Type")
// disk-backed JSON: stream-decode directly from the file to avoid
// materializing the entire payload back into a transient []byte
// (diskStorage.Bytes() would ReadFull the whole file into the heap).
if storage.IsDisk() && strings.HasPrefix(contentType, "application/json") {
if _, seekErr := storage.Seek(0, io.SeekStart); seekErr != nil {
return seekErr
}
if err := DecodeJson(storage, v); err != nil {
return err
}
if _, seekErr := storage.Seek(0, io.SeekStart); seekErr != nil {
return seekErr
}
c.Request.Body = io.NopCloser(storage)
return nil
}
requestBody, err := storage.Bytes()
if err != nil {
return err
}
contentType := c.Request.Header.Get("Content-Type")
if strings.HasPrefix(contentType, "application/json") {
err = Unmarshal(requestBody, v)
} else if strings.Contains(contentType, gin.MIMEPOSTForm) {
+4 -2
View File
@@ -102,6 +102,7 @@ func InitEnv() {
SyncFrequency = GetEnvOrDefault("SYNC_FREQUENCY", 60)
BatchUpdateInterval = GetEnvOrDefault("BATCH_UPDATE_INTERVAL", 5)
RelayTimeout = GetEnvOrDefault("RELAY_TIMEOUT", 0)
RelayIdleConnTimeout = GetEnvOrDefault("RELAY_IDLE_CONN_TIMEOUT", 90)
RelayMaxIdleConns = GetEnvOrDefault("RELAY_MAX_IDLE_CONNS", 500)
RelayMaxIdleConnsPerHost = GetEnvOrDefault("RELAY_MAX_IDLE_CONNS_PER_HOST", 100)
@@ -111,11 +112,11 @@ func InitEnv() {
// Initialize rate limit variables
GlobalApiRateLimitEnable = GetEnvOrDefaultBool("GLOBAL_API_RATE_LIMIT_ENABLE", true)
GlobalApiRateLimitNum = GetEnvOrDefault("GLOBAL_API_RATE_LIMIT", 180)
GlobalApiRateLimitNum = GetEnvOrDefault("GLOBAL_API_RATE_LIMIT", 360)
GlobalApiRateLimitDuration = int64(GetEnvOrDefault("GLOBAL_API_RATE_LIMIT_DURATION", 180))
GlobalWebRateLimitEnable = GetEnvOrDefaultBool("GLOBAL_WEB_RATE_LIMIT_ENABLE", true)
GlobalWebRateLimitNum = GetEnvOrDefault("GLOBAL_WEB_RATE_LIMIT", 60)
GlobalWebRateLimitNum = GetEnvOrDefault("GLOBAL_WEB_RATE_LIMIT", 120)
GlobalWebRateLimitDuration = int64(GetEnvOrDefault("GLOBAL_WEB_RATE_LIMIT_DURATION", 180))
CriticalRateLimitEnable = GetEnvOrDefaultBool("CRITICAL_RATE_LIMIT_ENABLE", true)
@@ -135,6 +136,7 @@ func initConstantEnv() {
constant.StreamScannerMaxBufferMB = GetEnvOrDefault("STREAM_SCANNER_MAX_BUFFER_MB", 128)
// MaxRequestBodyMB 请求体最大大小(解压后),用于防止超大请求/zip bomb导致内存暴涨
constant.MaxRequestBodyMB = GetEnvOrDefault("MAX_REQUEST_BODY_MB", 128)
constant.AnonymousRequestBodyLimitKB = GetEnvOrDefault("ANONYMOUS_REQUEST_BODY_LIMIT_KB", 512)
// ForceStreamOption 覆盖请求参数,强制返回usage信息
constant.ForceStreamOption = GetEnvOrDefaultBool("FORCE_STREAM_OPTION", true)
constant.CountToken = GetEnvOrDefaultBool("CountToken", true)
+13
View File
@@ -0,0 +1,13 @@
package common
import "github.com/QuantumNous/new-api/constant"
const defaultAnonymousRequestBodyLimitKB = 512
func GetAnonymousRequestBodyLimitBytes() int64 {
limitKB := constant.AnonymousRequestBodyLimitKB
if limitKB < 0 {
limitKB = defaultAnonymousRequestBodyLimitKB
}
return int64(limitKB) << 10
}
+11
View File
@@ -3,6 +3,7 @@ package common
import (
"encoding/base64"
"encoding/json"
"fmt"
"net/url"
"regexp"
"strconv"
@@ -20,6 +21,16 @@ var (
maskApiKeyPattern = regexp.MustCompile(`(['"]?)api_key:([^\s'"]+)(['"]?)`)
)
const LocalLogContentLimit = 2048
// LocalLogPreview limits log-only content unless debug logging is enabled.
func LocalLogPreview(content string) string {
if DebugEnabled || len(content) <= LocalLogContentLimit {
return content
}
return fmt.Sprintf("%s... [truncated, original_length=%d, limit=%d]", content[:LocalLogContentLimit], len(content), LocalLogContentLimit)
}
func GetStringIfEmpty(str string, defaultValue string) string {
if str == "" {
return defaultValue
+4
View File
@@ -206,4 +206,8 @@ var ChannelSpecialBases = map[string]ChannelSpecialBase{
ClaudeBaseURL: "https://ark.cn-beijing.volces.com/api/coding",
OpenAIBaseURL: "https://ark.cn-beijing.volces.com/api/coding/v3",
},
"tencent-coding-plan": {
ClaudeBaseURL: "https://api.lkeap.cloud.tencent.com/coding",
OpenAIBaseURL: "https://api.lkeap.cloud.tencent.com/coding/v3",
},
}
+1
View File
@@ -10,6 +10,7 @@ var GetMediaToken bool
var GetMediaTokenNotStream bool
var UpdateTask bool
var MaxRequestBodyMB int
var AnonymousRequestBodyLimitKB int
var AzureDefaultAPIVersion string
var NotifyLimitCount int
var NotificationLimitDurationMinute int
+34 -8
View File
@@ -57,7 +57,24 @@ func normalizeChannelTestEndpoint(channel *model.Channel, modelName, endpointTyp
return normalized
}
func testChannel(channel *model.Channel, testModel string, endpointType string, isStream bool) testResult {
func resolveChannelTestUserID(c *gin.Context) (int, error) {
if c != nil {
if userID := c.GetInt("id"); userID > 0 {
return userID, nil
}
}
var rootUser model.User
if err := model.DB.Select("id").Where("role = ?", common.RoleRootUser).First(&rootUser).Error; err != nil {
return 0, fmt.Errorf("failed to resolve channel test user: %w", err)
}
if rootUser.Id == 0 {
return 0, errors.New("failed to resolve channel test user")
}
return rootUser.Id, nil
}
func testChannel(channel *model.Channel, testUserID int, testModel string, endpointType string, isStream bool) testResult {
tik := time.Now()
var unsupportedTestChannelTypes = []int{
constant.ChannelTypeMidjourney,
@@ -143,7 +160,7 @@ func testChannel(channel *model.Channel, testModel string, endpointType string,
Header: make(http.Header),
}
cache, err := model.GetUserCache(1)
cache, err := model.GetUserCache(testUserID)
if err != nil {
return testResult{
localErr: err,
@@ -151,13 +168,13 @@ func testChannel(channel *model.Channel, testModel string, endpointType string,
}
}
cache.WriteContext(c)
c.Set("id", 1)
c.Set("id", testUserID)
//c.Request.Header.Set("Authorization", "Bearer "+channel.Key)
c.Request.Header.Set("Content-Type", "application/json")
c.Set("channel", channel.Type)
c.Set("base_url", channel.GetBaseURL())
group, _ := model.GetUserGroup(1, false)
group, _ := model.GetUserGroup(testUserID, false)
c.Set("group", group)
newAPIError := middleware.SetupContextForSelectedChannel(c, channel, testModel)
@@ -484,7 +501,7 @@ func testChannel(channel *model.Channel, testModel string, endpointType string,
milliseconds := tok.Sub(tik).Milliseconds()
consumedTime := float64(milliseconds) / 1000.0
other := buildTestLogOther(c, info, priceData, usage, tieredResult)
model.RecordConsumeLog(c, 1, model.RecordConsumeLogParams{
model.RecordConsumeLog(c, testUserID, model.RecordConsumeLogParams{
ChannelId: channel.Id,
PromptTokens: usage.PromptTokens,
CompletionTokens: usage.CompletionTokens,
@@ -797,7 +814,7 @@ func buildTestRequest(model string, endpointType string, channel *model.Channel,
testRequest.StreamOptions = &dto.StreamOptions{IncludeUsage: true}
}
if strings.HasPrefix(model, "o") {
if dto.IsOpenAIReasoningOModel(model) {
testRequest.MaxCompletionTokens = lo.ToPtr(uint(16))
} else if strings.Contains(model, "thinking") {
if !strings.Contains(model, "claude") {
@@ -834,8 +851,13 @@ func TestChannel(c *gin.Context) {
testModel := c.Query("model")
endpointType := c.Query("endpoint_type")
isStream, _ := strconv.ParseBool(c.Query("stream"))
testUserID, err := resolveChannelTestUserID(c)
if err != nil {
common.ApiError(c, err)
return
}
tik := time.Now()
result := testChannel(channel, testModel, endpointType, isStream)
result := testChannel(channel, testUserID, testModel, endpointType, isStream)
if result.localErr != nil {
resp := gin.H{
"success": false,
@@ -872,6 +894,10 @@ var testAllChannelsLock sync.Mutex
var testAllChannelsRunning bool = false
func testAllChannels(notify bool) error {
testUserID, err := resolveChannelTestUserID(nil)
if err != nil {
return err
}
testAllChannelsLock.Lock()
if testAllChannelsRunning {
@@ -902,7 +928,7 @@ func testAllChannels(notify bool) error {
}
isChannelEnabled := channel.Status == common.ChannelStatusEnabled
tik := time.Now()
result := testChannel(channel, "", "", shouldUseStreamForAutomaticChannelTest(channel))
result := testChannel(channel, testUserID, "", "", shouldUseStreamForAutomaticChannelTest(channel))
tok := time.Now()
milliseconds := tok.Sub(tik).Milliseconds()
+1 -1
View File
@@ -1218,7 +1218,7 @@ func CopyChannel(c *gin.Context) {
}
// insert
if err := model.BatchInsertChannels([]model.Channel{clone}); err != nil {
if err := clone.Insert(); err != nil {
common.SysError("failed to clone channel: " + err.Error())
c.JSON(http.StatusOK, gin.H{"success": false, "message": "复制渠道失败,请稍后重试"})
return
+11
View File
@@ -69,3 +69,14 @@ func TestBuildTestLogOtherInjectsTieredInfo(t *testing.T) {
require.Equal(t, "base", other["matched_tier"])
require.NotEmpty(t, other["expr_b64"])
}
func TestResolveChannelTestUserIDUsesRequestUser(t *testing.T) {
gin.SetMode(gin.TestMode)
ctx, _ := gin.CreateTestContext(httptest.NewRecorder())
ctx.Set("id", 2)
userID, err := resolveChannelTestUserID(ctx)
require.NoError(t, err)
require.Equal(t, 2, userID)
}
+5 -1
View File
@@ -312,7 +312,11 @@ func fetchChannelUpstreamModelIDs(channel *model.Channel) ([]string, error) {
url = fmt.Sprintf("%s/v1/models", baseURL)
}
default:
url = fmt.Sprintf("%s/v1/models", baseURL)
if plan, ok := constant.ChannelSpecialBases[baseURL]; ok && plan.OpenAIBaseURL != "" {
url = fmt.Sprintf("%s/models", plan.OpenAIBaseURL)
} else {
url = fmt.Sprintf("%s/v1/models", baseURL)
}
}
key, _, apiErr := channel.GetNextEnabledKey()
+256
View File
@@ -0,0 +1,256 @@
package controller
import (
"net/http"
"strconv"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/model"
"github.com/gin-gonic/gin"
)
// GetCategories 获取文档分类列表(公开)
func GetCategories(c *gin.Context) {
categories, err := model.GetDocumentCategories()
if err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, categories)
}
// CreateCategory 创建文档分类(管理员)
func CreateCategory(c *gin.Context) {
var category model.DocumentCategory
if err := c.ShouldBindJSON(&category); err != nil {
common.ApiError(c, err)
return
}
if category.Name == "" {
common.ApiErrorMsg(c, "分类名称不能为空")
return
}
if category.Slug == "" {
common.ApiErrorMsg(c, "分类标识不能为空")
return
}
if err := model.CreateDocumentCategory(&category); err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, &category)
}
// UpdateCategory 更新文档分类(管理员)
func UpdateCategory(c *gin.Context) {
id, err := strconv.Atoi(c.Param("id"))
if err != nil {
common.ApiError(c, err)
return
}
var category model.DocumentCategory
if err := c.ShouldBindJSON(&category); err != nil {
common.ApiError(c, err)
return
}
category.Id = id
if err := model.UpdateDocumentCategory(&category); err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, &category)
}
// DeleteCategory 删除文档分类(管理员)
func DeleteCategory(c *gin.Context) {
id, err := strconv.Atoi(c.Param("id"))
if err != nil {
common.ApiError(c, err)
return
}
if err := model.DeleteDocumentCategory(id); err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, nil)
}
// GetDocuments 获取文档列表(公开,根据认证状态过滤可见性)
func GetDocuments(c *gin.Context) {
keyword := c.Query("keyword")
categoryIdStr := c.Query("category_id")
var categoryId *int
if categoryIdStr != "" {
id, err := strconv.Atoi(categoryIdStr)
if err == nil {
categoryId = &id
}
}
pageInfo := common.GetPageQuery(c)
// 根据用户认证状态决定可见性过滤
visibility := c.Query("visibility")
role := c.GetInt("role")
var documents []*model.Document
var total int64
var err error
if role >= common.RoleAdminUser {
// 管理员可看所有
documents, total, err = model.GetDocuments(keyword, visibility, categoryId, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
} else if role >= common.RoleCommonUser {
// 普通用户只能看 public 和 auth
if visibility == "public" || visibility == "auth" {
documents, total, err = model.GetDocuments(keyword, visibility, categoryId, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
} else {
documents, total, err = model.GetDocumentsByVisibility(keyword, []string{"public", "auth"}, categoryId, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
}
} else {
// 未登录用户只能看 public
documents, total, err = model.GetDocuments(keyword, "public", categoryId, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
}
if err != nil {
common.ApiError(c, err)
return
}
pageInfo.SetTotal(int(total))
pageInfo.SetItems(documents)
common.ApiSuccess(c, pageInfo)
}
// GetDocument 获取单个文档(根据可见性检查权限)
func GetDocument(c *gin.Context) {
slug := c.Param("slug")
doc, err := model.GetDocumentBySlug(slug)
if err != nil {
common.ApiError(c, err)
return
}
// 检查可见性权限
role := c.GetInt("role")
switch doc.Visibility {
case "admin":
if role < common.RoleAdminUser {
c.JSON(http.StatusOK, gin.H{
"success": false,
"message": "无权访问该文档",
})
return
}
case "auth":
if role < common.RoleCommonUser {
c.JSON(http.StatusOK, gin.H{
"success": false,
"message": "请先登录后查看该文档",
})
return
}
}
common.ApiSuccess(c, doc)
}
// CreateDocument 创建文档(管理员)
func CreateDocument(c *gin.Context) {
var doc model.Document
if err := c.ShouldBindJSON(&doc); err != nil {
common.ApiError(c, err)
return
}
if doc.Title == "" {
common.ApiErrorMsg(c, "文档标题不能为空")
return
}
if doc.Slug == "" {
common.ApiErrorMsg(c, "文档标识不能为空")
return
}
if doc.Content == "" {
common.ApiErrorMsg(c, "文档内容不能为空")
return
}
if doc.Visibility == "" {
doc.Visibility = "public"
}
doc.AuthorId = c.GetInt("id")
if err := model.CreateDocument(&doc); err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, &doc)
}
// UpdateDocument 更新文档(管理员,自动创建版本记录)
func UpdateDocument(c *gin.Context) {
id, err := strconv.Atoi(c.Param("id"))
if err != nil {
common.ApiError(c, err)
return
}
var doc model.Document
if err := c.ShouldBindJSON(&doc); err != nil {
common.ApiError(c, err)
return
}
doc.Id = id
// 获取旧文档内容,自动创建版本记录
oldDoc, err := model.GetDocumentById(id)
if err != nil {
common.ApiError(c, err)
return
}
version := &model.DocumentVersion{
DocumentId: oldDoc.Id,
Content: oldDoc.Content,
AuthorId: oldDoc.AuthorId,
}
if err := model.CreateDocumentVersion(version); err != nil {
common.ApiError(c, err)
return
}
if err := model.UpdateDocument(&doc); err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, &doc)
}
// DeleteDocument 删除文档(管理员)
func DeleteDocument(c *gin.Context) {
id, err := strconv.Atoi(c.Param("id"))
if err != nil {
common.ApiError(c, err)
return
}
if err := model.DeleteDocument(id); err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, nil)
}
// GetDocumentVersions 获取文档版本历史(管理员)
func GetDocumentVersions(c *gin.Context) {
id, err := strconv.Atoi(c.Param("id"))
if err != nil {
common.ApiError(c, err)
return
}
pageInfo := common.GetPageQuery(c)
versions, total, err := model.GetDocumentVersions(id, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
if err != nil {
common.ApiError(c, err)
return
}
pageInfo.SetTotal(int(total))
pageInfo.SetItems(versions)
common.ApiSuccess(c, pageInfo)
}
+1
View File
@@ -88,6 +88,7 @@ func GetStatus(c *gin.Context) {
"demo_site_enabled": operation_setting.DemoSiteEnabled,
"self_use_mode_enabled": operation_setting.SelfUseModeEnabled,
"register_enabled": common.RegisterEnabled,
"password_login_enabled": common.PasswordLoginEnabled,
"password_register_enabled": common.PasswordRegisterEnabled,
"default_use_auto_group": setting.DefaultUseAutoGroup,
+120 -43
View File
@@ -3,6 +3,7 @@ package controller
import (
"fmt"
"net/http"
"strings"
"time"
"github.com/QuantumNous/new-api/common"
@@ -109,9 +110,102 @@ func init() {
})
}
func ListModels(c *gin.Context, modelType int) {
userOpenAiModels := make([]dto.OpenAIModels, 0)
func channelOwnerName(channelType int) string {
apiType, success := common.ChannelType2APIType(channelType)
if !success {
return strings.ToLower(constant.GetChannelTypeName(channelType))
}
adaptor := relay.GetAdaptor(apiType)
if adaptor == nil {
return strings.ToLower(constant.GetChannelTypeName(channelType))
}
adaptor.Init(&relaycommon.RelayInfo{ChannelMeta: &relaycommon.ChannelMeta{
ChannelType: channelType,
}})
if name := strings.TrimSpace(adaptor.GetChannelName()); name != "" {
return name
}
return strings.ToLower(constant.GetChannelTypeName(channelType))
}
func getPreferredModelOwners(modelNames []string, groups []string) map[string]string {
channelTypes, err := model.GetPreferredModelOwnerChannelTypes(modelNames, groups)
if err != nil {
common.SysLog(fmt.Sprintf("GetPreferredModelOwnerChannelTypes error: %v", err))
return map[string]string{}
}
ownerByChannelType := make(map[int]string)
owners := make(map[string]string, len(channelTypes))
for modelName, channelType := range channelTypes {
owner, ok := ownerByChannelType[channelType]
if !ok {
owner = channelOwnerName(channelType)
ownerByChannelType[channelType] = owner
}
if owner != "" {
owners[modelName] = owner
}
}
return owners
}
func buildOpenAIModel(modelName string, ownerByModel map[string]string) dto.OpenAIModels {
var oaiModel dto.OpenAIModels
if staticModel, ok := openAIModelsMap[modelName]; ok {
oaiModel = staticModel
} else {
oaiModel = dto.OpenAIModels{
Id: modelName,
Object: "model",
Created: 1626777600,
OwnedBy: "custom",
}
}
if owner, ok := ownerByModel[modelName]; ok && owner != "" {
oaiModel.OwnedBy = owner
}
oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(modelName)
return oaiModel
}
type modelListGroups struct {
userGroup string
tokenGroup string
ownerGroups []string
}
func getModelListGroups(c *gin.Context) (modelListGroups, error) {
tokenGroup := common.GetContextKeyString(c, constant.ContextKeyTokenGroup)
userGroup := common.GetContextKeyString(c, constant.ContextKeyUserGroup)
if userGroup == "" && (tokenGroup == "" || tokenGroup == "auto") {
var err error
userGroup, err = model.GetUserGroup(c.GetInt("id"), false)
if err != nil {
return modelListGroups{}, err
}
}
if tokenGroup == "auto" {
return modelListGroups{
userGroup: userGroup,
tokenGroup: tokenGroup,
ownerGroups: service.GetUserAutoGroup(userGroup),
}, nil
}
group := userGroup
if tokenGroup != "" {
group = tokenGroup
}
return modelListGroups{
userGroup: userGroup,
tokenGroup: tokenGroup,
ownerGroups: []string{group},
}, nil
}
func ListModels(c *gin.Context, modelType int) {
acceptUnsetRatioModel := operation_setting.SelfUseModeEnabled
if !acceptUnsetRatioModel {
userId := c.GetInt("id")
@@ -123,6 +217,16 @@ func ListModels(c *gin.Context, modelType int) {
}
}
userModelNames := make([]string, 0)
groups, err := getModelListGroups(c)
if err != nil {
c.JSON(http.StatusOK, gin.H{
"success": false,
"message": "get user group failed",
})
return
}
ownerGroups := groups.ownerGroups
modelLimitEnable := common.GetContextKeyBool(c, constant.ContextKeyTokenModelLimitEnabled)
if modelLimitEnable {
s, ok := common.GetContextKey(c, constant.ContextKeyTokenModelLimit)
@@ -138,37 +242,12 @@ func ListModels(c *gin.Context, modelType int) {
continue
}
}
if oaiModel, ok := openAIModelsMap[allowModel]; ok {
oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(allowModel)
userOpenAiModels = append(userOpenAiModels, oaiModel)
} else {
userOpenAiModels = append(userOpenAiModels, dto.OpenAIModels{
Id: allowModel,
Object: "model",
Created: 1626777600,
OwnedBy: "custom",
SupportedEndpointTypes: model.GetModelSupportEndpointTypes(allowModel),
})
}
userModelNames = append(userModelNames, allowModel)
}
} else {
userId := c.GetInt("id")
userGroup, err := model.GetUserGroup(userId, false)
if err != nil {
c.JSON(http.StatusOK, gin.H{
"success": false,
"message": "get user group failed",
})
return
}
group := userGroup
tokenGroup := common.GetContextKeyString(c, constant.ContextKeyTokenGroup)
if tokenGroup != "" {
group = tokenGroup
}
var models []string
if tokenGroup == "auto" {
for _, autoGroup := range service.GetUserAutoGroup(userGroup) {
if groups.tokenGroup == "auto" {
for _, autoGroup := range ownerGroups {
groupModels := model.GetGroupEnabledModels(autoGroup)
for _, g := range groupModels {
if !common.StringsContains(models, g) {
@@ -177,7 +256,7 @@ func ListModels(c *gin.Context, modelType int) {
}
}
} else {
models = model.GetGroupEnabledModels(group)
models = model.GetGroupEnabledModels(ownerGroups[0])
}
for _, modelName := range models {
if !acceptUnsetRatioModel {
@@ -185,21 +264,19 @@ func ListModels(c *gin.Context, modelType int) {
continue
}
}
if oaiModel, ok := openAIModelsMap[modelName]; ok {
oaiModel.SupportedEndpointTypes = model.GetModelSupportEndpointTypes(modelName)
userOpenAiModels = append(userOpenAiModels, oaiModel)
} else {
userOpenAiModels = append(userOpenAiModels, dto.OpenAIModels{
Id: modelName,
Object: "model",
Created: 1626777600,
OwnedBy: "custom",
SupportedEndpointTypes: model.GetModelSupportEndpointTypes(modelName),
})
}
userModelNames = append(userModelNames, modelName)
}
}
ownerByModel := map[string]string{}
if len(ownerGroups) > 0 {
ownerByModel = getPreferredModelOwners(userModelNames, ownerGroups)
}
userOpenAiModels := make([]dto.OpenAIModels, 0, len(userModelNames))
for _, modelName := range userModelNames {
userOpenAiModels = append(userOpenAiModels, buildOpenAIModel(modelName, ownerByModel))
}
switch modelType {
case constant.ChannelTypeAnthropic:
useranthropicModels := make([]dto.AnthropicModel, len(userOpenAiModels))
+85
View File
@@ -0,0 +1,85 @@
package controller
import (
"net/http/httptest"
"testing"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/constant"
"github.com/gin-gonic/gin"
"github.com/stretchr/testify/require"
)
func TestChannelOwnerNameUsesAdaptorChannelName(t *testing.T) {
tests := []struct {
name string
channelType int
expected string
}{
{
name: "openai",
channelType: constant.ChannelTypeOpenAI,
expected: "openai",
},
{
name: "codex",
channelType: constant.ChannelTypeCodex,
expected: "codex",
},
{
name: "openrouter",
channelType: constant.ChannelTypeOpenRouter,
expected: "openrouter",
},
{
name: "azure fallback",
channelType: constant.ChannelTypeAzure,
expected: "azure",
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
require.Equal(t, tt.expected, channelOwnerName(tt.channelType))
})
}
}
func TestBuildOpenAIModelOverridesOwnedBy(t *testing.T) {
modelItem := buildOpenAIModel("gpt-5.4", map[string]string{"gpt-5.4": "openai"})
require.Equal(t, "gpt-5.4", modelItem.Id)
require.Equal(t, "openai", modelItem.OwnedBy)
}
func TestBuildOpenAIModelFallsBackToCustomForUnknownModels(t *testing.T) {
modelItem := buildOpenAIModel("custom-test-model", nil)
require.Equal(t, "custom-test-model", modelItem.Id)
require.Equal(t, "custom", modelItem.OwnedBy)
}
func TestGetModelListGroupsUsesUserGroupWhenTokenGroupIsEmpty(t *testing.T) {
gin.SetMode(gin.TestMode)
ctx, _ := gin.CreateTestContext(httptest.NewRecorder())
common.SetContextKey(ctx, constant.ContextKeyUserGroup, "default")
groups, err := getModelListGroups(ctx)
require.NoError(t, err)
require.Equal(t, "default", groups.userGroup)
require.Empty(t, groups.tokenGroup)
require.Equal(t, []string{"default"}, groups.ownerGroups)
}
func TestGetModelListGroupsUsesExplicitTokenGroup(t *testing.T) {
gin.SetMode(gin.TestMode)
ctx, _ := gin.CreateTestContext(httptest.NewRecorder())
common.SetContextKey(ctx, constant.ContextKeyUserGroup, "default")
common.SetContextKey(ctx, constant.ContextKeyTokenGroup, "vip")
groups, err := getModelListGroups(ctx)
require.NoError(t, err)
require.Equal(t, "default", groups.userGroup)
require.Equal(t, "vip", groups.tokenGroup)
require.Equal(t, []string{"vip"}, groups.ownerGroups)
}
+2 -23
View File
@@ -7,7 +7,6 @@ import (
"strings"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/i18n"
