refactor: 将 MarkdownSplitter 移入 splitters 包,ingest.py 保留兼容 import

This commit is contained in:
2026-07-10 14:15:59 +08:00
parent 30c4d65cf0
commit 4dc51dccd2
5 changed files with 110 additions and 166 deletions
+22
View File
@@ -0,0 +1,22 @@
"""分块器注册表 — 按文件后缀名路由到对应 Splitter."""
SUPPORTED_SUFFIXES: dict[str, str] = {
".md": "markdown",
".txt": "text",
".html": "html",
".pdf": "pdf",
}
_registry: dict[str, type] = {}
def register_splitter(name: str, splitter_cls: type) -> None:
"""注册一个分块器."""
_registry[name] = splitter_cls
def get_splitter(name: str):
"""获取已注册的分块器类."""
if name not in _registry:
raise ValueError(f"未注册的分块器: {name}")
return _registry[name]