Commit Graph

82 Commits

Author SHA1 Message Date
4b75a88247 fix(ui): status pie — drop center number, carve clean donut hole
- remove the total-req number drawn in the pie center
- carve a center hole (donut) with the card background for a cleaner look; total still shown in the card header caption and legend
- deployed (bak .bak.20260811o), server active
2026-08-11 17:05:18 +08:00
f4289c025c fix(ui): active request card -> real-time QPS-style sparkline; remove pie slice outline
- drawActiveChart: rolling request-rate buffer (window.qpsSeries) sampled ~every 2s from records timestamps, drawn as an area+fine; header shows current active requests + 'reqs / 3s' subtitle; refreshes on the existing 3s paintStats poll
- status pie: remove the hard slice outline (no stroke), clean seamless slices
- active card now consistent with others: big value + subtitle + chart
- deployed (bak .bak.20260811n), server active
2026-08-11 17:03:26 +08:00
ac0caf9e0c fix(ui): status block = pie chart (no labels) with color legend below
- rewrite drawStatusChart to a Canvas pie (no per-slice labels), center shows total requests
- status card: title + total caption + tall pie area (110px) + color legend row below (swatch + code + count), each item has title tooltip with ok/err
- store status swatch palette and reuse across pie/legend for color consistency
- deployed (bak .bak.20260811m), server active
2026-08-11 16:58:44 +08:00
7f91985583 fix(ui): restore five uniform KPI cards — title + big value + subtitle + real canvas charts (no gray-shell placeholders)
Redo the KPI row properly: five identical cards each with title row, large metric value, subtitle, and an 86px canvas chart. Replaces the previous gray-box canvas placeholders (which dropped the numbers — a regression).

- Active requests: line sparkline (real-time refresh) + live count
- Total requests: 7-day vertical bar chart + current total
- Tokens: stacked bar of per-model token share + big total
- Avg latency: line sparkline (real-time refresh) + avg ms
- Status codes: stacked bar by code + chip subtitle (code/reqs/percent)

All charts drawn with plain Canvas 2D (sparkline/dailyBars helpers), no third-party lib. DPR-aware for crispness. Deployed (bak .bak.20260811l), server active.
2026-08-11 16:57:04 +08:00
82f90d5114 fix(ui): unify five KPI blocks into identical cards with canvas placeholders
- unify all five metric boxes into identical .kpi cards (same height, padding, flex layout, canvas content area)
- replace old numeric KPI blocks with uniform <canvas> content areas
- move status donut into 5th KPI slot (already done), now uses same .kpi-card layout
- add canvas placeholders (cv-active, cv-reqs, cv-tokens, cv-lat, cv-status) for future sparklines/charts
- draw stubs: active/reqs/tokens/latency/status stub renderers (drawActiveChart etc.)
- CSS unified: .kpi flex column, .k-hdr title row, .k-canvas flex:1 canvas content area
- deployed (bak .bak.20260811k), server active
2026-08-11 16:51:58 +08:00
5bafac1d02 fix(ui): status-code donut as 5th KPI block — crisp path arcs, no clipping, compact chips
- move status distribution out of the full-width standalone card into the top KPI row as a 5th small block (aligned with the four KPI boxes)
- swap stroke-dasharray circles for SVG path arcs inside a rotated <g>: keeps each slice crisp and prevents the thick stroke clipping against the viewBox edge
- compact 108px donut + wrap-able status chip legend (hover title shows code/reqs/percent)
- drop redundant dashStatus subtitle; simplify title
- deployed to /usr/local/bin/llmsproxy (bak .bak.20260811j), server active
2026-08-11 16:28:23 +08:00
675789693c fix(ui): status-code donut chart with hover tooltips + model usage % = share of total; plan.md P4-9/P4-10
- paintStatusTable: replace plain table with SVG ring (conic slices per status code, center total, clickable legend with title=value/reqs/percent/ok/err)