"github.com/QuantumNous/new-api/model"
"github.com/QuantumNous/new-api/setting"
"github.com/QuantumNous/new-api/setting/console_setting"
@@ -29,10 +28,6 @@ var completionRatioMetaOptionKeys = []string{
"AudioCompletionRatio",
}
func isPaymentComplianceOptionKey(key string) bool {
return strings.HasPrefix(key, "payment_setting.compliance_")
}
func isPositiveOptionValue(value string) bool {
intValue, err := strconv.Atoi(strings.TrimSpace(value))
if err == nil {
@@ -42,15 +37,6 @@ func isPositiveOptionValue(value string) bool {
return err == nil && floatValue > 0
}
func isVisiblePublicKeyOption(key string) bool {
switch key {
case "WaffoPancakeWebhookPublicKey", "WaffoPancakeWebhookTestKey":
return true
default:
return false
}
}
func collectModelNamesFromOptionValue(raw string, modelNames map[string]struct{}) {
if strings.TrimSpace(raw) == "" {
return
@@ -95,7 +81,7 @@ func GetOptions(c *gin.Context) {
strings.HasSuffix(k, "Key") ||
strings.HasSuffix(k, "secret") ||
strings.HasSuffix(k, "api_key")
if isSensitiveKey && !isVisiblePublicKeyOption(k) {
if isSensitiveKey {
continue
}
options = append(options, &model.Option{
@@ -148,15 +134,8 @@ func UpdateOption(c *gin.Context) {
}
switch option.Key {
case "QuotaForInviter", "QuotaForInvitee":
if isPositiveOptionValue(option.Value.(string)) && !operation_setting.IsPaymentComplianceConfirmed() {
common.ApiErrorI18n(c, i18n.MsgPaymentComplianceRequired)
return
}
// no compliance check needed
default:
if isPaymentComplianceOptionKey(option.Key) {
common.ApiErrorMsg(c, "合规确认字段不允许通过通用设置接口修改")
return
}
}
switch option.Key {
case "GitHubOAuthEnabled":
-82
View File
@@ -1,82 +0,0 @@
package controller
import (
"fmt"
"net/http"
"strconv"
"time"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/i18n"
"github.com/QuantumNous/new-api/logger"
"github.com/QuantumNous/new-api/model"
"github.com/QuantumNous/new-api/setting/operation_setting"
"github.com/gin-gonic/gin"
)
type PaymentComplianceRequest struct {
Confirmed bool `json:"confirmed"`
}
func requirePaymentCompliance(c *gin.Context) bool {
if !operation_setting.IsPaymentComplianceConfirmed() {
common.ApiErrorI18n(c, i18n.MsgPaymentComplianceRequired)
return false
}
return true
}
func ConfirmPaymentCompliance(c *gin.Context) {
if c.GetBool("use_access_token") {
c.JSON(http.StatusForbidden, gin.H{
"success": false,
"message": "This operation requires dashboard session authentication. API access token is not allowed.",
})
return
}
var req PaymentComplianceRequest
if err := common.DecodeJson(c.Request.Body, &req); err != nil {
common.ApiErrorMsg(c, "参数错误")
return
}
if !req.Confirmed {
common.ApiErrorMsg(c, "请确认合规声明")
return
}
now := time.Now().Unix()
userId := c.GetInt("id")
clientIP := c.ClientIP()
updates := map[string]string{
"payment_setting.compliance_confirmed": "true",
"payment_setting.compliance_terms_version": operation_setting.CurrentComplianceTermsVersion,
"payment_setting.compliance_confirmed_at": strconv.FormatInt(now, 10),
"payment_setting.compliance_confirmed_by": strconv.Itoa(userId),
"payment_setting.compliance_confirmed_ip": clientIP,
}
for key, value := range updates {
if err := model.UpdateOption(key, value); err != nil {
common.ApiError(c, err)
return
}
}
logger.LogInfo(c.Request.Context(), fmt.Sprintf(
"payment compliance confirmed user_id=%d ip=%s terms_version=%s confirmed_at=%d",
userId,
clientIP,
operation_setting.CurrentComplianceTermsVersion,
now,
))
common.ApiSuccess(c, gin.H{
"confirmed": true,
"terms_version": operation_setting.CurrentComplianceTermsVersion,
"confirmed_at": now,
"confirmed_by": userId,
})
}
+2 -32
View File
@@ -7,14 +7,7 @@ import (
"github.com/QuantumNous/new-api/setting/operation_setting"
)
func isPaymentComplianceConfirmed() bool {
return operation_setting.IsPaymentComplianceConfirmed()
}
func isStripeTopUpEnabled() bool {
if !isPaymentComplianceConfirmed() {
return false
}
return strings.TrimSpace(setting.StripeApiSecret) != "" &&
strings.TrimSpace(setting.StripeWebhookSecret) != "" &&
strings.TrimSpace(setting.StripePriceId) != ""
@@ -29,9 +22,6 @@ func isStripeWebhookEnabled() bool {
}
func isCreemTopUpEnabled() bool {
if !isPaymentComplianceConfirmed() {
return false
}
products := strings.TrimSpace(setting.CreemProducts)
return strings.TrimSpace(setting.CreemApiKey) != "" &&
products != "" &&
@@ -47,9 +37,6 @@ func isCreemWebhookEnabled() bool {
}
func isWaffoTopUpEnabled() bool {
if !isPaymentComplianceConfirmed() {
return false
}
if !setting.WaffoEnabled {
return false
}
@@ -74,27 +61,13 @@ func isWaffoWebhookEnabled() bool {
}
func isWaffoPancakeTopUpEnabled() bool {
if !isPaymentComplianceConfirmed() {
return false
}
if !setting.WaffoPancakeEnabled {
return false
}
return isWaffoPancakeWebhookConfigured() &&
strings.TrimSpace(setting.WaffoPancakeMerchantID) != "" &&
return strings.TrimSpace(setting.WaffoPancakeMerchantID) != "" &&
strings.TrimSpace(setting.WaffoPancakePrivateKey) != "" &&
strings.TrimSpace(setting.WaffoPancakeStoreID) != "" &&
strings.TrimSpace(setting.WaffoPancakeProductID) != ""
}
func isWaffoPancakeWebhookConfigured() bool {
currentWebhookKey := strings.TrimSpace(setting.WaffoPancakeWebhookPublicKey)
if setting.WaffoPancakeSandbox {
currentWebhookKey = strings.TrimSpace(setting.WaffoPancakeWebhookTestKey)
}
return currentWebhookKey != ""
return isWaffoPancakeTopUpEnabled()
}
func isWaffoPancakeWebhookEnabled() bool {
@@ -102,9 +75,6 @@ func isWaffoPancakeWebhookEnabled() bool {
}
func isEpayTopUpEnabled() bool {
if !isPaymentComplianceConfirmed() {
return false
}
return isEpayWebhookConfigured() && len(operation_setting.PayMethods) > 0
}
@@ -8,21 +8,7 @@ import (
"github.com/stretchr/testify/require"
)
func confirmPaymentComplianceForTest(t *testing.T) {
t.Helper()
paymentSetting := operation_setting.GetPaymentSetting()
originalConfirmed := paymentSetting.ComplianceConfirmed
originalTermsVersion := paymentSetting.ComplianceTermsVersion
t.Cleanup(func() {
paymentSetting.ComplianceConfirmed = originalConfirmed
paymentSetting.ComplianceTermsVersion = originalTermsVersion
})
paymentSetting.ComplianceConfirmed = true
paymentSetting.ComplianceTermsVersion = operation_setting.CurrentComplianceTermsVersion
}
func TestStripeWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
confirmPaymentComplianceForTest(t)
originalAPISecret := setting.StripeApiSecret
originalWebhookSecret := setting.StripeWebhookSecret
originalPriceID := setting.StripePriceId
@@ -45,7 +31,6 @@ func TestStripeWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
}
func TestCreemWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
confirmPaymentComplianceForTest(t)
originalAPIKey := setting.CreemApiKey
originalProducts := setting.CreemProducts
originalWebhookSecret := setting.CreemWebhookSecret
@@ -68,7 +53,6 @@ func TestCreemWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
}
func TestWaffoWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
confirmPaymentComplianceForTest(t)
originalEnabled := setting.WaffoEnabled
originalSandbox := setting.WaffoSandbox
originalAPIKey := setting.WaffoApiKey
@@ -113,52 +97,32 @@ func TestWaffoWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
}
func TestWaffoPancakeWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
confirmPaymentComplianceForTest(t)
originalEnabled := setting.WaffoPancakeEnabled
originalSandbox := setting.WaffoPancakeSandbox
originalMerchantID := setting.WaffoPancakeMerchantID
originalPrivateKey := setting.WaffoPancakePrivateKey
originalWebhookPublicKey := setting.WaffoPancakeWebhookPublicKey
originalWebhookTestKey := setting.WaffoPancakeWebhookTestKey
originalStoreID := setting.WaffoPancakeStoreID
originalProductID := setting.WaffoPancakeProductID
t.Cleanup(func() {
setting.WaffoPancakeEnabled = originalEnabled
setting.WaffoPancakeSandbox = originalSandbox
setting.WaffoPancakeMerchantID = originalMerchantID
setting.WaffoPancakePrivateKey = originalPrivateKey
setting.WaffoPancakeWebhookPublicKey = originalWebhookPublicKey
setting.WaffoPancakeWebhookTestKey = originalWebhookTestKey
setting.WaffoPancakeStoreID = originalStoreID
setting.WaffoPancakeProductID = originalProductID
})
setting.WaffoPancakeEnabled = true
setting.WaffoPancakeSandbox = false
setting.WaffoPancakeMerchantID = "merchant"
setting.WaffoPancakeMerchantID = ""
setting.WaffoPancakePrivateKey = "private"
setting.WaffoPancakeStoreID = "store"
setting.WaffoPancakeProductID = "product"
setting.WaffoPancakeWebhookPublicKey = ""
require.False(t, isWaffoPancakeWebhookEnabled())
setting.WaffoPancakeWebhookPublicKey = "public"
setting.WaffoPancakeMerchantID = "merchant"
require.True(t, isWaffoPancakeWebhookEnabled())
setting.WaffoPancakeEnabled = false
setting.WaffoPancakeProductID = ""
require.False(t, isWaffoPancakeWebhookEnabled())
setting.WaffoPancakeEnabled = true
setting.WaffoPancakeSandbox = true
setting.WaffoPancakeWebhookTestKey = ""
setting.WaffoPancakeProductID = "product"
setting.WaffoPancakePrivateKey = ""
require.False(t, isWaffoPancakeWebhookEnabled())
setting.WaffoPancakeWebhookTestKey = "test_public"
require.True(t, isWaffoPancakeWebhookEnabled())
}
func TestEpayWebhookEnabledRequiresTopUpAndWebhookConfig(t *testing.T) {
confirmPaymentComplianceForTest(t)
originalPayAddress := operation_setting.PayAddress
originalEpayID := operation_setting.EpayId
originalEpayKey := operation_setting.EpayKey
-6
View File
@@ -8,7 +8,6 @@ import (
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/i18n"
"github.com/QuantumNous/new-api/model"
"github.com/QuantumNous/new-api/setting/operation_setting"
"github.com/gin-gonic/gin"
)
@@ -60,11 +59,6 @@ func GetRedemption(c *gin.Context) {
}
func AddRedemption(c *gin.Context) {
if !operation_setting.IsPaymentComplianceConfirmed() {
common.ApiErrorI18n(c, i18n.MsgPaymentComplianceRequired)
return
}
redemption := model.Redemption{}
err := c.ShouldBindJSON(&redemption)
if err != nil {
+2 -2
View File
@@ -88,7 +88,7 @@ func Relay(c *gin.Context, relayFormat types.RelayFormat) {
defer func() {
if newAPIError != nil {
logger.LogError(c, fmt.Sprintf("relay error: %s", newAPIError.Error()))
logger.LogError(c, fmt.Sprintf("relay error: %s", common.LocalLogPreview(newAPIError.Error())))
newAPIError.SetMessage(common.MessageWithRequestId(newAPIError.Error(), requestId))
switch relayFormat {
case types.RelayFormatOpenAIRealtime:
@@ -354,7 +354,7 @@ func shouldRetry(c *gin.Context, openaiErr *types.NewAPIError, retryTimes int) b
}
func processChannelError(c *gin.Context, channelError types.ChannelError, err *types.NewAPIError) {
logger.LogError(c, fmt.Sprintf("channel error (channel #%d, status code: %d): %s", channelError.ChannelId, err.StatusCode, err.Error()))
logger.LogError(c, fmt.Sprintf("channel error (channel #%d, status code: %d): %s", channelError.ChannelId, err.StatusCode, common.LocalLogPreview(err.Error())))
// 不要使用context获取渠道信息,异步处理时可能会出现渠道信息不一致的情况
// do not use context to get channel info, there may be inconsistent channel info when processing asynchronously
if service.ShouldDisableChannel(err) && channelError.AutoBan {
+28 -26
View File
@@ -6,7 +6,6 @@ import (
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/model"
"github.com/QuantumNous/new-api/setting/operation_setting"
"github.com/QuantumNous/new-api/setting/ratio_setting"
"github.com/gin-gonic/gin"
"gorm.io/gorm"
@@ -22,14 +21,13 @@ type BillingPreferenceRequest struct {
BillingPreference string `json:"billing_preference"`
}
type SubscriptionBalancePayRequest struct {
PlanId int `json:"plan_id"`
}
// ---- User APIs ----
func GetSubscriptionPlans(c *gin.Context) {
if !operation_setting.IsPaymentComplianceConfirmed() {
common.ApiSuccess(c, []SubscriptionPlanDTO{})
return
}
var plans []model.SubscriptionPlan
if err := model.DB.Where("enabled = ?", true).Order("sort_order desc, id desc").Find(&plans).Error; err != nil {
common.ApiError(c, err)
@@ -37,6 +35,7 @@ func GetSubscriptionPlans(c *gin.Context) {
}
result := make([]SubscriptionPlanDTO, 0, len(plans))
for _, p := range plans {
p.NormalizeDefaults()
result = append(result, SubscriptionPlanDTO{
Plan: p,
})
@@ -92,6 +91,21 @@ func UpdateSubscriptionPreference(c *gin.Context) {
common.ApiSuccess(c, gin.H{"billing_preference": pref})
}
func SubscriptionRequestBalancePay(c *gin.Context) {
userId := c.GetInt("id")
var req SubscriptionBalancePayRequest
if err := c.ShouldBindJSON(&req); err != nil || req.PlanId <= 0 {
common.ApiErrorMsg(c, "参数错误")
return
}
if err := model.PurchaseSubscriptionWithBalance(userId, req.PlanId); err != nil {
common.ApiError(c, err)
return
}
common.ApiSuccess(c, nil)
}
// ---- Admin APIs ----
func AdminListSubscriptionPlans(c *gin.Context) {
@@ -102,6 +116,7 @@ func AdminListSubscriptionPlans(c *gin.Context) {
}
result := make([]SubscriptionPlanDTO, 0, len(plans))
for _, p := range plans {
p.NormalizeDefaults()
result = append(result, SubscriptionPlanDTO{
Plan: p,
})
@@ -114,10 +129,6 @@ type AdminUpsertSubscriptionPlanRequest struct {
}
func AdminCreateSubscriptionPlan(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
var req AdminUpsertSubscriptionPlanRequest
if err := c.ShouldBindJSON(&req); err != nil {
common.ApiErrorMsg(c, "参数错误")
@@ -140,6 +151,9 @@ func AdminCreateSubscriptionPlan(c *gin.Context) {
req.Plan.Currency = "USD"
}
req.Plan.Currency = "USD"
if req.Plan.AllowBalancePay == nil {
req.Plan.AllowBalancePay = common.GetPointer(true)
}
if req.Plan.DurationUnit == "" {
req.Plan.DurationUnit = model.SubscriptionDurationMonth
}
@@ -176,10 +190,6 @@ func AdminCreateSubscriptionPlan(c *gin.Context) {
}
func AdminUpdateSubscriptionPlan(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
id, _ := strconv.Atoi(c.Param("id"))
if id <= 0 {
common.ApiErrorMsg(c, "无效的ID")
@@ -248,6 +258,7 @@ func AdminUpdateSubscriptionPlan(c *gin.Context) {
"sort_order": req.Plan.SortOrder,
"stripe_price_id": req.Plan.StripePriceId,
"creem_product_id": req.Plan.CreemProductId,
"waffo_pancake_product_id": req.Plan.WaffoPancakeProductId,
"max_purchase_per_user": req.Plan.MaxPurchasePerUser,
"total_amount": req.Plan.TotalAmount,
"upgrade_group": req.Plan.UpgradeGroup,
@@ -255,6 +266,9 @@ func AdminUpdateSubscriptionPlan(c *gin.Context) {
"quota_reset_custom_seconds": req.Plan.QuotaResetCustomSeconds,
"updated_at": common.GetTimestamp(),
}
if req.Plan.AllowBalancePay != nil {
updateMap["allow_balance_pay"] = *req.Plan.AllowBalancePay
}
if err := tx.Model(&model.SubscriptionPlan{}).Where("id = ?", id).Updates(updateMap).Error; err != nil {
return err
}
@@ -273,10 +287,6 @@ type AdminUpdateSubscriptionPlanStatusRequest struct {
}
func AdminUpdateSubscriptionPlanStatus(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
id, _ := strconv.Atoi(c.Param("id"))
if id <= 0 {
common.ApiErrorMsg(c, "无效的ID")
@@ -301,10 +311,6 @@ type AdminBindSubscriptionRequest struct {
}
func AdminBindSubscription(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
var req AdminBindSubscriptionRequest
if err := c.ShouldBindJSON(&req); err != nil || req.UserId <= 0 || req.PlanId <= 0 {
common.ApiErrorMsg(c, "参数错误")
@@ -344,10 +350,6 @@ type AdminCreateUserSubscriptionRequest struct {
// AdminCreateUserSubscription creates a new user subscription from a plan (no payment).
func AdminCreateUserSubscription(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
userId, _ := strconv.Atoi(c.Param("id"))
if userId <= 0 {
common.ApiErrorMsg(c, "无效的用户ID")
-4
View File
@@ -21,10 +21,6 @@ type SubscriptionCreemPayRequest struct {
}
func SubscriptionRequestCreemPay(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
var req SubscriptionCreemPayRequest
// Keep body for debugging consistency (like RequestCreemPay)
-4
View File
@@ -22,10 +22,6 @@ type SubscriptionEpayPayRequest struct {
}
func SubscriptionRequestEpay(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
var req SubscriptionEpayPayRequest
if err := c.ShouldBindJSON(&req); err != nil || req.PlanId <= 0 {
common.ApiErrorMsg(c, "参数错误")
@@ -21,10 +21,6 @@ type SubscriptionStripePayRequest struct {
}
func SubscriptionRequestStripePay(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
var req SubscriptionStripePayRequest
if err := c.ShouldBindJSON(&req); err != nil || req.PlanId <= 0 {
common.ApiErrorMsg(c, "参数错误")
@@ -0,0 +1,126 @@
package controller
import (
"fmt"
"net/http"
"strings"
"time"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/logger"
"github.com/QuantumNous/new-api/model"
"github.com/QuantumNous/new-api/service"
"github.com/QuantumNous/new-api/setting"
"github.com/gin-gonic/gin"
"github.com/shopspring/decimal"
"github.com/thanhpk/randstr"
)
type SubscriptionWaffoPancakePayRequest struct {
PlanId int `json:"plan_id"`
}
func SubscriptionRequestWaffoPancakePay(c *gin.Context) {
var req SubscriptionWaffoPancakePayRequest
if err := c.ShouldBindJSON(&req); err != nil || req.PlanId <= 0 {
common.ApiErrorMsg(c, "参数错误")
return
}
plan, err := model.GetSubscriptionPlanById(req.PlanId)
if err != nil {
common.ApiError(c, err)
return
}
if !plan.Enabled {
common.ApiErrorMsg(c, "套餐未启用")
return
}
if strings.TrimSpace(plan.WaffoPancakeProductId) == "" {
common.ApiErrorMsg(c, "该套餐未配置 WaffoPancakeProductId")
return
}
// Plan targets its own Pancake product, so we only require credentials
// here — not the gateway-level WaffoPancakeProductID.
if strings.TrimSpace(setting.WaffoPancakeMerchantID) == "" ||
strings.TrimSpace(setting.WaffoPancakePrivateKey) == "" {
common.ApiErrorMsg(c, "Waffo Pancake 未配置或密钥无效")
return
}
userId := c.GetInt("id")
user, err := model.GetUserById(userId, false)
if err != nil {
common.ApiError(c, err)
return
}
if user == nil {
common.ApiErrorMsg(c, "用户不存在")
return
}
if plan.MaxPurchasePerUser > 0 {
count, err := model.CountUserSubscriptionsByPlan(userId, plan.Id)
if err != nil {
common.ApiError(c, err)
return
}
if count >= int64(plan.MaxPurchasePerUser) {
common.ApiErrorMsg(c, "已达到该套餐购买上限")
return
}
}
// WAFFO_PANCAKE_SUB- prefix (vs. wallet's WAFFO_PANCAKE-) drives webhook
// dispatch in WaffoPancakeWebhook.
tradeNo := fmt.Sprintf("WAFFO_PANCAKE_SUB-%d-%d-%s", userId, time.Now().UnixMilli(), randstr.String(6))
order := &model.SubscriptionOrder{
UserId: userId,
PlanId: plan.Id,
Money: plan.PriceAmount,
TradeNo: tradeNo,
PaymentMethod: model.PaymentMethodWaffoPancake,
PaymentProvider: model.PaymentProviderWaffoPancake,
CreateTime: time.Now().Unix(),
Status: common.TopUpStatusPending,
}
if err := order.Insert(); err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅订单创建失败 user_id=%d plan_id=%d trade_no=%s error=%q", userId, plan.Id, tradeNo, err.Error()))
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "创建订单失败"})
return
}
expiresInSeconds := 45 * 60
session, err := service.CreateWaffoPancakeCheckoutSession(c.Request.Context(), &service.WaffoPancakeCreateSessionParams{
ProductID: plan.WaffoPancakeProductId,
BuyerIdentity: service.WaffoPancakeBuyerIdentityFromUserID(user.Id),
PriceSnapshot: &service.WaffoPancakePriceSnapshot{
Amount: decimal.NewFromFloat(plan.PriceAmount).StringFixed(2),
TaxCategory: "saas",
},
BuyerEmail: getWaffoPancakeBuyerEmail(user),
ExpiresInSeconds: &expiresInSeconds,
OrderMerchantExternalID: tradeNo,
})
if err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅结账会话创建失败 user_id=%d plan_id=%d trade_no=%s error=%q", userId, plan.Id, tradeNo, err.Error()))
order.Status = common.TopUpStatusFailed
_ = order.Update()
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "拉起支付失败"})
return
}
logger.LogInfo(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅订单创建成功 user_id=%d plan_id=%d trade_no=%s session_id=%s money=%.2f", userId, plan.Id, tradeNo, session.SessionID, plan.PriceAmount))
c.JSON(http.StatusOK, gin.H{
"message": "success",
"data": gin.H{
"checkout_url": session.CheckoutURL,
"session_id": session.SessionID,
"expires_at": session.ExpiresAt,
"order_id": tradeNo,
"token": session.Token,
"token_expires_at": session.TokenExpiresAt,
},
})
}
+23 -28
View File
@@ -22,13 +22,8 @@ import (
)
func GetTopUpInfo(c *gin.Context) {
complianceConfirmed := operation_setting.IsPaymentComplianceConfirmed()
// 获取支付方式
payMethods := operation_setting.PayMethods
if !complianceConfirmed {
payMethods = []map[string]string{}
}
// 如果启用了 Stripe 支付,添加到支付方法列表
if isStripeTopUpEnabled() {
@@ -52,6 +47,27 @@ func GetTopUpInfo(c *gin.Context) {
}
}
// Waffo Pancake displayed above the legacy Waffo gateway.
enableWaffoPancake := isWaffoPancakeTopUpEnabled()
if enableWaffoPancake {
hasWaffoPancake := false
for _, method := range payMethods {
if method["type"] == model.PaymentMethodWaffoPancake {
hasWaffoPancake = true
break
}
}
if !hasWaffoPancake {
payMethods = append(payMethods, map[string]string{
"name": "Waffo Pancake",
"type": model.PaymentMethodWaffoPancake,
"color": "rgba(var(--semi-orange-5), 1)",
"min_topup": strconv.Itoa(setting.WaffoPancakeMinTopUp),
})
}
}
// 如果启用了 Waffo 支付,添加到支付方法列表
enableWaffo := isWaffoTopUpEnabled()
if enableWaffo {
@@ -74,35 +90,14 @@ func GetTopUpInfo(c *gin.Context) {
}
}
enableWaffoPancake := isWaffoPancakeTopUpEnabled()
if enableWaffoPancake {
hasWaffoPancake := false
for _, method := range payMethods {
if method["type"] == model.PaymentMethodWaffoPancake {
hasWaffoPancake = true
break
}
}
if !hasWaffoPancake {
payMethods = append(payMethods, map[string]string{
"name": "Waffo Pancake",
"type": model.PaymentMethodWaffoPancake,
"color": "rgba(var(--semi-orange-5), 1)",
"min_topup": strconv.Itoa(setting.WaffoPancakeMinTopUp),
})
}
}
data := gin.H{
"enable_online_topup": isEpayTopUpEnabled(),
"enable_stripe_topup": isStripeTopUpEnabled(),
"enable_creem_topup": isCreemTopUpEnabled(),
"enable_waffo_topup": enableWaffo,
"enable_waffo_pancake_topup": enableWaffoPancake,
"enable_redemption": complianceConfirmed,
"payment_compliance_confirmed": complianceConfirmed,
"payment_compliance_terms_version": operation_setting.CurrentComplianceTermsVersion,
"enable_redemption": true,
"payment_compliance_confirmed": true,
"waffo_pay_methods": func() interface{} {
if enableWaffo {
return setting.GetWaffoPayMethods()
+311 -33
View File
@@ -96,33 +96,257 @@ func getWaffoPancakeBuyerEmail(user *model.User) string {
if user != nil && strings.TrimSpace(user.Email) != "" {
return user.Email
}
if user != nil {
return fmt.Sprintf("%d@new-api.local", user.Id)
}
return ""
}
func getWaffoPancakeReturnURL() string {
if strings.TrimSpace(setting.WaffoPancakeReturnURL) != "" {
return setting.WaffoPancakeReturnURL
// The admin config endpoints below accept typed-but-not-yet-saved creds in
// the body and fall back to persisted creds when the body is blank (see
// resolveWaffoPancakeAdminCreds). Only SaveWaffoPancake writes to OptionMap.
type waffoPancakeCredsRequest struct {
MerchantID string `json:"merchant_id"`
PrivateKey string `json:"private_key"`
}
type saveWaffoPancakeRequest struct {
MerchantID string `json:"merchant_id"`
PrivateKey string `json:"private_key"`
ReturnURL string `json:"return_url"`
StoreID string `json:"store_id"`
ProductID string `json:"product_id"`
}
type createWaffoPancakePairRequest struct {
MerchantID string `json:"merchant_id"`
PrivateKey string `json:"private_key"`
ReturnURL string `json:"return_url"`
}
// SaveWaffoPancake atomically persists all five operator-controlled fields.
