первый коммит

app/modules/rag_session/retrieval/chunker.py

@@ -0,0 +1,20 @@
+class TextChunker:
+    def __init__(self, chunk_size: int = 900, overlap: int = 120) -> None:
+        self._chunk_size = chunk_size
+        self._overlap = overlap
+
+    def chunk(self, text: str) -> list[str]:
+        cleaned = text.replace("\r\n", "\n")
+        if not cleaned.strip():
+            return []
+        chunks: list[str] = []
+        start = 0
+        while start < len(cleaned):
+            end = min(len(cleaned), start + self._chunk_size)
+            piece = cleaned[start:end].strip()
+            if piece:
+                chunks.append(piece)
+            if end == len(cleaned):
+                break
+            start = max(0, end - self._overlap)
+        return chunks

app/modules/rag_session/retrieval/scoring.py

@@ -0,0 +1,12 @@
+import math
+
+
+def cosine_similarity(a: list[float], b: list[float]) -> float:
+    if not a or not b or len(a) != len(b):
+        return -1.0
+    dot = sum(x * y for x, y in zip(a, b))
+    norm_a = math.sqrt(sum(x * x for x in a))
+    norm_b = math.sqrt(sum(y * y for y in b))
+    if norm_a == 0 or norm_b == 0:
+        return -1.0
+    return dot / (norm_a * norm_b)