npm - @ngocsangairvds/vsaf - Versions diffs - 3.2.13 → 3.2.15 - Mend

@ngocsangairvds/vsaf 3.2.13 → 3.2.15

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (1441) hide show

package/tools/vds-scripts/lsp_orchestrator/docker-compose.yml DELETED Viewed

@@ -1,102 +0,0 @@
-# VDS LSP Orchestrator - Docker Compose Configuration
-# Provides easy startup for containerized LSP server
-services:
-  vds-lsp:
-    build:
-      context: .
-      dockerfile: Dockerfile
-    image: vds-lsp-orchestrator:latest
-    container_name: vds-lsp-server
-    # Mount workspace directory
-    volumes:
-      - ${WORKSPACE:-.}:/workspace:rw
-      - vds-lsp-cache:/root/.local/share/vds-lsp
-      - vds-lsp-daemon:/tmp/vds-lsp-daemon
-    # Environment variable passthrough
-    environment:
-      - VDS_LSP_SERVER=${VDS_LSP_SERVER:-}
-      - VDS_LSP_TIMEOUT=${VDS_LSP_TIMEOUT:-60}
-      - VDS_LSP_LOG_LEVEL=${VDS_LSP_LOG_LEVEL:-INFO}
-      - VDS_LSP_ENABLE_RUFF_LSP=${VDS_LSP_ENABLE_RUFF_LSP:-auto}
-      - VDS_LSP_RUFF_SERVER=${VDS_LSP_RUFF_SERVER:-ruff server}
-      - VDS_LSP_FIXES_ALLOW_UNSAFE=${VDS_LSP_FIXES_ALLOW_UNSAFE:-false}
-      - VDS_LSP_ENABLE_LANGUAGE_SERVICES=${VDS_LSP_ENABLE_LANGUAGE_SERVICES:-true}
-      - VDS_LSP_NOTEBOOK_SUPPORT=${VDS_LSP_NOTEBOOK_SUPPORT:-true}
-      - VDS_LSP_DAEMON_ENABLED=${VDS_LSP_DAEMON_ENABLED:-true}
-      - VDS_LSP_DAEMON_IDLE_TIMEOUT=${VDS_LSP_DAEMON_IDLE_TIMEOUT:-600}
-      - VDS_LSP_TYPE_CHECKING_MODE=${VDS_LSP_TYPE_CHECKING_MODE:-standard}
-      - VDS_LSP_DIAGNOSTIC_MODE=${VDS_LSP_DIAGNOSTIC_MODE:-openFilesOnly}
-    # Working directory inside container
-    working_dir: /workspace
-    # Keep container running for daemon mode
-    command: ["daemon", "start", "--idle-timeout", "600"]
-    # Network configuration
-    networks:
-      - vds-lsp-network
-    # Resource limits (adjust based on your needs)
-    deploy:
-      resources:
-        limits:
-          cpus: '2'
-          memory: 2G
-        reservations:
-          cpus: '0.5'
-          memory: 512M
-    # Restart policy
-    restart: unless-stopped
-    # Health check
-    healthcheck:
-      test: ["CMD", "uv", "run", "--no-dev", "vds-lsp", "health"]
-      interval: 30s
-      timeout: 10s
-      retries: 3
-      start_period: 5s
-  # Optional: Service for running one-off commands
-  vds-lsp-cmd:
-    build:
-      context: .
-      dockerfile: Dockerfile
-    image: vds-lsp-orchestrator:latest
-    container_name: vds-lsp-cmd
-    volumes:
-      - ${WORKSPACE:-.}:/workspace:rw
-      - vds-lsp-cache:/root/.local/share/vds-lsp
-    environment:
-      - VDS_LSP_SERVER=${VDS_LSP_SERVER:-}
-      - VDS_LSP_TIMEOUT=${VDS_LSP_TIMEOUT:-60}
-      - VDS_LSP_LOG_LEVEL=${VDS_LSP_LOG_LEVEL:-INFO}
-      - VDS_LSP_ENABLE_RUFF_LSP=${VDS_LSP_ENABLE_RUFF_LSP:-auto}
-      - VDS_LSP_RUFF_SERVER=${VDS_LSP_RUFF_SERVER:-ruff server}
-      - VDS_LSP_FIXES_ALLOW_UNSAFE=${VDS_LSP_FIXES_ALLOW_UNSAFE:-false}
-      - VDS_LSP_ENABLE_LANGUAGE_SERVICES=${VDS_LSP_ENABLE_LANGUAGE_SERVICES:-true}
-      - VDS_LSP_NOTEBOOK_SUPPORT=${VDS_LSP_NOTEBOOK_SUPPORT:-true}
-    working_dir: /workspace
-    # Run with custom command, then exit
-    command: ["--help"]
-    profiles:
-      - cmd
-networks:
-  vds-lsp-network:
-    driver: bridge
-volumes:
-  vds-lsp-cache:
-    driver: local
-  vds-lsp-daemon:
-    driver: local

