github/dbx
4
0 Fork 0
mirror of https://github.com/t8y2/dbx.git synced 2026-11-15 14:52:36 +00:00
Code Issues Packages Projects Releases Wiki Activity
2,968 Commits 35 Branches 497 Tags
5db0fc4fb14630bdf86e3f1c616f41d684600a0c
Commit Graph
260 Commits
Author SHA1 Message Date
t8y2 03e5e27c79 fix(ai): prevent marked parse error from crashing chat during streaming 2026-06-18 10:30:28 +08:00
Libon f1b8579994 style: improve UI interface 2026-06-18 02:01:00 +08:00
t8y2 6e7035823a fix(editor): use editor snapshot for SQL execution 2026-06-17 15:30:31 +08:00
Spencer.Chang 598b5dfef8 feat: add query editor view data context action 2026-06-17 14:42:15 +08:00
t8y2 5a08d42258 feat: make sidebar table page size configurable in settings 2026-06-17 02:00:15 +08:00
t8y2 412bb2be87 fix: add word-wrap toggle to danger confirm dialog 2026-06-16 23:00:46 +08:00
haipengno1 f14973eaef fix(redis): make completion menu follow typing 
The result used filter:false + validFor, which CodeMirror treats as
mutually exclusive — validFor is ignored, so the list froze and the
selection did not move while typing. Build the result with the default
built-in filter so typing narrows the list and moves the selection.
 2026-06-16 22:46:24 +08:00
haipengno1 1e96937455 feat(redis): add Redis command completion to query editor 
- add completion module: command names, two-stage subcommands, and key names
- key-name completion is async with a per-db cache capped at 1000 entries, cleared on disconnect
- gate key completion by argument position: stop after a single-key command is filled, keep suggesting for variadic key-list commands
- reuse the diagnostics command table as the data source, showing group/arity/safety
 2026-06-16 22:03:48 +08:00
Abeautifulsnow ec28005f20 feat(agent): safety check display 
* fix(i18n): add missing agent tool step translations for zh-TW, pt-BR, it, ja

* propagate explainData to agent step items for ExplainPlanViewer

* feat(ai): show safety check result for execute_query tool steps
 2026-06-16 13:56:16 +08:00
haipengno1 80a390a4c0 feat(redis): add Redis syntax checking for query editor 
- Built-in metadata for ~200 Redis commands (arity/group/safety), available offline
- Add diagnostic builder: unknown command name, arity errors, unclosed quotes, danger/write command highlighting
- Reuse existing diagnostic pipeline and integrate underlines + hover tips in QueryEditor
 2026-06-16 10:31:09 +08:00
Abeautifulsnow a1f15f6c45 fix(ai): propagate explainData and add missing agent i18n keys 
Thanks @Abeautifulsnow!
 2026-06-16 10:28:35 +08:00
t8y2 5f870be0d9 feat(editor): open table ddl panel from SQL 2026-06-16 00:51:34 +08:00
t8y2 cce7fcc600 fix(mysql): complete database-qualified columns 2026-06-15 23:35:11 +08:00
Libon 1991bf118b feat(infinite-scroll): implement infinite scroll feature with customizable max rows 2026-06-15 23:05:35 +08:00
t8y2 59fa888149 fix(dialog): keep drop database confirmation loading 2026-06-15 22:02:59 +08:00
Abeautifulsnow c300336281 feat(ai): context window management and token usage tracking 
* @
feat(ai): add conversation context management and token usage tracking (Phase 4)

## Background
Phase 4 of AI agent loop upgrade: token usage extraction from LLM responses,
cost estimation display, and automatic context window management (summarization
when approaching limits).

