first commit

legacy_reference/workflows/actions/rag/helpers.py

@@ -0,0 +1,81 @@
+from __future__ import annotations
+
+from typing import Any
+
+def _normalize_rag_queries(params: dict[str, Any]) -> list[str]:
+    raw_queries = params.get("queries")
+    queries: list[str] = []
+    if isinstance(raw_queries, (list, tuple)):
+        for item in raw_queries:
+            text = str(item or "").strip()
+            if text:
+                queries.append(text)
+    elif isinstance(raw_queries, str) and raw_queries.strip():
+        queries.append(raw_queries.strip())
+
+    if not queries:
+        query = str(params.get("query", "")).strip()
+        if query:
+            queries.append(query)
+    return queries
+
+
+def _extract_rag_documents(payload: Any) -> list[dict[str, Any]]:
+    documents: list[dict[str, Any]] = []
+    if not isinstance(payload, dict):
+        return documents
+    raw_documents = payload.get("documents")
+    if isinstance(raw_documents, (list, tuple)):
+        for item in raw_documents:
+            if isinstance(item, dict):
+                documents.append(dict(item))
+    return documents
+
+
+def _rag_document_title(doc: dict[str, Any]) -> str:
+    return str(
+        doc.get("title_proc")
+        or doc.get("title")
+        or doc.get("chunk_texto")
+        or ""
+    ).strip()
+
+
+def _rag_document_text(doc: dict[str, Any]) -> str:
+    return str(
+        doc.get("chunk_texto")
+        or doc.get("text")
+        or doc.get("title_proc")
+        or doc.get("title")
+        or ""
+    ).strip()
+
+
+def _is_selected_rag_document(doc: dict[str, Any]) -> bool:
+    try:
+        return float(doc.get("distance", 1.0)) <= 0.6
+    except (TypeError, ValueError):
+        return False
+
+
+def _build_rag_answer_item(query: str, documents: list[dict[str, Any]]) -> str:
+    if not documents:
+        return f"{query}: Nao encontrei trechos relevantes para essa busca."
+    snippets = []
+    for doc in documents:
+        text = _rag_document_text(doc)
+        if text:
+            snippets.append(text)
+    if not snippets:
+        return f"{query}: Encontrei trecho(s), mas sem texto disponivel."
+    return f"{query}: {' '.join(snippets)}"
+
+
+__all__ = [
+    '_normalize_rag_queries',
+    '_extract_rag_documents',
+    '_rag_document_title',
+    '_rag_document_text',
+    '_is_selected_rag_document',
+    '_build_rag_answer_item',
+]