- paintModelTable: percentage now = model share of total requests (was relative-to-max, so the top model always showed 100%/97%); bar stays max-relative
- plan.md: record P4-9 (visualization) and P4-10 (claude 400 'extra usage' = Anthropic account quota, not fingerprint; qijiar+openai serves claude fine, no adapter disguise needed)
- deployed to /usr/local/bin/llmsproxy (bak .bak.20260811i), server active
2026-08-11 16:08:43 +08:00
20d2268546 fix: close P10 audit items — P10-1 sources API admin guard, P10-2 scope prefix strip, P10-3 runtime source timeout w/ stream-safe clients
- api.go: handleSourcesAPI now requires admin role (GET leaks upstream api_keys, POST/DELETE mutate routing)
- chat.go: hasScopeModel made a Gateway method that strips source-model/:// prefix strictly via Registry.EffectiveModel (only when the prefix names a real source serving the bare model) so dash-bearing ids like deepseek-v4-flash-free are never corrupted; +TestHasScopeModelWithSourcePrefix
- config.go: DefaultSourceTimeout/QueueTimeout/Concurrency constants shared by YAML ApplyDefaults and runtime sources
- core.go: mergedSources applies the same defaults to runtime sources (JSON never persisted timeout fields); a dead upstream can no longer hold a concurrency slot forever
- provider.go: split non-streaming client{Timeout} vs stream client{} sharing a Transport with ResponseHeaderTimeout, so long SSE bodies are not cut by client.Timeout; ChatStream uses doRawStream
- plan.md: mark P4-4/5/6 done, record P4-7/8 (tier-order, audit export, UI key view, zen upstream diagnosis)
- online verified: user key -> /api/sources 403 (GET+POST), admin 200, AUTO stream/non-stream healthy
2026-08-11 15:22:20 +08:00
854b3e2e37 fix: AUTO tier order ascending + stats/CSV export completeness + UI key-view toggle & exit affordance
- scheduler: BuildChain sorts tiers ascending so tier 1 (highest priority) is tried first; previously descending inverted the chain (P10-1..P10-5)
- stats/api: handleStatsAPI limit→20000; CSV reads full audit via new Stats.AuditRecords (includes *.old rotation); key_names mapped by keyID() masked key; non-admin filter and keys-csv use masked keys
- server: ring buffer NewStats(10000)
- ui: dash-row card scroll area moved to table container (fix overflow below card); key view toggle (re-click same key returns to global) + kpi-exit affordance; rec-exit span
- chat: chain-failure record now surfaces first failed tier; AUTO comment sync
2026-08-11 13:36:23 +08:00
7647b7dff8 fix: WebUI chat crashes on streaming reasoning — sendChat never passed the reason flag to addMsg, so .think/.tk-hint were null and first reasoning_content chunk threw 'Cannot set properties of null'; pass reason=true and hide the empty box until thinking arrives 2026-08-11 09:52:12 +08:00
e8ef73321c fix: restart wipes stats history — LoadAudit replayed only the last 3000 lines (92% of which are 3s access events) and fed access rows into aggregates; now replays every real request into totals/quota windows, skips event rows, tolerates >64KB lines; +TestLoadAuditFullReplay 2026-08-11 08:42:13 +08:00
6221502adc fix: right-click ctx menu never closes — showCtx appended the menu but never stored it in ctxEl, so hideCtx (item click / outside press) was always a no-op; priority page and keys page affected 2026-08-11 00:15:16 +08:00
88802f9ef6 feat: AUTO chain rewrite — silent failover+busy skip+pref round-robin+503 tier summary; chain edits reset slot cooldowns (P0/P1); stats by_status + audit jsonl rotation; UI priority-page health badges & status-code card; ctx-menu capture-phase close (outside-press guard); main.go ops warnings; local bundled-Lua verified tests (3 latent bugs fixed); plan.md 2026-08-11 00:03:11 +08:00
63c2b13b4b fix: WebUI 密钥用量导出 CSV + 修复 Stats API key 过滤 bug
