minor fix

tybalex · tybalex · commit 0a91cc3c2b8a · 2024-01-09T20:12:28.000-08:00
diff --git a/llama_cpp/server/app.py b/llama_cpp/server/app.py
@@ -49,9 +49,10 @@
 
 _server_settings: Optional[ServerSettings] = None
 
-DEFAULT_GRAMMAR_FILE = os.getenv("DEFAULT_GRAMMAR_FILE", "./json_grammar.gbnf")
-with open(DEFAULT_GRAMMAR_FILE) as f:
-    default_grammar_text = f.read()
+DEFAULT_GRAMMAR_FILE = os.getenv("GRAMMAR_FILE", "")
+if DEFAULT_GRAMMAR_FILE != "":
+    with open(DEFAULT_GRAMMAR_FILE) as f:
+        default_grammar_text = f.read()
 
 def set_server_settings(server_settings: ServerSettings):
     global _server_settings
@@ -308,7 +309,8 @@ async def create_chat_completion(
     if body.grammar is not None:
         kwargs["grammar"] = llama_cpp.LlamaGrammar.from_string(body.grammar)
     else:
-        kwargs["grammar"] = llama_cpp.LlamaGrammar.from_string(default_grammar_text)
+        if DEFAULT_GRAMMAR_FILE:
+            kwargs["grammar"] = llama_cpp.LlamaGrammar.from_string(default_grammar_text)
     iterator_or_completion: Union[
         llama_cpp.ChatCompletion, Iterator[llama_cpp.ChatCompletionChunk]
     ] = await run_in_threadpool(llama.create_chat_completion, **kwargs)
diff --git a/vendor/llama.cpp b/vendor/llama.cpp
@@ -1 +1 @@
-Subproject commit 6efb8eb30e7025b168f3fda3ff83b9b386428ad6
+Subproject commit 1fc2f265ff9377a37fd2c61eae9cd813a3491bea