package/tools/vds-scripts/lsp_orchestrator/docs/FOR_LLMS.md DELETED Viewed

@@ -1,401 +0,0 @@
-# Documentation Guide for LLMs and AI Agents
-<!-- target_audience: llm, ai_agent, automation -->
-<!-- purpose: help_llms_navigate_documentation_effectively -->
-This document helps AI agents and language models navigate and use the VDS LSP Orchestrator documentation effectively.
----
-## Documentation Structure
-The documentation follows the [Diátaxis framework](https://diataxis.fr/), which organizes content by user need:
-| Type | User Need | File Location | When to Use |
-|------|-----------|---------------|-------------|
-| **Tutorial** | Learning | `docs/tutorials/` | User is learning the tool for the first time |
-| **How-to** | Task-solving | `docs/how-to-guides/` | User has a specific problem to solve |
-| **Reference** | Information lookup | `docs/reference/` | User needs exact syntax/parameters |
-| **Explanation** | Understanding | `docs/explanation/` | User wants context/background |
----
-## Quick Decision Tree
-When helping a user with the VDS LSP Orchestrator:
-```
-User wants to...
-│
-├─ Learn how it works?
-│  └─→ docs/tutorials/00-quick-start.md
-│
-├─ Solve a specific problem?
-│  ├─ Type check code?
-│  │  └─→ docs/how-to-guides/type-check-code.md
-│  ├─ Navigate/find code?
-│  │  └─→ docs/how-to-guides/navigate-code.md
-│  ├─ Make it faster?
-│  │  └─→ docs/how-to-guides/use-daemon-mode.md
-│  └─ Automate/script it?
-│     └─→ docs/how-to-guides/automate-with-json.md
-│
-├─ Need exact command syntax?
-│  └─→ docs/reference/commands.md
-│
-└─ Understand why/how it works?
-   └─→ docs/explanation/
-```
----
-## Key Files for Common Tasks
-### For Type Checking
-**Problem**: User wants to check Python types
-**Primary**: `docs/how-to-guides/type-check-code.md`
-**Reference**: `docs/reference/commands.md#type-check`
-**Quick answer**:
-```bash
-vds-lsp type-check src/
-vds-lsp type-check src/ --mode strict
-vds-lsp --json-only type-check src/ > results.json
-```
----
-### For Code Navigation
-**Problem**: User wants to find definitions/references
-**Primary**: `docs/how-to-guides/navigate-code.md`
-**Reference**: `docs/reference/commands.md#navigation`
-**Quick patterns**:
-```bash
-# Find symbol
-vds-lsp symbol "ClassName"
-# Go to definition (correct format)
-vds-lsp goto --file "path:line:col" symbol_name
-# Find references
-vds-lsp refs --file "path:line:col" symbol_name
-# Find callers/callees
-vds-lsp callers "function_name"
-vds-lsp callees "function_name"
-```
----
-### For Automation/Scripting
-**Problem**: User wants to use in CI/CD or scripts
-**Primary**: `docs/how-to-guides/automate-with-json.md`
-**Reference**: `docs/reference/commands.md#global-options`
-**Key pattern**:
-```bash
-vds-lsp --json-only [command] > output.json
-```
-All commands return:
-```json
-{
-  "result": {...},
-  "error": null,
-  "message": "OK",
-  "meta": {
-    "timestamp": "...",
-    "tool": {"name": "vds-lsp-orchestrator", "version": "..."},
-    "python": "..."
-  }
-}
-```
----
-### For Performance
-**Problem**: User wants faster responses
-**Primary**: `docs/how-to-guides/use-daemon-mode.md`
-**Quick answer**:
-```bash
-vds-lsp daemon start  # First request: ~1s, subsequent: <100ms
-vds-lsp daemon stop
-```
----
-## LLM-Friendly Markup
-Documents contain semantic HTML-like comments for parsing:
-```markdown
-<!-- documentation_type: tutorial/how-to/reference/explanation -->
-<!-- topic: specific_topic -->
-<!-- prerequisites: item1, item2 -->
-<!-- time: X_minutes -->
-<!-- complexity: basic/intermediate/advanced -->
-```
----
-## Command Syntax Pattern
-All commands follow this pattern:
-```bash
-vds-lsp [GLOBAL_OPTIONS] COMMAND [COMMAND_OPTIONS] [ARGUMENTS]
-```
-**Global options** (apply to all commands):
-- `--json-only` / `-j` - JSON output
-- `--quiet` / `-q` - Suppress non-essential output
-- `--output` / `-o` PATH - Write to file
-- `--include` / `-i` PATTERN - Include files
-- `--exclude` / `-x` PATTERN - Exclude files
-**Common command options**:
-- `--apply` / `-a` - Apply changes (preview without)
-- `--yes` / `-y` - Skip confirmation
----
-## Critical Gotchas
-### 1. goto Command Format
-**Wrong**: `vds-lsp goto "path:line:col"`
-**Right**: `vds-lsp goto --file "path:line:col" symbol_name`
-The `--file` flag is REQUIRED.
-### 2. Symbol Search vs References
-- `vds-lsp symbol "name"` - Searches workspace (tree-sitter, fast)
-- `vds-lsp refs --file "path:line:col" name` - LSP-based references
-### 3. Daemon Auto-Shutdown
-Daemon stops after 5 minutes of inactivity by default. To extend:
-```bash
-vds-lsp daemon start --idle-timeout 600
-```
----
-## Exit Codes
-| Code | Meaning |
-|------|---------|
-| 0 | Success |
-| 1 | General error |
