Initial submission of FAQ-бот: add src/utils.py

src/utils.py

@@ -0,0 +1,42 @@
+import os
+from pathlib import Path
+from langchain_ollama import OllamaEmbeddings
+from langchain_chroma import Chroma
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+
+CHROMA_PATH = Path("./chroma_faq")
+
+
+def load_faq_to_chroma(md_dir: str = "data"):
+    """Load all .md files from md_dir into a persistent Chroma store.
+    The function will create or update the store at CHROMA_PATH.
+    """
+    # Ensure directory exists
+    Path(md_dir).mkdir(parents=True, exist_ok=True)
+    # Gather all markdown files
+    md_files = list(Path(md_dir).glob("*.md"))
+    if not md_files:
+        raise FileNotFoundError(f"No .md files found in {md_dir}")
+
+    # Read and split documents
+    texts = []
+    for md_file in md_files:
+        text = md_file.read_text(encoding="utf-8")
+        splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
+        texts.extend(splitter.split_text(text))
+
+    # Create embeddings
+    embeddings = OllamaEmbeddings(model="nomic-embed-text")
+
+    # Persist to Chroma
+    chroma = Chroma(persist_directory=str(CHROMA_PATH), embedding_function=embeddings)
+    chroma.add_texts(texts)
+    chroma.persist()
+    return chroma
+
+
+def search_course_docs(query: str, k: int = 3):
+    """Search the persistent Chroma store for the top k documents matching query."""
+    chroma = Chroma(persist_directory=str(CHROMA_PATH), embedding_function=OllamaEmbeddings(model="nomic-embed-text"))
+    retriever = chroma.as_retriever(search_kwargs={"k": k})
+    return retriever.get_relevant_documents(query)