From de022639b6b40c8888c1f9ff2c312d815838c674 Mon Sep 17 00:00:00 2001 From: Yiorgis Gozadinos Date: Tue, 20 Jan 2026 09:50:40 +0200 Subject: [PATCH] Correctly lookup docs for the orb dataset --- evaluations/evaluations/benchmark.py | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/evaluations/evaluations/benchmark.py b/evaluations/evaluations/benchmark.py index b239a704..a2dcbacd 100644 --- a/evaluations/evaluations/benchmark.py +++ b/evaluations/evaluations/benchmark.py @@ -92,7 +92,13 @@ async def populate_db( progress.advance(task) continue - existing = await rag.get_document_by_uri(payload.uri) + # Use the actual URI that will be stored in the database + if payload.source_path is not None: + lookup_uri = payload.source_path.absolute().as_uri() + else: + lookup_uri = payload.uri + + existing = await rag.get_document_by_uri(lookup_uri) if existing is not None: assert existing.id chunks = await rag.chunk_repository.get_by_document_id(existing.id)