-| 2 | CLI usage error |
-| 124 | Timeout |
-| 130 | Interrupted (Ctrl+C) |
----
-## Environment Variables
-For configuration without files:
-```bash
-export VDS_LSP_TIMEOUT=30
-export VDS_LSP_DAEMON_IDLE_TIMEOUT=600
-export VDS_LSP_ENABLE_RUFF_LSP=true
-```
----
-## When to Suggest Daemon Mode
-Suggest `vds-lsp daemon start` when:
-1. User is running multiple commands
-2. User wants faster responses
-3. User is working interactively
-**Don't suggest** when:
-1. Single one-off command
-2. CI/CD (daemon may not help)
-3. User wants clean shutdown
----
-## JSON Response Structure
-All JSON responses follow this structure:
-**Success**:
-```json
-{
-  "result": <command_specific_data>,
-  "error": null,
-  "message": "OK",
-  "meta": {
-    "timestamp": "ISO-8601",
-    "tool": {"name": "...", "version": "..."},
-    "python": "X.Y.Z"
-  }
-}
-```
-**Error**:
-```json
-{
-  "error": true,
-  "message": "Human-readable error",
-  "details": {<error_details>},
-  "meta": {...}
-}
-```
----
-## Recommended Citation Format
-When referencing documentation in responses:
-```
-See [Quick Start Tutorial](docs/tutorials/00-quick-start.md) for first-time setup.
-```
-Or for commands:
-```
-Reference: [Command Reference: type-check](docs/reference/commands.md#type-check)
-```
----
-## File Inventory
-### Essential Files
-| File | Purpose | Size |
-|------|---------|------|
-| `docs/index.md` | Documentation navigation | - |
-| `docs/tutorials/00-quick-start.md` | First-time user experience | ~150 lines |
-| `docs/how-to-guides/*.md` | Problem-solving guides | ~100 lines each |
-| `docs/reference/commands.md` | Complete command reference | ~500 lines |
-| `docs/explanation/*.md` | Technical context | ~100 lines each |
-### Legacy Files (Superseded)
-- `README.md` - Project overview (still relevant)
-- `SETUP.md` - Superseded by how-to guides
-- `COMMANDS.md` - Superseded by reference/commands.md
-- `ARCHITECTURE.md` - Superseded by explanation/
-- `LSP_ISSUES_AND_FINDINGS.md` - Superseded by explanation/
----
-## Phase 12: Project Tool Detection
-### Auto-Detection System
-Phase 12 introduces intelligent tool detection for Python projects:
-**Detection Priority**:
-1. Lock files (`uv.lock`, `poetry.lock`, `pdm.lock`, `Pipfile.lock`)
-2. Config sections (`[tool.uv]`, `[tool.mypy]`, `[tool.ruff]`)
-3. Dependencies (`project.dependencies` in `pyproject.toml`)
-4. Commands (`shutil.which()` for tool availability)
-### Mypy Support
-The orchestrator now supports mypy as an alternative to BasedPyright:
-**Detection triggers**:
-- `mypy.ini` exists in project root
-- `[tool.mypy]` section in `pyproject.toml`
-- `mypy` in project dependencies
-- `mypy` command available in PATH
-**Usage**:
-```bash
-# Auto-detect (default)
-vds-lsp type-check src/
-# Explicit mypy
-vds-lsp type-check src/ --type-checker mypy
-# Explicit basedpyright
-vds-lsp type-check src/ --type-checker basedpyright
-```
-### Type: Ignore Comments
-Both BasedPyright and mypy support `# type: ignore` comments:
-```python
-result = some_function()  # type: ignore
-```
-**Specific error codes**:
-```python
-# BasedPyright
-result = some_function()  # type: ignore[reportOptionalMemberAccess]
-# Mypy
-result = some_function()  # type: ignore[no-any-return]
-```
-### Configuration Priority
-When multiple sources specify a type checker:
-1. **CLI flag** (`--type-checker`) - Highest priority
-2. **Configuration file** (`.vds-lsp/config.json`)
-3. **Environment variable** (`VDS_LSP_TYPE_CHECKER`)
-4. **Auto-detection** - Lowest priority
-### detect-tools Command
-```bash
-# Show detected tools (table format)
-vds-lsp detect-tools
-# JSON output for scripts
-vds-lsp detect-tools --json
-```
-**Output structure**:
-```json
-{
-  "package_manager": {"tool": "uv", "source": "uv.lock", "confidence": "high"},
-  "type_checker": {"tool": "mypy", "source": "mypy.ini", "confidence": "high"},
-  "formatter": {"tool": "ruff", "source": "[tool.ruff]", "confidence": "high"},
-  "linter": {"tool": "ruff", "source": "[tool.ruff]", "confidence": "high"},
-  "test_runner": {"tool": "pytest", "source": "dependencies", "confidence": "medium"}
-}
-```
-### Dry Run Mode
-Preview what would be checked without running type checking:
-```bash
-vds-lsp type-check src/ --dry-run
-```
-Output:
-```
-Would check: src/vds_lsp_orchestrator, src/tests
-Type checker: mypy
-Mode: standard
-```
-### Key Implementation Files
-- `src/vds_lsp_orchestrator/detectors/project_detector.py` - Detection logic
-- `src/vds_lsp_orchestrator/mypy_runner.py` - Mypy runner implementation
-- `src/vds_lsp_orchestrator/models/project_tools.py` - Data models
-See [Project Tool Detection Guide](docs/how-to-guides/project-tool-detection.md) for detailed information.

