From 35f85066a4a26fa2052db53dd66c552108d0cdde Mon Sep 17 00:00:00 2001
From: ClaudioDrews <claudio@drews.com.br>
Date: Wed, 3 Jun 2026 15:59:28 -0300
Subject: [PATCH] =?UTF-8?q?fix:=20semantic=5Fdedup.py=20=E2=80=94=20adicio?=
 =?UTF-8?q?nar=20using=3D'dense'=20ao=20QueryRequest?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Collection knowledge_base usa named vectors (dense + sparse). O Qdrant
exige o parâmetro 'using' em collections com múltiplos vetores nomeados.
Sem ele, query_batch_points retornava 'Not existing vector name error'
em 100% das batches (44/44). O script reportava falsos 0 near-duplicates.

Correção: 1 linha — using='dense' em models.QueryRequest (linha 176).

Validado com --dry-run --threshold 0.95 contra 8886 pontos reais:
0 erros, 1776 near-duplicate pairs detectados.
---
 scripts/semantic_dedup.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/scripts/semantic_dedup.py b/scripts/semantic_dedup.py
index 3ffcea2..9a11f6f 100755
--- a/scripts/semantic_dedup.py
+++ b/scripts/semantic_dedup.py
@@ -175,6 +175,7 @@ def find_near_duplicates(chunks: List[Dict],
             requests_list.append(
                 models.QueryRequest(
                     query=vec,
+                    using="dense",
                     limit=1,
                     score_threshold=threshold,
                     filter=models.Filter(