## Changes

### Backend (Rust)
- New token_usage::TokenUsage — cross-turn token accumulation
- New agent_loop::maybe_compact — proactive context compaction: token estimation
  → cut-point selection → LLM summarization → fallback summary
  → tool-pair integrity protection → drain/insert
- New agent_loop::context_window_for_model — auto-detect context window by model
- New agent_loop::is_context_length_error — detect upstream context-length errors
- New agent_loop::compact_tool_result_for_context — JSON/text sampling for large results
- Modified agent_loop::run_agent_loop — pre-turn compaction check, retry on context error
- Modified agent_events::AgentEnd — split total_tokens into input_tokens/output_tokens
- New agent_events::ContextCompacted — compaction event with metrics
- New AiConfig.context_window — user-configurable context window size
- New AiChatMessage.kind — context summary message marker
- Modified stream_with_tools & 3 provider stream functions — extract TokenUsage
  - Claude: message_start(input) + message_delta(output)
  - OpenAI: stream_options { include_usage } → final usage chunk
  - Gemini: usageMetadata per chunk, keep last

### Frontend (TypeScript/Vue)
- AiAssistant.vue: token + cost display row, estimateCost() with 7 model families,
  contextCompacted UI step, pendingCompaction deferred apply,
  messagesForAgentHistory post-compaction history reconstruction
- EditorSettingsDialog.vue: contextWindow input (min=1000, auto-detect)
- settingsStore.ts: AiConfig.contextWindow field
- tauri.ts: AgentEvent type updates, AiChatMessage.kind
- i18n: 6 new keys in en.ts and zh-CN.ts

## Fixed Issues
- messages[0] dropped from LLM context after compaction
- estimate_text_tokens hard threshold causing 2.8x estimate jump
- context_compacted event field semantics mismatch with frontend insert position
- validate_summary English-only keywords causing fallback on non-English LLMs
- Stale agentTokens after conversation switch
@

* remove cost computation logic
 2026-06-15 18:55:22 +08:00
Abeautifulsnow d845a318c6 feat(ai-agent): Phase 3 — explain_query tool, parallel/sequential execution, DB gating (#1187) 
- Add explain_query tool with structured EXPLAIN result, gated on supports_explain_plan()
- Add supports_sql_query() to gate execute_query/get_sample_data for non-SQL DBs (Redis, ES, MongoDB, etc.)
- Split agent loop tool dispatch: execute_query runs sequentially, all others run in parallel via join_all
- Wire explain_query result button to ExplainPlanViewer in AiAssistant + App.vue
- Fix O(n²) agentSteps rebuild: incremental O(1) push per event during streaming
- finally block falls back to full rebuild only when streaming push did not occur
 2026-06-15 13:14:59 +08:00
echocdeandEchocde 7dc27f8f6d fix(ui): fix appearance setting page display with text truncation tooltips 
* chore: test main release token push

* fix(desktop): unify toolbar dropdown tooltips

* chore: test main release token push

* fix(desktop): unify toolbar dropdown tooltips

* fix(ui) Appearance setting page displayed incorrectly.

* fix(ui) Appearance setting page displayed incorrectly.

* fix(ui) Appearance setting page displayed incorrectly.

* Remove unused import

---------

Co-authored-by: Echocde <udixt@outlook.com>
 2026-06-15 11:08:27 +08:00
t8y2 8af25343cc feat(ui): add password visibility toggle to all password inputs 2026-06-15 02:34:19 +08:00
t8y2 6c57169162 feat(mcp): add one-click install/upgrade button 2026-06-13 20:05:46 +08:00
t8y2 6a87f31fdc feat(ai): add resizable textarea with drag handle 2026-06-13 15:13:19 +08:00
dbin0123 2ef29c004a fix(editor): keep cursor at end of inserted completion text (#1136) 
Thanks @dbin0123!
 2026-06-13 11:42:03 +08:00
skyler abacac3c07 feat(ai): agent loop Phase 2 — SQL execution tools, risk classification, and streaming tool calls 
feat(ai): agent loop Phase 2 — SQL execution tools, risk classification, and streaming tool calls
 2026-06-12 19:56:36 +08:00
t8y2 484658c6b7 feat(ai): streaming probe for connection test with latency measurement 2026-06-12 19:29:55 +08:00
t8y2 fa4dbb9104 feat(completion): cache foreign key metadata 2026-06-12 19:27:16 +08:00
runstone 094412d956 fix error 2026-06-12 19:00:27 +08:00
runstone 40f0f98540 Merge branch 'main' of https://github.com/Abeautifulsnow/dbx into feat/ai-agent-loop-phase1 2026-06-12 18:52:05 +08:00
runstone 7df7fd9235 feat(ai): add tool result collapsible UI and fix agent mode prompt 
AiAssistant.vue:
- Tool call steps now support click-to-expand/collapse for SQL args
  and results (expandedSteps set); chevron rotates on open
- Remove fixed-open result display (was always visible, no toggle)
- Max result height raised to 48 (max-h-48) with scroll