// Catalog / pair endpoints are transient — only this one writes the OptionMap.
func SaveWaffoPancake(c *gin.Context) {
var req saveWaffoPancakeRequest
if err := c.ShouldBindJSON(&req); err != nil {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
return
}
return paymentReturnPath("/console/topup?show_history=true")
if err := service.SaveWaffoPancakeConfig(
c.Request.Context(),
req.MerchantID,
req.PrivateKey,
req.ReturnURL,
req.StoreID,
req.ProductID,
); err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake 保存配置失败 store_id=%q product_id=%q error=%q",
req.StoreID, req.ProductID, err.Error(),
))
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "保存配置失败"})
return
}
c.JSON(http.StatusOK, gin.H{
"message": "success",
"data": gin.H{
"product_id": setting.WaffoPancakeProductID,
"store_id": setting.WaffoPancakeStoreID,
},
})
}
// resolveWaffoPancakeAdminCreds prefers body creds (typed-but-not-yet-saved
// values, for verification) and falls back to persisted creds when the body
// is blank (so returning admins don't have to re-paste the private key,
// which is stripped from GET /api/option/).
func resolveWaffoPancakeAdminCreds(bodyMerchantID, bodyPrivateKey string) (string, string) {
m := strings.TrimSpace(bodyMerchantID)
k := strings.TrimSpace(bodyPrivateKey)
if m == "" && k == "" {
return setting.WaffoPancakeMerchantID, setting.WaffoPancakePrivateKey
}
return m, k
}
// CreateWaffoPancakePair mints a Store + OnetimeProduct pair in one round-
// trip. Surfaces an orphan-store flag when the product half fails so the
// frontend can preselect / retry without losing context.
func CreateWaffoPancakePair(c *gin.Context) {
var req createWaffoPancakePairRequest
if c.Request.ContentLength > 0 {
if err := c.ShouldBindJSON(&req); err != nil {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
return
}
}
merchantID, privateKey := resolveWaffoPancakeAdminCreds(req.MerchantID, req.PrivateKey)
if merchantID == "" || privateKey == "" {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 凭证未配置"})
return
}
result, err := service.CreateWaffoPancakePrimaryPair(
c.Request.Context(), merchantID, privateKey, req.ReturnURL,
)
if err != nil {
orphan := result != nil && result.OrphanStore
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake 创建店铺与产品失败 orphan_store=%t store_id=%q error=%q",
orphan, func() string {
if result == nil {
return ""
}
return result.StoreID
}(), err.Error(),
))
data := gin.H{"error": err.Error()}
if orphan {
data["store_id"] = result.StoreID
data["store_name"] = result.StoreName
data["orphan_store"] = true
}
c.JSON(http.StatusOK, gin.H{"message": "error", "data": data})
return
}
c.JSON(http.StatusOK, gin.H{
"message": "success",
"data": gin.H{
"store_id": result.StoreID,
"store_name": result.StoreName,
"product_id": result.ProductID,
"product_name": result.ProductName,
},
})
}
// ListWaffoPancakeCatalog returns the merchant's Stores + OnetimeProducts.
// Doubles as a credential probe (a successful 200 proves the resolved creds
// authenticate). See resolveWaffoPancakeAdminCreds for credential resolution.
func ListWaffoPancakeCatalog(c *gin.Context) {
var req waffoPancakeCredsRequest
// An empty body means "use persisted creds"; only fail on malformed JSON.
if c.Request.ContentLength > 0 {
if err := c.ShouldBindJSON(&req); err != nil {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
return
}
}
merchantID, privateKey := resolveWaffoPancakeAdminCreds(req.MerchantID, req.PrivateKey)
if merchantID == "" || privateKey == "" {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 凭证未配置"})
return
}
catalog, err := service.ListWaffoPancakeCatalog(c.Request.Context(), merchantID, privateKey)
if err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake 拉取店铺与产品目录失败 error=%q", err.Error(),
))
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "拉取目录失败"})
return
}
c.JSON(http.StatusOK, gin.H{"message": "success", "data": catalog})
}
type createWaffoPancakeSubscriptionProductRequest struct {
Name string `json:"name"`
Amount string `json:"amount"`
}
// CreateWaffoPancakeSubscriptionProduct mints an OnetimeProduct (not
// SubscriptionProduct — see service.CreateWaffoPancakeProductForPlan)
// sized to a plan's `name` + `amount`, using persisted Pancake credentials
// + StoreID. Reads from the form, not the plan row, so newly-typed unsaved
// plans can mint a product too.
func CreateWaffoPancakeSubscriptionProduct(c *gin.Context) {
var req createWaffoPancakeSubscriptionProductRequest
if c.Request.ContentLength > 0 {
if err := c.ShouldBindJSON(&req); err != nil {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "参数错误"})
return
}
}
if strings.TrimSpace(req.Name) == "" {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "套餐名称不能为空"})
return
}
if strings.TrimSpace(req.Amount) == "" {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "套餐价格不能为空"})
return
}
merchantID, privateKey := resolveWaffoPancakeAdminCreds("", "")
storeID := strings.TrimSpace(setting.WaffoPancakeStoreID)
if merchantID == "" || privateKey == "" || storeID == "" {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 未完成配置,请先在支付设置中完成网关绑定"})
return
}
productID, err := service.CreateWaffoPancakeProductForPlan(
c.Request.Context(),
merchantID,
privateKey,
storeID,
req.Name,
req.Amount,
setting.WaffoPancakeReturnURL,
)
if err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake 创建套餐产品失败 store_id=%q name=%q amount=%q error=%q",
storeID, req.Name, req.Amount, err.Error(),
))
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "创建套餐产品失败"})
return
}
c.JSON(http.StatusOK, gin.H{
"message": "success",
"data": gin.H{
"product_id": productID,
"product_name": req.Name,
"store_id": storeID,
},
})
}
// ListWaffoPancakeSubscriptionProductOptions returns the OnetimeProducts
// in the saved Pancake store, for the subscription-plan dropdown. The name
// reflects new-api's plan concept; under the hood it's still OnetimeProducts.
func ListWaffoPancakeSubscriptionProductOptions(c *gin.Context) {
merchantID, privateKey := resolveWaffoPancakeAdminCreds("", "")
storeID := strings.TrimSpace(setting.WaffoPancakeStoreID)
if merchantID == "" || privateKey == "" || storeID == "" {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 未完成配置,请先在支付设置中完成网关绑定"})
return
}
catalog, err := service.ListWaffoPancakeCatalog(c.Request.Context(), merchantID, privateKey)
if err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake 拉取订阅产品列表失败 store_id=%q error=%q", storeID, err.Error(),
))
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "拉取产品列表失败"})
return
}
products := []service.WaffoPancakeCatalogProduct{}
for _, store := range catalog.Stores {
if store.ID == storeID {
products = store.OnetimeProducts
break
}
}
c.JSON(http.StatusOK, gin.H{
"message": "success",
"data": gin.H{
"store_id": storeID,
"products": products,
},
})
}
func getWaffoPancakeBuyerIdentity(user *model.User) string {
if user == nil {
return ""
}
return service.WaffoPancakeBuyerIdentityFromUserID(user.Id)
}
func RequestWaffoPancakePay(c *gin.Context) {
if !setting.WaffoPancakeEnabled {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 支付未启用"})
return
}
currentWebhookKey := setting.WaffoPancakeWebhookPublicKey
if setting.WaffoPancakeSandbox {
currentWebhookKey = setting.WaffoPancakeWebhookTestKey
}
if strings.TrimSpace(setting.WaffoPancakeMerchantID) == "" ||
strings.TrimSpace(setting.WaffoPancakePrivateKey) == "" ||
strings.TrimSpace(currentWebhookKey) == "" ||
strings.TrimSpace(setting.WaffoPancakeStoreID) == "" ||
strings.TrimSpace(setting.WaffoPancakeProductID) == "" {
if !isWaffoPancakeTopUpEnabled() {
c.JSON(http.StatusOK, gin.H{"message": "error", "data": "Waffo Pancake 配置不完整"})
return
}
@@ -175,18 +399,15 @@ func RequestWaffoPancakePay(c *gin.Context) {
expiresInSeconds := 45 * 60
session, err := service.CreateWaffoPancakeCheckoutSession(c.Request.Context(), &service.WaffoPancakeCreateSessionParams{
StoreID: setting.WaffoPancakeStoreID,
ProductID: setting.WaffoPancakeProductID,
ProductType: "onetime",
Currency: strings.ToUpper(strings.TrimSpace(setting.WaffoPancakeCurrency)),
ProductID: setting.WaffoPancakeProductID,
BuyerIdentity: getWaffoPancakeBuyerIdentity(user),
PriceSnapshot: &service.WaffoPancakePriceSnapshot{
Amount: formatWaffoPancakeAmount(payMoney),
TaxIncluded: false,
TaxCategory: "saas",
},
BuyerEmail: getWaffoPancakeBuyerEmail(user),
SuccessURL: getWaffoPancakeReturnURL(),
ExpiresInSeconds: &expiresInSeconds,
BuyerEmail: getWaffoPancakeBuyerEmail(user),
ExpiresInSeconds: &expiresInSeconds,
OrderMerchantExternalID: tradeNo,
})
if err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 创建结账会话失败 user_id=%d trade_no=%s error=%q", id, tradeNo, err.Error()))
@@ -200,10 +421,12 @@ func RequestWaffoPancakePay(c *gin.Context) {
c.JSON(http.StatusOK, gin.H{
"message": "success",
"data": gin.H{
"checkout_url": session.CheckoutURL,
"session_id": session.SessionID,
"expires_at": session.ExpiresAt,
"order_id": tradeNo,
"checkout_url": session.CheckoutURL,
"session_id": session.SessionID,
"expires_at": session.ExpiresAt,
"order_id": tradeNo,
"token": session.Token,
"token_expires_at": session.TokenExpiresAt,
},
})
}
@@ -215,6 +438,19 @@ func WaffoPancakeWebhook(c *gin.Context) {
return
}
// :env splits test vs prod traffic at the routing layer — operator
// registers each URL in the matching webhook slot in Pancake's dashboard.
// We then enforce event.mode == expectedEnv to catch mis-registrations.
expectedEnv := strings.TrimSpace(c.Param("env"))
if expectedEnv != "test" && expectedEnv != "prod" {
logger.LogWarn(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake webhook 路径环境段无效 env=%q path=%q client_ip=%s",
expectedEnv, c.Request.RequestURI, c.ClientIP(),
))
c.String(http.StatusNotFound, "unknown env")
return
}
bodyBytes, err := io.ReadAll(c.Request.Body)
if err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake webhook 读取请求体失败 path=%q client_ip=%s error=%q", c.Request.RequestURI, c.ClientIP(), err.Error()))
@@ -232,15 +468,57 @@ func WaffoPancakeWebhook(c *gin.Context) {
return
}
if !strings.EqualFold(strings.TrimSpace(event.Mode), expectedEnv) {
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake webhook 环境不匹配 expected=%q actual_mode=%q event_id=%s order_id=%s client_ip=%s",
expectedEnv, event.Mode, event.ID, event.Data.OrderID, c.ClientIP(),
))
c.String(http.StatusOK, "OK")
return
}
logger.LogInfo(c.Request.Context(), fmt.Sprintf("Waffo Pancake webhook 验签成功 event_type=%s event_id=%s order_id=%s client_ip=%s", event.NormalizedEventType(), event.ID, event.Data.OrderID, c.ClientIP()))
if event.NormalizedEventType() != "order.completed" {
c.String(http.StatusOK, "OK")
return
}
// Dispatch by trade_no prefix. OrderMerchantExternalID = our trade_no;
// OrderID is Pancake's internal ORD_* (logs only).
rawTradeNo := strings.TrimSpace(event.Data.OrderMerchantExternalID)
isSubscription := strings.HasPrefix(rawTradeNo, "WAFFO_PANCAKE_SUB-")
if isSubscription {
tradeNo, err := service.ResolveWaffoPancakeSubscriptionTradeNo(event)
if err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake webhook 订阅订单解析失败 event_id=%s order_id=%s buyer_identity=%q client_ip=%s error=%q",
event.ID, event.Data.OrderID, event.Data.MerchantProvidedBuyerIdentity, c.ClientIP(), err.Error(),
))
c.String(http.StatusOK, "OK")
return
}
LockOrder(tradeNo)
defer UnlockOrder(tradeNo)
if err := model.CompleteSubscriptionOrder(tradeNo, string(bodyBytes), model.PaymentProviderWaffoPancake, ""); err != nil {
logger.LogError(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅完成失败 trade_no=%s event_id=%s order_id=%s client_ip=%s error=%q", tradeNo, event.ID, event.Data.OrderID, c.ClientIP(), err.Error()))
c.String(http.StatusInternalServerError, "retry")
return
}
logger.LogInfo(c.Request.Context(), fmt.Sprintf("Waffo Pancake 订阅完成 trade_no=%s event_id=%s order_id=%s client_ip=%s", tradeNo, event.ID, event.Data.OrderID, c.ClientIP()))
c.String(http.StatusOK, "OK")
return
}
tradeNo, err := service.ResolveWaffoPancakeTradeNo(event)
if err != nil {
logger.LogWarn(c.Request.Context(), fmt.Sprintf("Waffo Pancake webhook 订单号映射失败 event_id=%s order_id=%s error=%q", event.ID, event.Data.OrderID, err.Error()))
// LogError (not LogWarn): covers order-not-found and buyer-identity
// mismatch — both warrant human attention. 200 OK so Waffo doesn't
// retry a permanently-unresolvable webhook.
logger.LogError(c.Request.Context(), fmt.Sprintf(
"Waffo Pancake webhook 订单解析失败 event_id=%s order_id=%s buyer_identity=%q client_ip=%s error=%q",
event.ID, event.Data.OrderID, event.Data.MerchantProvidedBuyerIdentity, c.ClientIP(), err.Error(),
))
c.String(http.StatusOK, "OK")
return
}
+13 -11
View File
@@ -17,7 +17,6 @@ import (
"github.com/QuantumNous/new-api/model"
"github.com/QuantumNous/new-api/service"
"github.com/QuantumNous/new-api/setting"
"github.com/QuantumNous/new-api/setting/operation_setting"
"github.com/QuantumNous/new-api/constant"
@@ -251,8 +250,20 @@ func GetAllUsers(c *gin.Context) {
func SearchUsers(c *gin.Context) {
keyword := c.Query("keyword")
group := c.Query("group")
var role *int
if roleStr := c.Query("role"); roleStr != "" {
if parsed, err := strconv.Atoi(roleStr); err == nil {
role = &parsed
}
}
var status *int
if statusStr := c.Query("status"); statusStr != "" {
if parsed, err := strconv.Atoi(statusStr); err == nil {
status = &parsed
}
}
pageInfo := common.GetPageQuery(c)
users, total, err := model.SearchUsers(keyword, group, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
users, total, err := model.SearchUsers(keyword, group, role, status, pageInfo.GetStartIdx(), pageInfo.GetPageSize())
if err != nil {
common.ApiError(c, err)
return
@@ -332,10 +343,6 @@ type TransferAffQuotaRequest struct {
}
func TransferAffQuota(c *gin.Context) {
if !requirePaymentCompliance(c) {
return
}
id := c.GetInt("id")
user, err := model.GetUserById(id, true)
if err != nil {
@@ -1092,11 +1099,6 @@ func getTopUpLock(userID int) *topUpTryLock {
}
func TopUp(c *gin.Context) {
if !operation_setting.IsPaymentComplianceConfirmed() {
common.ApiErrorI18n(c, i18n.MsgPaymentComplianceRequired)
return
}
id := c.GetInt("id")
lock := getTopUpLock(id)
if !lock.TryLock() {
+2 -1
View File
@@ -16,7 +16,7 @@ version: '3.4' # For compatibility with older Docker versions
services:
new-api:
image: calciumion/new-api:latest
image: git.viaeon.com/admin/new-api:latest
container_name: new-api
restart: always
command: --log-dir /app/logs
@@ -34,6 +34,7 @@ services:
- BATCH_UPDATE_ENABLED=true # 是否启用批量更新 (Whether to enable batch update)
- NODE_NAME=new-api-node-1 # 节点名称,用于审计日志中标识节点身份;多节点/容器部署时建议设置 (Node name used in audit logs; recommended when running multiple instances or in containers)
# - STREAMING_TIMEOUT=300 # 流模式无响应超时时间,单位秒,默认120秒,如果出现空补全可以尝试改为更大值 (Streaming timeout in seconds, default is 120s. Increase if experiencing empty completions
# - RELAY_IDLE_CONN_TIMEOUT=90 # Relay HTTP 客户端空闲连接超时时间,单位秒,默认跟随 Go 标准库,设置为0表示不限制 (Relay HTTP client idle keep-alive timeout in seconds, defaults to Go standard library; set 0 to disable)
# - SESSION_SECRET=random_string # 多机部署时设置,必须修改这个随机字符串!! (multi-node deployment, set this to a random string!!!!!!!
# - SYNC_FREQUENCY=60 # Uncomment if regular database syncing is needed
# - GOOGLE_ANALYTICS_ID=G-XXXXXXXXXX # Google Analytics 的测量 ID (Google Analytics Measurement ID)
+6 -6
View File
@@ -26,11 +26,11 @@ type ImageRequest struct {
OutputFormat json.RawMessage `json:"output_format,omitempty"`
OutputCompression json.RawMessage `json:"output_compression,omitempty"`
PartialImages json.RawMessage `json:"partial_images,omitempty"`
// Stream bool `json:"stream,omitempty"`
Images json.RawMessage `json:"images,omitempty"`
Mask json.RawMessage `json:"mask,omitempty"`
InputFidelity json.RawMessage `json:"input_fidelity,omitempty"`
Watermark *bool `json:"watermark,omitempty"`
Stream *bool `json:"stream,omitempty"`
Images json.RawMessage `json:"images,omitempty"`
Mask json.RawMessage `json:"mask,omitempty"`
InputFidelity json.RawMessage `json:"input_fidelity,omitempty"`
Watermark *bool `json:"watermark,omitempty"`
// zhipu 4v
WatermarkEnabled json.RawMessage `json:"watermark_enabled,omitempty"`
UserId json.RawMessage `json:"user_id,omitempty"`
@@ -163,7 +163,7 @@ func (i *ImageRequest) GetTokenCountMeta() *types.TokenCountMeta {
}
func (i *ImageRequest) IsStream(c *gin.Context) bool {
return false
return i.Stream != nil && *i.Stream
}
func (i *ImageRequest) SetModelName(modelName string) {
+12 -2
View File
@@ -213,12 +213,22 @@ func (r *GeneralOpenAIRequest) ToMap() map[string]any {
return result
}
func IsOpenAIReasoningOModel(modelName string) bool {
return strings.HasPrefix(modelName, "o1") ||
strings.HasPrefix(modelName, "o3") ||
strings.HasPrefix(modelName, "o4")
}
func IsOpenAIGPT5Model(modelName string) bool {
return strings.HasPrefix(modelName, "gpt-5")
}
func (r *GeneralOpenAIRequest) GetSystemRoleName() string {
if strings.HasPrefix(r.Model, "o") {
if IsOpenAIReasoningOModel(r.Model) {
if !strings.HasPrefix(r.Model, "o1-mini") && !strings.HasPrefix(r.Model, "o1-preview") {
return "developer"
}
} else if strings.HasPrefix(r.Model, "gpt-5") {
} else if IsOpenAIGPT5Model(r.Model) {
return "developer"
}
return "system"
+24
View File
@@ -71,3 +71,27 @@ func TestOpenAIResponsesRequestPreserveExplicitZeroValues(t *testing.T) {
require.True(t, gjson.GetBytes(encoded, "stream").Exists())
require.True(t, gjson.GetBytes(encoded, "top_p").Exists())
}
func TestGeneralOpenAIRequestGetSystemRoleName(t *testing.T) {
tests := []struct {
name string
model string
want string
}{
{name: "o1 uses developer", model: "o1", want: "developer"},
{name: "o3 family uses developer", model: "o3-mini-high", want: "developer"},
{name: "o4 family uses developer", model: "o4-mini", want: "developer"},
{name: "o1 mini stays system", model: "o1-mini", want: "system"},
{name: "o1 preview stays system", model: "o1-preview", want: "system"},
{name: "gpt 5 uses developer", model: "gpt-5", want: "developer"},
{name: "omni is not o series", model: "omni-moderation-latest", want: "system"},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
req := GeneralOpenAIRequest{Model: tt.model}
require.Equal(t, tt.want, req.GetSystemRoleName())
})
}
}
+2 -2
View File
@@ -18,10 +18,10 @@
"openai",
"claude"
],
"author": "QuantumNous",
"author": "modelstoken",
"repository": {
"type": "git",
"url": "https://github.com/QuantumNous/new-api"
"url": "https://git.viaeon.com/admin/new-api"
},
"devDependencies": {
"cross-env": "^7.0.3",
+2
View File
@@ -60,6 +60,8 @@ require (
gorm.io/gorm v1.25.2
)
require github.com/waffo-com/waffo-pancake-sdk-go v0.3.1
require (
github.com/DmitriyVTitov/size v1.5.0 // indirect
github.com/anknown/darts v0.0.0-20151216065714-83ff685239e6 // indirect
+6
View File
@@ -308,6 +308,12 @@ github.com/ugorji/go/codec v1.2.12 h1:9LC83zGrHhuUA9l16C9AHXAqEV/2wBQ4nkvumAE65E
github.com/ugorji/go/codec v1.2.12/go.mod h1:UNopzCgEMSXjBc6AOMqYvWC1ktqTAfzJZUZgYf6w6lg=
github.com/waffo-com/waffo-go v1.3.1 h1:NCYD3oQ59DTJj1bwS5T/659LI4h8PuAIW4Qj/w7fKPw=
github.com/waffo-com/waffo-go v1.3.1/go.mod h1:IaXVYq6mmYtrLFFsLxPslNwuIZx0mIadWWjhe+eWb0g=
github.com/waffo-com/waffo-pancake-sdk-go v0.1.1 h1:YOI7+3zTBlTB7Ou6+ZXnJV2JvW/ag9d7CwE/TxH3Hls=
github.com/waffo-com/waffo-pancake-sdk-go v0.1.1/go.mod h1:5MBCGH/nqRRA5sHO/lQB/96r4BTAqy8QpWxn53m9htI=
github.com/waffo-com/waffo-pancake-sdk-go v0.2.0 h1:cCSgccM66p7feTtgRqUUGT50tYQOhahsoPXavd+ib1U=
github.com/waffo-com/waffo-pancake-sdk-go v0.2.0/go.mod h1:5MBCGH/nqRRA5sHO/lQB/96r4BTAqy8QpWxn53m9htI=
github.com/waffo-com/waffo-pancake-sdk-go v0.3.1 h1:ngQSN/oVB35xTwFPLfg++bxPC+SptcF145Mb6c62YCc=
github.com/waffo-com/waffo-pancake-sdk-go v0.3.1/go.mod h1:OB2MyFIQaefoPO0FV3J+yu9sDP8RVFQ+sbFsXqGuObc=
github.com/x448/float16 v0.8.4 h1:qLwI1I70+NjRFUR3zs1JPUCgaCXSh3SW62uAKT1mSBM=
github.com/x448/float16 v0.8.4/go.mod h1:14CWIYCyZA/cWjXOioeEpHeN/83MdbZDRQHoFcYsOfg=
github.com/xyproto/randomstring v1.0.5 h1:YtlWPoRdgMu3NZtP45drfy1GKoojuR7hmRcnhZqKjWU=
-1
View File
@@ -152,7 +152,6 @@ const (
MsgPaymentWebhookNotConfig = "payment.webhook_not_configured"
MsgPaymentPriceIdNotConfig = "payment.price_id_not_configured"
MsgPaymentCreemNotConfig = "payment.creem_not_configured"
MsgPaymentComplianceRequired = "payment.compliance_required"
)
// Topup related messages
+1 -1
View File
@@ -164,7 +164,7 @@ func main() {
common.SysLog(fmt.Sprintf("panic detected: %v", err))
c.JSON(http.StatusInternalServerError, gin.H{
"error": gin.H{
"message": fmt.Sprintf("Panic detected, error: %v. Please submit a issue here: https://github.com/Calcium-Ion/new-api", err),
"message": fmt.Sprintf("Panic detected, error: %v. Please submit a issue here: https://git.viaeon.com/admin/new-api/issues", err),
"type": "new_api_panic",
},
})
+32 -5
View File
@@ -1,6 +1,8 @@
FRONTEND_DIR = ./web/default
FRONTEND_CLASSIC_DIR = ./web/classic
BACKEND_DIR = .
DEV_FRONTEND_DEFAULT_PORT ?= 5173
DEV_FRONTEND_CLASSIC_PORT ?= 5174
DEV_COMPOSE_FILE = docker-compose.dev.yml
DEV_POSTGRES_SERVICE = postgres
DEV_BACKEND_SERVICE = new-api
@@ -14,11 +16,13 @@ all: build-all-frontends start-backend
build-frontend:
@echo "Building default frontend..."
@cd $(FRONTEND_DIR) && bun install && DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
@cd ./web && bun install --frozen-lockfile
@cd $(FRONTEND_DIR) && DISABLE_ESLINT_PLUGIN='true' VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
build-frontend-classic:
@echo "Building classic frontend..."
@cd $(FRONTEND_CLASSIC_DIR) && bun install && VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
@cd ./web && bun install --frozen-lockfile
@cd $(FRONTEND_CLASSIC_DIR) && VITE_REACT_APP_VERSION=$(cat ../../VERSION) bun run build
build-all-frontends: build-frontend build-frontend-classic
@@ -35,12 +39,35 @@ dev-api-rebuild:
@docker compose -f $(DEV_COMPOSE_FILE) up -d --build $(DEV_BACKEND_SERVICE)
dev-web:
@echo "Starting frontend dev server..."
