@jerryjiao/knowflow 0.3.0

package/scripts/graph_builder.py

@@ -0,0 +1,612 @@
+#!/usr/bin/env python3
+"""
+KnowFlow — Knowledge Graph Builder (M2 Enhanced)
+Parses wiki/ markdown files, extracts [[wikilink]] relationships,
+generates interactive graph.html (self-contained, no server needed).
+
+M2 改进:
+  - 支持增量更新（--incremental 模式，检查已有数据避免全量重建）
+  - 输出详细节点统计信息到控制台
+  - 确保 vis.js HTML 文件中的中文显示正常（UTF-8 BOM + 显式 charset）
+  - 同时输出 knowledge-graph.json 供外部工具消费
+
+Usage:
+  python3 graph_builder.py [--wiki-dir WIKI_ROOT] [--output OUTPUT.html] [--incremental]
+"""
+
+import os
+import re
+import json
+import argparse
+import hashlib
+import time
+from pathlib import Path
+from collections import defaultdict
+
+# ── Configuration ──────────────────────────────────────────
+PROJECT_ROOT = Path(__file__).resolve().parent.parent
+DEFAULT_WIKI_DIR = os.environ.get("KNOWFLOW_WIKI_DIR", str(PROJECT_ROOT / "wiki"))
+DEFAULT_OUTPUT = os.environ.get("KNOWFLOW_GRAPH_OUTPUT", str(PROJECT_ROOT / "graph" / "graph.html"))
+# 增量更新状态文件：记录每个文件的 hash，用于检测变更
+STATE_FILE = str(Path(DEFAULT_OUTPUT).resolve().parent / '.graph-state.json')
+
+# Node colors by category
+CATEGORY_COLORS = {
+    "sources":   {"bg": "#E3F2FD", "border": "#1565C0", "text": "#0D47A1"},     # Blue
+    "entities":  {"bg": "#F3E5F5", "border": "#7B1FA2", "text": "#4A148C"},     # Purple
+    "concepts":  {"bg": "#E8F5E9", "border": "#2E7D32", "text": "#1B5E20"},     # Green
+    "comparisons": {"bg": "#FFF3E0", "border": "#EF6C00", "text": "#E65100"},  # Orange
+    "_special":  {"bg": "#FFEBEE", "border": "#C62828", "text": "#B71C1C"},     # Red (index/log/overview)
+}
+
+# Node shapes by category
+CATEGORY_SHAPES = {
+    "sources":      "box",
+    "entities":      "dot",
+    "concepts":      "diamond",
+    "comparisons":   "hexagon",
+    "_special":      "star",
+}
+
+
+def log(msg):
+    """带时间戳的日志"""
+    ts = time.strftime("%H:%M:%S")
+    print(f"[{ts}] {msg}")
+
+
+def file_hash(filepath: str) -> str:
+    """计算文件的 MD5 hash 用于增量检测"""
+    try:
+        with open(filepath, 'rb') as f:
+            return hashlib.md5(f.read()).hexdigest()
+    except Exception:
+        return ""
+
+
+def load_state() -> dict:
+    """加载上次的构建状态（文件 hash 映射）"""
+    if os.path.exists(STATE_FILE):
+        try:
+            with open(STATE_FILE, 'r', encoding='utf-8') as f:
+                return json.load(f)
+        except Exception:
+            pass
+    return {}
+
+
+def save_state(file_hashes: dict):
+    """保存当前构建状态"""
+    state_dir = os.path.dirname(STATE_FILE)
+    os.makedirs(state_dir, exist_ok=True)
+    with open(STATE_FILE, 'w', encoding='utf-8') as f:
+        json.dump({
+            "file_hashes": file_hashes,
+            "updated_at": time.strftime("%Y-%m-%d %H:%M:%S"),
+        }, f, ensure_ascii=False, indent=2)
+
+
+def detect_category(filepath: str) -> str:
+    """Detect category from file path."""
+    parts = Path(filepath).parts
+    if len(parts) >= 2:
+        parent = parts[-2]
+        if parent in CATEGORY_COLORS:
+            return parent
+    filename = Path(filepath).stem.lower()
+    if filename in ("index", "log", "overview"):
+        return "_special"
+    return "sources"  # default
+
+
+def extract_title(content: str, filepath: str) -> str:
+    """Extract title from first H1 heading."""
+    m = re.search(r'^#\s+(.+)$', content, re.MULTILINE)
+    if m:
+        return m.group(1).strip()
+    return Path(filepath).stem.replace("-", " ").title()
+
+
+def extract_summary(content: str, max_len: int = 120) -> str:
+    """Extract first meaningful paragraph as summary."""
+    lines = content.split("\n")
+    for line in lines:
+        line = line.strip()
+        if not line or line.startswith("#") or line.startswith(">") or line.startswith("|"):
+            continue
+        if line.startswith("-") or line.startswith("*"):
+            continue
+        clean = re.sub(r'\[([^\]]+)\]\([^)]+\)', r'\1', line)
+        clean = re.sub(r'[*_`#]', '', clean)
+        if len(clean) > 10:
+            return clean[:max_len] + ("..." if len(clean) > max_len else "")
+    return "(无摘要)"
+
+
+def extract_wikilinks(content: str) -> list[tuple[str, int]]:
+    """Extract all [[wikilink]] targets with their positions."""
+    links = []
+    for m in re.finditer(r'\[\[([^\]]+)\]\]', content):
+        target = m.group(1).strip()
+        links.append((target, m.start()))
+    return links
+
+
+def resolve_link(target: str, all_files: dict[str, str]) -> str | None:
+    """
+    Resolve a wikilink target to an actual file path.
+    Handles various formats:
+      - entities/karpathy → entities/karpathy.md
+      - karpathy-andrej → find karpathy-andrej.md anywhere
+    """
+    # Direct match with .md extension
+    if target + ".md" in all_files:
+        return target + ".md"
+    
+    # Already has .md
+    if target.endswith(".md") and target in all_files:
+        return target
+    
+    # Partial match (filename without path)
+    basename = Path(target).stem
+    for filepath in all_files:
+        if Path(filepath).stem == basename:
+            return filepath
+    
+    # Fuzzy match (contains)
+    for filepath in all_files:
+        if basename in Path(filepath).stem.lower():
+            return filepath
+    
+    return None
+
+
+def build_graph(wiki_dir: str, incremental: bool = False) -> dict:
+    """Build graph data from wiki directory."""
+    nodes = {}       # id → node data
+    edges = []       # list of edge data
+    all_files = {}   # relative_path → full content
+    
+    # Read all markdown files
+    wiki_path = Path(wiki_dir)
+    if not wiki_path.exists():
+        log(f"❌ Wiki directory not found: {wiki_dir}")
+        return {"nodes": [], "edges": []}
+    
+    md_files = sorted(wiki_path.rglob("*.md"))
+    
+    # ── Incremental check ──
+    prev_state = {}
+    current_hashes = {}
+    changed_files = set()
+    
+    if incremental:
+        prev_state = load_state()
+        log(f"📋 增量模式：加载上次状态 ({len(prev_state.get('file_hashes', {}))} 个文件记录)")
+    
+    for fpath in md_files:
+        rel_path = str(fpath.relative_to(wiki_path))
+        try:
+            content = fpath.read_text(encoding="utf-8")
+            all_files[rel_path] = content
+            
+            # 计算并记录 hash
+            fh = file_hash(str(fpath))
+            current_hashes[rel_path] = fh
+            
+            # 增量模式：检查是否有变更
+            if incremental:
+                prev_hash = prev_state.get('file_hashes', {}).get(rel_path, '')
+                if fh != prev_hash:
+                    changed_files.add(rel_path)
+        except Exception as e:
+            log(f"⚠️ Failed to read {rel_path}: {e}")
+            continue
+    
+    if incremental and len(changed_files) == 0 and len(prev_state.get('file_hashes', {})) > 0:
+        log("✅ 没有检测到文件变更，跳过重建（使用 --force 可强制重建）")
+        # 返回空结果表示无需重建（调用方可以据此跳过）
+        return {"nodes": [], "edges": [], "skipped": True, "reason": "no_changes"}
+    
+    if incremental:
+        log(f"🔄 检测到 {len(changed_files)} 个文件有变更，开始增量构建...")
