Support for array minItems and maxItems constraints

[Jason-CKY]

currently xgrammar does not support json schema with the following constraints:

xgrammar/cpp/json_schema_converter.cc

Lines 975 to 992 in c1b6492

	std::string JSONSchemaConverter::VisitArray(
	const picojson::object& schema, const std::string& rule_name
	) {
	XGRAMMAR_CHECK(
	(schema.count("type") && schema.at("type").get<std::string>() == "array") ||
	schema.count("items") || schema.count("prefixItems") || schema.count("unevaluatedItems")
	);
	WarnUnsupportedKeywords(
	schema,
	{
	"uniqueItems",
	"contains",
	"minContains",
	"maxContains",
	"minItems",
	"maxItems",
	}
	);

relevant PR from vllm

reproducing the issue

with vllm/vllm-openai:v0.6.6.post1 docker image:

from openai import OpenAI
from pydantic import BaseModel

client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="-",
)

class Person(BaseModel):
  names: list[str]: Field(..., min_length=2, max_length=2)

print(Person.model_json_schema())  # {'properties': {'names': {'items': {'type': 'string'}, 'maxItems': 2, 'minItems': 2, 'title': 'Names', 'type': 'array'}}, 'required': ['names'], 'title': 'Person', 'type': 'object'}

response = client.chat.completions.create(
  model='aya-23-35b',
  messages=[
    {
      'role': 'user',
      'content': 'Generate 4 names. Respond in json format.'
    }
  ],
  extra_body={"guided_json": Person.model_json_schema()}
)
print(response.choices[0].message.content)

• 2 names should be generated due to JSON constraint, but 4 names is still generated

[Ubospica]

Hi @Jason-CKY, thanks for raising the issue about supporting JSON Schema! I apologize for the late response—I was occupied with some paper-related work.

The minItems and maxItems are indeed not supported yet. We have plans to enhance the JSON Schema converter soon, aiming to cover most features in JSON Schema, including these. Please stay tuned for that!