@cd $(FRONTEND_DIR) && bun install && bun run dev
@echo "Starting both frontend dev servers..."
@echo "Default frontend: http://localhost:$(DEV_FRONTEND_DEFAULT_PORT)"
@echo "Classic frontend: http://localhost:$(DEV_FRONTEND_CLASSIC_PORT)"
@cd ./web && bun install
@(cd $(FRONTEND_DIR) && bun run dev -- --host 0.0.0.0 --port $(DEV_FRONTEND_DEFAULT_PORT)) & \
default_pid=$$!; \
(cd $(FRONTEND_CLASSIC_DIR) && bun run dev -- --host 0.0.0.0 --port $(DEV_FRONTEND_CLASSIC_PORT)) & \
classic_pid=$$!; \
trap 'kill $$default_pid $$classic_pid 2>/dev/null; wait $$default_pid $$classic_pid 2>/dev/null; exit 130' INT TERM; \
while kill -0 $$default_pid 2>/dev/null && kill -0 $$classic_pid 2>/dev/null; do \
sleep 1; \
done; \
if ! kill -0 $$default_pid 2>/dev/null; then \
wait $$default_pid; \
status=$$?; \
kill $$classic_pid 2>/dev/null; \
wait $$classic_pid 2>/dev/null; \
exit $$status; \
fi; \
wait $$classic_pid; \
status=$$?; \
kill $$default_pid 2>/dev/null; \
wait $$default_pid 2>/dev/null; \
exit $$status
dev-web-classic:
@echo "Starting classic frontend dev server..."
@cd $(FRONTEND_CLASSIC_DIR) && bun install && bun run dev
@cd ./web && bun install
@cd $(FRONTEND_CLASSIC_DIR) && bun run dev -- --host 0.0.0.0 --port $(DEV_FRONTEND_CLASSIC_PORT)
dev: dev-api dev-web
+4
View File
@@ -163,6 +163,10 @@ func TryUserAuth() func(c *gin.Context) {
if id != nil {
c.Set("id", id)
}
role := session.Get("role")
if role != nil {
c.Set("role", role)
}
c.Next()
}
}
+89 -7
View File
@@ -3,6 +3,7 @@ package middleware
import (
"errors"
"fmt"
"io"
"net/http"
"slices"
"strconv"
@@ -20,6 +21,7 @@ import (
"github.com/QuantumNous/new-api/types"
"github.com/gin-gonic/gin"
"github.com/tidwall/gjson"
)
type ModelRequest struct {
@@ -100,14 +102,10 @@ func Distribute() func(c *gin.Context) {
}
if preferredChannelID, found := service.GetPreferredChannelByAffinity(c, modelRequest.Model, usingGroup); found {
affinityUsable := false
preferred, err := model.CacheGetChannel(preferredChannelID)
if err == nil && preferred != nil {
if preferred.Status != common.ChannelStatusEnabled {
if service.ShouldSkipRetryAfterChannelAffinityFailure(c) {
abortWithOpenAiMessage(c, http.StatusForbidden, i18n.T(c, i18n.MsgDistributorAffinityChannelDisabled))
return
}
} else if usingGroup == "auto" {
if err == nil && preferred != nil && preferred.Status == common.ChannelStatusEnabled {
if usingGroup == "auto" {
userGroup := common.GetContextKeyString(c, constant.ContextKeyUserGroup)
autoGroups := service.GetUserAutoGroup(userGroup)
for _, g := range autoGroups {
@@ -115,6 +113,7 @@ func Distribute() func(c *gin.Context) {
selectGroup = g
common.SetContextKey(c, constant.ContextKeyAutoGroup, g)
channel = preferred
affinityUsable = true
service.MarkChannelAffinityUsed(c, g, preferred.Id)
break
}
@@ -122,9 +121,13 @@ func Distribute() func(c *gin.Context) {
} else if model.IsChannelEnabledForGroupModel(usingGroup, modelRequest.Model, preferred.Id) {
channel = preferred
selectGroup = usingGroup
affinityUsable = true
service.MarkChannelAffinityUsed(c, usingGroup, preferred.Id)
}
}
if !affinityUsable && !service.ShouldKeepChannelAffinityOnChannelDisabled() {
service.ClearCurrentChannelAffinityCache(c)
}
}
if channel == nil {
@@ -170,6 +173,14 @@ func Distribute() func(c *gin.Context) {
// - application/x-www-form-urlencoded
// - multipart/form-data
func getModelFromRequest(c *gin.Context) (*ModelRequest, error) {
if strings.HasPrefix(c.Request.Header.Get("Content-Type"), "application/json") {
modelRequest, err := getModelFromJSONBody(c)
if err != nil {
return nil, errors.New(i18n.T(c, i18n.MsgDistributorInvalidRequest, map[string]any{"Error": err.Error()}))
}
return modelRequest, nil
}
var modelRequest ModelRequest
err := common.UnmarshalBodyReusable(c, &modelRequest)
if err != nil {
@@ -178,6 +189,50 @@ func getModelFromRequest(c *gin.Context) (*ModelRequest, error) {
return &modelRequest, nil
}
func getModelFromJSONBody(c *gin.Context) (*ModelRequest, error) {
storage, err := common.GetBodyStorage(c)
if err != nil {
return nil, err
}
requestBody, err := storage.Bytes()
if err != nil {
return nil, err
}
if !gjson.ValidBytes(requestBody) {
return nil, errors.New("invalid JSON request body")
}
values := gjson.GetManyBytes(requestBody, "model", "group")
model, err := getJSONStringValue(values[0], "model")
if err != nil {
return nil, err
}
group, err := getJSONStringValue(values[1], "group")
if err != nil {
return nil, err
}
if _, seekErr := storage.Seek(0, io.SeekStart); seekErr != nil {
return nil, seekErr
}
c.Request.Body = io.NopCloser(storage)
return &ModelRequest{
Model: model,
Group: group,
}, nil
}
func getJSONStringValue(result gjson.Result, field string) (string, error) {
if !result.Exists() || result.Type == gjson.Null {
return "", nil
}
if result.Type != gjson.String {
return "", fmt.Errorf("field %s must be a string", field)
}
return result.String(), nil
}
func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
var modelRequest ModelRequest
shouldSelectChannel := true
@@ -244,6 +299,7 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
} else if c.Request.Method == http.MethodGet {
relayMode = relayconstant.RelayModeVideoFetchByID
shouldSelectChannel = false
modelRequest.Model = getTaskOriginModelName(c)
}
c.Set("relay_mode", relayMode)
} else if strings.Contains(c.Request.URL.Path, "/v1/video/generations") {
@@ -258,6 +314,7 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
} else if c.Request.Method == http.MethodGet {
relayMode = relayconstant.RelayModeVideoFetchByID
shouldSelectChannel = false
modelRequest.Model = getTaskOriginModelName(c)
}
if _, ok := c.Get("relay_mode"); !ok {
c.Set("relay_mode", relayMode)
@@ -342,6 +399,31 @@ func getModelRequest(c *gin.Context) (*ModelRequest, bool, error) {
return &modelRequest, shouldSelectChannel, nil
}
// 修复 #4834: GET /v1/video/generations/:task_id && /v1/video/:task_id 此前不解析 model
// 当 token 启用「可用模型限制」时,下游 modelLimitEnable 校验会因
// modelRequest.Model 为空而误报 "This token has no access to model"。
// 从已存储的任务记录中回填 OriginModelName 即可让校验走在正确的模型上。
func getTaskOriginModelName(c *gin.Context) string {
if !common.GetContextKeyBool(c, constant.ContextKeyTokenModelLimitEnabled) {
return ""
}
taskId := c.Param("task_id")
if taskId == "" {
// jimeng adapter
taskId = c.GetString("task_id")
}
if taskId == "" {
return ""
}
userId := c.GetInt("id")
if task, exist, err := model.GetByTaskId(userId, taskId); err == nil && exist && task != nil {
return task.Properties.OriginModelName
}
return ""
}
func SetupContextForSelectedChannel(c *gin.Context, channel *model.Channel, modelName string) *types.NewAPIError {
c.Set("original_model", modelName) // for retry
if channel == nil {
+1 -1
View File
@@ -17,7 +17,7 @@ func RelayPanicRecover() gin.HandlerFunc {
common.SysLog(fmt.Sprintf("stacktrace from panic: %s", string(debug.Stack())))
c.JSON(http.StatusInternalServerError, gin.H{
"error": gin.H{
"message": fmt.Sprintf("Panic detected, error: %v. Please submit a issue here: https://github.com/Calcium-Ion/new-api", err),
"message": fmt.Sprintf("Panic detected, error: %v. Please submit a issue here: https://git.viaeon.com/admin/new-api/issues", err),
"type": "new_api_panic",
},
})
+47
View File
@@ -0,0 +1,47 @@
package middleware
import (
"bytes"
"io"
"net/http"
"github.com/QuantumNous/new-api/common"
"github.com/gin-gonic/gin"
)
func AnonymousRequestBodyLimit() gin.HandlerFunc {
return func(c *gin.Context) {
maxBytes := common.GetAnonymousRequestBodyLimitBytes()
if maxBytes <= 0 || c.Request.Body == nil {
c.Next()
return
}
originalBody := c.Request.Body
limitedBody, err := readAnonymousRequestBody(originalBody, maxBytes)
_ = originalBody.Close()
if err != nil {
if common.IsRequestBodyTooLargeError(err) {
c.AbortWithStatus(http.StatusRequestEntityTooLarge)
return
}
c.AbortWithStatus(http.StatusBadRequest)
return
}
c.Request.Body = io.NopCloser(bytes.NewReader(limitedBody))
c.Request.ContentLength = int64(len(limitedBody))
c.Next()
}
}
func readAnonymousRequestBody(body io.Reader, maxBytes int64) ([]byte, error) {
data, err := io.ReadAll(io.LimitReader(body, maxBytes+1))
if err != nil {
return nil, err
}
if int64(len(data)) > maxBytes {
return nil, common.ErrRequestBodyTooLarge
}
return data, nil
}
+33 -2
View File
@@ -643,13 +643,25 @@ func handlerMultiKeyUpdate(channel *Channel, usingKey string, status int, reason
if len(keys) == 0 {
channel.Status = status
} else {
var keyIndex int
keyIndex := -1
for i, key := range keys {
if key == usingKey {
keyIndex = i
break
}
}
if keyIndex < 0 {
if usingKey != "" {
common.SysLog(fmt.Sprintf("failed to update multi-key status: channel_id=%d, using key not found", channel.Id))
return
}
channel.Status = status
info := channel.GetOtherInfo()
info["status_reason"] = reason
info["status_time"] = common.GetTimestamp()
channel.SetOtherInfo(info)
return
}
if channel.ChannelInfo.MultiKeyStatusList == nil {
channel.ChannelInfo.MultiKeyStatusList = make(map[int]int)
}
@@ -666,16 +678,31 @@ func handlerMultiKeyUpdate(channel *Channel, usingKey string, status int, reason
channel.ChannelInfo.MultiKeyDisabledReason[keyIndex] = reason
channel.ChannelInfo.MultiKeyDisabledTime[keyIndex] = common.GetTimestamp()
}
if len(channel.ChannelInfo.MultiKeyStatusList) >= channel.ChannelInfo.MultiKeySize {
if !hasEnabledMultiKey(keys, channel.ChannelInfo.MultiKeyStatusList) {
channel.Status = common.ChannelStatusAutoDisabled
info := channel.GetOtherInfo()
info["status_reason"] = "All keys are disabled"
info["status_time"] = common.GetTimestamp()
channel.SetOtherInfo(info)
} else if status == common.ChannelStatusEnabled {
channel.Status = common.ChannelStatusEnabled
}
}
}
func hasEnabledMultiKey(keys []string, statusList map[int]int) bool {
for i := range keys {
if statusList == nil {
return true
}
status, ok := statusList[i]
if !ok || status == common.ChannelStatusEnabled {
return true
}
}
return false
}
func UpdateChannelStatus(channelId int, usingKey string, status int, reason string) bool {
if common.MemoryCacheEnabled {
channelStatusLock.Lock()
@@ -687,11 +714,15 @@ func UpdateChannelStatus(channelId int, usingKey string, status int, reason stri
}
if channelCache.ChannelInfo.IsMultiKey {
// Use per-channel lock to prevent concurrent map read/write with GetNextEnabledKey
beforeStatus := channelCache.Status
pollingLock := GetChannelPollingLock(channelId)
pollingLock.Lock()
// 如果是多Key模式,更新缓存中的状态
handlerMultiKeyUpdate(channelCache, usingKey, status, reason)
pollingLock.Unlock()
if beforeStatus != channelCache.Status {
CacheUpdateChannelStatus(channelId, channelCache.Status)
}
//CacheUpdateChannel(channelCache)
//return true
} else {
+96
View File
@@ -0,0 +1,96 @@
package model
import (
"time"
)
type Document struct {
Id int `json:"id" gorm:"primaryKey"`
Title string `json:"title" gorm:"not null"`
Slug string `json:"slug" gorm:"type:varchar(255);uniqueIndex;not null"`
Content string `json:"content" gorm:"type:text;not null"`
CategoryId *int `json:"category_id" gorm:"index"`
Visibility string `json:"visibility" gorm:"default:'public'"` // public, auth, admin
SortOrder int `json:"sort_order" gorm:"default:0"`
AuthorId int `json:"author_id" gorm:"not null"`
CreatedAt time.Time `json:"created_at" gorm:"autoCreateTime"`
UpdatedAt time.Time `json:"updated_at" gorm:"autoUpdateTime"`
}
func GetDocuments(keyword string, visibility string, categoryId *int, startIdx int, num int) ([]*Document, int64, error) {
query := DB.Model(&Document{})
if keyword != "" {
like := "%" + keyword + "%"
query = query.Where("title LIKE ? OR content LIKE ?", like, like)
}
if visibility != "" {
query = query.Where("visibility = ?", visibility)
}
if categoryId != nil {
query = query.Where("category_id = ?", *categoryId)
}
var total int64
if err := query.Count(&total).Error; err != nil {
return nil, 0, err
}
var documents []*Document
if err := query.Order("sort_order ASC, id DESC").Offset(startIdx).Limit(num).Find(&documents).Error; err != nil {
return nil, 0, err
}
return documents, total, nil
}
func GetDocumentsByVisibility(keyword string, visibilities []string, categoryId *int, startIdx int, num int) ([]*Document, int64, error) {
query := DB.Model(&Document{})
if keyword != "" {
like := "%" + keyword + "%"
query = query.Where("title LIKE ? OR content LIKE ?", like, like)
}
if len(visibilities) > 0 {
query = query.Where("visibility IN ?", visibilities)
}
if categoryId != nil {
query = query.Where("category_id = ?", *categoryId)
}
var total int64
if err := query.Count(&total).Error; err != nil {
return nil, 0, err
}
var documents []*Document
if err := query.Order("sort_order ASC, id DESC").Offset(startIdx).Limit(num).Find(&documents).Error; err != nil {
return nil, 0, err
}
return documents, total, nil
}
func GetDocumentBySlug(slug string) (*Document, error) {
var doc Document
err := DB.Where("slug = ?", slug).First(&doc).Error
if err != nil {
return nil, err
}
return &doc, nil
}
func GetDocumentById(id int) (*Document, error) {
var doc Document
err := DB.First(&doc, id).Error
if err != nil {
return nil, err
}
return &doc, nil
}
func CreateDocument(doc *Document) error {
return DB.Create(doc).Error
}
func UpdateDocument(doc *Document) error {
return DB.Model(doc).Select("title", "slug", "content", "category_id", "visibility", "sort_order").Updates(doc).Error
}
func DeleteDocument(id int) error {
// Delete associated versions first
DB.Where("document_id = ?", id).Delete(&DocumentVersion{})
return DB.Delete(&Document{}, id).Error
}
+38
View File
@@ -0,0 +1,38 @@
package model
import (
"time"
)
type DocumentCategory struct {
Id int `json:"id" gorm:"primaryKey"`
Name string `json:"name" gorm:"not null"`
Slug string `json:"slug" gorm:"type:varchar(255);uniqueIndex;not null"`
ParentId *int `json:"parent_id" gorm:"index"`
SortOrder int `json:"sort_order" gorm:"default:0"`
CreatedAt time.Time `json:"created_at" gorm:"autoCreateTime"`
}
func GetDocumentCategories() ([]*DocumentCategory, error) {
var categories []*DocumentCategory
err := DB.Order("sort_order ASC, id ASC").Find(&categories).Error
return categories, err
}
func GetDocumentCategoryTree() ([]*DocumentCategory, error) {
var categories []*DocumentCategory
err := DB.Order("sort_order ASC, id ASC").Find(&categories).Error
return categories, err
}
func CreateDocumentCategory(category *DocumentCategory) error {
return DB.Create(category).Error
}
func UpdateDocumentCategory(category *DocumentCategory) error {
return DB.Model(category).Select("name", "slug", "parent_id", "sort_order").Updates(category).Error
}
func DeleteDocumentCategory(id int) error {
return DB.Delete(&DocumentCategory{}, id).Error
}
+30
View File
@@ -0,0 +1,30 @@
package model
import (
"time"
)
type DocumentVersion struct {
Id int `json:"id" gorm:"primaryKey"`
DocumentId int `json:"document_id" gorm:"index;not null"`
Content string `json:"content" gorm:"type:text;not null"`
AuthorId int `json:"author_id" gorm:"not null"`
CreatedAt time.Time `json:"created_at" gorm:"autoCreateTime"`
}
func GetDocumentVersions(documentId int, startIdx int, num int) ([]*DocumentVersion, int64, error) {
query := DB.Model(&DocumentVersion{}).Where("document_id = ?", documentId)
var total int64
if err := query.Count(&total).Error; err != nil {
return nil, 0, err
}
var versions []*DocumentVersion
if err := query.Order("id DESC").Offset(startIdx).Limit(num).Find(&versions).Error; err != nil {
return nil, 0, err
}
return versions, total, nil
}
func CreateDocumentVersion(version *DocumentVersion) error {
return DB.Create(version).Error
}
+65 -49
View File
@@ -17,25 +17,39 @@ import (
"gorm.io/gorm"
)
func applyExplicitLogTextFilter(tx *gorm.DB, column string, value string) (*gorm.DB, error) {
if value == "" {
return tx, nil
}
if strings.Contains(value, "%") {
pattern, err := sanitizeLikePattern(value)
if err != nil {
return nil, err
}
return tx.Where(column+" LIKE ? ESCAPE '!'", pattern), nil
}
return tx.Where(column+" = ?", value), nil
}
type Log struct {
Id int `json:"id" gorm:"index:idx_created_at_id,priority:1;index:idx_user_id_id,priority:2"`
UserId int `json:"user_id" gorm:"index;index:idx_user_id_id,priority:1"`
CreatedAt int64 `json:"created_at" gorm:"bigint;index:idx_created_at_id,priority:2;index:idx_created_at_type"`
Type int `json:"type" gorm:"index:idx_created_at_type"`
Content string `json:"content"`
Username string `json:"username" gorm:"index;index:index_username_model_name,priority:2;default:''"`
TokenName string `json:"token_name" gorm:"index;default:''"`
ModelName string `json:"model_name" gorm:"index;index:index_username_model_name,priority:1;default:''"`
Quota int `json:"quota" gorm:"default:0"`
PromptTokens int `json:"prompt_tokens" gorm:"default:0"`
CompletionTokens int `json:"completion_tokens" gorm:"default:0"`
UseTime int `json:"use_time" gorm:"default:0"`
IsStream bool `json:"is_stream"`
ChannelId int `json:"channel" gorm:"index"`
ChannelName string `json:"channel_name" gorm:"->"`
TokenId int `json:"token_id" gorm:"default:0;index"`
Group string `json:"group" gorm:"index"`
Ip string `json:"ip" gorm:"index;default:''"`
Id int `json:"id" gorm:"index:idx_created_at_id,priority:2;index:idx_user_id_id,priority:2"`
UserId int `json:"user_id" gorm:"index;index:idx_user_id_id,priority:1"`
CreatedAt int64 `json:"created_at" gorm:"bigint;index:idx_created_at_id,priority:1;index:idx_created_at_type"`
Type int `json:"type" gorm:"index:idx_created_at_type"`
Content string `json:"content"`
Username string `json:"username" gorm:"index;index:index_username_model_name,priority:2;default:''"`
TokenName string `json:"token_name" gorm:"index;default:''"`
ModelName string `json:"model_name" gorm:"index;index:index_username_model_name,priority:1;default:''"`
Quota int `json:"quota" gorm:"default:0"`
PromptTokens int `json:"prompt_tokens" gorm:"default:0"`
CompletionTokens int `json:"completion_tokens" gorm:"default:0"`
UseTime int `json:"use_time" gorm:"default:0"`
IsStream bool `json:"is_stream"`
ChannelId int `json:"channel" gorm:"index"`
ChannelName string `json:"channel_name" gorm:"->"`
TokenId int `json:"token_id" gorm:"default:0;index"`
Group string `json:"group" gorm:"index"`
Ip string `json:"ip" gorm:"index;default:''"`
RequestId string `json:"request_id,omitempty" gorm:"type:varchar(64);index:idx_logs_request_id;default:''"`
UpstreamRequestId string `json:"upstream_request_id,omitempty" gorm:"type:varchar(128);index:idx_logs_upstream_request_id;default:''"`
Other string `json:"other"`
@@ -146,7 +160,7 @@ func RecordTopupLog(userId int, content string, callerIp string, paymentMethod s
func RecordErrorLog(c *gin.Context, userId int, channelId int, modelName string, tokenName string, content string, tokenId int, useTimeSeconds int,
isStream bool, group string, other map[string]interface{}) {
logger.LogInfo(c, fmt.Sprintf("record error log: userId=%d, channelId=%d, modelName=%s, tokenName=%s, content=%s", userId, channelId, modelName, tokenName, content))
logger.LogInfo(c, fmt.Sprintf("record error log: userId=%d, channelId=%d, modelName=%s, tokenName=%s, content=%s", userId, channelId, modelName, tokenName, common.LocalLogPreview(content)))
username := c.GetString("username")
requestId := c.GetString(common.RequestIdKey)
upstreamRequestId := c.GetString(common.UpstreamRequestIdKey)
@@ -309,9 +323,15 @@ func GetAllLogs(logType int, startTimestamp int64, endTimestamp int64, modelName
tx = LOG_DB.Where("logs.type = ?", logType)
}
tx = applyLogContainsFilter(tx, "logs.model_name", modelName)
tx = applyLogContainsFilter(tx, "logs.username", username)
tx = applyLogContainsFilter(tx, "logs.token_name", tokenName)
if tx, err = applyExplicitLogTextFilter(tx, "logs.model_name", modelName); err != nil {
return nil, 0, err
}
if tx, err = applyExplicitLogTextFilter(tx, "logs.username", username); err != nil {
return nil, 0, err
}
if tokenName != "" {
tx = tx.Where("logs.token_name = ?", tokenName)
}
if requestId != "" {
tx = tx.Where("logs.request_id = ?", requestId)
}
@@ -334,7 +354,7 @@ func GetAllLogs(logType int, startTimestamp int64, endTimestamp int64, modelName
if err != nil {
return nil, 0, err
}
err = tx.Order("logs.id desc").Limit(num).Offset(startIdx).Find(&logs).Error
err = tx.Order("logs.created_at desc, logs.id desc").Limit(num).Offset(startIdx).Find(&logs).Error
if err != nil {
return nil, 0, err
}
@@ -392,8 +412,12 @@ func GetUserLogs(userId int, logType int, startTimestamp int64, endTimestamp int
tx = LOG_DB.Where("logs.user_id = ? and logs.type = ?", userId, logType)
}
tx = applyLogContainsFilter(tx, "logs.model_name", modelName)
tx = applyLogContainsFilter(tx, "logs.token_name", tokenName)
if tx, err = applyExplicitLogTextFilter(tx, "logs.model_name", modelName); err != nil {
return nil, 0, err
}
if tokenName != "" {
tx = tx.Where("logs.token_name = ?", tokenName)
}
if requestId != "" {
tx = tx.Where("logs.request_id = ?", requestId)
}
@@ -430,42 +454,34 @@ type Stat struct {
Tpm int `json:"tpm"`
}
func logContainsPattern(input string) (string, bool) {
input = strings.TrimSpace(input)
if input == "" {
return "", false
}
replacer := strings.NewReplacer("!", "!!", "%", "!%", "_", "!_")
return "%" + replacer.Replace(input) + "%", true
}
func applyLogContainsFilter(tx *gorm.DB, column string, value string) *gorm.DB {
pattern, ok := logContainsPattern(value)
if !ok {
return tx
}
return tx.Where(column+" LIKE ? ESCAPE '!'", pattern)
}
func SumUsedQuota(logType int, startTimestamp int64, endTimestamp int64, modelName string, username string, tokenName string, channel int, group string) (stat Stat, err error) {
tx := LOG_DB.Table("logs").Select("sum(quota) quota")
// 为rpm和tpm创建单独的查询
rpmTpmQuery := LOG_DB.Table("logs").Select("count(*) rpm, sum(prompt_tokens) + sum(completion_tokens) tpm")
tx = applyLogContainsFilter(tx, "username", username)
rpmTpmQuery = applyLogContainsFilter(rpmTpmQuery, "username", username)
tx = applyLogContainsFilter(tx, "token_name", tokenName)
rpmTpmQuery = applyLogContainsFilter(rpmTpmQuery, "token_name", tokenName)
if tx, err = applyExplicitLogTextFilter(tx, "username", username); err != nil {
return stat, err
}
if rpmTpmQuery, err = applyExplicitLogTextFilter(rpmTpmQuery, "username", username); err != nil {
return stat, err
}
if tokenName != "" {
tx = tx.Where("token_name = ?", tokenName)
rpmTpmQuery = rpmTpmQuery.Where("token_name = ?", tokenName)
}
if startTimestamp != 0 {
tx = tx.Where("created_at >= ?", startTimestamp)
}
if endTimestamp != 0 {
tx = tx.Where("created_at <= ?", endTimestamp)
}
tx = applyLogContainsFilter(tx, "model_name", modelName)
rpmTpmQuery = applyLogContainsFilter(rpmTpmQuery, "model_name", modelName)
if tx, err = applyExplicitLogTextFilter(tx, "model_name", modelName); err != nil {
return stat, err
}
if rpmTpmQuery, err = applyExplicitLogTextFilter(rpmTpmQuery, "model_name", modelName); err != nil {
return stat, err
}
if channel != 0 {
tx = tx.Where("channel_id = ?", channel)
rpmTpmQuery = rpmTpmQuery.Where("channel_id = ?", channel)
+10
View File
@@ -281,6 +281,9 @@ func migrateDB() error {
&CustomOAuthProvider{},
&UserOAuthBinding{},
&PerfMetric{},
&DocumentCategory{},
&Document{},
&DocumentVersion{},
)
if err != nil {
return err
@@ -330,6 +333,9 @@ func migrateDBFast() error {
{&CustomOAuthProvider{}, "CustomOAuthProvider"},
{&UserOAuthBinding{}, "UserOAuthBinding"},
{&PerfMetric{}, "PerfMetric"},
{&DocumentCategory{}, "DocumentCategory"},
{&Document{}, "Document"},
{&DocumentVersion{}, "DocumentVersion"},
}
// 动态计算migration数量,确保errChan缓冲区足够大
errChan := make(chan error, len(migrations))
@@ -397,8 +403,10 @@ func ensureSubscriptionPlanTableSQLite() error {
` + "`custom_seconds`" + ` bigint NOT NULL DEFAULT 0,
` + "`enabled`" + ` numeric DEFAULT 1,
` + "`sort_order`" + ` integer DEFAULT 0,
` + "`allow_balance_pay`" + ` numeric DEFAULT 1,
` + "`stripe_price_id`" + ` varchar(128) DEFAULT '',
` + "`creem_product_id`" + ` varchar(128) DEFAULT '',
` + "`waffo_pancake_product_id`" + ` varchar(128) DEFAULT '',
` + "`max_purchase_per_user`" + ` integer DEFAULT 0,
` + "`upgrade_group`" + ` varchar(64) DEFAULT '',
` + "`total_amount`" + ` bigint NOT NULL DEFAULT 0,
@@ -430,8 +438,10 @@ PRIMARY KEY (` + "`id`" + `)
{Name: "custom_seconds", DDL: "`custom_seconds` bigint NOT NULL DEFAULT 0"},
{Name: "enabled", DDL: "`enabled` numeric DEFAULT 1"},
{Name: "sort_order", DDL: "`sort_order` integer DEFAULT 0"},
{Name: "allow_balance_pay", DDL: "`allow_balance_pay` numeric DEFAULT 1"},
{Name: "stripe_price_id", DDL: "`stripe_price_id` varchar(128) DEFAULT ''"},
{Name: "creem_product_id", DDL: "`creem_product_id` varchar(128) DEFAULT ''"},
{Name: "waffo_pancake_product_id", DDL: "`waffo_pancake_product_id` varchar(128) DEFAULT ''"},
{Name: "max_purchase_per_user", DDL: "`max_purchase_per_user` integer DEFAULT 0"},
{Name: "upgrade_group", DDL: "`upgrade_group` varchar(64) DEFAULT ''"},
{Name: "total_amount", DDL: "`total_amount` bigint NOT NULL DEFAULT 0"},
+57
View File
@@ -2,6 +2,7 @@ package model
import (
"strconv"
"strings"
"github.com/QuantumNous/new-api/common"
@@ -135,6 +136,62 @@ func GetBoundChannelsByModelsMap(modelNames []string) (map[string][]BoundChannel
return result, nil
}
func normalizeLookupValues(values []string) []string {
seen := make(map[string]struct{}, len(values))
normalized := make([]string, 0, len(values))
for _, value := range values {
value = strings.TrimSpace(value)
if value == "" {
continue
}
if _, ok := seen[value]; ok {
continue
}
seen[value] = struct{}{}
normalized = append(normalized, value)
}
return normalized
}
func GetPreferredModelOwnerChannelTypes(modelNames []string, groups []string) (map[string]int, error) {
result := make(map[string]int)
modelNames = normalizeLookupValues(modelNames)
if len(modelNames) == 0 {
return result, nil
}
type row struct {
Model string
ChannelType int
}
var rows []row
query := DB.Table("abilities").
Select("abilities.model as model, channels.type as channel_type").
Joins("JOIN channels ON abilities.channel_id = channels.id").
Where("abilities.model IN ? AND abilities.enabled = ? AND channels.status = ?", modelNames, true, common.ChannelStatusEnabled).
Order("COALESCE(abilities.priority, 0) DESC").
Order("abilities.weight DESC").
Order("abilities.channel_id ASC")
groups = normalizeLookupValues(groups)
if len(groups) > 0 {
query = query.Where("abilities."+commonGroupCol+" IN ?", groups)
}
if err := query.Scan(&rows).Error; err != nil {
return nil, err
}
for _, r := range rows {
if _, ok := result[r.Model]; ok {
continue
}
result[r.Model] = r.ChannelType
}
return result, nil
}
func SearchModels(keyword string, vendor string, offset int, limit int) ([]*Model, int64, error) {
var models []*Model
db := DB.Model(&Model{})
+141
View File
@@ -0,0 +1,141 @@
package model
import (
"fmt"
"testing"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/constant"
"github.com/stretchr/testify/require"
)
func clearPreferredOwnerTables(t *testing.T) {
t.Helper()
require.NoError(t, DB.Exec("DELETE FROM abilities").Error)
require.NoError(t, DB.Exec("DELETE FROM channels").Error)
}
func insertPreferredOwnerCandidate(
t *testing.T,
channelID int,
modelName string,
group string,
channelType int,
priority int64,
weight uint,
channelStatus int,
abilityEnabled bool,
) {
t.Helper()
require.NoError(t, DB.Create(&Channel{
Id: channelID,
Type: channelType,
Key: fmt.Sprintf("key-%d", channelID),
Status: channelStatus,
Name: fmt.Sprintf("channel-%d", channelID),
}).Error)
require.NoError(t, DB.Create(&Ability{
Group: group,
Model: modelName,
ChannelId: channelID,
Enabled: abilityEnabled,
Priority: &priority,
Weight: weight,
}).Error)
}
func TestGetPreferredModelOwnerChannelTypes(t *testing.T) {
const modelName = "gpt-5.4"
tests := []struct {
name string
setup func(t *testing.T)
groups []string
expected int
found bool
}{
{
name: "openai only",
setup: func(t *testing.T) {
insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 0, 0, common.ChannelStatusEnabled, true)
},
groups: []string{"default"},
expected: constant.ChannelTypeOpenAI,
found: true,
},
{
name: "codex only",
setup: func(t *testing.T) {
insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeCodex, 0, 0, common.ChannelStatusEnabled, true)
},
groups: []string{"default"},
expected: constant.ChannelTypeCodex,
found: true,
},
{
name: "priority wins",
setup: func(t *testing.T) {
insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 1, 100, common.ChannelStatusEnabled, true)
insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeCodex, 2, 0, common.ChannelStatusEnabled, true)
},
groups: []string{"default"},
expected: constant.ChannelTypeCodex,
found: true,
},
{
name: "weight wins when priority is equal",
setup: func(t *testing.T) {
insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 1, 10, common.ChannelStatusEnabled, true)
insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeCodex, 1, 20, common.ChannelStatusEnabled, true)
},
groups: []string{"default"},
expected: constant.ChannelTypeCodex,
found: true,
},
{
name: "channel id stabilizes exact ties",
setup: func(t *testing.T) {
insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeCodex, 1, 10, common.ChannelStatusEnabled, true)
insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeOpenAI, 1, 10, common.ChannelStatusEnabled, true)
},
groups: []string{"default"},
expected: constant.ChannelTypeOpenAI,
found: true,
},
{
name: "group filter excludes other groups",
setup: func(t *testing.T) {
insertPreferredOwnerCandidate(t, 1, modelName, "vip", constant.ChannelTypeCodex, 10, 100, common.ChannelStatusEnabled, true)
insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeOpenAI, 1, 0, common.ChannelStatusEnabled, true)
},
groups: []string{"default"},
expected: constant.ChannelTypeOpenAI,
found: true,
},
{
name: "disabled candidates are ignored",
setup: func(t *testing.T) {
insertPreferredOwnerCandidate(t, 1, modelName, "default", constant.ChannelTypeCodex, 10, 100, common.ChannelStatusEnabled, false)
insertPreferredOwnerCandidate(t, 2, modelName, "default", constant.ChannelTypeOpenAI, 1, 0, common.ChannelStatusManuallyDisabled, true)
},
groups: []string{"default"},
found: false,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
clearPreferredOwnerTables(t)
tt.setup(t)
owners, err := GetPreferredModelOwnerChannelTypes([]string{modelName}, tt.groups)
require.NoError(t, err)
got, ok := owners[modelName]
require.Equal(t, tt.found, ok)
if tt.found {
require.Equal(t, tt.expected, got)
}
})
}
}
+38 -19
View File
@@ -12,6 +12,7 @@ import (
"github.com/QuantumNous/new-api/setting/performance_setting"
"github.com/QuantumNous/new-api/setting/ratio_setting"
"github.com/QuantumNous/new-api/setting/system_setting"
"gorm.io/gorm"
)
type Option struct {
@@ -106,18 +107,13 @@ func InitOptionMap() {
common.OptionMap["WaffoUnitPrice"] = strconv.FormatFloat(setting.WaffoUnitPrice, 'f', -1, 64)
common.OptionMap["WaffoMinTopUp"] = strconv.Itoa(setting.WaffoMinTopUp)
common.OptionMap["WaffoPayMethods"] = setting.WaffoPayMethods2JsonString()
common.OptionMap["WaffoPancakeEnabled"] = strconv.FormatBool(setting.WaffoPancakeEnabled)
common.OptionMap["WaffoPancakeSandbox"] = strconv.FormatBool(setting.WaffoPancakeSandbox)
common.OptionMap["WaffoPancakeMerchantID"] = setting.WaffoPancakeMerchantID
common.OptionMap["WaffoPancakePrivateKey"] = setting.WaffoPancakePrivateKey
common.OptionMap["WaffoPancakeWebhookPublicKey"] = setting.WaffoPancakeWebhookPublicKey
common.OptionMap["WaffoPancakeWebhookTestKey"] = setting.WaffoPancakeWebhookTestKey
common.OptionMap["WaffoPancakeStoreID"] = setting.WaffoPancakeStoreID
common.OptionMap["WaffoPancakeProductID"] = setting.WaffoPancakeProductID
common.OptionMap["WaffoPancakeReturnURL"] = setting.WaffoPancakeReturnURL
common.OptionMap["WaffoPancakeCurrency"] = setting.WaffoPancakeCurrency
common.OptionMap["WaffoPancakeUnitPrice"] = strconv.FormatFloat(setting.WaffoPancakeUnitPrice, 'f', -1, 64)
common.OptionMap["WaffoPancakeMinTopUp"] = strconv.Itoa(setting.WaffoPancakeMinTopUp)
common.OptionMap["WaffoPancakeStoreID"] = setting.WaffoPancakeStoreID
common.OptionMap["WaffoPancakeProductID"] = setting.WaffoPancakeProductID
common.OptionMap["TopupGroupRatio"] = common.TopupGroupRatio2JSONString()
common.OptionMap["Chats"] = setting.Chats2JsonString()
common.OptionMap["AutoGroups"] = setting.AutoGroups2JsonString()
@@ -222,6 +218,39 @@ func UpdateOption(key string, value string) error {
return updateOptionMap(key, value)
}
// UpdateOptionsBulk persists multiple key/value pairs in a single database
// transaction, then dispatches them through updateOptionMap in one pass. If
// any DB write fails the whole transaction rolls back and no in-memory state
// is touched — safe for callers that must commit a set of related options
// atomically (e.g. payment gateway binding).
func UpdateOptionsBulk(values map[string]string) error {
if len(values) == 0 {
return nil
}
err := DB.Transaction(func(tx *gorm.DB) error {
for k, v := range values {
option := Option{Key: k}
if err := tx.FirstOrCreate(&option, Option{Key: k}).Error; err != nil {
return err
}
option.Value = v
if err := tx.Save(&option).Error; err != nil {
return err
}
}
return nil
})
if err != nil {
return err
}
for k, v := range values {
if err := updateOptionMap(k, v); err != nil {
return err
}
}
return nil
}
func updateOptionMap(key string, value string) (err error) {
common.OptionMapRWMutex.Lock()
defer common.OptionMapRWMutex.Unlock()
@@ -419,26 +448,16 @@ func updateOptionMap(key string, value string) (err error) {
setting.WaffoUnitPrice, _ = strconv.ParseFloat(value, 64)
case "WaffoMinTopUp":
setting.WaffoMinTopUp, _ = strconv.Atoi(value)
case "WaffoPancakeEnabled":
setting.WaffoPancakeEnabled = value == "true"
case "WaffoPancakeSandbox":
setting.WaffoPancakeSandbox = value == "true"
case "WaffoPancakeMerchantID":
setting.WaffoPancakeMerchantID = value
case "WaffoPancakePrivateKey":
setting.WaffoPancakePrivateKey = value
case "WaffoPancakeWebhookPublicKey":
setting.WaffoPancakeWebhookPublicKey = value
case "WaffoPancakeWebhookTestKey":
setting.WaffoPancakeWebhookTestKey = value
case "WaffoPancakeReturnURL":
setting.WaffoPancakeReturnURL = value
case "WaffoPancakeStoreID":
setting.WaffoPancakeStoreID = value
case "WaffoPancakeProductID":
setting.WaffoPancakeProductID = value
case "WaffoPancakeReturnURL":
setting.WaffoPancakeReturnURL = value
case "WaffoPancakeCurrency":
setting.WaffoPancakeCurrency = value
case "WaffoPancakeUnitPrice":
setting.WaffoPancakeUnitPrice, _ = strconv.ParseFloat(value, 64)
case "WaffoPancakeMinTopUp":
+117 -2
View File
@@ -11,6 +11,7 @@ import (
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/pkg/cachex"
"github.com/samber/hot"
"github.com/shopspring/decimal"
"gorm.io/gorm"
)
@@ -159,8 +160,11 @@ type SubscriptionPlan struct {
Enabled bool `json:"enabled" gorm:"default:true"`
SortOrder int `json:"sort_order" gorm:"type:int;default:0"`
StripePriceId string `json:"stripe_price_id" gorm:"type:varchar(128);default:''"`
CreemProductId string `json:"creem_product_id" gorm:"type:varchar(128);default:''"`
AllowBalancePay *bool `json:"allow_balance_pay" gorm:"default:true"`
StripePriceId string `json:"stripe_price_id" gorm:"type:varchar(128);default:''"`
CreemProductId string `json:"creem_product_id" gorm:"type:varchar(128);default:''"`
WaffoPancakeProductId string `json:"waffo_pancake_product_id" gorm:"type:varchar(128);default:''"`
// Max purchases per user (0 = unlimited)
MaxPurchasePerUser int `json:"max_purchase_per_user" gorm:"type:int;default:0"`
@@ -191,6 +195,12 @@ func (p *SubscriptionPlan) BeforeUpdate(tx *gorm.DB) error {
return nil
}
func (p *SubscriptionPlan) NormalizeDefaults() {
if p.AllowBalancePay == nil {
p.AllowBalancePay = common.GetPointer(true)
}
}
// Subscription order (payment -> webhook -> create UserSubscription)
type SubscriptionOrder struct {
Id int `json:"id"`
@@ -358,6 +368,7 @@ func getSubscriptionPlanByIdTx(tx *gorm.DB, id int) (*SubscriptionPlan, error) {
key := subscriptionPlanCacheKey(id)
if key != "" {
if cached, found, err := getSubscriptionPlanCache().Get(key); err == nil && found {
cached.NormalizeDefaults()
return &cached, nil
}
}
@@ -369,6 +380,7 @@ func getSubscriptionPlanByIdTx(tx *gorm.DB, id int) (*SubscriptionPlan, error) {
if err := query.Where("id = ?", id).First(&plan).Error; err != nil {
return nil, err
}
plan.NormalizeDefaults()
_ = getSubscriptionPlanCache().SetWithTTL(key, plan, subscriptionPlanCacheTTL())
return &plan, nil
}
@@ -664,6 +676,109 @@ func AdminBindSubscription(userId int, planId int, sourceNote string) (string, e
return "", nil
}
func calcSubscriptionBalanceQuota(priceAmount float64) (int, error) {
if priceAmount <= 0 {
return 0, nil
}
if common.QuotaPerUnit <= 0 {
return 0, errors.New("额度单位配置错误")
}
quota := decimal.NewFromFloat(priceAmount).
Mul(decimal.NewFromFloat(common.QuotaPerUnit)).
Ceil().
IntPart()
return int(quota), nil
}
// PurchaseSubscriptionWithBalance creates a subscription by deducting the user's wallet quota.
func PurchaseSubscriptionWithBalance(userId int, planId int) error {
if userId <= 0 || planId <= 0 {
return errors.New("invalid userId or planId")
}
var logPlanTitle string
var logMoney float64
var chargedQuota int
var upgradeGroup string
err := DB.Transaction(func(tx *gorm.DB) error {
plan, err := getSubscriptionPlanByIdTx(tx, planId)
if err != nil {
return err
}
if !plan.Enabled {
return errors.New("套餐未启用")
}
if plan.PriceAmount < 0 {
return errors.New("套餐价格不能为负数")
}
if plan.AllowBalancePay != nil && !*plan.AllowBalancePay {
return errors.New("该套餐不允许使用余额兑换")
}
requiredQuota, err := calcSubscriptionBalanceQuota(plan.PriceAmount)
if err != nil {
return err
}
var user User
if err := tx.Set("gorm:query_option", "FOR UPDATE").Where("id = ?", userId).First(&user).Error; err != nil {
return err
}
if requiredQuota > 0 && user.Quota < requiredQuota {
return errors.New("余额不足")
}
if requiredQuota > 0 {
if err := tx.Model(&User{}).Where("id = ?", userId).
Update("quota", gorm.Expr("quota - ?", requiredQuota)).Error; err != nil {
return err
}
}
if _, err := CreateUserSubscriptionFromPlanTx(tx, userId, plan, PaymentMethodBalance); err != nil {
return err
}
now := common.GetTimestamp()
tradeNo := fmt.Sprintf("SUBBALUSR%dNO%s%d", userId, common.GetRandomString(6), time.Now().UnixNano())
order := &SubscriptionOrder{
UserId: userId,
PlanId: plan.Id,
Money: plan.PriceAmount,
TradeNo: tradeNo,
PaymentMethod: PaymentMethodBalance,
PaymentProvider: PaymentProviderBalance,
Status: common.TopUpStatusSuccess,
CreateTime: now,
CompleteTime: now,
ProviderPayload: fmt.Sprintf("charged_quota=%d", requiredQuota),
}
if err := tx.Create(order).Error; err != nil {
return err
}
logPlanTitle = plan.Title
logMoney = plan.PriceAmount
chargedQuota = requiredQuota
upgradeGroup = strings.TrimSpace(plan.UpgradeGroup)
return nil
})
if err != nil {
return err
}
if chargedQuota > 0 {
if err := cacheDecrUserQuota(userId, int64(chargedQuota)); err != nil {
common.SysLog("failed to decrease user quota cache after subscription balance purchase: " + err.Error())
}
}
if upgradeGroup != "" {
_ = UpdateUserGroupCache(userId, upgradeGroup)
}
msg := fmt.Sprintf("使用余额购买订阅成功,套餐: %s,支付金额: %.2f,扣除额度: %d", logPlanTitle, logMoney, chargedQuota)
RecordLog(userId, LogTypeTopup, msg)
return nil
}
// GetAllActiveUserSubscriptions returns all active subscriptions for a user.
func GetAllActiveUserSubscriptions(userId int) ([]SubscriptionSummary, error) {
if userId <= 0 {
+2
View File
@@ -40,6 +40,7 @@ func TestMain(m *testing.M) {
&Token{},
&Log{},
&Channel{},
&Ability{},
&TopUp{},
&SubscriptionPlan{},
&SubscriptionOrder{},
@@ -60,6 +61,7 @@ func truncateTables(t *testing.T) {
DB.Exec("DELETE FROM tokens")
DB.Exec("DELETE FROM logs")
DB.Exec("DELETE FROM channels")
DB.Exec("DELETE FROM abilities")
DB.Exec("DELETE FROM top_ups")
DB.Exec("DELETE FROM subscription_orders")
DB.Exec("DELETE FROM subscription_plans")
+2
View File
@@ -29,6 +29,7 @@ const (
PaymentMethodCreem = "creem"
PaymentMethodWaffo = "waffo"
PaymentMethodWaffoPancake = "waffo_pancake"
PaymentMethodBalance = "balance"
)
const (
@@ -37,6 +38,7 @@ const (
PaymentProviderCreem = "creem"
PaymentProviderWaffo = "waffo"
PaymentProviderWaffoPancake = "waffo_pancake"
PaymentProviderBalance = "balance"
)
var (
+33 -20
View File
@@ -11,7 +11,6 @@ import (
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/dto"
"github.com/QuantumNous/new-api/logger"
"github.com/QuantumNous/new-api/setting/operation_setting"
"github.com/bytedance/gopkg/util/gopool"
"gorm.io/gorm"
@@ -225,7 +224,7 @@ func GetAllUsers(pageInfo *common.PageInfo) (users []*User, total int64, err err
return users, total, nil
}
func SearchUsers(keyword string, group string, startIdx int, num int) ([]*User, int64, error) {
func SearchUsers(keyword string, group string, role *int, status *int, startIdx int, num int) ([]*User, int64, error) {
var users []*User
var total int64
var err error
@@ -246,28 +245,25 @@ func SearchUsers(keyword string, group string, startIdx int, num int) ([]*User,
// 构建搜索条件
likeCondition := "username LIKE ? OR email LIKE ? OR display_name LIKE ?"
likeArgs := []interface{}{"%" + keyword + "%", "%" + keyword + "%", "%" + keyword + "%"}
// 尝试将关键字转换为整数ID
keywordInt, err := strconv.Atoi(keyword)
if err == nil {
// 如果是数字,同时搜索ID和其他字段
likeCondition = "id = ? OR " + likeCondition
if group != "" {
query = query.Where("("+likeCondition+") AND "+commonGroupCol+" = ?",
keywordInt, "%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%", group)
} else {
query = query.Where(likeCondition,
keywordInt, "%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%")
}
} else {
// 非数字关键字,只搜索字符串字段
if group != "" {
query = query.Where("("+likeCondition+") AND "+commonGroupCol+" = ?",
"%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%", group)
} else {
query = query.Where(likeCondition,
"%"+keyword+"%", "%"+keyword+"%", "%"+keyword+"%")
}
likeArgs = append([]interface{}{keywordInt}, likeArgs...)
}
query = query.Where("("+likeCondition+")", likeArgs...)
if group != "" {
query = query.Where(commonGroupCol+" = ?", group)
}
if role != nil {
query = query.Where("role = ?", *role)
}
if status != nil {
query = query.Where("status = ?", *status)
}
// 获取总数
@@ -421,7 +417,7 @@ func (user *User) Insert(inviterId int) error {
if common.QuotaForNewUser > 0 {
RecordLog(user.Id, LogTypeSystem, fmt.Sprintf("新用户注册赠送 %s", logger.LogQuota(common.QuotaForNewUser)))
}
if inviterId != 0 && operation_setting.IsPaymentComplianceConfirmed() {
if inviterId != 0 {
if common.QuotaForInvitee > 0 {
_ = IncreaseUserQuota(user.Id, common.QuotaForInvitee, true)
RecordLog(user.Id, LogTypeSystem, fmt.Sprintf("使用邀请码赠送 %s", logger.LogQuota(common.QuotaForInvitee)))
@@ -482,7 +478,7 @@ func (user *User) FinalizeOAuthUserCreation(inviterId int) {
if common.QuotaForNewUser > 0 {
RecordLog(user.Id, LogTypeSystem, fmt.Sprintf("新用户注册赠送 %s", logger.LogQuota(common.QuotaForNewUser)))
}
if inviterId != 0 && operation_setting.IsPaymentComplianceConfirmed() {
if inviterId != 0 {
if common.QuotaForInvitee > 0 {
_ = IncreaseUserQuota(user.Id, common.QuotaForInvitee, true)
RecordLog(user.Id, LogTypeSystem, fmt.Sprintf("使用邀请码赠送 %s", logger.LogQuota(common.QuotaForInvitee)))
@@ -987,6 +983,23 @@ func updateUserUsedQuotaAndRequestCount(id int, quota int, count int) {
//}
}
func updateUserQuotaUsedQuotaAndRequestCount(id int, quota int, usedQuota int, requestCount int) {
if quota == 0 && usedQuota == 0 && requestCount == 0 {
return
}
err := DB.Model(&User{}).Where("id = ?", id).Updates(
map[string]interface{}{
"quota": gorm.Expr("quota + ?", quota),
"used_quota": gorm.Expr("used_quota + ?", usedQuota),
"request_count": gorm.Expr("request_count + ?", requestCount),
},
).Error
if err != nil {
common.SysLog("failed to batch update user quota, used quota and request count: " + err.Error())
}
}
func updateUserUsedQuota(id int, quota int) {
err := DB.Model(&User{}).Where("id = ?", id).Updates(
map[string]interface{}{
+26 -11
View File
@@ -67,33 +67,48 @@ func batchUpdate() {
}
common.SysLog("batch update started")
stores := make([]map[int]int, BatchUpdateTypeCount)
for i := 0; i < BatchUpdateTypeCount; i++ {
batchUpdateLocks[i].Lock()
store := batchUpdateStores[i]
stores[i] = batchUpdateStores[i]
batchUpdateStores[i] = make(map[int]int)
batchUpdateLocks[i].Unlock()
// TODO: maybe we can combine updates with same key?
}
for i, store := range stores {
if i == BatchUpdateTypeUserQuota || i == BatchUpdateTypeUsedQuota || i == BatchUpdateTypeRequestCount {
continue
}
for key, value := range store {
switch i {
case BatchUpdateTypeUserQuota:
err := increaseUserQuota(key, value)
if err != nil {
common.SysLog("failed to batch update user quota: " + err.Error())
}
case BatchUpdateTypeTokenQuota:
err := increaseTokenQuota(key, value)
if err != nil {
common.SysLog("failed to batch update token quota: " + err.Error())
}
case BatchUpdateTypeUsedQuota:
updateUserUsedQuota(key, value)
case BatchUpdateTypeRequestCount:
updateUserRequestCount(key, value)
case BatchUpdateTypeChannelUsedQuota:
updateChannelUsedQuota(key, value)
}
}
}
userQuotaStore := stores[BatchUpdateTypeUserQuota]
usedQuotaStore := stores[BatchUpdateTypeUsedQuota]
requestCountStore := stores[BatchUpdateTypeRequestCount]
userIDs := make(map[int]struct{}, len(userQuotaStore)+len(usedQuotaStore)+len(requestCountStore))
for key := range userQuotaStore {
userIDs[key] = struct{}{}
}
for key := range usedQuotaStore {
userIDs[key] = struct{}{}
}
for key := range requestCountStore {
userIDs[key] = struct{}{}
}
for key := range userIDs {
updateUserQuotaUsedQuotaAndRequestCount(key, userQuotaStore[key], usedQuotaStore[key], requestCountStore[key])
}
common.SysLog("batch update finished")
}
+20
View File
@@ -25,6 +25,23 @@ import (
"github.com/gorilla/websocket"
)
// applyUpstreamContentLength populates req.ContentLength when the upstream
// body is wrapped in a BodyStorage (see relay/common/outbound_body.go).