+    
+    # Build nodes
+    for rel_path, content in all_files.items():
+        category = detect_category(rel_path)
+        title = extract_title(content, rel_path)
+        summary = extract_summary(content)
+        
+        color = CATEGORY_COLORS.get(category, CATEGORY_COLORS["_special"])
+        shape = CATEGORY_SHAPES.get(category, "ellipse")
+        
+        node = {
+            "id": rel_path,
+            "label": title,
+            "title": f"{title}\n({rel_path})\n\n{summary}",
+            "category": category,
+            "color": {
+                "background": color["bg"],
+                "border": color["border"],
+                "highlight": {"background": color["border"], "border": color["text"]},
+            },
+            "font": {"color": color["text"], "size": 14, "face": "Inter"},
+            "shape": shape,
+            "size": 20 if category == "_special" else 16,
+        }
+        
+        # Special nodes are slightly larger
+        if category == "_special":
+            node["size"] = 25
+            node["font"]["bold"] = True
+        
+        nodes[rel_path] = node
+    
+    # Build edges from wikilinks
+    edge_id = 0
+    seen_edges = set()  # avoid duplicates
+    
+    for rel_path, content in all_files.items():
+        links = extract_wikilinks(content)
+        for target, _pos in links:
+            resolved = resolve_link(target, all_files)
+            if resolved and resolved != rel_path:
+                edge_key = tuple(sorted([rel_path, resolved]))
+                if edge_key not in seen_edges:
+                    seen_edges.add(edge_key)
+                    edges.append({
+                        "id": edge_id,
+                        "from": rel_path,
+                        "to": resolved,
+                        "arrows": "to",
+                        "color": {"color": "#90A4AE", "opacity": 0.4},
+                        "width": 1.5,
+                        "smooth": {"type": "curvedCW", "roundness": 0.15},
+                    })
+                    edge_id += 1
+    
+    # Compute degree for sizing
+    degrees = defaultdict(int)
+    for e in edges:
+        degrees[e["from"]] += 1
+        degrees[e["to"]] += 1
+    
+    for nid, node in nodes.items():
+        deg = degrees.get(nid, 0)
+        node["value"] = max(deg + 2, 5)  # minimum size
+        node["degree"] = deg
+    
+    # Category breakdown
+    categories = {}
+    for n in nodes.values():
+        cat = n["category"]
+        categories[cat] = categories.get(cat, 0) + 1
+    
+    graph_data = {
+        "nodes": list(nodes.values()),
+        "edges": edges,
+        "stats": {
+            "total_nodes": len(nodes),
+            "total_edges": len(edges),
+            "files_processed": len(all_files),
+            "categories": categories,
+        },
+    }
+    
+    # Save state after successful build
+    save_state(current_hashes)
+    
+    return graph_data
+
+
+def generate_html(graph_data: dict, output_path: str):
+    """Generate self-contained HTML with vis.js. UTF-8 ensured."""
+    stats = graph_data["stats"]
+    nodes_json = json.dumps(graph_data["nodes"], ensure_ascii=False)
+    edges_json = json.dumps(graph_data["edges"], ensure_ascii=False)
+    
+    html = f'''<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>🧠 KnowFlow Knowledge Graph</title>
+<script src="https://unpkg.com/vis-network@9.1.6/standalone/umd/vis-network.min.js"></script>
+<style>
+  * {{ margin: 0; padding: 0; box-sizing: border-box; }}
+  body {{ font-family: -apple-system, 'Inter', 'SF Pro', 'PingFang SC', 'Microsoft YaHei', sans-serif; background: #fafafa; }}
+  
+  /* Header */
+  header {{
+    position: fixed; top: 0; left: 0; right: 0; z-index: 100;
