Map DateTime to google.protobuf.Timestamp and document temporal string formats

elasticgraph-proto_ingestion/README.md

@@ -94,27 +94,50 @@ ElasticGraph.define_schema do |schema|
 end
 ```
 
+A custom scalar can also map to an externally defined proto type by passing `import:` with the
+proto file that defines it, and `comment:` documents the expected format on each generated field
+(useful when the proto type is wider than the ElasticGraph type):
+
+```ruby
+# in config/schema/phone_number.rb
+
+ElasticGraph.define_schema do |schema|
+  schema.scalar_type "PhoneNumber" do |t|
+    t.mapping type: "keyword"
+    t.json_schema type: "string"
+    t.protobuf type: "string", comment: "E.164 phone number"
+  end
+end
+```
+
 ## Type Mappings
 
 The generated `schema.proto` uses these built-in scalar mappings:
 
-| ElasticGraph Type | Protobuf Type |
-|-------------------|------------|
-| `Boolean`         | `bool`     |
-| `Cursor`          | `string`   |
-| `Date`            | `string`   |
-| `DateTime`        | `string`   |
-| `Float`           | `double`   |
-| `ID`              | `string`   |
-| `Int`             | `int32`    |
-| `JsonSafeLong`    | `int64`    |
-| `LocalTime`       | `string`   |
-| `LongString`      | `int64`    |
-| `String`          | `string`   |
-| `TimeZone`        | `string`   |
-| `Untyped`         | `string`   |
+| ElasticGraph Type | Protobuf Type               |
+|-------------------|-----------------------------|
+| `Boolean`         | `bool`                      |
+| `Cursor`          | `string`                    |
+| `Date`            | `string`                    |
+| `DateTime`        | `google.protobuf.Timestamp` |
+| `Float`           | `double`                    |
+| `ID`              | `string`                    |
+| `Int`             | `int32`                     |
+| `JsonSafeLong`    | `int64`                     |
+| `LocalTime`       | `string`                    |
+| `LongString`      | `int64`                     |
+| `String`          | `string`                    |
+| `TimeZone`        | `string`                    |
+| `Untyped`         | `string`                    |
 
 Additionally:
+- `DateTime` uses the [well-known `Timestamp` type](https://protobuf.dev/reference/protobuf/google.protobuf/#timestamp);
+  `schema.proto` imports `google/protobuf/timestamp.proto` automatically. Note that a `Timestamp`
+  is a UTC instant, so a publisher's original UTC offset is not preserved.
+- `string`-typed temporal scalars (`Date`, `LocalTime`, `TimeZone`) are wider than the
+  ElasticGraph types they carry, so generated fields of these types document the expected format
+  in a comment (e.g. `// ISO 8601 date, e.g. "2024-11-25"`). Values are validated when events
+  are ingested, just as with JSON ingestion.
 - List types become `repeated` fields.
 - Lists of lists (e.g. `[[Float!]!]!`) are not supported because Protocol Buffers cannot represent
   them directly. Schema artifact generation raises an error identifying the unsupported field.

