fix(xinference): support string reranker documents

apps/models_provider/impl/xinference_model_provider/model/reranker.py

@@ -27,6 +27,13 @@ def new_instance(model_type, model_name, model_credential: Dict[str, object], **
         return XInferenceReranker(server_url=model_credential.get('server_url'), model_uid=model_name,
                                   api_key=model_credential.get('api_key'), top_n=model_kwargs.get('top_n', 3))
 
+    @staticmethod
+    def _get_document_text(result: Dict[str, Any]) -> str:
+        document = result.get('document')
+        if isinstance(document, dict):
+            return document.get('text', '')
+        return document or ''
+
     top_n: Optional[int] = 3
 
     def compress_documents(self, documents: Sequence[Document], query: str, callbacks: Optional[Callbacks] = None) -> \
@@ -50,5 +57,5 @@ def compress_documents(self, documents: Sequence[Document], query: str, callback
         client = RESTfulClient(self.server_url, self.api_key)
         model: RESTfulRerankModelHandle = client.get_model(self.model_uid)
         res = model.rerank([document.page_content for document in documents], query, self.top_n, return_documents=True)
-        return [Document(page_content=d.get('document', {}).get('text'),
+        return [Document(page_content=self._get_document_text(d),
                          metadata={'relevance_score': d.get('relevance_score')}) for d in res.get('results', [])]

apps/models_provider/tests.py

@@ -2,8 +2,10 @@
 from unittest.mock import patch
 
 from django.test import SimpleTestCase
+from langchain_core.documents import Document
 
 from models_provider.impl.vllm_model_provider.model.whisper_sst import VllmWhisperSpeechToText
+from models_provider.impl.xinference_model_provider.model.reranker import XInferenceReranker
 
 
 class VllmWhisperSpeechToTextTest(SimpleTestCase):
@@ -24,3 +26,27 @@ def test_normalizes_trailing_slash_in_v1_base_url(self, openai_mock):
             base_url='https://vllm.example/v1',
         )
         self.assertEqual(result, 'transcript')
+
+
+class XInferenceRerankerTest(SimpleTestCase):
+    @patch('xinference_client.RESTfulClient')
+    def test_compress_documents_accepts_string_documents(self, client_mock):
+        client_mock.return_value.get_model.return_value.rerank.return_value = {
+            'results': [{'document': 'current result', 'relevance_score': 0.9}]
+        }
+        model = XInferenceReranker(server_url='http://localhost', model_uid='reranker', api_key=None)
+
+        result = model.compress_documents([Document(page_content='query text')], 'query')
+
+        self.assertEqual(result[0].page_content, 'current result')
+        self.assertEqual(result[0].metadata['relevance_score'], 0.9)
+
+    def test_extracts_text_from_legacy_document_object(self):
+        result = {'document': {'text': 'legacy result'}}
+
+        self.assertEqual(XInferenceReranker._get_document_text(result), 'legacy result')
+
+    def test_extracts_text_from_current_document_string(self):
+        result = {'document': 'current result'}
+
+        self.assertEqual(XInferenceReranker._get_document_text(result), 'current result')