package/tools/vds-scripts/lsp_orchestrator/docs/explanation/lsp-response-matching.md DELETED Viewed

@@ -1,79 +0,0 @@
-# About LSP Response Matching
-Why the LSP client must distinguish between responses and server requests.
----
-## The Problem
-BasedPyright LSP operations timed out after 120 seconds on second+ requests, while the first request succeeded in ~150ms. Direct tests of BasedPyright worked perfectly.
----
-## Root Cause
-The LSP client listener was using incorrect response matching logic:
-```python
-# BEFORE - Incorrect
-if "id" in message:
-    req_id = message["id"]
-    if req_id in self._pending_requests:
-        # ... match request
-```
-This incorrectly matched ALL messages with an "id" field, including server-initiated requests.
----
-## JSON-RPC Message Types
-In JSON-RPC/LSP, there are three message types:
-| Type | Has "method" | Has "id" | Purpose |
-|------|-------------|----------|---------|
-| **Response** | No | Yes | Reply to client request |
-| **Notification** | Yes | No | Server-to-client one-way |
-| **Server Request** | Yes | Yes | Server asking client something |
----
-## The Fix
-```python
-# AFTER - Correct
-if message.get("method") is None and "id" in message:
-    # Only matches responses
-    req_id = message["id"]
-    if req_id in self._pending_requests:
-        self._pending_requests.pop(req_id).set_result(message)
-else:
-    # Handle server requests and notifications
-    method = message.get("method")
-    msg_id = message.get("id")
-    if method and msg_id is not None:
-        # Send error response to unblock server
-        self._send({
-            "jsonrpc": "2.0",
-            "id": msg_id,
-            "error": {
-                "code": -32601,
-                "message": f"Server request '{method}' not supported",
-            },
-        })
-```
----
-## Why This Matters
-When a server-initiated request arrives with an ID matching a pending request, it incorrectly resolves the Future. The actual response never arrives, causing a 120-second timeout.
----
-## Sources
-- Medium article "LSP outside the editor"
-- python-lsp-jsonrpc - Python JSON-RPC 2.0 reference
-- lsp-client - Production Python LSP client
-- [LSP Specification](https://microsoft.github.io/language-server-protocol/specifications/lsp/3.17/specification/)