Merge pull request #108 from runpod-workers/cmpl-htf

[Bug]fix oai completion api error

src/engine.py

@@ -176,7 +176,8 @@ class OpenAIvLLMEngine(vLLMEngine):
             yield create_error_response(str(e)).model_dump()
             return
         
-        response_generator = await generator_function(request, raw_request=None)
+        dummy_request = DummyRequest()
+        response_generator = await generator_function(request, raw_request=dummy_request)
 
         if not openai_request.openai_input.get("stream") or isinstance(response_generator, ErrorResponse):
             yield response_generator.model_dump()

src/utils.py

@@ -49,9 +49,11 @@ class JobInput:
         self.openai_input = job.get("openai_input")
 
 class DummyRequest:
+    def __init__(self):
+        self.headers = {}
     async def is_disconnected(self):
         return False
-    
+
 class BatchSize:
     def __init__(self, max_batch_size, min_batch_size, batch_size_growth_factor):
         self.max_batch_size = max_batch_size