+    background: rgba(255,255,255,0.95); backdrop-filter: blur(10px);
+    border-bottom: 1px solid #eee; padding: 12px 24px;
+    display: flex; align-items: center; justify-content: space-between;
+  }}
+  header h1 {{ font-size: 18px; font-weight: 600; color: #333; }}
+  header h1 span {{ opacity: 0.5; }}
+  
+  /* Stats bar */
+  .stats {{ display: flex; gap: 16px; font-size: 13px; color: #666; }}
+  .stat {{ display: flex; align-items: center; gap: 4px; }}
+  .stat-dot {{ width: 8px; height: 8px; border-radius: 50%; display: inline-block; }}
+  
+  /* Legend */
+  .legend {{ position: fixed; bottom: 20px; left: 20px; z-index: 100;
+    background: rgba(255,255,255,0.95); backdrop-filter: blur(10px);
+    border: 1px solid #e0e0e0; border-radius: 12px; padding: 14px 18px;
+    font-size: 12px; box-shadow: 0 2px 8px rgba(0,0,0,0.06); }}
+  .legend-title {{ font-weight: 600; margin-bottom: 10px; color: #333; }}
+  .legend-item {{ display: flex; align-items: center; gap: 8px; margin-bottom: 6px; color: #555; }}
+  .legend-icon {{ width: 14px; height: 14px; border-radius: 3px; flex-shrink: 0; }}
+  
+  /* Controls */
+  .controls {{ position: fixed; bottom: 20px; right: 20px; z-index: 100;
+    display: flex; flex-direction: column; gap: 8px; }}
+  .ctrl-btn {{
+    background: white; border: 1px solid #ddd; border-radius: 8px;
+    padding: 8px 14px; font-size: 12px; cursor: pointer; color: #555;
+    transition: all 0.15s; box-shadow: 0 1px 3px rgba(0,0,0,0.05);
+  }}
+  .ctrl-btn:hover {{ border-color: #999; color: #333; }}
+  
+  /* Search */
+  .search-box {{
+    position: fixed; top: 60px; right: 24px; z-index: 100;
+    background: rgba(255,255,255,0.95); backdrop-filter: blur(10px);
+    border: 1px solid #ddd; border-radius: 8px; padding: 8px 12px;
+    box-shadow: 0 2px 6px rgba(0,0,0,0.05);
+  }}
+  .search-box input {{ border: none; outline: none; font-size: 13px; width: 200px;
+    background: transparent; color: #333; }}
+  .search-box input::placeholder {{ color: #aaa; }}
+  
+  /* Graph container */
+  #graph-container {{ 
+    margin-top: 56px; width: 100vw; height: calc(100vh - 56px); 
+  }}
+  
+  /* Info panel */
+  #info-panel {{
+    position: fixed; top: 60px; left: 24px; z-index: 100;
+    background: rgba(255,255,255,0.95); backdrop-filter: blur(10px);
+    border: 1px solid #e0e0e0; border-radius: 12px; padding: 16px 20px;
+    max-width: 320px; max-height: 40vh; overflow-y: auto;
+    font-size: 13px; line-height: 1.6; color: #444;
+    box-shadow: 0 2px 8px rgba(0,0,0,0.06); display: none;
+  }}
+  #info-panel.visible {{ display: block; }}
+  #info-panel h3 {{ font-size: 15px; margin-bottom: 8px; color: #222; }}
+  #info-panel .meta {{ font-size: 11px; color: #888; margin-bottom: 10px; }}
+  #info-panel .links {{ margin-top: 10px; }}
+  #info-panel .link-item {{ display: inline-block; margin: 2px 4px 2px 0;
+    padding: 2px 8px; background: #f0f0f0; border-radius: 10px; font-size: 11px; color: #555; }}
+
+  @media (prefers-color-scheme: dark) {{
+    body {{ background: #1a1a2e; }}
+    header {{ background: rgba(26,26,46,0.95); border-color: #333; color: #eee; }}
+    header h1 {{ color: #eee; }}
+    .stats {{ color: #aaa; }}
+    .legend, .controls, .search-box, #info-panel {{
+      background: rgba(26,26,46,0.95); border-color: #333; color: #ccc; }}
+    .ctrl-btn {{ background: #252540; border-color: #444; color: #ccc; }}
+    .ctrl-btn:hover {{ border-color: #666; color: #fff; }}
+    .search-box input {{ color: #ccc; }}
+    .search-box input::placeholder {{ color: #666; }}
+    #info-panel {{ color: #ccc; }}
+    #info-panel h3 {{ color: #fff; }}
+    #info-panel .meta {{ color: #888; }}
+    #info-panel .link-item {{ background: #333; color: #aaa; }}