elasticgraph-proto_ingestion/lib/elastic_graph/proto_ingestion/schema_definition/schema.rb

@@ -45,6 +45,7 @@ def to_proto
           sections = [
             %(syntax = "proto3";),
             "package #{@package_name};",
+            *render_imports(types),
             render_definitions(types)
           ]
 
@@ -130,6 +131,14 @@ def render_definitions(types)
             .join("\n\n")
         end
 
+        def render_imports(types)
+          imports = types.filter_map do |type|
+            type.protobuf_import if type.respond_to?(:protobuf_import)
+          end.uniq.sort
+
+          imports.empty? ? [] : [imports.map { |import| %(import "#{import}";) }.join("\n")]
+        end
+
         def validate_unique_enum_value_prefixes(types)
           enum_type_by_prefix = {} # : ::Hash[::String, untyped]
 

elasticgraph-proto_ingestion/lib/elastic_graph/proto_ingestion/schema_definition/schema_elements/object_interface_and_union_extension.rb

@@ -75,7 +75,7 @@ def render_proto_message(schema, message_name, package_name)
             active_field_names = fields.map { |schema_field, _| schema_field.name }
             documentation = ProtoDocumentation.comment_lines_for(doc_comment).map { |line| "#{line}\n" }.join
             field_definitions = fields.map do |schema_field, field|
-              repeated, field_type = proto_field_type_for(
+              repeated, field_type, type_comment = proto_field_type_for(
                 field.type,
                 package_name: package_name,
                 context_field_name: field.name
@@ -87,6 +87,7 @@ def render_proto_message(schema, message_name, package_name)
               )
               label = "repeated " if repeated
               line = "  #{label}#{field_type} #{schema_field.name} = #{field_number};"
+              line += " // #{type_comment}" if type_comment
               field_documentation = ProtoDocumentation
                 .comment_lines_for(schema_field.doc_comment, indent: "  ")
                 .map { |comment_line| "#{comment_line}\n" }
@@ -160,7 +161,8 @@ def proto_field_type_for(type_ref, package_name:, context_field_name:)
             end
 
             proto_type = _ = base_type_ref.resolved
-            [list_depth == 1, proto_type.proto_type_reference(package_name)]
+            type_comment = (ScalarTypeExtension === proto_type) ? proto_type.protobuf_comment : nil
+            [list_depth == 1, proto_type.proto_type_reference(package_name), type_comment]
           end
         end
       end

elasticgraph-proto_ingestion/lib/elastic_graph/proto_ingestion/schema_definition/schema_elements/scalar_type_extension.rb

@@ -14,33 +14,45 @@ module SchemaDefinition
       module SchemaElements
         # Extends ScalarType with proto field type conversion.
         module ScalarTypeExtension
-          # Default protobuf types applied to ElasticGraph's built-in scalar types as they are constructed.
-          BUILT_IN_SCALAR_PROTO_TYPES_BY_NAME = {
-            "Boolean" => "bool",
-            "Cursor" => "string",
-            "Date" => "string",
-            "DateTime" => "string",
-            "Float" => "double",
-            "ID" => "string",
-            "Int" => "int32",
-            "JsonSafeLong" => "int64",
-            "LocalTime" => "string",
-            "LongString" => "int64",
-            "String" => "string",
-            "TimeZone" => "string",
-            "Untyped" => "string"
+          # Default protobuf options applied to ElasticGraph's built-in scalar types as they are constructed.
+          BUILT_IN_SCALAR_PROTO_OPTIONS_BY_NAME = {
+            "Boolean" => {type: "bool"},
+            "Cursor" => {type: "string"},
+            "Date" => {type: "string", comment: %(ISO 8601 date, e.g. "2024-11-25")},
+            "DateTime" => {type: "google.protobuf.Timestamp", import: "google/protobuf/timestamp.proto"},
+            "Float" => {type: "double"},
+            "ID" => {type: "string"},
+            "Int" => {type: "int32"},
+            "JsonSafeLong" => {type: "int64"},
+            "LocalTime" => {type: "string", comment: %(ISO 8601 local time, e.g. "14:23:12")},
+            "LongString" => {type: "int64"},
+            "String" => {type: "string"},
+            "TimeZone" => {type: "string", comment: %(IANA time zone identifier, e.g. "America/Los_Angeles")},
+            "Untyped" => {type: "string"}
           }.freeze
 
           # Configured protobuf type (e.g. string, int64, bool).
           # @dynamic protobuf_type
           attr_reader :protobuf_type
 
+          # Proto file to import for the configured protobuf type, if it is externally defined.
+          # @dynamic protobuf_import
+          attr_reader :protobuf_import
+
+          # Comment rendered on generated proto fields of this scalar type.