- api.go: 非 admin 用户过滤改用完整 key;keyNames 映射用完整 key;keys-csv 导出支持 key 查询参数过滤,安全类型断言
- stats.go: 新增 StatsRow 类型和 rows() 函数供导出使用
- server.go: handleStatusAPI 返回当前用户 key(已存在逻辑)
- index.html: 密钥用量卡片右上角添加导出 CSV 按钮(与请求记录一致)
2026-08-10 14:52:30 +08:00
41c9b0e14a feat: source-model routing disambiguation (source-model/:/ prefix); same-tier round-robin load balancing; public_base_url for copy config; fmtTok(B/M/K) unit scaling; status column reorder (reachability first); drop emoji from seed-warn modal; fix tests for first-run adapter seeding; install lua5.1 dev lib 2026-08-10 12:55:22 +08:00
fe06e0861a feat(seed-warn): dynamic seed key detection in UI modal; feat(https): TLS config (tls_cert_file/tls_key_file) with dynamic base_url; docs: rotate admin key reminder 2026-08-10 12:11:32 +08:00
f46b02089c feat(delete): real deletes — adapters seeded once into adapter_dir (existing dir is authoritative), sources removed from config.yaml on delete; drop tombstone mechanism for both. Docs: multi-key architecture, gateway_keys as seed 2026-08-10 11:52:51 +08:00
e2dd4d9727 fix(ui): clipboard fallback for non-secure contexts; touch-friendly delete button on key scope blocks; append named model after AUTO when adding (+ docs sync: encrypted storage, AUTO chain, live probing, image endpoint) 2026-08-10 11:10:59 +08:00
62c8a07b88 fix(ui): sources page model tags wrap 4 per row (same grid as status page) 2026-08-10 10:37:57 +08:00
839f33ba01 feat(secrets): encrypt sensitive store fields (AES-GCM, master.key 0600, api_key_env) + fast models-endpoint probing (fix zen backlog + source status) + UI cleanup (drop redundant parens labels, grid models 4/row) 2026-08-10 10:33:03 +08:00
e0978db150 fix(ui): logout power icon was italic (<i> default style); hide it on wide screens, normalize font-style 2026-08-10 09:38:33 +08:00
fa65dcbe70 fix(ui): narrow-screen header (hide brand name, icon-only nav, scrollable), wrap wide tables for horizontal scroll, add favicon 2026-08-10 09:35:21 +08:00
597babe331 fix(stats): drop empty-model records from model aggregates; clean stale blank audit entries 2026-08-10 08:57:33 +08:00
d48b993010 feat(auto): AUTO-only priority chain with tiered slots + live source probing; CSV export w/ key names; audit persistence; fix prompt token accounting & deepseek thinking 2026-08-10 00:10:19 +08:00
859d310ad3 feat(auto): quota-aware AUTO slot scheduling (model tiers, hour/week/month resets) + key scope periods; fix key/slot create/delete UX 2026-08-09 11:07:38 +08:00
3408c9cb1f feat(keys): role-based gateway keys with admin management UI and per-user model scope 2026-08-09 10:01:40 +08:00
dec03238dd feat(stats): gateway request stats/audit API + polish scratch sort animations (link chain, reset, cleanup) 2026-08-08 13:34:40 +08:00
31eefe9288 feat(webui): scratch jigsaw-block canvas priority sorting (same-value prio, image excluded) 2026-08-08 11:41:46 +08:00
b0f8bc8ffd feat(webui): scratch-style drag sorting of model priority per source 2026-08-08 11:34:57 +08:00
5d50b69153 fix: tool call anchor & wire format, streaming chunk passthrough, WebUI narrow-screen, docs bilingual 2026-08-08 11:26:58 +08:00
ccbcede581 feat: LuaJIT worker-pool VM, multimodal/disable-thinking passthrough, WebUI redesign, DeepSeek V4 2026-08-07 13:34:16 +08:00
f7f76e097d feat: ModelRouter — unified OpenAI-compatible multi-source LLM gateway
- Lua adapters per upstream (transform_request/response/stream_chunk, build_headers signing hooks)
- AUTO priority routing with per-model kind (chat/image), explicit source/model routing
- Per-source concurrency caps with queueing, exponential backoff, AUTO failover
- OpenAI-compatible API: chat completions, SSE streaming, image generations, models
- Gateway key auth, web UI for adapter/source management, runtime persistence
- e2e test running the real binary against mocked upstreams
2026-08-05 15:25:47 +08:00