Merge pull request #105 from heetch/logical-type/duration

README.md

@@ -38,6 +38,7 @@ When the `avrogo` command generates Go datatypes from Avro schemas, it uses the
   as `time.Time` type
 - `{"type": "string", "logicalType": "uuid"}` is represented as
   [github.com/google/uuid.UUID](https://pkg.go.dev/github.com/google/uuid#UUID) type.
+- `{"type": "long", "name": "duration-nanos"}` is represented as `time.Duration` type.
 
 If a definition has a `go.package` annotation the type from that package will be used instead of generating a Go type. The type must be compatible with the Avro schema (it may contain extra fields, but all fields in common must be compatible).
 

analyze.go

@@ -16,9 +16,10 @@ import (
 )
 
 var (
-	timeType = reflect.TypeOf(time.Time{})
-	byteType = reflect.TypeOf(byte(0))
-	uuidType = reflect.TypeOf(gouuid.UUID{})
+	timeType     = reflect.TypeOf(time.Time{})
+	durationType = reflect.TypeOf(time.Duration(0))
+	byteType     = reflect.TypeOf(byte(0))
+	uuidType     = reflect.TypeOf(gouuid.UUID{})
 )
 
 type decodeProgram struct {
@@ -492,7 +493,7 @@ func canAssignVMType(operand int, dstType reflect.Type) bool {
 	case vm.Boolean:
 		return dstKind == reflect.Bool
 	case vm.Int, vm.Long:
-		return dstType == timeType || reflect.Int <= dstKind && dstKind <= reflect.Int64
+		return dstType == timeType || dstType == durationType || reflect.Int <= dstKind && dstKind <= reflect.Int64
 	case vm.Float, vm.Double:
 		return dstKind == reflect.Float64 || dstKind == reflect.Float32
 	case vm.Bytes:

cmd/avrogo/generate.go

@@ -15,6 +15,7 @@ import (
 )
 
 const (
+	durationNanos   = "duration-nanos"
 	timestampMicros = "timestamp-micros"
 	timestampMillis = "timestamp-millis"
 	uuid            = "uuid"
@@ -476,11 +477,15 @@ func (gc *generateContext) GoTypeOf(t schema.AvroType) typeInfo {
 		// Note: Go int is at least 32 bits.
 		info.GoType = "int"
 	case *schema.LongField:
-		// TODO support timestampMillis. https://github.com/heetch/avro/issues/3
-		if logicalType(t) == timestampMicros {
+		switch logicalType(t) {
+		case timestampMicros:
+			// TODO support timestampMillis. https://github.com/heetch/avro/issues/3
 			info.GoType = "time.Time"
 			gc.addImport("time")
-		} else {
+		case durationNanos:
+			info.GoType = "time.Duration"
+			gc.addImport("time")
+		default:
 			info.GoType = "int64"
 		}
 	case *schema.FloatField:

cmd/avrogo/testdata/logicaltype.cue

@@ -51,3 +51,20 @@ tests: invalidUUID: {
        outData: null
        expectError: unmarshal: "invalid UUID in Avro encoding: invalid UUID length: 12"
 }
+
+tests: durationNanos: {
+	inSchema: {
+		type: "record"
+		name: "R"
+		fields: [{
+			name: "D"
+			type: {
+				type:        "long"
+				logicalType: "duration-nanos"
+			}
+		}]
+	}
+	outSchema: inSchema
+	inData: D: 15000000000
+	outData: inData
+}

cue.mod/pkg/github.com/heetch/cue-schema/avro/schema.cue

@@ -91,7 +91,7 @@ LogicalType :: {
 	logicalType: string
 }
 
-LogicalType :: DecimalBytes | DecimalFixed | UUID | Date | *TimeMillis | *TimeMicros | TimestampMillis | TimestampMicros
+LogicalType :: DecimalBytes | DecimalFixed | UUID | Date | *TimeMillis | *TimeMicros | TimestampMillis | TimestampMicros | DurationNanos
 
 DecimalBytes :: {
 	type:        "bytes"
@@ -139,3 +139,8 @@ TimestampMicros :: {
 	type:        "long"
 	logicalType: "timestamp-micros"
 }
+
+DurationNanos :: {
+	type:        "long"
+	logicalType: "duration-nanos"
+}

decode.go

@@ -149,12 +149,16 @@ func (d *decoder) eval(target reflect.Value) {
 				// need more information from the VM to be able to
 				// do that, so support only timestamp-micros for now.
 				// See https://github.com/heetch/avro/issues/3
-				if target.Type() == timeType {
+				switch target.Type() {
+				case timeType:
 					// timestamp-micros
 					target.Set(reflect.ValueOf(time.Unix(frame.Int/1e6, frame.Int%1e6*1e3)))
-					break
+				case durationType:
+					// duration-nanos
+					target.Set(reflect.ValueOf(time.Duration(frame.Int)))
+				default:
+					target.SetInt(frame.Int)
 				}
-				target.SetInt(frame.Int)
 			case vm.Int:
 				target.SetInt(frame.Int)
 			case vm.Float, vm.Double:

encode.go

@@ -219,15 +219,23 @@ func (b *encoderBuilder) typeEncoder(at schema.AvroType, t reflect.Type, info ty
 	case *schema.NullField:
 		return nullEncoder
 	case *schema.LongField:
-		if t == timeType {
+		switch t {
+		case timeType:
 			if lt := logicalType(at); lt == timestampMicros {
 				return timestampMicrosEncoder
 			} else {
 				// TODO timestamp-millis support.
 				return errorEncoder(fmt.Errorf("cannot encode time.Time as long with logical type %q", lt))
 			}
+		case durationType:
+			if lt := logicalType(at); lt == durationNanos {
+				return durationNanosEncoder
+			} else {
+				return errorEncoder(fmt.Errorf("cannot encode %t as long with logical type %q", t, lt))
+			}
+		default:
+			return longEncoder
 		}
-		return longEncoder
 	case *schema.StringField:
 		if t == uuidType {
 			if lt := logicalType(at); lt == uuid {
@@ -277,6 +285,11 @@ func uuidEncoder(e *encodeState, v reflect.Value) {
 	}
 }
 
+func durationNanosEncoder(e *encodeState, v reflect.Value) {
+	d := v.Interface().(time.Duration)
+	e.writeLong(d.Nanoseconds())
+}
+
 type fixedEncoder struct {
 	size int
 }

gotype.go

@@ -13,6 +13,7 @@ import (
 )
 
 const (
+	durationNanos   = "duration-nanos"
 	timestampMicros = "timestamp-micros"
 	timestampMillis = "timestamp-millis"
 	uuid            = "uuid"
@@ -45,6 +46,7 @@ type errorSchema struct {
 //	- float64 encodes as "double"
 //	- string encodes as "string"
 //	- Null{} encodes as "null"
+//	- time.Duration encodes as {"type": "long", "logicalType": "duration-nanos"}
 //	- time.Time encodes as {"type": "long", "logicalType": "timestamp-micros"}
 //	- github.com/google/uuid.UUID encodes as {"type": "string", "logicalType": "string"}
 //	- [N]byte encodes as {"type": "fixed", "name": "go.FixedN", "size": N}
@@ -123,7 +125,7 @@ func avroTypeOfUncached(names *Names, t reflect.Type) (*Type, error) {
 
 type goTypeDef struct {
 	// name holds the Avro name for the Go type.
-	name   string
+	name string
 	// schema holds the JSON-marshalable schema for the type.
 	schema interface{}
 }
@@ -134,7 +136,7 @@ type goTypeSchema struct {
 	names *Names
 	// defs maps from Go type to Avro definition for all
 	// types being traversed by schemaForGoType..
-	defs  map[reflect.Type]goTypeDef
+	defs map[reflect.Type]goTypeDef
 }
 
 func (gts *goTypeSchema) schemaForGoType(t reflect.Type) (interface{}, error) {
@@ -164,6 +166,12 @@ func (gts *goTypeSchema) schemaForGoType(t reflect.Type) (interface{}, error) {
 	case reflect.String:
 		return "string", nil
 	case reflect.Int, reflect.Int64, reflect.Uint32:
+		if t == durationType {
+			return map[string]interface{}{
+				"type":        "long",
+				"logicalType": durationNanos,
+			}, nil
+		}
 		return "long", nil
 	case reflect.Int32, reflect.Int16, reflect.Uint16, reflect.Int8, reflect.Uint8:
 		return "int", nil

gotype_test.go

@@ -251,6 +251,51 @@ func TestGoTypeWithUUID(t *testing.T) {
 
 }
 
+func TestGoTypeWithDuration(t *testing.T) {
+	c := qt.New(t)
+	type R struct {
+		D time.Duration
+	}
+
+	data, wType, err := avro.Marshal(R{
+		D: 15 * time.Second,
+	})
+	c.Assert(err, qt.IsNil)
+	var x R
+	_, err = avro.Unmarshal(data, &x, wType)
+	c.Assert(err, qt.IsNil)
+	c.Assert(x, qt.DeepEquals, R{
+		D: 15 * time.Second,
+	})
+
+	c.Assert(mustTypeOf(R{}).String(), qt.JSONEquals, json.RawMessage(`{
+		"type": "record",
+		"name": "R",
+		"fields": [{
+			"name": "D",
+			"default": 0,
+			"type": {
+				"logicalType": "duration-nanos",
+				"type": "long"
+			}
+		}]
+        }`))
+
+	c.Run("zero", func(c *qt.C) {
+		data, wType, err := avro.Marshal(R{})
+		c.Assert(err, qt.IsNil)
+		{
+			type R struct {
+				D int64
+			}
+			var x R
+			_, err = avro.Unmarshal(data, &x, wType)
+			c.Assert(err, qt.IsNil)
+			c.Assert(x, qt.DeepEquals, R{})
+		}
+	})
+}
+
 func TestGoTypeWithStructField(t *testing.T) {
 	c := qt.New(t)
 	type F2 struct {