+          # @dynamic protobuf_comment
+          attr_reader :protobuf_comment
+
           # Configures the protobuf type for this scalar type.
           #
-          # @param type [String] protobuf scalar type name
+          # @param type [String] protobuf type name
+          # @param import [String, nil] proto file to import for an externally defined type
+          # @param comment [String, nil] comment rendered on generated fields of this type
           # @return [void]
-          def protobuf(type:)
+          def protobuf(type:, import: nil, comment: nil)
             @protobuf_type = type
+            @protobuf_import = import
+            @protobuf_comment = comment
           end
 
           # Applies any built-in protobuf type, yields for further configuration, and validates the result.
@@ -50,8 +62,12 @@ def protobuf(type:)
           # @raise [Errors::SchemaError] when a protobuf type is missing
           def initialize_proto_extension
             original_name = type_ref.with_reverted_override.name
-            if (proto_type = BUILT_IN_SCALAR_PROTO_TYPES_BY_NAME[original_name])
-              protobuf type: proto_type
+            if (proto_options = BUILT_IN_SCALAR_PROTO_OPTIONS_BY_NAME[original_name])
+              protobuf(
+                type: proto_options.fetch(:type),
+                import: proto_options[:import],
+                comment: proto_options[:comment]
+              )
             end
 
             yield

elasticgraph-proto_ingestion/sig/elastic_graph/proto_ingestion/schema_definition/schema.rbs

@@ -32,6 +32,7 @@ module ElasticGraph
 
         def proto_types: () -> ::Array[untyped]
         def render_definitions: (::Array[untyped] types) -> ::String
+        def render_imports: (::Array[untyped] types) -> ::Array[::String]
         def validate_unique_enum_value_prefixes: (::Array[untyped] types) -> void
         def previous_field_names_for: (::String, ::String) -> ::Array[::String]
         def previous_field_names_by_type_name_and_field_name: () -> ::Hash[::String, ::Hash[::String, ::Array[::String]]]

elasticgraph-proto_ingestion/sig/elastic_graph/proto_ingestion/schema_definition/schema_elements/object_interface_and_union_extension.rbs

@@ -28,7 +28,7 @@ module ElasticGraph
             ::ElasticGraph::SchemaDefinition::SchemaElements::TypeReference,
             package_name: ::String,
             context_field_name: ::String
-          ) -> [bool, ::String]
+          ) -> [bool, ::String, ::String?]
         end
       end
     end

elasticgraph-proto_ingestion/sig/elastic_graph/proto_ingestion/schema_definition/schema_elements/scalar_type_extension.rbs

@@ -3,11 +3,13 @@ module ElasticGraph
     module SchemaDefinition
       module SchemaElements
         module ScalarTypeExtension: ::ElasticGraph::SchemaDefinition::SchemaElements::ScalarType
-          BUILT_IN_SCALAR_PROTO_TYPES_BY_NAME: ::Hash[::String, ::String]
+          BUILT_IN_SCALAR_PROTO_OPTIONS_BY_NAME: ::Hash[::String, ::Hash[::Symbol, ::String]]
 
           attr_reader protobuf_type: ::String?
+          attr_reader protobuf_import: ::String?
+          attr_reader protobuf_comment: ::String?
 
-          def protobuf: (type: ::String) -> void
+          def protobuf: (type: ::String, ?import: ::String?, ?comment: ::String?) -> void
           def initialize_proto_extension: () { () -> void } -> void
           def proto_name: () -> ::String
           def proto_type_reference: (::String package_name) -> ::String

elasticgraph-proto_ingestion/spec/unit/elastic_graph/proto_ingestion/schema_definition/api_extension_spec.rb

@@ -32,20 +32,22 @@ module SchemaDefinition
         end
 
         it "maps every built-in scalar to a proto field type" do
+          built_in_scalar_options = SchemaElements::ScalarTypeExtension::BUILT_IN_SCALAR_PROTO_OPTIONS_BY_NAME
           field_types = []
           proto = define_proto_schema do |s|
             s.object_type "Widget" do |t|
-              SchemaElements::ScalarTypeExtension::BUILT_IN_SCALAR_PROTO_TYPES_BY_NAME.each_key do |type_name|
+              built_in_scalar_options.each_key do |type_name|
                 t.field type_name.downcase, type_name
               end
               field_types = t.graphql_fields_by_name.values.map { |field| field.type.name }
               t.index "widgets"
             end
           end
 