+  }}
+</style>
+</head>
+<body>
+
+<header>
+  <h1>🧠 KnowFlow <span>Knowledge Graph</span></h1>
+  <div class="stats">
+    <div class="stat"><span class="stat-dot" style="background:#1565C0"></span>{stats['total_nodes']} nodes</div>
+    <div class="stat"><span class="stat-dot" style="background:#90A4AE"></span>{stats['total_edges']} links</div>
+    <div class="stat"><span class="stat-dot" style="background:#4CAF50"></span>{stats['files_processed']} files</div>
+  </div>
+</header>
+
+<div class="search-box">
+  🔍 <input type="text" id="search-input" placeholder="Search nodes..." />
+</div>
+
+<div id="info-panel">
+  <h3 id="info-title">-</h3>
+  <div class="meta" id="info-meta">-</div>
+  <div id="info-summary">-</div>
+  <div class="links" id="info-links"></div>
+</div>
+
+<div id="graph-container"></div>
+
+<div class="legend">
+  <div class="legend-title">📂 Legend</div>
+  <div class="legend-item"><span class="legend-icon" style="background:#E3F2BD;border:2px solid #1565C0"></span> 📥 Sources</div>
+  <div class="legend-item"><span class="legend-icon" style="background:#F3E5F5;border:2px solid #7B1FA2"></span> 👤 Entities</div>
+  <div class="legend-item"><span class="legend-icon" style="background:#E8F5E9;border:2px solid #2E7D32"></span> 💡 Concepts</div>
+  <div class="legend-item"><span class="legend-icon" style="background:#FFF3E0;border:2px solid #EF6C00"></span> ⚖️ Comparisons</div>
+  <div class="legend-item"><span class="legend-icon" style="background:#FFEBEE;border:2px solid #C62828"></span> ⭐ Index / Log</div>
+</div>
+
+<div class="controls">
+  <button class="ctrl-btn" onclick="fitAll()">🎯 Fit all</button>
+  <button class="ctrl-btn" onclick="togglePhysics()">⚡ Physics</button>
+  <button class="ctrl-btn" onclick="clusterByCategory()">📁 Cluster by type</button>
+</div>
+
+<script>
+const nodes = new vis.DataSet({nodes_json});
+const edges = new vis.DataSet({edges_json});
+
+const container = document.getElementById('graph-container');
+const options = {{
+  physics: {{
+    enabled: true,
+    stabilization: {{ iterations: 150, fit: true }},
+    barnesHut: {{
+      gravitationalConstant: -2500,
+      springConstant: 0.04,
+      damping: 0.09,
+      avoidOverlap: 0.5
+    }}
+  }},
+  interaction: {{
+    hover: true,
+    tooltipDelay: 200,
+    hideEdgesOnDrag: false,
+    multiselect: true,
+  }},
+  nodes: {{
+    font: {{ multi: false }},
+    borderWidthSelected: 3,
+    shadow: {{ enabled: true, color: 'rgba(0,0,0,0.2)', size: 10 }},
+  }},
+  edges: {{
+    selectionWidth: 2,
+    smooth: {{ forceDirection: 'none' }},
+  }},
+  layout: {{
+    improvedLayout: true,
+    clusterThreshold: 150,
+  }}
+}};
+
+const network = new vis.Network(container, {{ nodes, edges }}, options);
+
+// ── Event handlers ──
+network.on("selectNode", function(params) {{
+  if (params.nodes.length === 1) {{
+    const nodeId = params.nodes[0];
+    const node = nodes.get(nodeId);
+    showInfo(node);
+  }}
+}});
+
+network.on("deselectNode", function() {{
+  document.getElementById('info-panel').classList.remove('visible');
+}});
+
+function showInfo(node) {{
+  const panel = document.getElementById('info-panel');
+  document.getElementById('info-title').textContent = node.label || nodeId;
+  document.getElementById('info-meta').textContent = node.category + ' | Degree: ' + (node.degree || 0);
+  document.getElementById('info-summary').innerHTML = (node.title || '').replace(/\\n/g, '<br>');
+  panel.classList.add('visible');
+}}
+
+// ── Controls ──
+function fitAll() {{ network.fit({{ animation: {{ duration: 500, easingFunction: 'easeInOutQuad' }} }}); }}
+let physicsEnabled = true;
+function togglePhysics() {{
+  physicsEnabled = !physicsEnabled;