ai.ts (buildModePromptLines agent branch):
- Replace vague "prioritize SQL" hint with explicit tool-use
  instruction: must call execute_query, not just emit SQL text
- List available tools (list_tables, get_columns, execute_query,
  get_sample_data) so LLM knows what it can call
- Clarify allowed statement types for execute_query
 2026-06-12 18:27:28 +08:00
runstone 100124da3e feat(ai): Phase 2 SQL 风险分类 + Agent 模式工具调用结果渲染 
后端：
  - 新增 sql_risk.rs：基于 sqlparser AST 的 SQL 风险分类（ReadOnly/Write/Ddl/Transaction）
  - agent_tools.rs：新增 execute_query 和 get_sample_data 工具，写操作/DDL 被风险分类阻止
  - agent_loop.rs：支持 is_agent_mode 参数，Agent 模式下暴露全部工具，Ask 模式仅暴露只读工具
  - agent_loop/web/tauri：mode 参数全链路传递（前端→HTTP→Tauri→Rust）

  前端：
  - AiAgentStepItem 接口扩展：toolName/toolArgs/toolResult/isError 字段
  - AiAssistant.vue：工具调用结果实时+最终渲染，SQL 查询结果/错误信息展示
  - ai.ts/http.ts/tauri.ts：mode 参数透传到后端
 2026-06-12 17:42:23 +08:00
polempandSam ef5ed05ae7 feat(schema-diff): professional DDL diff view with deployment-aware colors and custom themes (#1112) 
Co-authored-by: Sam <14344444@@qq.com>
 2026-06-12 17:14:05 +08:00
t8y2 2395727b6b fix(ai): auto-add /v1 prefix for bare-origin OpenAI-compatible endpoints 2026-06-12 16:10:42 +08:00
Abeautifulsnow 082e3bb4eb feat(ai): add agent tool calling loop 2026-06-12 13:18:36 +08:00
runstone 2d75667b9a feat(ai): upgrade agent loop tool calling to true SSE streaming 
Move tool calling from non-streaming (complete) API calls to real
streaming SSE, so text, reasoning, and tool call arguments all arrive
incrementally — users see the AI think and act in real time.

- Extract ~370 lines of non-streaming tool-call functions from
  agent_loop.rs (call_with_tools, call_openai_with_tools,
  call_claude_with_tools, call_gemini_with_tools) and replace with
  a single ai::stream_with_tools() dispatch call.

- Add streaming-with-tools infrastructure to ai.rs (~550 lines):
  - StreamToolEvent enum: Chunk, ToolCallStart, ToolCallDelta,
    ToolCallComplete — provider-agnostic, feeds into the
    StreamingToolCallAccumulator
  - stream_claude_with_tools(): parse SSE content_block_start/delta/
    stop events, incremental input_json_delta for tool arguments
  - stream_openai_with_tools(): parse SSE delta.tool_calls with
    incremental function.arguments
  - stream_gemini_with_tools(): parse SSE functionCall (Gemini
    sends complete objects, not deltas — emit as one chunk)
  - Cancellation support via tokio::select! in every streaming loop
  - Extended timeout for reasoning models (600s vs 120s)

- StreamingToolCallAccumulator: collect partial tool call fragments
  from streaming deltas, then deserialize into complete ToolCall
  objects in index order once the stream ends.
 2026-06-12 11:43:09 +08:00
runstone 91f1206783 Merge branch 'main' of https://github.com/Abeautifulsnow/dbx into feat/ai-agent-loop-phase1 2026-06-12 08:48:50 +08:00
haipengno1 7de0de2313 实现 Redis 的查询功能 2026-06-12 00:20:17 +08:00
t8y2 c9d655324d fix(editor): preserve IME composition in comments 2026-06-11 22:55:16 +08:00
runstone 9ab41795d6 @ 
feat(ai): add agent loop with tool calling, fix connection routing and Gemini compatibility