//
// net/http.NewRequest only auto-detects ContentLength for *bytes.Reader,
// *bytes.Buffer and *strings.Reader. When the body is a type-erased io.Reader
// (which is the case for ReaderOnly(BodyStorage)), the Content-Length header
// would otherwise be omitted, forcing chunked transfer encoding and breaking
// some upstreams that require an explicit Content-Length.
func applyUpstreamContentLength(req *http.Request, info *common.RelayInfo) {
if info == nil {
return
}
if info.UpstreamRequestBodySize > 0 && req.ContentLength <= 0 {
req.ContentLength = info.UpstreamRequestBodySize
}
}
func SetupApiRequestHeader(info *common.RelayInfo, c *gin.Context, req *http.Header) {
if info.RelayMode == constant.RelayModeAudioTranscription || info.RelayMode == constant.RelayModeAudioTranslation {
// multipart/form-data
@@ -297,6 +314,7 @@ func DoApiRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBody
if err != nil {
return nil, fmt.Errorf("new request failed: %w", err)
}
applyUpstreamContentLength(req, info)
headers := req.Header
err = a.SetupRequestHeader(c, &headers, info)
if err != nil {
@@ -326,6 +344,7 @@ func DoFormRequest(a Adaptor, c *gin.Context, info *common.RelayInfo, requestBod
if err != nil {
return nil, fmt.Errorf("new request failed: %w", err)
}
applyUpstreamContentLength(req, info)
// set form data
req.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
headers := req.Header
@@ -522,6 +541,7 @@ func DoTaskApiRequest(a TaskAdaptor, c *gin.Context, info *common.RelayInfo, req
if err != nil {
return nil, fmt.Errorf("new request failed: %w", err)
}
applyUpstreamContentLength(req, info)
req.GetBody = func() (io.ReadCloser, error) {
return io.NopCloser(requestBody), nil
}
+6
View File
@@ -19,6 +19,7 @@ var awsModelIDMap = map[string]string{
"claude-opus-4-5-20251101": "anthropic.claude-opus-4-5-20251101-v1:0",
"claude-opus-4-6": "anthropic.claude-opus-4-6-v1",
"claude-opus-4-7": "anthropic.claude-opus-4-7",
"claude-opus-4-8": "anthropic.claude-opus-4-8",
// Nova models
"nova-micro-v1:0": "amazon.nova-micro-v1:0",
"nova-lite-v1:0": "amazon.nova-lite-v1:0",
@@ -97,6 +98,11 @@ var awsModelCanCrossRegionMap = map[string]map[string]bool{
"ap": true,
"eu": true,
},
"anthropic.claude-opus-4-8": {
"us": true,
"ap": true,
"eu": true,
},
"anthropic.claude-haiku-4-5-20251001-v1:0": {
"us": true,
"ap": true,
+7
View File
@@ -33,6 +33,13 @@ var ModelList = []string{
"claude-opus-4-7-medium",
"claude-opus-4-7-low",
"claude-opus-4-7-thinking",
"claude-opus-4-8",
"claude-opus-4-8-max",
"claude-opus-4-8-xhigh",
"claude-opus-4-8-high",
"claude-opus-4-8-medium",
"claude-opus-4-8-low",
"claude-opus-4-8-thinking",
}
var ChannelName = "claude"
+10 -9
View File
@@ -154,14 +154,17 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
}
if baseModel, effortLevel, ok := reasoning.TrimEffortSuffix(textRequest.Model); ok && effortLevel != "" &&
(strings.HasPrefix(textRequest.Model, "claude-opus-4-6") || strings.HasPrefix(textRequest.Model, "claude-opus-4-7")) {
(strings.HasPrefix(textRequest.Model, "claude-opus-4-6") ||
strings.HasPrefix(textRequest.Model, "claude-opus-4-7") ||
strings.HasPrefix(textRequest.Model, "claude-opus-4-8")) {
claudeRequest.Model = baseModel
claudeRequest.Thinking = &dto.Thinking{
Type: "adaptive",
}
claudeRequest.OutputConfig = json.RawMessage(fmt.Sprintf(`{"effort":"%s"}`, effortLevel))
if strings.HasPrefix(baseModel, "claude-opus-4-7") {
// Opus 4.7 rejects non-default temperature/top_p/top_k with 400
if strings.HasPrefix(baseModel, "claude-opus-4-7") ||
strings.HasPrefix(baseModel, "claude-opus-4-8") {
// Opus 4.7/4.8 reject non-default temperature/top_p/top_k with 400
// and defaults display to "omitted"; restore the 4.6 visible summary.
claudeRequest.Thinking.Display = "summarized"
claudeRequest.Temperature = nil
@@ -175,8 +178,9 @@ func RequestOpenAI2ClaudeMessage(c *gin.Context, textRequest dto.GeneralOpenAIRe
strings.HasSuffix(textRequest.Model, "-thinking") {
trimmedModel := strings.TrimSuffix(textRequest.Model, "-thinking")
if strings.HasPrefix(trimmedModel, "claude-opus-4-7") {
// Opus 4.7 rejects thinking.type="enabled"; use adaptive at high effort.
if strings.HasPrefix(trimmedModel, "claude-opus-4-7") ||
strings.HasPrefix(trimmedModel, "claude-opus-4-8") {
// Opus 4.7/4.8 reject thinking.type="enabled"; use adaptive at high effort.
claudeRequest.Thinking = &dto.Thinking{Type: "adaptive", Display: "summarized"}
claudeRequest.OutputConfig = json.RawMessage(`{"effort":"high"}`)
claudeRequest.Temperature = nil
@@ -442,10 +446,7 @@ func StreamResponseClaude2OpenAI(claudeResponse *dto.ClaudeResponse) *dto.ChatCo
tools := make([]dto.ToolCallResponse, 0)
fcIdx := 0
if claudeResponse.Index != nil {
fcIdx = *claudeResponse.Index - 1
if fcIdx < 0 {
fcIdx = 0
}
fcIdx = *claudeResponse.Index
}
var choice dto.ChatCompletionsStreamResponseChoice
if claudeResponse.Type == "message_start" {
+56
View File
@@ -9,6 +9,10 @@ import (
"github.com/stretchr/testify/require"
)
func commonPointer[T any](value T) *T {
return &value
}
func TestFormatClaudeResponseInfo_MessageStart(t *testing.T) {
claudeInfo := &ClaudeResponseInfo{
Usage: &dto.Usage{},
@@ -310,6 +314,58 @@ func TestRequestOpenAI2ClaudeMessage_IgnoresUnsupportedFileContent(t *testing.T)
require.Equal(t, "see attachment", *content[0].Text)
}
func TestRequestOpenAI2ClaudeMessage_ClaudeOpus48HighUsesAdaptiveThinking(t *testing.T) {
request := dto.GeneralOpenAIRequest{
Model: "claude-opus-4-8-high",
Temperature: commonPointer(0.7),
TopP: commonPointer(0.9),
TopK: commonPointer(40),
Messages: []dto.Message{
{
Role: "user",
Content: "hello",
},
},
}
claudeRequest, err := RequestOpenAI2ClaudeMessage(nil, request)
require.NoError(t, err)
require.Equal(t, "claude-opus-4-8", claudeRequest.Model)
require.NotNil(t, claudeRequest.Thinking)
require.Equal(t, "adaptive", claudeRequest.Thinking.Type)
require.Equal(t, "summarized", claudeRequest.Thinking.Display)
require.JSONEq(t, `{"effort":"high"}`, string(claudeRequest.OutputConfig))
require.Nil(t, claudeRequest.Temperature)
require.Nil(t, claudeRequest.TopP)
require.Nil(t, claudeRequest.TopK)
}
func TestRequestOpenAI2ClaudeMessage_ClaudeOpus48ThinkingUsesAdaptiveHighEffort(t *testing.T) {
request := dto.GeneralOpenAIRequest{
Model: "claude-opus-4-8-thinking",
Temperature: commonPointer(0.7),
TopP: commonPointer(0.9),
TopK: commonPointer(40),
Messages: []dto.Message{
{
Role: "user",
Content: "hello",
},
},
}
claudeRequest, err := RequestOpenAI2ClaudeMessage(nil, request)
require.NoError(t, err)
require.Equal(t, "claude-opus-4-8", claudeRequest.Model)
require.NotNil(t, claudeRequest.Thinking)
require.Equal(t, "adaptive", claudeRequest.Thinking.Type)
require.Equal(t, "summarized", claudeRequest.Thinking.Display)
require.JSONEq(t, `{"effort":"high"}`, string(claudeRequest.OutputConfig))
require.Nil(t, claudeRequest.Temperature)
require.Nil(t, claudeRequest.TopP)
require.Nil(t, claudeRequest.TopK)
}
func TestRequestOpenAI2ClaudeMessage_SupportsPDFFileContent(t *testing.T) {
request := dto.GeneralOpenAIRequest{
Model: "claude-3-5-sonnet",
+1 -1
View File
@@ -30,7 +30,7 @@ func convertCf2CompletionsRequest(textRequest dto.GeneralOpenAIRequest) *CfReque
}
func cfStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*types.NewAPIError, *dto.Usage) {
scanner := bufio.NewScanner(resp.Body)
scanner := helper.NewStreamScanner(resp.Body)
scanner.Split(bufio.ScanLines)
helper.SetEventStreamHeaders(c)
+4 -2
View File
@@ -1,7 +1,6 @@
package cohere
import (
"bufio"
"encoding/json"
"io"
"net/http"
@@ -86,7 +85,7 @@ func cohereStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
createdTime := common.GetTimestamp()
usage := &dto.Usage{}
responseText := ""
scanner := bufio.NewScanner(resp.Body)
scanner := helper.NewStreamScanner(resp.Body)
scanner.Split(func(data []byte, atEOF bool) (advance int, token []byte, err error) {
if atEOF && len(data) == 0 {
return 0, nil, nil
@@ -106,6 +105,9 @@ func cohereStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
data := scanner.Text()
dataChan <- data
}
if err := scanner.Err(); err != nil {
common.SysLog("error reading stream: " + err.Error())
}
stopChan <- true
}()
helper.SetEventStreamHeaders(c)
+1 -1
View File
@@ -98,7 +98,7 @@ func cozeChatHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Res
}
func cozeChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
scanner := bufio.NewScanner(resp.Body)
scanner := helper.NewStreamScanner(resp.Body)
scanner.Split(bufio.ScanLines)
helper.SetEventStreamHeaders(c)
id := helper.GetResponseID(c)
+8 -3
View File
@@ -159,9 +159,14 @@ func requestOpenAI2Dify(c *gin.Context, info *relaycommon.RelayInfo, request dto
media := mediaContent.GetImageMedia()
var file *DifyFile
if media.IsRemoteImage() {
file.Type = media.MimeType
file.TransferMode = "remote_url"
file.URL = media.Url
// 修复 #2083: 远程图片分支此前未初始化 file,
// 导致 file.Type = ... 触发 nil pointer dereference
// 而 panic500: "invalid memory address or nil pointer dereference")。
file = &DifyFile{
Type: media.MimeType,
TransferMode: "remote_url",
URL: media.Url,
}
} else {
file = uploadDifyFile(c, info, difyReq.User, mediaContent)
}
+103 -18
View File
@@ -1079,17 +1079,47 @@ func responseGeminiChat2OpenAI(c *gin.Context, response *dto.GeminiChatResponse)
FinishReason: constant.FinishReasonStop,
}
if len(candidate.Content.Parts) > 0 {
var texts []string
// 使用 strings.Builder 直接累积最终 content,避免:
// 1) 每张 inline image 生成一次中间 "![image](...)" 字符串
// 2) 末尾 strings.Join 再分配一份等大缓冲
// Gemini 图片返回时 InlineData.Data 可能是数 MB 的 base64
// 上述两份临时分配在高并发下会显著放大堆驻留。
var content strings.Builder
var inlineGrow int
for _, part := range candidate.Content.Parts {
if part.InlineData != nil {
inlineGrow += len(part.InlineData.MimeType) + len(part.InlineData.Data) + 32
}
}
if inlineGrow > 0 {
content.Grow(inlineGrow)
}
appended := 0
writeSep := func() {
if appended > 0 {
content.WriteByte('\n')
}
appended++
}
var toolCalls []dto.ToolCallResponse
for _, part := range candidate.Content.Parts {
if part.InlineData != nil {
// 媒体内容
if strings.HasPrefix(part.InlineData.MimeType, "image") {
imgText := "![image](data:" + part.InlineData.MimeType + ";base64," + part.InlineData.Data + ")"
texts = append(texts, imgText)
writeSep()
content.WriteString("![image](data:")
content.WriteString(part.InlineData.MimeType)
content.WriteString(";base64,")
content.WriteString(part.InlineData.Data)
content.WriteByte(')')
} else {
// 其他媒体类型,直接显示链接
texts = append(texts, fmt.Sprintf("[media](data:%s;base64,%s)", part.InlineData.MimeType, part.InlineData.Data))
writeSep()
content.WriteString("[media](data:")
content.WriteString(part.InlineData.MimeType)
content.WriteString(";base64,")
content.WriteString(part.InlineData.Data)
content.WriteByte(')')
}
} else if part.FunctionCall != nil {
choice.FinishReason = constant.FinishReasonToolCalls
@@ -1100,13 +1130,22 @@ func responseGeminiChat2OpenAI(c *gin.Context, response *dto.GeminiChatResponse)
choice.Message.ReasoningContent = &part.Text
} else {
if part.ExecutableCode != nil {
texts = append(texts, "```"+part.ExecutableCode.Language+"\n"+part.ExecutableCode.Code+"\n```")
writeSep()
content.WriteString("```")
content.WriteString(part.ExecutableCode.Language)
content.WriteByte('\n')
content.WriteString(part.ExecutableCode.Code)
content.WriteString("\n```")
} else if part.CodeExecutionResult != nil {
texts = append(texts, "```output\n"+part.CodeExecutionResult.Output+"\n```")
writeSep()
content.WriteString("```output\n")
content.WriteString(part.CodeExecutionResult.Output)
content.WriteString("\n```")
} else {
// 过滤掉空行
if part.Text != "\n" {
texts = append(texts, part.Text)
writeSep()
content.WriteString(part.Text)
}
}
}
@@ -1115,7 +1154,7 @@ func responseGeminiChat2OpenAI(c *gin.Context, response *dto.GeminiChatResponse)
choice.Message.SetToolCalls(toolCalls)
isToolCall = true
}
choice.Message.SetStringContent(strings.Join(texts, "\n"))
choice.Message.SetStringContent(content.String())
}
if candidate.FinishReason != nil {
@@ -1169,7 +1208,25 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *dto.GeminiChatResponse) (*d
//Role: "assistant",
},
}
var texts []string
// 使用 strings.Builder 直接累积 delta content,避免每张 image / 每个
// 文本片段都先 `+` 拼出一份临时 string,再 strings.Join 再拷贝一遍。
var content strings.Builder
var inlineGrow int
for _, part := range candidate.Content.Parts {
if part.InlineData != nil {
inlineGrow += len(part.InlineData.MimeType) + len(part.InlineData.Data) + 32
}
}
if inlineGrow > 0 {
content.Grow(inlineGrow)
}
appended := 0
writeSep := func() {
if appended > 0 {
content.WriteByte('\n')
}
appended++
}
isTools := false
isThought := false
if candidate.FinishReason != nil {
@@ -1207,8 +1264,12 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *dto.GeminiChatResponse) (*d
for _, part := range candidate.Content.Parts {
if part.InlineData != nil {
if strings.HasPrefix(part.InlineData.MimeType, "image") {
imgText := "![image](data:" + part.InlineData.MimeType + ";base64," + part.InlineData.Data + ")"
texts = append(texts, imgText)
writeSep()
content.WriteString("![image](data:")
content.WriteString(part.InlineData.MimeType)
content.WriteString(";base64,")
content.WriteString(part.InlineData.Data)
content.WriteByte(')')
}
} else if part.FunctionCall != nil {
isTools = true
@@ -1219,23 +1280,33 @@ func streamResponseGeminiChat2OpenAI(geminiResponse *dto.GeminiChatResponse) (*d
} else if part.Thought {
isThought = true
texts = append(texts, part.Text)
writeSep()
content.WriteString(part.Text)
} else {
if part.ExecutableCode != nil {
texts = append(texts, "```"+part.ExecutableCode.Language+"\n"+part.ExecutableCode.Code+"\n```\n")
writeSep()
content.WriteString("```")
content.WriteString(part.ExecutableCode.Language)
content.WriteByte('\n')
content.WriteString(part.ExecutableCode.Code)
content.WriteString("\n```\n")
} else if part.CodeExecutionResult != nil {
texts = append(texts, "```output\n"+part.CodeExecutionResult.Output+"\n```\n")
writeSep()
content.WriteString("```output\n")
content.WriteString(part.CodeExecutionResult.Output)
content.WriteString("\n```\n")
} else {
if part.Text != "\n" {
texts = append(texts, part.Text)
writeSep()
content.WriteString(part.Text)
}
}
}
}
if isThought {
choice.Delta.SetReasoningContent(strings.Join(texts, "\n"))
choice.Delta.SetReasoningContent(content.String())
} else {
choice.Delta.SetContentString(strings.Join(texts, "\n"))
choice.Delta.SetContentString(content.String())
}
if isTools {
choice.FinishReason = &constant.FinishReasonToolCalls
@@ -1339,6 +1410,14 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
response.Id = id
response.Created = createAt
response.Model = info.UpstreamModelName
if response.IsToolCall() {
finishReason = constant.FinishReasonToolCalls
if info.RelayFormat == types.RelayFormatClaude {
for choiceIdx := range response.Choices {
response.Choices[choiceIdx].FinishReason = nil
}
}
}
for choiceIdx := range response.Choices {
choiceKey := response.Choices[choiceIdx].Index
for toolIdx := range response.Choices[choiceIdx].Delta.ToolCalls {
@@ -1399,7 +1478,9 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
logger.LogError(c, err.Error())
}
if isStop {
_ = handleStream(c, info, helper.GenerateStopResponse(id, createAt, info.UpstreamModelName, finishReason))
if info.RelayFormat != types.RelayFormatClaude {
_ = handleStream(c, info, helper.GenerateStopResponse(id, createAt, info.UpstreamModelName, finishReason))
}
}
return true
})
@@ -1409,6 +1490,10 @@ func GeminiChatStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *
}
response := helper.GenerateFinalUsageResponse(id, createAt, info.UpstreamModelName, *usage)
if info.RelayFormat == types.RelayFormatClaude && info.ClaudeConvertInfo != nil && !info.ClaudeConvertInfo.Done {
response = helper.GenerateStopResponse(id, createAt, info.UpstreamModelName, finishReason)
response.Usage = usage
}
handleErr := handleFinalStream(c, info, response)
if handleErr != nil {
common.SysLog("send final response failed: " + handleErr.Error())
+16
View File
@@ -5,7 +5,9 @@ import (
"fmt"
"io"
"net/http"
"strings"
"github.com/QuantumNous/new-api/common"
channelconstant "github.com/QuantumNous/new-api/constant"
"github.com/QuantumNous/new-api/dto"
"github.com/QuantumNous/new-api/relay/channel"
@@ -79,9 +81,23 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, req *http.Header, info *rel
}
func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayInfo, request *dto.GeneralOpenAIRequest) (any, error) {
if request.Temperature != nil && isTemperatureOneOnlyModel(getUpstreamModelName(info, request.Model)) && *request.Temperature != 1.0 {
request.Temperature = common.GetPointer[float64](1.0)
}
return request, nil
}
func getUpstreamModelName(info *relaycommon.RelayInfo, fallback string) string {
if info != nil && info.ChannelMeta != nil && info.UpstreamModelName != "" {
return info.UpstreamModelName
}
return fallback
}
func isTemperatureOneOnlyModel(model string) bool {
return strings.EqualFold(model, "kimi-k2.6")
}
func (a *Adaptor) ConvertOpenAIResponsesRequest(c *gin.Context, info *relaycommon.RelayInfo, request dto.OpenAIResponsesRequest) (any, error) {
// TODO implement me
return nil, errors.New("not implemented")
+68
View File
@@ -0,0 +1,68 @@
package moonshot
import (
"testing"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/dto"
relaycommon "github.com/QuantumNous/new-api/relay/common"
"github.com/stretchr/testify/require"
)
func TestConvertOpenAIRequestKimiK26UsesOnlyAllowedTemperature(t *testing.T) {
request := &dto.GeneralOpenAIRequest{
Model: "kimi-k2.6",
Temperature: common.GetPointer[float64](0.7),
}
info := &relaycommon.RelayInfo{
ChannelMeta: &relaycommon.ChannelMeta{
UpstreamModelName: "kimi-k2.6",
},
}
converted, err := (&Adaptor{}).ConvertOpenAIRequest(nil, info, request)
require.NoError(t, err)
convertedRequest, ok := converted.(*dto.GeneralOpenAIRequest)
require.True(t, ok)
require.NotNil(t, convertedRequest.Temperature)
require.Equal(t, 1.0, *convertedRequest.Temperature)
}
func TestConvertOpenAIRequestKimiK26KeepsOmittedTemperatureOmitted(t *testing.T) {
request := &dto.GeneralOpenAIRequest{
Model: "kimi-k2.6",
}
info := &relaycommon.RelayInfo{
ChannelMeta: &relaycommon.ChannelMeta{
UpstreamModelName: "kimi-k2.6",
},
}
converted, err := (&Adaptor{}).ConvertOpenAIRequest(nil, info, request)
require.NoError(t, err)
convertedRequest, ok := converted.(*dto.GeneralOpenAIRequest)
require.True(t, ok)
require.Nil(t, convertedRequest.Temperature)
}
func TestConvertOpenAIRequestOtherMoonshotModelKeepsTemperature(t *testing.T) {
request := &dto.GeneralOpenAIRequest{
Model: "kimi-k2.5",
Temperature: common.GetPointer[float64](0.7),
}
info := &relaycommon.RelayInfo{
ChannelMeta: &relaycommon.ChannelMeta{
UpstreamModelName: "kimi-k2.5",
},
}
converted, err := (&Adaptor{}).ConvertOpenAIRequest(nil, info, request)
require.NoError(t, err)
convertedRequest, ok := converted.(*dto.GeneralOpenAIRequest)
require.True(t, ok)
require.NotNil(t, convertedRequest.Temperature)
require.Equal(t, 0.7, *convertedRequest.Temperature)
}
+2 -2
View File
@@ -1,7 +1,6 @@
package ollama
import (
"bufio"
"encoding/json"
"fmt"
"io"
@@ -12,6 +11,7 @@ import (
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/dto"
relaycommon "github.com/QuantumNous/new-api/relay/common"
"github.com/QuantumNous/new-api/relay/helper"
"github.com/QuantumNous/new-api/service"
"github.com/QuantumNous/new-api/types"
@@ -397,7 +397,7 @@ func PullOllamaModelStream(baseURL, apiKey, modelName string, progressCallback f
}
// 读取流式响应
scanner := bufio.NewScanner(response.Body)
scanner := helper.NewStreamScanner(response.Body)
successful := false
for scanner.Scan() {
line := scanner.Text()
+1 -2
View File
@@ -1,7 +1,6 @@
package ollama
import (
"bufio"
"encoding/json"
"fmt"
"io"
@@ -70,7 +69,7 @@ func ollamaStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http
defer service.CloseResponseBodyGracefully(resp)
helper.SetEventStreamHeaders(c)
scanner := bufio.NewScanner(resp.Body)
scanner := helper.NewStreamScanner(resp.Body)
usage := &dto.Usage{}
var model = info.UpstreamModelName
var responseId = common.GetUUID()
+32 -8
View File
@@ -9,6 +9,7 @@ import (
"mime/multipart"
"net/http"
"net/textproto"
"net/url"
"path/filepath"
"strings"
@@ -163,6 +164,20 @@ func (a *Adaptor) GetRequestURL(info *relaycommon.RelayInfo) (string, error) {
url = strings.Replace(url, "{model}", info.UpstreamModelName, -1)
return url, nil
default:
// Handle coding plan special base URLs
if specialPlan, ok := constant.ChannelSpecialBases[info.ChannelBaseUrl]; ok && specialPlan.OpenAIBaseURL != "" {
if info.RelayFormat == types.RelayFormatClaude {
return fmt.Sprintf("%s/v1/messages", specialPlan.ClaudeBaseURL), nil
}
switch info.RelayMode {
case relayconstant.RelayModeEmbeddings:
return fmt.Sprintf("%s/embeddings", specialPlan.OpenAIBaseURL), nil
case relayconstant.RelayModeImagesGenerations:
return fmt.Sprintf("%s/images/generations", specialPlan.OpenAIBaseURL), nil
default:
return fmt.Sprintf("%s/chat/completions", specialPlan.OpenAIBaseURL), nil
}
}
if (info.RelayFormat == types.RelayFormatClaude || info.RelayFormat == types.RelayFormatGemini) &&
info.RelayMode != relayconstant.RelayModeResponses &&
info.RelayMode != relayconstant.RelayModeResponsesCompact {
@@ -220,7 +235,7 @@ func (a *Adaptor) SetupRequestHeader(c *gin.Context, header *http.Header, info *
header.Set("HTTP-Referer", "https://www.newapi.ai")
}
if header.Get("X-OpenRouter-Title") == "" {
header.Set("X-OpenRouter-Title", "New API")
header.Set("X-OpenRouter-Title", "ModelsToken")
}
}
return nil
@@ -310,18 +325,20 @@ func (a *Adaptor) ConvertOpenAIRequest(c *gin.Context, info *relaycommon.RelayIn
}
}
if strings.HasPrefix(info.UpstreamModelName, "o") || strings.HasPrefix(info.UpstreamModelName, "gpt-5") {
isOModel := dto.IsOpenAIReasoningOModel(info.UpstreamModelName)
isGPT5Model := dto.IsOpenAIGPT5Model(info.UpstreamModelName)
if isOModel || isGPT5Model {
if lo.FromPtrOr(request.MaxCompletionTokens, uint(0)) == 0 && lo.FromPtrOr(request.MaxTokens, uint(0)) != 0 {
request.MaxCompletionTokens = request.MaxTokens
request.MaxTokens = nil
}
if strings.HasPrefix(info.UpstreamModelName, "o") {
if isOModel {
request.Temperature = nil
}
// gpt-5系列模型适配 归零不再支持的参数
if strings.HasPrefix(info.UpstreamModelName, "gpt-5") {
if isGPT5Model {
request.Temperature = nil
request.TopP = nil
request.LogProbs = nil
@@ -437,10 +454,13 @@ func (a *Adaptor) ConvertImageRequest(c *gin.Context, info *relaycommon.RelayInf
// 使用已解析的 multipart 表单,避免重复解析
mf := c.Request.MultipartForm
if mf == nil {
if _, err := c.MultipartForm(); err != nil {
return nil, errors.New("failed to parse multipart form")
form, err := common.ParseMultipartFormReusable(c)
if err != nil {
return nil, fmt.Errorf("failed to parse multipart form: %w", err)
}
mf = c.Request.MultipartForm
c.Request.MultipartForm = form
c.Request.PostForm = url.Values(form.Value)
mf = form
}
// 写入所有非文件字段
@@ -623,7 +643,11 @@ func (a *Adaptor) DoResponse(c *gin.Context, resp *http.Response, info *relaycom
case relayconstant.RelayModeAudioTranscription:
err, usage = OpenaiSTTHandler(c, resp, info, a.ResponseFormat)
case relayconstant.RelayModeImagesGenerations, relayconstant.RelayModeImagesEdits:
usage, err = OpenaiHandlerWithUsage(c, info, resp)
if info.IsStream {
usage, err = OpenaiImageStreamHandler(c, info, resp)
} else {
usage, err = OpenaiImageHandler(c, info, resp)
}
case relayconstant.RelayModeRerank:
usage, err = common_handler.RerankHandler(c, info, resp)
case relayconstant.RelayModeResponses:
+14 -65
View File
@@ -1,7 +1,6 @@
package openai
import (
"encoding/json"
"strings"
"github.com/QuantumNous/new-api/common"
@@ -92,78 +91,28 @@ func ProcessStreamResponse(streamResponse dto.ChatCompletionsStreamResponse, res
return nil
}
func processTokens(relayMode int, streamItems []string, responseTextBuilder *strings.Builder, toolCount *int) error {
streamResp := "[" + strings.Join(streamItems, ",") + "]"
func processTokenData(relayMode int, data string, responseTextBuilder *strings.Builder, toolCount *int) error {
switch relayMode {
case relayconstant.RelayModeChatCompletions:
return processChatCompletions(streamResp, streamItems, responseTextBuilder, toolCount)
var streamResponse dto.ChatCompletionsStreamResponse
if err := common.UnmarshalJsonStr(data, &streamResponse); err != nil {
return err
}
return ProcessStreamResponse(streamResponse, responseTextBuilder, toolCount)
case relayconstant.RelayModeCompletions:
return processCompletions(streamResp, streamItems, responseTextBuilder)
var streamResponse dto.CompletionsStreamResponse
if err := common.UnmarshalJsonStr(data, &streamResponse); err != nil {
return err
}
processCompletionsStreamResponse(streamResponse, responseTextBuilder)
}
return nil
}
func processChatCompletions(streamResp string, streamItems []string, responseTextBuilder *strings.Builder, toolCount *int) error {
var streamResponses []dto.ChatCompletionsStreamResponse
if err := json.Unmarshal(common.StringToByteSlice(streamResp), &streamResponses); err != nil {
// 一次性解析失败,逐个解析
common.SysLog("error unmarshalling stream response: " + err.Error())
for _, item := range streamItems {
var streamResponse dto.ChatCompletionsStreamResponse
if err := json.Unmarshal(common.StringToByteSlice(item), &streamResponse); err != nil {
return err
}
if err := ProcessStreamResponse(streamResponse, responseTextBuilder, toolCount); err != nil {
common.SysLog("error processing stream response: " + err.Error())
}
}
return nil
func processCompletionsStreamResponse(streamResponse dto.CompletionsStreamResponse, responseTextBuilder *strings.Builder) {
for _, choice := range streamResponse.Choices {
responseTextBuilder.WriteString(choice.Text)
}
// 批量处理所有响应
for _, streamResponse := range streamResponses {
for _, choice := range streamResponse.Choices {
responseTextBuilder.WriteString(choice.Delta.GetContentString())
responseTextBuilder.WriteString(choice.Delta.GetReasoningContent())
if choice.Delta.ToolCalls != nil {
if len(choice.Delta.ToolCalls) > *toolCount {
*toolCount = len(choice.Delta.ToolCalls)
}
for _, tool := range choice.Delta.ToolCalls {
responseTextBuilder.WriteString(tool.Function.Name)
responseTextBuilder.WriteString(tool.Function.Arguments)
}
}
}
}
return nil
}
func processCompletions(streamResp string, streamItems []string, responseTextBuilder *strings.Builder) error {
var streamResponses []dto.CompletionsStreamResponse
if err := json.Unmarshal(common.StringToByteSlice(streamResp), &streamResponses); err != nil {
// 一次性解析失败,逐个解析
common.SysLog("error unmarshalling stream response: " + err.Error())
for _, item := range streamItems {
var streamResponse dto.CompletionsStreamResponse
if err := json.Unmarshal(common.StringToByteSlice(item), &streamResponse); err != nil {
continue
}
for _, choice := range streamResponse.Choices {
responseTextBuilder.WriteString(choice.Text)
}
}
return nil
}
// 批量处理所有响应
for _, streamResponse := range streamResponses {
for _, choice := range streamResponse.Choices {
responseTextBuilder.WriteString(choice.Text)
}
}
return nil
}
func handleLastResponse(lastStreamData string, responseId *string, createAt *int64,
+98
View File
@@ -0,0 +1,98 @@
package openai
import (
"bytes"
"io"
"mime/multipart"
"net/http"
"net/http/httptest"
"testing"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/dto"
relaycommon "github.com/QuantumNous/new-api/relay/common"
relayconstant "github.com/QuantumNous/new-api/relay/constant"
"github.com/gin-gonic/gin"
"github.com/stretchr/testify/require"
)
// TestConvertImageEditRequestMultipart verifies that ConvertImageRequest
// re-serializes multipart image edit requests with all fields (including
// stream) and the file intact, both when the form was already parsed and when
// it must be re-parsed from the reusable body.
