[V1] Update structured output (#16812)

Signed-off-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: reidliu41 <reid201711@gmail.com>

benchmarks/benchmark_serving_structured_output.py

@@ -150,17 +150,17 @@ def sample_requests(tokenizer: PreTrainedTokenizerBase,
 
     elif args.dataset == "grammar":
         schema = """
-            ?start: select_statement
+        root ::= select_statement
 
-            ?select_statement: "SELECT " column_list " FROM " table_name
+        select_statement ::= "SELECT " column " from " table " where " condition
 
-            ?column_list: column_name ("," column_name)*
+        column ::= "col_1 " | "col_2 "
 
-            ?table_name: identifier
+        table ::= "table_1 " | "table_2 "
 
-            ?column_name: identifier
+        condition ::= column "= " number
 
-            ?identifier: /[a-zA-Z_][a-zA-Z0-9_]*/
+        number ::= "1 " | "2 "
         """
         prompt = "Generate an SQL query to show the 'username' \
             and 'email' from the 'users' table."

docs/source/features/structured_outputs.md

@@ -50,7 +50,7 @@ completion = client.chat.completions.create(
             "content": "Generate an example email address for Alan Turing, who works in Enigma. End in .com and new line. Example result: alan.turing@enigma.com\n",
         }
     ],
-    extra_body={"guided_regex": "\w+@\w+\.com\n", "stop": ["\n"]},
+    extra_body={"guided_regex": r"\w+@\w+\.com\n", "stop": ["\n"]},
 )
 print(completion.choices[0].message.content)
 ```
@@ -105,17 +105,17 @@ It works by using a context free EBNF grammar, which for example we can use to d
 
 ```python
 simplified_sql_grammar = """
-    ?start: select_statement
+    root ::= select_statement
 
-    ?select_statement: "SELECT " column_list " FROM " table_name
+    select_statement ::= "SELECT " column " from " table " where " condition
 
-    ?column_list: column_name ("," column_name)*
+    column ::= "col_1 " | "col_2 "
 
-    ?table_name: identifier
+    table ::= "table_1 " | "table_2 "
 
-    ?column_name: identifier
+    condition ::= column "= " number
 
-    ?identifier: /[a-zA-Z_][a-zA-Z0-9_]*/
+    number ::= "1 " | "2 "
 """
 
 completion = client.chat.completions.create(

examples/online_serving/openai_chat_completion_structured_outputs.py

@@ -33,7 +33,7 @@ completion = client.chat.completions.create(
         "content": prompt,
     }],
     extra_body={
-        "guided_regex": "\w+@\w+\.com\n",
+        "guided_regex": r"\w+@\w+\.com\n",
         "stop": ["\n"]
     },
 )
@@ -70,17 +70,17 @@ print(completion.choices[0].message.content)
 
 # Guided decoding by Grammar
 simplified_sql_grammar = """
-    ?start: select_statement
+    root ::= select_statement
 
-    ?select_statement: "SELECT " column_list " FROM " table_name
+    select_statement ::= "SELECT " column " from " table " where " condition
 
-    ?column_list: column_name ("," column_name)*
+    column ::= "col_1 " | "col_2 "
 
-    ?table_name: identifier
+    table ::= "table_1 " | "table_2 "
 
-    ?column_name: identifier
+    condition ::= column "= " number
 
-    ?identifier: /[a-zA-Z_][a-zA-Z0-9_]*/
+    number ::= "1 " | "2 "
 """
 
 prompt = ("Generate an SQL query to show the 'username' and 'email'"
@@ -110,7 +110,7 @@ try:
             "content": prompt,
         }],
         extra_body={
-            "guided_regex": "\w+@\w+\.com\n",
+            "guided_regex": r"\w+@\w+\.com\n",
             "stop": ["\n"],
             "guided_decoding_backend": "xgrammar:no-fallback"
         },

examples/online_serving/openai_chat_completion_structured_outputs_with_reasoning.py

@@ -101,17 +101,17 @@ print("content: ", completion.choices[0].message.content)
 
 # Guided decoding by Grammar
 simplified_sql_grammar = """
-    ?start: select_statement
+    root ::= select_statement
 
-    ?select_statement: "SELECT " column_list " FROM " table_name
+    select_statement ::= "SELECT " column " from " table " where " condition
 
-    ?column_list: column_name ("," column_name)*
+    column ::= "col_1 " | "col_2 "
 
-    ?table_name: identifier
+    table ::= "table_1 " | "table_2 "
 
-    ?column_name: identifier
+    condition ::= column "= " number
 
-    ?identifier: /[a-zA-Z_][a-zA-Z0-9_]*/
+    number ::= "1 " | "2 "
 """
 
 # This may be very slow https://github.com/vllm-project/vllm/issues/12122