## 背景说明
将 dbx AI 助手从单次 SQL 文本生成器升级为具备 tool calling 的数据库 Agent。
Phase 1 实现 Agent Loop 框架 + 只读工具（list_tables、get_columns）。
参考：Pi agent-core，.trellis/tasks/06-11-feat-ai-agent-loop-tool-calling-upgrade

## 修改内容
- 新增 agent_loop.rs：Agent Loop 核心，支持 OpenAI/Claude/Gemini Provider 的 tool calling，最大 10 轮，Ollama 自动降级为 text-only
- 新增 agent_tools.rs：read_only_tools (list_tables, get_columns) 的工具定义与执行器
- 新增 agent_events.rs：AgentEvent/ToolCall/ToolResult/ToolDefinition 类型定义
- 修改 ai.rs：AiMessage 新增 tool_call_id 字段
- 修改 lib.rs：注册 agent_events、agent_loop、agent_tools 模块
- 修改 routes/ai.rs & commands/ai.rs：新增 ai_agent_stream SSE 端点
- 修改 tauri.ts & http.ts：新增 AgentEvent 类型与 aiAgentStream 函数
- 修改 api.ts：注册 aiAgentStream 转发
- 修改 ai.ts：新增 runAgentStream，AiContext 增加 connectionId 字段
- 修改 AiAssistant.vue：接入 runAgentStream，解析 AgentEvent 渲染工具调用步骤
- 修改 en.ts / zh-CN.ts：新增 agentSteps 国际化文案

## 修复的问题
- [#I01] connectionName 误传为 connectionId → add connectionId to AiContext
- [#I02] agent loop 取消操作无效 → add cancel checks before each turn
- [#C01] TODO 缺少 issue 引用 → add PRD reference to TODO
- [#C02] stream_with_tools 命名不准确 → add clarifying doc comment
- [#C03] Gemini functionResponse.name 解析错误 → fix tool_call_id format and extraction

## 测试建议
- execute_get_columns：合法表名/空表名/SQL 注入字符拒绝
- call_openai_with_tools：空 choices/空 tool_calls/非 JSON arguments
- call_gemini_with_tools：空 candidates/parts 无 functionCall
- runAgentStream：有效 connectionId 正常调用
@
 2026-06-11 18:27:24 +08:00
t8y2 21fd98d0ee feat: configurable toolbar visibility with settings UI 2026-06-11 13:32:39 +08:00
t8y2 4a264d638c feat: reorganize toolbar with more dropdown and add theme to appearance settings 2026-06-11 13:13:23 +08:00
SuLea-IT 89aa315a2d Merge pull request #1037 
Thanks!
 2026-06-11 11:12:05 +08:00
SuLea-IT 634e26e8e7 Merge PR #1024 2026-06-11 10:47:42 +08:00
t8y2 e3b4aac84c fix(desktop): tighten frontend transitions 2026-06-10 19:20:30 +08:00
t8y2 500d7c614b fix: resolve vue-tsc type errors in SqlPreviewPanel and DataGrid 2026-06-10 12:41:24 +08:00
t8y2 8df9484447 feat(ai): support auth methods 2026-06-10 12:14:36 +08:00
t8y2 8599ab8200 feat(data-grid): add SQL preview panel for pending data changes 2026-06-10 12:06:36 +08:00
t8y2 410e3bbfcb feat(completion): suggest database qualifiers 2026-06-10 11:08:13 +08:00
t8y2 3a52d961fc fix(completion): avoid table fallback for aliases 2026-06-10 10:51:41 +08:00
t8y2 51fe5de98b fix(shortcuts): allow clearing keyboard shortcuts in settings 2026-06-10 10:25:06 +08:00
t8y2 edba8214b5 feat: show username/password fields for Access database connections 2026-06-10 08:37:18 +08:00
t8y2 9ffa4c41d9 feat: add SQL library panel with folder/file management, rename, and sync support 2026-06-10 02:46:10 +08:00