feat(grpc): adding service metadata (#3278)

src/bentoml/_internal/server/grpc/servicer/v1/__init__.py

@@ -1,12 +1,13 @@
 from __future__ import annotations
 
 import sys
+import typing as t
 import asyncio
 import logging
 from typing import TYPE_CHECKING
 
 import anyio
-
+from .....utils import LazyLoader
 from ......exceptions import InvalidArgument
 from ......exceptions import BentoMLException
 from ......grpc.utils import import_grpc
@@ -20,14 +21,15 @@
     from logging import _ExcInfoType as ExcInfoType  # type: ignore (private warning)
 
     import grpc
-
+    from google.protobuf import struct_pb2
     from ......grpc.v1 import service_pb2 as pb
     from ......grpc.v1 import service_pb2_grpc as services
     from ......grpc.types import BentoServicerContext
     from .....service.service import Service
 else:
     grpc, _ = import_grpc()
     pb, services = import_generated_stubs(version="v1")
+    struct_pb2 = LazyLoader("struct_pb2", globals(), "google.protobuf.struct_pb2")
 
 
 def log_exception(request: pb.Request, exc_info: ExcInfoType) -> None:
@@ -97,4 +99,62 @@ async def Call(  # type: ignore (no async types) # pylint: disable=invalid-overr
                 )
             return response
 
+        async def ServiceMetadata(  # type: ignore (no async types) # pylint: disable=invalid-overridden-method
+            self,
+            request: pb.ServiceMetadataRequest,  # pylint: disable=unused-argument
+            context: BentoServicerContext,  # pylint: disable=unused-argument
+        ) -> pb.ServiceMetadataResponse:
+            return pb.ServiceMetadataResponse(
+                name=service.name,
+                docs=service.doc,
+                apis=[
+                    pb.ServiceMetadataResponse.InferenceAPI(
+                        name=api.name,
+                        docs=api.doc,
+                        input=make_descriptor_spec(
+                            api.input.to_spec(), pb.ServiceMetadataResponse
+                        ),
+                        output=make_descriptor_spec(
+                            api.output.to_spec(), pb.ServiceMetadataResponse
+                        ),
+                    )
+                    for api in service.apis.values()
+                ],
+            )
+
     return BentoServiceImpl()
+
+
+if TYPE_CHECKING:
+    NestedDictStrAny = dict[str, dict[str, t.Any] | t.Any]
+    TupleAny = tuple[t.Any, ...]
+
+
+def _tuple_converter(d: NestedDictStrAny | None) -> NestedDictStrAny | None:
+    # handles case for struct_pb2.Value where nested items are tuple.
+    # if that is the case, then convert to list.
+    # This dict is only one level deep, as we don't allow nested Multipart.
+    if d is not None:
+        for key, value in d.items():
+            if isinstance(value, tuple):
+                d[key] = list(t.cast("TupleAny", value))
+            elif isinstance(value, dict):
+                d[key] = _tuple_converter(t.cast("NestedDictStrAny", value))
+    return d
+
+
+def make_descriptor_spec(
+    spec: dict[str, t.Any], pb: type[pb.ServiceMetadataResponse]
+) -> pb.ServiceMetadataResponse.DescriptorMetadata:
+    from .....io_descriptors.json import parse_dict_to_proto
+
+    descriptor_id = spec.pop("id")
+    return pb.DescriptorMetadata(
+        descriptor_id=descriptor_id,
+        attributes=struct_pb2.Struct(
+            fields={
+                key: parse_dict_to_proto(_tuple_converter(value), struct_pb2.Value())
+                for key, value in spec.items()
+            }
+        ),
+    )

src/bentoml/grpc/v1/service.proto

@@ -18,6 +18,43 @@ option py_generic_services = true;
 service BentoService {
   // Call handles methodcaller of given API entrypoint.
   rpc Call(Request) returns (Response) {}
+  // ServiceMetadata returns metadata of bentoml.Service.
+  rpc ServiceMetadata(ServiceMetadataRequest) returns (ServiceMetadataResponse) {}
+}
+
+// ServiceMetadataRequest message doesn't take any arguments.
+message ServiceMetadataRequest {}
+
+// ServiceMetadataResponse returns metadata of bentoml.Service.
+// Currently it includes name, version, apis, and docs.
+message ServiceMetadataResponse {
+  // DescriptorMetadata is a metadata of any given IODescriptor.
+  message DescriptorMetadata {
+    // descriptor_id describes the given ID of the descriptor, which matches with our OpenAPI definition.
+    optional string descriptor_id = 1;
+
+    // attributes is the kwargs of the given descriptor.
+    google.protobuf.Struct attributes = 2;
+  }
+  // InferenceAPI is bentoml._internal.service.inferece_api.InferenceAPI
+  // that is exposed to gRPC client.
+  // There is no way for reflection to get information of given @svc.api.
+  message InferenceAPI {
+    // name is the name of the API.
+    string name = 1;
+    // input is the input descriptor of the API.
+    DescriptorMetadata input = 2;
+    // output is the output descriptor of the API.
+    DescriptorMetadata output = 3;
+    // docs is the optional documentation of the API.
+    optional string docs = 4;
+  }
+  // name is the service name.
+  string name = 1;
+  // apis holds a list of InferenceAPI of the service.
+  repeated InferenceAPI apis = 2;
+  // docs is the documentation of the service.
+  string docs = 3;
 }
 
 // Request message for incoming Call.