From d3cb88525cbd26e764b53af2e1b0357740f59a10 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E5=88=98=E8=88=AA=E5=AE=87?= <3364451258@qq.com>
Date: Sun, 5 Jul 2026 02:14:17 +0800
Subject: [PATCH] feat: multi-file, stdin, and glob support for ingest; fix
 stderr UTF-8

---
 src/cli/main.py | 49 +++++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 45 insertions(+), 4 deletions(-)

diff --git a/src/cli/main.py b/src/cli/main.py
index 6755133..fb2132a 100644
--- a/src/cli/main.py
+++ b/src/cli/main.py
@@ -10,6 +10,7 @@ import uvicorn
 
 if sys.stdout.encoding != "utf-8":
     sys.stdout = io.TextIOWrapper(sys.stdout.buffer, encoding="utf-8", errors="replace")
+    sys.stderr = io.TextIOWrapper(sys.stderr.buffer, encoding="utf-8", errors="replace")
 
 sys.path.insert(0, str(Path(__file__).parent.parent))
 
@@ -66,17 +67,54 @@ CollectionOpt = Annotated[
 # --- 命令 ---
 
 
-@app.command(help="入库单个 Markdown 文件.")
+@app.command(help="入库 Markdown 文件. 支持多文件、通配符、标准输入.")
 def ingest(
-    file_path: Annotated[str, typer.Argument(help="Markdown 文件路径")],
+    file_paths: Annotated[
+        list[str] | None,
+        typer.Argument(help="Markdown 文件路径 (可多个, 或 - 从标准输入读取)"),
+    ] = None,
+    name: Annotated[str | None, typer.Option("--name", help="标准输入模式下的虚拟文件名")] = None,
     config: ConfigOpt = DEFAULT_CONFIG_PATH,
     collection: CollectionOpt = None,
 ):
     _init_shared(config)
     from src.core.ingest import DocumentIngestor
     ingestor = DocumentIngestor(_db, _embedder, _resolve_collection(collection))
-    count = ingestor.ingest_file(file_path)
-    typer.echo(f"✅ 已入库: {file_path} → {count} chunks [{ingestor.collection_name}]")
+
+    # 标准输入模式
+    if file_paths and file_paths[0] == "-":
+        content = sys.stdin.read()
+        file_name = name or "stdin.md"
+        count = ingestor.ingest_content(content, file_name)
+        typer.echo(f"✅ 已入库: stdin → {count} chunks [{ingestor.collection_name}]")
+        return
+
+    # 多文件模式
+    if file_paths:
+        total = 0
+        for fp in file_paths:
+            # 支持通配符 (shell 展开或 Python glob)
+            from pathlib import Path as _Path
+            p = _Path(fp)
+            if "*" in fp or "?" in fp:
+                import glob as _glob
+                matches = _glob.glob(fp, recursive=True)
+                for m in matches:
+                    c = ingestor.ingest_file(m)
+                    typer.echo(f"  📄 {m}: {c} chunks")
+                    total += c
+            elif p.is_file():
+                c = ingestor.ingest_file(fp)
+                typer.echo(f"  📄 {fp}: {c} chunks")
+                total += c
+            else:
+                typer.echo(f"⚠️ 跳过 (非文件): {fp}", err=True)
+        typer.echo(f"✅ 共入库 {total} chunks [{ingestor.collection_name}]")
+        return
+
+    # 无参数 → 显示帮助
+    typer.echo("用法: md-vector-db ingest <文件1> [文件2 ...]  或  echo '内容' | md-vector-db ingest - --name doc.md", err=True)
+    raise typer.Exit(code=1)
 
 
 @app.command(help="批量入库目录下所有 .md 文件 (递归).")
@@ -89,6 +127,9 @@ def ingest_dir(
     from src.core.ingest import DocumentIngestor
     ingestor = DocumentIngestor(_db, _embedder, _resolve_collection(collection))
     results = ingestor.ingest_directory(dir_path)
+    if not results:
+        typer.echo(f"⚠️ 目录中未找到 .md 文件: {dir_path}")
+        return
     total = sum(results.values())
     for name, count in results.items():
         typer.echo(f"  📄 {name}: {count} chunks")