-          expect(field_types).to match_array(SchemaElements::ScalarTypeExtension::BUILT_IN_SCALAR_PROTO_TYPES_BY_NAME.keys)
-          SchemaElements::ScalarTypeExtension::BUILT_IN_SCALAR_PROTO_TYPES_BY_NAME.each.with_index(1) do |(type_name, proto_type), field_number|
-            expect(proto).to include("#{proto_type} #{type_name.downcase} = #{field_number};")
+          expect(field_types).to match_array(built_in_scalar_options.keys)
+          expect(proto).to include('import "google/protobuf/timestamp.proto";')
+          built_in_scalar_options.each.with_index(1) do |(type_name, options), field_number|
+            expect(proto).to include("#{options.fetch(:type)} #{type_name.downcase} = #{field_number};")
           end
         end
 

elasticgraph-proto_ingestion/spec/unit/elastic_graph/proto_ingestion/schema_definition/schema_spec.rb

@@ -249,6 +249,96 @@ module SchemaDefinition
           expect(proto_type_def_from(proto, "Event")).to include("int64 occurred_at = 2;")
         end
 
+        it "maps `DateTime` fields to `google.protobuf.Timestamp`, importing its proto file once" do
+          proto = define_proto_schema do |s|
+            s.object_type "Event" do |t|
+              t.field "id", "ID"
+              t.field "created_at", "DateTime"
+              t.field "updated_at", "DateTime"
+              t.index "events"
+            end
+          end
+
+          expect(proto).to eq(<<~PROTO)
+            syntax = "proto3";
+
+            package elasticgraph;
+
+            import "google/protobuf/timestamp.proto";
+
+            message Event {
+              string id = 1;
+              google.protobuf.Timestamp created_at = 2;
+              google.protobuf.Timestamp updated_at = 3;
+              // Next field number: 4
+            }
+          PROTO
+        end
+
+        it "sorts and de-duplicates imports shared by distinct protobuf types" do
+          proto = define_proto_schema do |s|
+            s.scalar_type "FirstZType" do |t|
+              t.mapping type: "keyword"
+              t.protobuf type: "example.FirstZType", import: "z/types.proto"
+            end
+
+            s.scalar_type "SecondZType" do |t|
+              t.mapping type: "keyword"
+              t.protobuf type: "example.SecondZType", import: "z/types.proto"
+            end
+
+            s.scalar_type "AType" do |t|
+              t.mapping type: "keyword"
+              t.protobuf type: "example.AType", import: "a/types.proto"
+            end
+
+            s.object_type "Event" do |t|
+              t.field "id", "ID"
+              t.field "first_z", "FirstZType"
+              t.field "second_z", "SecondZType"
+              t.field "a", "AType"
+              t.index "events"
+            end
+          end
+
+          expect(proto.lines.grep(/\Aimport /).map(&:chomp)).to eq([
+            %(import "a/types.proto";),
+            %(import "z/types.proto";)
+          ])
+        end
+
+        it "renders format comments on fields whose scalar type documents one" do
+          proto = define_proto_schema do |s|
+            s.object_type "Person" do |t|
+              t.field "id", "ID"
+              t.field "birth_date", "Date"
+              t.index "people"
+            end
+          end
+
+          expect(proto).to include(
+            %(string birth_date = 2; // ISO 8601 date, e.g. "2024-11-25")
+          )
+        end
+
+        it "supports `import:` and `comment:` on custom scalar types" do
+          proto = define_proto_schema do |s|
+            s.scalar_type "Money" do |t|
+              t.mapping type: "keyword"
+              t.protobuf type: "myapp.types.Money", import: "myapp/types/money.proto", comment: "amount + currency"
+            end
+
+            s.object_type "Order" do |t|
+              t.field "id", "ID"
+              t.field "total", "Money"
+              t.index "orders"
+            end
+          end
+
+          expect(proto).to include('import "myapp/types/money.proto";')
+          expect(proto).to include("myapp.types.Money total = 2; // amount + currency")
+        end
+
         it "assigns a new field the stored `next_number` rather than filling an earlier gap" do
           # A cursor with gaps below it can only arise from a hand-edited artifact, so this test
           # must seed raw mappings instead of results from a prior dump.