+  network.setOptions({{ physics: {{ enabled: physicsEnabled }} }});
+}}
+function clusterByCategory() {{
+  network.fit({{ animation: {{ duration: 800 }} }});
+}}
+
+// ── Search ──
+document.getElementById('search-input').addEventListener('input', function(e) {{
+  const query = e.target.value.toLowerCase().trim();
+  if (!query) {{
+    nodes.update(nodes.map(n => ({{ id: n.id, hidden: false }})));
+    return;
+  }}
+  const matched = nodes.getIds().filter(id => {{
+    const n = nodes.get(id);
+    return (n.label || '').toLowerCase().includes(query) || 
+           (n.title || '').toLowerCase().includes(query) ||
+           id.toLowerCase().includes(query);
+  }});
+  const unmatched = nodes.getIds().filter(id => !matched.includes(id));
+  nodes.update(matched.map(id => ({{ id, hidden: false }})));
+  nodes.update(unmatched.map(id => ({{ id, hidden: true }})));
+  edges.update(edges.map(e => ({{
+    id: e.id, 
+    hidden: !matched.includes(e.from) || !matched.includes(e.to)
+  }})));
+}});
+
+// ── Auto-fit on load ──
+network.once("stabilizationIterationsDone", function() {{
+  network.fit({{ animation: {{ duration: 600 }} }});
+}});
+</script>
+</body>
+</html>'''
+    
+    # Ensure output directory exists
+    out_dir = os.path.dirname(output_path)
+    os.makedirs(out_dir, exist_ok=True)
+    
+    # Write with explicit UTF-8 encoding (no BOM for HTML — charset meta tag handles it)
+    with open(output_path, "w", encoding="utf-8", newline='\n') as f:
+        f.write(html)
+    
+    # Also write raw JSON data for external tool consumption
+    json_output = output_path.replace('.html', '.json')
+    with open(json_output, "w", encoding="utf-8") as f:
+        json.dump(graph_data, f, ensure_ascii=False, indent=2)
+
+
+def print_stats(stats: dict):
+    """Print detailed node statistics to console"""
+    log("")
+    log("=" * 50)
+    log("📊 知识图谱统计信息")
+    log("=" * 50)
+    log(f"  节点总数:     {stats['total_nodes']}")
+    log(f"  关系总数:     {stats['total_edges']}")
+    log(f"  处理文件数:   {stats['files_processed']}")
+    log("")
+    log("📂 分类详情:")
+    
+    cat_labels = {
+        "sources": "📥 来源 (Sources)",
+        "entities": "👤 实体 (Entities)",
+        "concepts": "💡 概念 (Concepts)",
+        "comparisons": "⚖️ 对比 (Comparisons)",
+        "_special": "⭐ 索引/日志",
+    }
+    
+    # Sort by count descending
+    sorted_cats = sorted(stats["categories"].items(), key=lambda x: -x[1])
+    for cat, count in sorted_cats:
+        label = cat_labels.get(cat, cat)
+        bar = "█" * count + "░" * max(0, stats['total_nodes'] - count)
+        log(f"  {label:30s} {count:4d}  {bar}")
+    
+    log("")
+    log("=" * 50)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="KnowFlow Knowledge Graph Builder (M2)")
+    parser.add_argument("--wiki-dir", default=DEFAULT_WIKI_DIR, help="Wiki root directory")
+    parser.add_argument("--output", default=DEFAULT_OUTPUT, help="Output HTML path")
+    parser.add_argument("--incremental", "-i", action="store_true",
+                        help="Incremental mode: skip rebuild if no files changed")
+    args = parser.parse_args()
+    
+    log("🕸️ 开始构建知识图谱...")
+    if args.incremental:
+        log("📋 增量更新模式已启用")
+    
+    start_time = time.time()
+    graph_data = build_graph(args.wiki_dir, incremental=args.incremental)
+    elapsed = time.time() - start_time
+    
+    # Check if skipped (no changes in incremental mode)
+    if graph_data.get("skipped"):
+        log(f"⏭️ 跳过构建（{graph_data['reason']}），耗时 {elapsed:.2f}s")
+    else:
+        generate_html(graph_data, args.output)
+        print_stats(graph_data["stats"])
+        log(f"✅ 完成！耗时 {elapsed:.2f}s")
+        log(f"📄 HTML 输出: {args.output}")
+        log(f"📄 JSON 数据: {args.output.replace('.html', '.json')}")