feat: ingest.py 使用 registry 自动选择 Splitter，ingest_directory 支持多格式

src/core/ingest.py

@@ -7,6 +7,7 @@ from src.core.db import VectorDB
 from src.core.embedder import Embedder, batch_embed
 from src.core.splitters.markdown import MarkdownSplitter  # 兼容旧 import 路径
 from src.core.splitters.base import Splitter  # 兼容旧 import 路径
+from src.core.splitters.registry import SUPPORTED_SUFFIXES, get_splitter
 
 logger = logging.getLogger("md-vector-db")
 
@@ -31,35 +32,44 @@ class DocumentIngestor:
         return self.db.get_or_create_collection(self.collection_name)
 
     def ingest_file(self, file_path: str) -> int:
-        """入库单个 Markdown 文件, 返回 chunk 数量.
+        """入库单个文件, 返回 chunk 数量.
 
-        使用文件路径的 SHA256 前 12 位 + 文件名作为唯一标识，
-        避免不同目录下同名文件冲突.
+        根据文件扩展名自动选择 Splitter（.md→MarkdownSplitter, .txt→TextSplitter, .pdf→PDFSplitter 等）。
+        使用文件路径的 SHA256 前 12 位 + 文件名作为唯一标识。
         """
         import hashlib
         path = Path(file_path).resolve()
-        content = path.read_text(encoding="utf-8")
-        # 用路径 hash 保证同名文件在不同目录下不冲突
         path_hash = hashlib.sha256(str(path).encode()).hexdigest()[:12]
         file_name = f"{path_hash}_{path.name}"
 
-        return self.ingest_content(content, file_name)
+        splitter = self.splitter or get_splitter(
+            file_path, max_size=1000, overlap=100
+        )
 
-    def ingest_content(self, content: str, file_name: str) -> int:
-        """入库 Markdown 内容（无需实际文件）."""
-        # 去重：先删旧 chunks
+        # PDF 文件特殊处理：splitter 内部读取二进制内容
+        suffix = path.suffix.lower()
+        if suffix in (".pdf",):
+            chunks = splitter.split(str(path), source_file=file_name)
+            return self._add_chunks(chunks, file_name)
+
+        content = path.read_text(encoding="utf-8")
+        return self._ingest_with_splitter(content, file_name, splitter)
+
+    def _ingest_with_splitter(self, content: str, file_name: str, splitter) -> int:
+        """分块 + 嵌入 + 入库（文本文件通用路径）."""
         self._remove_by_source(file_name)
 
-        # 分块
-        chunks = self.splitter.split(content, source_file=file_name)
+        chunks = splitter.split(content, source_file=file_name)
+        return self._add_chunks(chunks, file_name)
+
+    def _add_chunks(self, chunks: list[dict], file_name: str) -> int:
+        """嵌入 + 写入 ChromaDB（分块已完成）."""
         if not chunks:
             return 0
 
-        # 分批嵌入 (避免大文档 OOM)
         texts = [c["content"] for c in chunks]
         embeddings = batch_embed(self.embedder, texts)
 
-        # 入库
         ids = [f"{file_name}_{i}" for i in range(len(chunks))]
         metadatas = [
             {
@@ -81,12 +91,18 @@ class DocumentIngestor:
 
         return len(chunks)
 
+    def ingest_content(self, content: str, file_name: str) -> int:
+        """入库 Markdown 内容（无需实际文件）。若未指定 splitter，默认用 MarkdownSplitter."""
+        splitter = self.splitter or MarkdownSplitter()
+        return self._ingest_with_splitter(content, file_name, splitter)
+
     def ingest_directory(self, dir_path: str) -> dict[str, int]:
-        """入库目录下所有 Markdown 文件."""
+        """入库目录下所有支持的文档格式."""
         results = {}
-        for md_file in Path(dir_path).rglob("*.md"):
-            count = self.ingest_file(str(md_file))
-            results[md_file.name] = count
+        for f in Path(dir_path).rglob("*"):
+            if f.suffix.lower() in SUPPORTED_SUFFIXES:
+                count = self.ingest_file(str(f))
+                results[f.name] = count
         return results
 
     def _remove_by_source(self, file_name: str) -> None: