From ce208efc5a3d374a111aa81c57e644725c2fa9f6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=D0=94=D0=B0=D0=BD=D0=B8=D0=B8=D0=BB=20=D0=92=D0=B8=D0=BA?=
 =?UTF-8?q?=D1=82=D0=BE=D1=80=D0=BE=D0=B2?= <mr.face2002@gmail.com>
Date: Thu, 2 Jul 2026 08:35:11 +0000
Subject: [PATCH] =?UTF-8?q?fix:=20init=5Fdocs.py=20=E2=80=94=20=D0=90?=
 =?UTF-8?q?=D0=B3=D0=B5=D0=BD=D1=82=20=D1=81=20RAG-=D0=BF=D0=B0=D0=BC?=
 =?UTF-8?q?=D1=8F=D1=82=D1=8C=D1=8E?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 init_docs.py | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
 create mode 100644 init_docs.py

diff --git a/init_docs.py b/init_docs.py
new file mode 100644
index 0000000..fb5fd8f
--- /dev/null
+++ b/init_docs.py
@@ -0,0 +1,24 @@
+import os
+from rag_store import KnowledgeBase
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+from langchain_core.documents import Document
+
+def load_documents_from_dir(directory: str, kb: KnowledgeBase):
+    splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+    for root, _, files in os.walk(directory):
+        for file in files:
+            if file.lower().endswith(('.txt', '.md')):
+                path = os.path.join(root, file)
+                try:
+                    with open(path, 'r', encoding='utf-8') as f:
+                        text = f.read()
+                    docs = [Document(page_content=text, metadata={"title": file})]
+                    chunks = splitter.split_documents(docs)
+                    kb.add_documents(chunks)
+                    print(f"Loaded {file} into KB.")
+                except Exception as e:
+                    print(f"Failed to load {file}: {e}")
+
+if __name__ == "__main__":
+    kb = KnowledgeBase()
+    load_documents_from_dir("./data", kb)
\ No newline at end of file