test: long timeout

Author: YoungVor

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "fenic"
-version = "0.5.0"
+version = "0.5.1"
 authors = [{ name = "typedef", email = "[email protected]" }]
 description = """
 fenic is a Python DataFrame library for processing text data with APIs inspired by PySpark.

src/fenic/_inference/common_openai/openai_chat_completions_core.py

@@ -94,8 +94,9 @@ async def make_single_request(
             }
 
             max_completion_tokens = self.get_max_output_token_request_limit(request, profile_configuration)
-            if max_completion_tokens is not None:
-                common_params["max_completion_tokens"] = max_completion_tokens
+            # Don't set any limit for lineage builder
+            #if max_completion_tokens is not None:
+            #    common_params["max_completion_tokens"] = max_completion_tokens
 
             if request.temperature:
                 common_params.update({"temperature": request.temperature})

src/fenic/_inference/model_client.py

@@ -26,7 +26,7 @@
     RateLimitStrategy,
     TokenEstimate,
 )
-from fenic._inference.request_utils import DEFAULT_CLIENT_TIMEOUT
+from fenic._inference.request_utils import MAX_CLIENT_TIMEOUT
 from fenic._inference.token_counter import (
     TokenCounter,
     Tokenizable,
@@ -579,7 +579,7 @@ async def _process_single_request(self, queue_item: QueueItem[RequestT]):
         """
         try:
             try:
-                timeout = queue_item.request.request_timeout or DEFAULT_CLIENT_TIMEOUT
+                timeout = MAX_CLIENT_TIMEOUT
                 maybe_response = await asyncio.wait_for(
                     self.make_single_request(queue_item.request),
                     timeout=timeout,