func TestConvertImageEditRequestMultipart(t *testing.T) {
gin.SetMode(gin.TestMode)
newMultipartContext := func(t *testing.T, prompt string) *gin.Context {
var body bytes.Buffer
writer := multipart.NewWriter(&body)
require.NoError(t, writer.WriteField("model", "gpt-image-1"))
require.NoError(t, writer.WriteField("prompt", prompt))
require.NoError(t, writer.WriteField("stream", "true"))
require.NoError(t, writer.WriteField("partial_images", "3"))
part, err := writer.CreateFormFile("image", "input.png")
require.NoError(t, err)
_, err = part.Write([]byte("fake image"))
require.NoError(t, err)
require.NoError(t, writer.Close())
c, _ := gin.CreateTestContext(httptest.NewRecorder())
c.Request = httptest.NewRequest(http.MethodPost, "/v1/images/edits", &body)
c.Request.Header.Set("Content-Type", writer.FormDataContentType())
return c
}
convertAndReplay := func(t *testing.T, c *gin.Context, prompt string) {
info := &relaycommon.RelayInfo{
RelayMode: relayconstant.RelayModeImagesEdits,
}
request := dto.ImageRequest{
Model: "gpt-image-1",
Prompt: prompt,
Stream: common.GetPointer(true),
}
converted, err := (&Adaptor{}).ConvertImageRequest(c, info, request)
require.NoError(t, err)
convertedBody, ok := converted.(*bytes.Buffer)
require.True(t, ok)
replayedRequest := httptest.NewRequest(http.MethodPost, "/v1/images/edits", bytes.NewReader(convertedBody.Bytes()))
replayedRequest.Header.Set("Content-Type", c.Request.Header.Get("Content-Type"))
require.NoError(t, replayedRequest.ParseMultipartForm(32<<20))
require.Equal(t, "gpt-image-1", replayedRequest.PostForm.Get("model"))
require.Equal(t, prompt, replayedRequest.PostForm.Get("prompt"))
require.Equal(t, "true", replayedRequest.PostForm.Get("stream"))
require.Equal(t, "3", replayedRequest.PostForm.Get("partial_images"))
require.Len(t, replayedRequest.MultipartForm.File["image"], 1)
file, err := replayedRequest.MultipartForm.File["image"][0].Open()
require.NoError(t, err)
defer file.Close()
fileBytes, err := io.ReadAll(file)
require.NoError(t, err)
require.Equal(t, []byte("fake image"), fileBytes)
}
t.Run("with pre-parsed form", func(t *testing.T) {
prompt := "edit this image"
c := newMultipartContext(t, prompt)
require.NoError(t, c.Request.ParseMultipartForm(32<<20))
convertAndReplay(t, c, prompt)
})
t.Run("re-parses reusable body when form is missing", func(t *testing.T) {
prompt := "edit without pre-parsed form"
c := newMultipartContext(t, prompt)
storage, err := common.GetBodyStorage(c)
require.NoError(t, err)
c.Request.Body = io.NopCloser(storage)
c.Request.MultipartForm = nil
c.Request.PostForm = nil
convertAndReplay(t, c, prompt)
})
}
+173
View File
@@ -0,0 +1,173 @@
package openai
import (
"io"
"net/http"
"net/http/httptest"
"strings"
"testing"
"github.com/QuantumNous/new-api/constant"
relaycommon "github.com/QuantumNous/new-api/relay/common"
"github.com/gin-gonic/gin"
"github.com/stretchr/testify/require"
)
func newImageTestContext(t *testing.T, body, contentType string, isStream bool) (*gin.Context, *httptest.ResponseRecorder, *http.Response, *relaycommon.RelayInfo) {
t.Helper()
recorder := httptest.NewRecorder()
c, _ := gin.CreateTestContext(recorder)
c.Request = httptest.NewRequest(http.MethodPost, "/v1/images/generations", nil)
resp := &http.Response{
StatusCode: http.StatusOK,
Body: io.NopCloser(strings.NewReader(body)),
Header: http.Header{"Content-Type": []string{contentType}},
}
info := &relaycommon.RelayInfo{
ChannelMeta: &relaycommon.ChannelMeta{},
IsStream: isStream,
}
return c, recorder, resp, info
}
// TestOpenaiImageStreamHandlerForwardsSSEAndUsage covers the core SSE path:
// chunks are forwarded with rebuilt event lines, usage is extracted and
// normalized (input_tokens -> prompt_tokens with details), and [DONE] is
// re-emitted to the client.
func TestOpenaiImageStreamHandlerForwardsSSEAndUsage(t *testing.T) {
oldMode := gin.Mode()
gin.SetMode(gin.TestMode)
t.Cleanup(func() { gin.SetMode(oldMode) })
oldTimeout := constant.StreamingTimeout
constant.StreamingTimeout = 30
t.Cleanup(func() { constant.StreamingTimeout = oldTimeout })
body := strings.Join([]string{
`event: image_generation.partial_image`,
`data: {"type":"image_generation.partial_image","b64_json":"partial"}`,
``,
`data: {"usage":{"input_tokens":3,"output_tokens":4,"total_tokens":7,"input_tokens_details":{"image_tokens":2,"text_tokens":1}}}`,
``,
`data: [DONE]`,
``,
}, "\n")
c, recorder, resp, info := newImageTestContext(t, body, "text/event-stream", true)
usage, err := OpenaiImageStreamHandler(c, info, resp)
require.Nil(t, err)
require.Equal(t, 3, usage.PromptTokens)
require.Equal(t, 4, usage.CompletionTokens)
require.Equal(t, 7, usage.TotalTokens)
require.Equal(t, 2, usage.PromptTokensDetails.ImageTokens)
require.Equal(t, 1, usage.PromptTokensDetails.TextTokens)
require.Contains(t, recorder.Body.String(), `event: image_generation.partial_image`)
require.Contains(t, recorder.Body.String(), `data: {"type":"image_generation.partial_image","b64_json":"partial"}`)
require.Contains(t, recorder.Body.String(), `data: {"usage":{"input_tokens":3,"output_tokens":4,"total_tokens":7,"input_tokens_details":{"image_tokens":2,"text_tokens":1}}}`)
require.Contains(t, recorder.Body.String(), `data: [DONE]`)
require.Equal(t, "text/event-stream", recorder.Header().Get("Content-Type"))
}
// TestOpenaiImageStreamHandlerWrapsJSONResponse covers the non-SSE fallback:
// a JSON upstream response is wrapped into pseudo-SSE completed events.
func TestOpenaiImageStreamHandlerWrapsJSONResponse(t *testing.T) {
oldMode := gin.Mode()
gin.SetMode(gin.TestMode)
t.Cleanup(func() { gin.SetMode(oldMode) })
body := `{"created":1710000000,"data":[{"b64_json":"final","revised_prompt":"draw a cat"}],"usage":{"input_tokens":3,"output_tokens":4,"total_tokens":7,"input_tokens_details":{"image_tokens":2,"text_tokens":1}}}`
c, recorder, resp, info := newImageTestContext(t, body, "application/json", true)
usage, err := OpenaiImageStreamHandler(c, info, resp)
require.Nil(t, err)
require.Equal(t, 3, usage.PromptTokens)
require.Equal(t, 4, usage.CompletionTokens)
require.Equal(t, 7, usage.TotalTokens)
require.Equal(t, 2, usage.PromptTokensDetails.ImageTokens)
require.Equal(t, 1, usage.PromptTokensDetails.TextTokens)
require.Equal(t, "text/event-stream", recorder.Header().Get("Content-Type"))
require.Empty(t, recorder.Header().Get("Content-Length"))
require.Contains(t, recorder.Body.String(), `event: image_generation.completed`)
require.Contains(t, recorder.Body.String(), `"type":"image_generation.completed"`)
require.Contains(t, recorder.Body.String(), `"b64_json":"final"`)
require.Contains(t, recorder.Body.String(), `"revised_prompt":"draw a cat"`)
require.Contains(t, recorder.Body.String(), `data: [DONE]`)
}
// TestOpenaiImageHandlersReturnJSONError covers JSON error responses for both
// entry points: the non-streaming handler and the stream handler's non-SSE
// fallback. Neither must leak the error body to the client.
func TestOpenaiImageHandlersReturnJSONError(t *testing.T) {
oldMode := gin.Mode()
gin.SetMode(gin.TestMode)
t.Cleanup(func() { gin.SetMode(oldMode) })
body := `{"error":{"message":"content moderation failed","type":"upstream_error","code":"content_moderation_failed","status":502}}`
t.Run("non-streaming handler", func(t *testing.T) {
c, recorder, resp, info := newImageTestContext(t, body, "application/json", false)
usage, err := OpenaiImageHandler(c, info, resp)
require.Nil(t, usage)
require.NotNil(t, err)
require.Equal(t, http.StatusOK, err.StatusCode)
oaiError := err.ToOpenAIError()
require.Equal(t, "content moderation failed", oaiError.Message)
require.Equal(t, "upstream_error", oaiError.Type)
require.Equal(t, "content_moderation_failed", oaiError.Code)
require.Empty(t, recorder.Body.String())
})
t.Run("stream handler JSON fallback", func(t *testing.T) {
c, recorder, resp, info := newImageTestContext(t, body, "application/json", true)
usage, err := OpenaiImageStreamHandler(c, info, resp)
require.Nil(t, usage)
require.NotNil(t, err)
require.Equal(t, http.StatusOK, err.StatusCode)
require.Equal(t, "content moderation failed", err.ToOpenAIError().Message)
require.Empty(t, recorder.Body.String())
})
}
// TestOpenaiImageStreamHandlerRecordsUpstreamErrorEvent verifies that an error
// event inside the SSE stream is recorded as a soft error while the payload is
// still forwarded to the client.
func TestOpenaiImageStreamHandlerRecordsUpstreamErrorEvent(t *testing.T) {
oldMode := gin.Mode()
gin.SetMode(gin.TestMode)
t.Cleanup(func() { gin.SetMode(oldMode) })
oldTimeout := constant.StreamingTimeout
constant.StreamingTimeout = 30
t.Cleanup(func() { constant.StreamingTimeout = oldTimeout })
body := strings.Join([]string{
`event: image_generation.partial_image`,
`data: {"type":"image_generation.partial_image","b64_json":"partial"}`,
``,
`event: error`,
`data: {"type":"upstream_error","error":{"message":"stream error: stream ID 77; INTERNAL_ERROR; received from peer"}}`,
``,
}, "\n")
c, recorder, resp, info := newImageTestContext(t, body, "text/event-stream", true)
usage, err := OpenaiImageStreamHandler(c, info, resp)
require.Nil(t, err)
require.NotNil(t, usage)
require.NotNil(t, info.StreamStatus)
require.Equal(t, relaycommon.StreamEndReasonEOF, info.StreamStatus.EndReason)
require.True(t, info.StreamStatus.HasErrors())
require.Equal(t, 1, info.StreamStatus.TotalErrorCount())
require.Contains(t, info.StreamStatus.Errors[0].Message, "INTERNAL_ERROR")
// The scanner strips the upstream "event: error" line; the event name is
// rebuilt from the JSON "type" field (upstream_error). The error message
// is still forwarded in the data: payload (stream ID 77).
require.Contains(t, recorder.Body.String(), `event: upstream_error`)
require.Contains(t, recorder.Body.String(), `stream ID 77`)
}
+4 -428
View File
@@ -14,12 +14,9 @@ import (
relaycommon "github.com/QuantumNous/new-api/relay/common"
"github.com/QuantumNous/new-api/relay/helper"
"github.com/QuantumNous/new-api/service"
"github.com/QuantumNous/new-api/types"
"github.com/bytedance/gopkg/util/gopool"
"github.com/gin-gonic/gin"
"github.com/gorilla/websocket"
)
func sendStreamData(c *gin.Context, info *relaycommon.RelayInfo, data string, forceFormat bool, thinkToContent bool) error {
@@ -119,7 +116,6 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
var responseTextBuilder strings.Builder
var toolCount int
var usage = &dto.Usage{}
var streamItems []string // store stream items
var lastStreamData string
var secondLastStreamData string // 存储倒数第二个stream data,用于音频模型
@@ -140,7 +136,10 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
}
lastStreamData = data
streamItems = append(streamItems, data)
if err := processTokenData(info.RelayMode, data, &responseTextBuilder, &toolCount); err != nil {
logger.LogError(c, "error processing stream token data: "+err.Error())
sr.Error(err)
}
}
})
@@ -175,11 +174,6 @@ func OaiStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Re
}
}
// 处理token计算
if err := processTokens(info.RelayMode, streamItems, &responseTextBuilder, &toolCount); err != nil {
logger.LogError(c, "error processing tokens: "+err.Error())
}
if !containStreamUsage {
usage = service.ResponseText2Usage(c, responseTextBuilder.String(), info.UpstreamModelName, info.GetEstimatePromptTokens())
usage.CompletionTokens += toolCount * 7
@@ -296,421 +290,3 @@ func OpenaiHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Respo
return &simpleResponse.Usage, nil
}
func streamTTSResponse(c *gin.Context, resp *http.Response) {
c.Writer.WriteHeaderNow()
flusher, ok := c.Writer.(http.Flusher)
if !ok {
logger.LogWarn(c, "streaming not supported")
_, err := io.Copy(c.Writer, resp.Body)
if err != nil {
logger.LogWarn(c, err.Error())
}
return
}
buffer := make([]byte, 4096)
for {
n, err := resp.Body.Read(buffer)
//logger.LogInfo(c, fmt.Sprintf("streamTTSResponse read %d bytes", n))
if n > 0 {
if _, writeErr := c.Writer.Write(buffer[:n]); writeErr != nil {
logger.LogError(c, writeErr.Error())
break
}
flusher.Flush()
}
if err != nil {
if err != io.EOF {
logger.LogError(c, err.Error())
}
break
}
}
}
func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.NewAPIError, *dto.RealtimeUsage) {
if info == nil || info.ClientWs == nil || info.TargetWs == nil {
return types.NewError(fmt.Errorf("invalid websocket connection"), types.ErrorCodeBadResponse), nil
}
info.IsStream = true
clientConn := info.ClientWs
targetConn := info.TargetWs
clientClosed := make(chan struct{})
targetClosed := make(chan struct{})
sendChan := make(chan []byte, 100)
receiveChan := make(chan []byte, 100)
errChan := make(chan error, 2)
usage := &dto.RealtimeUsage{}
localUsage := &dto.RealtimeUsage{}
sumUsage := &dto.RealtimeUsage{}
gopool.Go(func() {
defer func() {
if r := recover(); r != nil {
errChan <- fmt.Errorf("panic in client reader: %v", r)
}
}()
for {
select {
case <-c.Done():
return
default:
_, message, err := clientConn.ReadMessage()
if err != nil {
if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
errChan <- fmt.Errorf("error reading from client: %v", err)
}
close(clientClosed)
return
}
realtimeEvent := &dto.RealtimeEvent{}
err = common.Unmarshal(message, realtimeEvent)
if err != nil {
errChan <- fmt.Errorf("error unmarshalling message: %v", err)
return
}
if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdate {
if realtimeEvent.Session != nil {
if realtimeEvent.Session.Tools != nil {
info.RealtimeTools = realtimeEvent.Session.Tools
}
}
}
textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
if err != nil {
errChan <- fmt.Errorf("error counting text token: %v", err)
return
}
logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
localUsage.TotalTokens += textToken + audioToken
localUsage.InputTokens += textToken + audioToken
localUsage.InputTokenDetails.TextTokens += textToken
localUsage.InputTokenDetails.AudioTokens += audioToken
err = helper.WssString(c, targetConn, string(message))
if err != nil {
errChan <- fmt.Errorf("error writing to target: %v", err)
return
}
select {
case sendChan <- message:
default:
}
}
}
})
gopool.Go(func() {
defer func() {
if r := recover(); r != nil {
errChan <- fmt.Errorf("panic in target reader: %v", r)
}
}()
for {
select {
case <-c.Done():
return
default:
_, message, err := targetConn.ReadMessage()
if err != nil {
if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
errChan <- fmt.Errorf("error reading from target: %v", err)
}
close(targetClosed)
return
}
info.SetFirstResponseTime()
realtimeEvent := &dto.RealtimeEvent{}
err = common.Unmarshal(message, realtimeEvent)
if err != nil {
errChan <- fmt.Errorf("error unmarshalling message: %v", err)
return
}
if realtimeEvent.Type == dto.RealtimeEventTypeResponseDone {
realtimeUsage := realtimeEvent.Response.Usage
if realtimeUsage != nil {
usage.TotalTokens += realtimeUsage.TotalTokens
usage.InputTokens += realtimeUsage.InputTokens
usage.OutputTokens += realtimeUsage.OutputTokens
usage.InputTokenDetails.AudioTokens += realtimeUsage.InputTokenDetails.AudioTokens
usage.InputTokenDetails.CachedTokens += realtimeUsage.InputTokenDetails.CachedTokens
usage.InputTokenDetails.TextTokens += realtimeUsage.InputTokenDetails.TextTokens
usage.OutputTokenDetails.AudioTokens += realtimeUsage.OutputTokenDetails.AudioTokens
usage.OutputTokenDetails.TextTokens += realtimeUsage.OutputTokenDetails.TextTokens
err := preConsumeUsage(c, info, usage, sumUsage)
if err != nil {
errChan <- fmt.Errorf("error consume usage: %v", err)
return
}
// 本次计费完成,清除
usage = &dto.RealtimeUsage{}
localUsage = &dto.RealtimeUsage{}
} else {
textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
if err != nil {
errChan <- fmt.Errorf("error counting text token: %v", err)
return
}
logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
localUsage.TotalTokens += textToken + audioToken
info.IsFirstRequest = false
localUsage.InputTokens += textToken + audioToken
localUsage.InputTokenDetails.TextTokens += textToken
localUsage.InputTokenDetails.AudioTokens += audioToken
err = preConsumeUsage(c, info, localUsage, sumUsage)
if err != nil {
errChan <- fmt.Errorf("error consume usage: %v", err)
return
}
// 本次计费完成,清除
localUsage = &dto.RealtimeUsage{}
// print now usage
}
logger.LogInfo(c, fmt.Sprintf("realtime streaming sumUsage: %v", sumUsage))
logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
} else if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdated || realtimeEvent.Type == dto.RealtimeEventTypeSessionCreated {
realtimeSession := realtimeEvent.Session
if realtimeSession != nil {
// update audio format
info.InputAudioFormat = common.GetStringIfEmpty(realtimeSession.InputAudioFormat, info.InputAudioFormat)
info.OutputAudioFormat = common.GetStringIfEmpty(realtimeSession.OutputAudioFormat, info.OutputAudioFormat)
}
} else {
textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
if err != nil {
errChan <- fmt.Errorf("error counting text token: %v", err)
return
}
logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
localUsage.TotalTokens += textToken + audioToken
localUsage.OutputTokens += textToken + audioToken
localUsage.OutputTokenDetails.TextTokens += textToken
localUsage.OutputTokenDetails.AudioTokens += audioToken
}
err = helper.WssString(c, clientConn, string(message))
if err != nil {
errChan <- fmt.Errorf("error writing to client: %v", err)
return
}
select {
case receiveChan <- message:
default:
}
}
}
})
select {
case <-clientClosed:
case <-targetClosed:
case err := <-errChan:
//return service.OpenAIErrorWrapper(err, "realtime_error", http.StatusInternalServerError), nil
logger.LogError(c, "realtime error: "+err.Error())
case <-c.Done():
}
if usage.TotalTokens != 0 {
_ = preConsumeUsage(c, info, usage, sumUsage)
}
if localUsage.TotalTokens != 0 {
_ = preConsumeUsage(c, info, localUsage, sumUsage)
}
// check usage total tokens, if 0, use local usage
return nil, sumUsage
}
func preConsumeUsage(ctx *gin.Context, info *relaycommon.RelayInfo, usage *dto.RealtimeUsage, totalUsage *dto.RealtimeUsage) error {
if usage == nil || totalUsage == nil {
return fmt.Errorf("invalid usage pointer")
}
totalUsage.TotalTokens += usage.TotalTokens
totalUsage.InputTokens += usage.InputTokens
totalUsage.OutputTokens += usage.OutputTokens
totalUsage.InputTokenDetails.CachedTokens += usage.InputTokenDetails.CachedTokens
totalUsage.InputTokenDetails.TextTokens += usage.InputTokenDetails.TextTokens
totalUsage.InputTokenDetails.AudioTokens += usage.InputTokenDetails.AudioTokens
totalUsage.OutputTokenDetails.TextTokens += usage.OutputTokenDetails.TextTokens
totalUsage.OutputTokenDetails.AudioTokens += usage.OutputTokenDetails.AudioTokens
// clear usage
err := service.PreWssConsumeQuota(ctx, info, usage)
return err
}
func OpenaiHandlerWithUsage(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
defer service.CloseResponseBodyGracefully(resp)
responseBody, err := io.ReadAll(resp.Body)
if err != nil {
return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
}
var usageResp dto.SimpleResponse
err = common.Unmarshal(responseBody, &usageResp)
if err != nil {
return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
}
// 写入新的 response body
service.IOCopyBytesGracefully(c, resp, responseBody)
// Once we've written to the client, we should not return errors anymore
// because the upstream has already consumed resources and returned content
// We should still perform billing even if parsing fails
// format
if usageResp.InputTokens > 0 {
usageResp.PromptTokens += usageResp.InputTokens
}
if usageResp.OutputTokens > 0 {
usageResp.CompletionTokens += usageResp.OutputTokens
}
if usageResp.InputTokensDetails != nil {
usageResp.PromptTokensDetails.ImageTokens += usageResp.InputTokensDetails.ImageTokens
usageResp.PromptTokensDetails.TextTokens += usageResp.InputTokensDetails.TextTokens
}
applyUsagePostProcessing(info, &usageResp.Usage, responseBody)
return &usageResp.Usage, nil
}
func applyUsagePostProcessing(info *relaycommon.RelayInfo, usage *dto.Usage, responseBody []byte) {
if info == nil || usage == nil {
return
}
switch info.ChannelType {
case constant.ChannelTypeDeepSeek:
if usage.PromptTokensDetails.CachedTokens == 0 && usage.PromptCacheHitTokens != 0 {
usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
}
case constant.ChannelTypeZhipu_v4:
// 智普的cached_tokens在标准位置: usage.prompt_tokens_details.cached_tokens
if usage.PromptTokensDetails.CachedTokens == 0 {
if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
} else if usage.PromptCacheHitTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
}
}
case constant.ChannelTypeMoonshot:
// Moonshot的cached_tokens在非标准位置: choices[].usage.cached_tokens
if usage.PromptTokensDetails.CachedTokens == 0 {
if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
} else if cachedTokens, ok := extractMoonshotCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
} else if usage.PromptCacheHitTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
}
}
case constant.ChannelTypeOpenAI:
if usage.PromptTokensDetails.CachedTokens == 0 {
if cachedTokens, ok := extractLlamaCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
}
}
}
}
func extractCachedTokensFromBody(body []byte) (int, bool) {
if len(body) == 0 {
return 0, false
}
var payload struct {
Usage struct {
PromptTokensDetails struct {
CachedTokens *int `json:"cached_tokens"`
} `json:"prompt_tokens_details"`
CachedTokens *int `json:"cached_tokens"`
PromptCacheHitTokens *int `json:"prompt_cache_hit_tokens"`
} `json:"usage"`
}
if err := common.Unmarshal(body, &payload); err != nil {
return 0, false
}
if payload.Usage.PromptTokensDetails.CachedTokens != nil {
return *payload.Usage.PromptTokensDetails.CachedTokens, true
}
if payload.Usage.CachedTokens != nil {
return *payload.Usage.CachedTokens, true
}
if payload.Usage.PromptCacheHitTokens != nil {
return *payload.Usage.PromptCacheHitTokens, true
}
return 0, false
}
// extractMoonshotCachedTokensFromBody 从Moonshot的非标准位置提取cached_tokens
// Moonshot的流式响应格式: {"choices":[{"usage":{"cached_tokens":111}}]}
func extractMoonshotCachedTokensFromBody(body []byte) (int, bool) {
if len(body) == 0 {
return 0, false
}
var payload struct {
Choices []struct {
Usage struct {
CachedTokens *int `json:"cached_tokens"`
} `json:"usage"`
} `json:"choices"`
}
if err := common.Unmarshal(body, &payload); err != nil {
return 0, false
}
// 遍历choices查找cached_tokens
for _, choice := range payload.Choices {
if choice.Usage.CachedTokens != nil && *choice.Usage.CachedTokens > 0 {
return *choice.Usage.CachedTokens, true
}
}
return 0, false
}
// extractLlamaCachedTokensFromBody 从llama.cpp的非标准位置提取cache_n
func extractLlamaCachedTokensFromBody(body []byte) (int, bool) {
if len(body) == 0 {
return 0, false
}
var payload struct {
Timings struct {
CachedTokens *int `json:"cache_n"`
} `json:"timings"`
}
if err := common.Unmarshal(body, &payload); err != nil {
return 0, false
}
if payload.Timings.CachedTokens == nil {
return 0, false
}
return *payload.Timings.CachedTokens, true
}
+287
View File
@@ -0,0 +1,287 @@
package openai
import (
"encoding/json"
"fmt"
"io"
"net/http"
"strings"
"time"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/dto"
"github.com/QuantumNous/new-api/logger"
relaycommon "github.com/QuantumNous/new-api/relay/common"
"github.com/QuantumNous/new-api/relay/helper"
"github.com/QuantumNous/new-api/service"
"github.com/QuantumNous/new-api/types"
"github.com/gin-gonic/gin"
)
// OpenaiImageHandler handles non-streaming OpenAI image responses
// (generations/edits), returning the parsed usage for billing.
func OpenaiImageHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
defer service.CloseResponseBodyGracefully(resp)
responseBody, err := io.ReadAll(resp.Body)
if err != nil {
return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
}
var usageResp dto.SimpleResponse
err = common.Unmarshal(responseBody, &usageResp)
if err != nil {
return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
}
if oaiError := usageResp.GetOpenAIError(); oaiError != nil && oaiError.Type != "" {
return nil, types.WithOpenAIError(*oaiError, resp.StatusCode)
}
// 写入新的 response body
service.IOCopyBytesGracefully(c, resp, responseBody)
normalizeOpenAIUsage(&usageResp.Usage)
applyUsagePostProcessing(info, &usageResp.Usage, responseBody)
return &usageResp.Usage, nil
}
// normalizeOpenAIUsage maps the OpenAI Images usage shape (input_tokens /
// output_tokens / input_tokens_details) onto the canonical prompt/completion
// fields. It is used only on the OpenAI image relay paths (generations/edits,
// streaming and non-streaming): the image API never returns prompt_tokens /
// completion_tokens, so the overwrite (=) semantics here are equivalent to the
// previous additive (+=) behavior while avoiding any future double-counting if
// both field sets are ever populated. Do not reuse this on chat/embedding paths
// without revisiting the overwrite semantics.
func normalizeOpenAIUsage(usage *dto.Usage) {
if usage == nil {
return
}
if usage.InputTokens != 0 {
usage.PromptTokens = usage.InputTokens
}
if usage.OutputTokens != 0 {
usage.CompletionTokens = usage.OutputTokens
}
if usage.InputTokensDetails != nil {
usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
usage.PromptTokensDetails.CachedCreationTokens = usage.InputTokensDetails.CachedCreationTokens
usage.PromptTokensDetails.ImageTokens = usage.InputTokensDetails.ImageTokens
usage.PromptTokensDetails.TextTokens = usage.InputTokensDetails.TextTokens
usage.PromptTokensDetails.AudioTokens = usage.InputTokensDetails.AudioTokens
}
if usage.TotalTokens == 0 {
usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
}
}
func OpenaiImageStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
if resp == nil || resp.Body == nil {
logger.LogError(c, "invalid image stream response")
return nil, types.NewOpenAIError(fmt.Errorf("invalid response"), types.ErrorCodeBadResponse, http.StatusInternalServerError)
}
contentType := strings.ToLower(resp.Header.Get("Content-Type"))
if resp.StatusCode < http.StatusOK || resp.StatusCode >= http.StatusMultipleChoices {
return OpenaiImageHandler(c, info, resp)
}
if !strings.Contains(contentType, "text/event-stream") {
return OpenaiImageJSONAsStreamHandler(c, info, resp)
}
// Reuse the shared streaming engine (helper.StreamScannerHandler) so the
// image streaming path gets the same ping keepalive, streaming-timeout
// watchdog, client-disconnect detection, panic recovery and goroutine
// cleanup as every other relay stream. The scanner delivers only the
// "data:" payload, so the SSE "event:" line is rebuilt from the JSON "type"
// field (real OpenAI image events keep event == type).
usage := &dto.Usage{}
var lastStreamData []byte
helper.StreamScannerHandler(c, resp, info, func(data string, sr *helper.StreamResult) {
raw := common.StringToByteSlice(data)
lastStreamData = raw
if isOpenAIImageStreamErrorEvent(raw) {
// Record the error as a soft error; the scanner drives the final
// EndReason. HasErrors() flags the failure for logging/handling.
sr.Error(fmt.Errorf("%s", extractOpenAIImageStreamErrorMessage(raw)))
}
var usageResp dto.SimpleResponse
if err := common.Unmarshal(raw, &usageResp); err == nil {
normalizeOpenAIUsage(&usageResp.Usage)
if service.ValidUsage(&usageResp.Usage) {
usage = &usageResp.Usage
}
}
writeOpenaiImageStreamChunk(c, raw)
})
// StreamScannerHandler consumes the upstream [DONE]; re-emit it so the
// client still receives a terminal data: [DONE].
if info != nil && info.StreamStatus != nil && info.StreamStatus.EndReason == relaycommon.StreamEndReasonDone {
helper.Done(c)
}
applyUsagePostProcessing(info, usage, lastStreamData)
return usage, nil
}
// writeOpenaiImageStreamChunk rebuilds the SSE frame for an image stream chunk:
// it emits an "event:" line derived from the JSON "type" field (when present)
// followed by the verbatim "data:" payload, mirroring helper.ResponseChunkData.
func writeOpenaiImageStreamChunk(c *gin.Context, data []byte) {
var payload struct {
Type string `json:"type"`
}
_ = common.Unmarshal(data, &payload)
if eventName := strings.TrimSpace(payload.Type); eventName != "" {
c.Render(-1, common.CustomEvent{Data: fmt.Sprintf("event: %s\n", eventName)})
}
c.Render(-1, common.CustomEvent{Data: "data: " + string(data)})
_ = helper.FlushWriter(c)
}
// isOpenAIImageStreamErrorEvent detects upstream error chunks by JSON content
// only ("type" of error/upstream_error, or a non-empty "error" field). The SSE
// "event:" line is not available here: StreamScannerHandler delivers only the
// "data:" payload. A payload carrying just a "message" key is deliberately NOT
// treated as an error to avoid false positives.
func isOpenAIImageStreamErrorEvent(data []byte) bool {
if !json.Valid(data) {
return false
}
var payload struct {
Type string `json:"type"`
Error json.RawMessage `json:"error"`
}
if err := common.Unmarshal(data, &payload); err != nil {
return false
}
payloadType := strings.ToLower(strings.TrimSpace(payload.Type))
return payloadType == "error" || payloadType == "upstream_error" || len(payload.Error) > 0
}
func extractOpenAIImageStreamErrorMessage(data []byte) string {
if len(data) == 0 || !json.Valid(data) {
return "upstream image stream returned error event"
}
var payload struct {
Message string `json:"message"`
Error json.RawMessage `json:"error"`
}
if err := common.Unmarshal(data, &payload); err != nil {
return "upstream image stream returned error event"
}
if msg := strings.TrimSpace(payload.Message); msg != "" {
return msg
}
if len(payload.Error) > 0 {
var nested struct {
Message string `json:"message"`
}
if err := common.Unmarshal(payload.Error, &nested); err == nil {
if msg := strings.TrimSpace(nested.Message); msg != "" {
return msg
}
}
if msg := strings.TrimSpace(common.JsonRawMessageToString(payload.Error)); msg != "" {
return msg
}
}
return "upstream image stream returned error event"
}
func OpenaiImageJSONAsStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
defer service.CloseResponseBodyGracefully(resp)
responseBody, err := io.ReadAll(resp.Body)
if err != nil {
return nil, types.NewOpenAIError(err, types.ErrorCodeReadResponseBodyFailed, http.StatusInternalServerError)
}
var imageResp dto.ImageResponse
if err := common.Unmarshal(responseBody, &imageResp); err != nil {
return nil, types.NewOpenAIError(err, types.ErrorCodeBadResponseBody, http.StatusInternalServerError)
}
var usageResp dto.SimpleResponse
_ = common.Unmarshal(responseBody, &usageResp)
if oaiError := usageResp.GetOpenAIError(); oaiError != nil && oaiError.Type != "" {
return nil, types.WithOpenAIError(*oaiError, resp.StatusCode)
}
normalizeOpenAIUsage(&usageResp.Usage)
applyUsagePostProcessing(info, &usageResp.Usage, responseBody)
helper.SetEventStreamHeaders(c)
c.Status(http.StatusOK)
created := imageResp.Created
if created == 0 {
created = time.Now().Unix()
}
if info != nil {
info.SetFirstResponseTime()
}
for _, image := range imageResp.Data {
payload := map[string]any{
"type": "image_generation.completed",
"created_at": created,
}
if image.Url != "" {
payload["url"] = image.Url
}
if image.B64Json != "" {
payload["b64_json"] = image.B64Json
}
if image.RevisedPrompt != "" {
payload["revised_prompt"] = image.RevisedPrompt
}
if service.ValidUsage(&usageResp.Usage) {
payload["usage"] = usageResp.Usage
}
if err := writeOpenaiImageStreamPayload(c, "image_generation.completed", payload); err != nil {
if info != nil && info.StreamStatus != nil {
info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonClientGone, err)
}
return &usageResp.Usage, nil
}
}
if err := writeOpenaiImageStreamDone(c); err != nil {
if info != nil && info.StreamStatus != nil {
info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonClientGone, err)
}
return &usageResp.Usage, nil
}
if info != nil {
info.ReceivedResponseCount += len(imageResp.Data)
if info.StreamStatus == nil {
info.StreamStatus = relaycommon.NewStreamStatus()
}
info.StreamStatus.SetEndReason(relaycommon.StreamEndReasonDone, nil)
}
return &usageResp.Usage, nil
}
func writeOpenaiImageStreamPayload(c *gin.Context, eventName string, payload any) error {
data, err := common.Marshal(payload)
if err != nil {
return err
}
if eventName != "" {
if _, err := fmt.Fprintf(c.Writer, "event: %s\n", eventName); err != nil {
return err
}
}
if _, err := fmt.Fprintf(c.Writer, "data: %s\n\n", data); err != nil {
return err
}
return helper.FlushWriter(c)
}
func writeOpenaiImageStreamDone(c *gin.Context) error {
if _, err := fmt.Fprint(c.Writer, "data: [DONE]\n\n"); err != nil {
return err
}
return helper.FlushWriter(c)
}
+242
View File
@@ -0,0 +1,242 @@
package openai
import (
"fmt"
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/dto"
"github.com/QuantumNous/new-api/logger"
relaycommon "github.com/QuantumNous/new-api/relay/common"
"github.com/QuantumNous/new-api/relay/helper"
"github.com/QuantumNous/new-api/service"
"github.com/QuantumNous/new-api/types"
"github.com/bytedance/gopkg/util/gopool"
"github.com/gin-gonic/gin"
"github.com/gorilla/websocket"
)
func OpenaiRealtimeHandler(c *gin.Context, info *relaycommon.RelayInfo) (*types.NewAPIError, *dto.RealtimeUsage) {
if info == nil || info.ClientWs == nil || info.TargetWs == nil {
return types.NewError(fmt.Errorf("invalid websocket connection"), types.ErrorCodeBadResponse), nil
}
info.IsStream = true
clientConn := info.ClientWs
targetConn := info.TargetWs
clientClosed := make(chan struct{})
targetClosed := make(chan struct{})
sendChan := make(chan []byte, 100)
receiveChan := make(chan []byte, 100)
errChan := make(chan error, 2)
usage := &dto.RealtimeUsage{}
localUsage := &dto.RealtimeUsage{}
sumUsage := &dto.RealtimeUsage{}
gopool.Go(func() {
defer func() {
if r := recover(); r != nil {
errChan <- fmt.Errorf("panic in client reader: %v", r)
}
}()
for {
select {
case <-c.Done():
return
default:
_, message, err := clientConn.ReadMessage()
if err != nil {
if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
errChan <- fmt.Errorf("error reading from client: %v", err)
}
close(clientClosed)
return
}
realtimeEvent := &dto.RealtimeEvent{}
err = common.Unmarshal(message, realtimeEvent)
if err != nil {
errChan <- fmt.Errorf("error unmarshalling message: %v", err)
return
}
if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdate {
if realtimeEvent.Session != nil {
if realtimeEvent.Session.Tools != nil {
info.RealtimeTools = realtimeEvent.Session.Tools
}
}
}
textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
if err != nil {
errChan <- fmt.Errorf("error counting text token: %v", err)
return
}
logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
localUsage.TotalTokens += textToken + audioToken
localUsage.InputTokens += textToken + audioToken
localUsage.InputTokenDetails.TextTokens += textToken
localUsage.InputTokenDetails.AudioTokens += audioToken
err = helper.WssString(c, targetConn, string(message))
if err != nil {
errChan <- fmt.Errorf("error writing to target: %v", err)
return
}
select {
case sendChan <- message:
default:
}
}
}
})
gopool.Go(func() {
defer func() {
if r := recover(); r != nil {
errChan <- fmt.Errorf("panic in target reader: %v", r)
}
}()
for {
select {
case <-c.Done():
return
default:
_, message, err := targetConn.ReadMessage()
if err != nil {
if !websocket.IsCloseError(err, websocket.CloseNormalClosure, websocket.CloseGoingAway) {
errChan <- fmt.Errorf("error reading from target: %v", err)
}
close(targetClosed)
return
}
info.SetFirstResponseTime()
realtimeEvent := &dto.RealtimeEvent{}
err = common.Unmarshal(message, realtimeEvent)
if err != nil {
errChan <- fmt.Errorf("error unmarshalling message: %v", err)
return
}
if realtimeEvent.Type == dto.RealtimeEventTypeResponseDone {
realtimeUsage := realtimeEvent.Response.Usage
if realtimeUsage != nil {
usage.TotalTokens += realtimeUsage.TotalTokens
usage.InputTokens += realtimeUsage.InputTokens
usage.OutputTokens += realtimeUsage.OutputTokens
usage.InputTokenDetails.AudioTokens += realtimeUsage.InputTokenDetails.AudioTokens
usage.InputTokenDetails.CachedTokens += realtimeUsage.InputTokenDetails.CachedTokens
usage.InputTokenDetails.TextTokens += realtimeUsage.InputTokenDetails.TextTokens
usage.OutputTokenDetails.AudioTokens += realtimeUsage.OutputTokenDetails.AudioTokens
usage.OutputTokenDetails.TextTokens += realtimeUsage.OutputTokenDetails.TextTokens
err := preConsumeUsage(c, info, usage, sumUsage)
if err != nil {
errChan <- fmt.Errorf("error consume usage: %v", err)
return
}
// 本次计费完成,清除
usage = &dto.RealtimeUsage{}
localUsage = &dto.RealtimeUsage{}
} else {
textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
if err != nil {
errChan <- fmt.Errorf("error counting text token: %v", err)
return
}
logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
localUsage.TotalTokens += textToken + audioToken
info.IsFirstRequest = false
localUsage.InputTokens += textToken + audioToken
localUsage.InputTokenDetails.TextTokens += textToken
localUsage.InputTokenDetails.AudioTokens += audioToken
err = preConsumeUsage(c, info, localUsage, sumUsage)
if err != nil {
errChan <- fmt.Errorf("error consume usage: %v", err)
return
}
// 本次计费完成,清除
localUsage = &dto.RealtimeUsage{}
// print now usage
}
logger.LogInfo(c, fmt.Sprintf("realtime streaming sumUsage: %v", sumUsage))
logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
logger.LogInfo(c, fmt.Sprintf("realtime streaming localUsage: %v", localUsage))
} else if realtimeEvent.Type == dto.RealtimeEventTypeSessionUpdated || realtimeEvent.Type == dto.RealtimeEventTypeSessionCreated {
realtimeSession := realtimeEvent.Session
if realtimeSession != nil {
// update audio format
info.InputAudioFormat = common.GetStringIfEmpty(realtimeSession.InputAudioFormat, info.InputAudioFormat)
info.OutputAudioFormat = common.GetStringIfEmpty(realtimeSession.OutputAudioFormat, info.OutputAudioFormat)
}
} else {
textToken, audioToken, err := service.CountTokenRealtime(info, *realtimeEvent, info.UpstreamModelName)
if err != nil {
errChan <- fmt.Errorf("error counting text token: %v", err)
return
}
logger.LogInfo(c, fmt.Sprintf("type: %s, textToken: %d, audioToken: %d", realtimeEvent.Type, textToken, audioToken))
localUsage.TotalTokens += textToken + audioToken
localUsage.OutputTokens += textToken + audioToken
localUsage.OutputTokenDetails.TextTokens += textToken
localUsage.OutputTokenDetails.AudioTokens += audioToken
}
err = helper.WssString(c, clientConn, string(message))
if err != nil {
errChan <- fmt.Errorf("error writing to client: %v", err)
return
}
select {
case receiveChan <- message:
default:
}
}
}
})
select {
case <-clientClosed:
case <-targetClosed:
case err := <-errChan:
//return service.OpenAIErrorWrapper(err, "realtime_error", http.StatusInternalServerError), nil
logger.LogError(c, "realtime error: "+err.Error())
case <-c.Done():
}
if usage.TotalTokens != 0 {
_ = preConsumeUsage(c, info, usage, sumUsage)
}
if localUsage.TotalTokens != 0 {
_ = preConsumeUsage(c, info, localUsage, sumUsage)
}
// check usage total tokens, if 0, use local usage
return nil, sumUsage
}
func preConsumeUsage(ctx *gin.Context, info *relaycommon.RelayInfo, usage *dto.RealtimeUsage, totalUsage *dto.RealtimeUsage) error {
if usage == nil || totalUsage == nil {
return fmt.Errorf("invalid usage pointer")
}
totalUsage.TotalTokens += usage.TotalTokens
totalUsage.InputTokens += usage.InputTokens
totalUsage.OutputTokens += usage.OutputTokens
totalUsage.InputTokenDetails.CachedTokens += usage.InputTokenDetails.CachedTokens
totalUsage.InputTokenDetails.TextTokens += usage.InputTokenDetails.TextTokens
totalUsage.InputTokenDetails.AudioTokens += usage.InputTokenDetails.AudioTokens
totalUsage.OutputTokenDetails.TextTokens += usage.OutputTokenDetails.TextTokens
totalUsage.OutputTokenDetails.AudioTokens += usage.OutputTokenDetails.AudioTokens
// clear usage
err := service.PreWssConsumeQuota(ctx, info, usage)
return err
}
+133
View File
@@ -0,0 +1,133 @@
package openai
import (
"github.com/QuantumNous/new-api/common"
"github.com/QuantumNous/new-api/constant"
"github.com/QuantumNous/new-api/dto"
relaycommon "github.com/QuantumNous/new-api/relay/common"
)
func applyUsagePostProcessing(info *relaycommon.RelayInfo, usage *dto.Usage, responseBody []byte) {
if info == nil || usage == nil {
return
}
switch info.ChannelType {
case constant.ChannelTypeDeepSeek:
if usage.PromptTokensDetails.CachedTokens == 0 && usage.PromptCacheHitTokens != 0 {
usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
}
case constant.ChannelTypeZhipu_v4:
// 智普的cached_tokens在标准位置: usage.prompt_tokens_details.cached_tokens
if usage.PromptTokensDetails.CachedTokens == 0 {
if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
} else if usage.PromptCacheHitTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
}
}
case constant.ChannelTypeMoonshot:
// Moonshot的cached_tokens在非标准位置: choices[].usage.cached_tokens
if usage.PromptTokensDetails.CachedTokens == 0 {
if usage.InputTokensDetails != nil && usage.InputTokensDetails.CachedTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.InputTokensDetails.CachedTokens
} else if cachedTokens, ok := extractMoonshotCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
} else if cachedTokens, ok := extractCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
} else if usage.PromptCacheHitTokens > 0 {
usage.PromptTokensDetails.CachedTokens = usage.PromptCacheHitTokens
}
}
case constant.ChannelTypeOpenAI:
if usage.PromptTokensDetails.CachedTokens == 0 {
if cachedTokens, ok := extractLlamaCachedTokensFromBody(responseBody); ok {
usage.PromptTokensDetails.CachedTokens = cachedTokens
}
}
}
}
func extractCachedTokensFromBody(body []byte) (int, bool) {
if len(body) == 0 {
return 0, false
}
var payload struct {
Usage struct {
PromptTokensDetails struct {
CachedTokens *int `json:"cached_tokens"`
} `json:"prompt_tokens_details"`
CachedTokens *int `json:"cached_tokens"`
PromptCacheHitTokens *int `json:"prompt_cache_hit_tokens"`
} `json:"usage"`
}
if err := common.Unmarshal(body, &payload); err != nil {
return 0, false
}
if payload.Usage.PromptTokensDetails.CachedTokens != nil {
return *payload.Usage.PromptTokensDetails.CachedTokens, true
}
if payload.Usage.CachedTokens != nil {
return *payload.Usage.CachedTokens, true
}
if payload.Usage.PromptCacheHitTokens != nil {
return *payload.Usage.PromptCacheHitTokens, true
}
return 0, false
}
// extractMoonshotCachedTokensFromBody 从Moonshot的非标准位置提取cached_tokens
// Moonshot的流式响应格式: {"choices":[{"usage":{"cached_tokens":111}}]}
func extractMoonshotCachedTokensFromBody(body []byte) (int, bool) {
if len(body) == 0 {
return 0, false
}
var payload struct {
Choices []struct {
Usage struct {
CachedTokens *int `json:"cached_tokens"`
} `json:"usage"`
} `json:"choices"`
}
if err := common.Unmarshal(body, &payload); err != nil {
return 0, false
}
// 遍历choices查找cached_tokens
for _, choice := range payload.Choices {
if choice.Usage.CachedTokens != nil && *choice.Usage.CachedTokens > 0 {
return *choice.Usage.CachedTokens, true
}
}
return 0, false
}
// extractLlamaCachedTokensFromBody 从llama.cpp的非标准位置提取cache_n
func extractLlamaCachedTokensFromBody(body []byte) (int, bool) {
if len(body) == 0 {
return 0, false
}
var payload struct {
Timings struct {
CachedTokens *int `json:"cache_n"`
} `json:"timings"`
}
if err := common.Unmarshal(body, &payload); err != nil {
return 0, false
}
if payload.Timings.CachedTokens == nil {
return 0, false
}
return *payload.Timings.CachedTokens, true
}
+1 -1
View File
@@ -92,7 +92,7 @@ func streamResponseTencent2OpenAI(TencentResponse *TencentChatResponse) *dto.Cha
func tencentStreamHandler(c *gin.Context, info *relaycommon.RelayInfo, resp *http.Response) (*dto.Usage, *types.NewAPIError) {
var responseText string
scanner := bufio.NewScanner(resp.Body)
scanner := helper.NewStreamScanner(resp.Body)
scanner.Split(bufio.ScanLines)
helper.SetEventStreamHeaders(c)
+1
View File
@@ -45,6 +45,7 @@ var claudeModelMap = map[string]string{
"claude-opus-4-5-20251101": "claude-opus-4-5@20251101",
"claude-opus-4-6": "claude-opus-4-6",
"claude-opus-4-7": "claude-opus-4-7",
"claude-opus-4-8": "claude-opus-4-8",
}
const anthropicVersion = "vertex-2023-10-16"

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