The Bond logo: a stylized glue gun


Bond is an extensible framework for working with schematized data. It is suitable for scenarios ranging from service communications to Big Data storage and processing.

Bond defines a rich type system and schema evolution rules which allow forward and backward compatibility. The core Bond features include high performance serialization/deserialization and a very powerful, generic data transform mechanism. The framework is highly extensible via pluggable serialization protocols, data streams, user defined type aliases and more.

By design Bond is language and platform independent and is currently supported for C++, C#, Java, and Python on Linux, macOS, and Windows.

Bond is published on GitHub at

Basic example

In Bond data schemas are defined using idl-like syntax:

namespace examples

struct Record
    0: string name;
    1: vector<double> constants;

In order to use the schema in a Java program, it needs to be compiled using the Bond compiler. This step is sometimes also referred to as code generation (or codegen) because the compilation generates Java code corresponding to the schema definition.

gbc java

Using the generated Java code, we can write a simple program that will serialize and deserialize an instance of the Record schema using the Compact Binary protocol:

package examples;

import org.bondlib.*;


public class Example {
    public static void main(String[] args) throws IOException {
        final Record src = new Record(); = "FooBar";

        final ByteArrayOutputStream output = new ByteArrayOutputStream();
        final CompactBinaryWriter writer = new CompactBinaryWriter(output, 1);

        final Serializer<Record> serializer = new Serializer<>();
        serializer.serialize(src, writer);

        final ByteArrayInputStream input = new ByteArrayInputStream(output.toByteArray());
        final CompactBinaryReader reader = new CompactBinaryReader(input, 1);

        final Deserializer<Record> deserializer = new Deserializer<>(Record.BOND_TYPE);
        final Record dst = deserializer.deserialize(reader);

Code generation

In order to use a Bond schema in a Java program, it needs to be compiled using the Bond compiler gbc. The compiler generates Java classes that represent the schema. Schema fields are represented by public fields, and collection fields will be automatically initialized to an empty instance.

The mapping between the Bond and Java type systems is mostly obvious, but it is worth noting that, unlike Java reference types, Bond types are not nullable. This means that string in Bond IDL will be mapped to java.lang.String, which is a reference type, but the value null will not be valid. In order to allow null values, a type must be declared as nullable, e.g.:

struct Foo
    0: list<nullable<string>> listOfNullableStrings;

The value null is also legal for fields declared in Bond IDL to have a default of nothing, e.g.:

struct Bar
    0: string str = nothing;


The Bond serialization API is provided by the Serializer class. It is a generic class parameterized with a single Bond-generated type:


The constructor of the Serializer class takes a BondType, which contains the information necessary to serialize the generated type:

new Serializer<Record>(Record.BOND_TYPE)
// or
new Serializer<Record>(someRecord.getBondType())

serializer.serialize(obj, writer);


The Bond deserialization API is provided by the Deserializer class. It is a generic class parameterized with a single Bond-generated type:


The constructor of the Deserializer class takes a BondType, which contains the information necessary to deserialize the generated type:

deserializer = new Deserializer<Record>(Record.BOND_TYPE)
// or
deserializer = new Deserializer<Record>(someRecord.getBondType())

record = deserializer.deserialize(reader);

Deserializing from a payload encoded in an untagged protocol like Simple Binary requires specifying the schema of the payload. To address this scenario, Deserializer.deserialize() has an overload that takes a RuntimeSchema as an argument:

RuntimeSchema schema;
// ...
deserializer = new Deserializer<Record>(Record.BOND_TYPE);
record = deserializer.deserialize(reader, schema);

See also the following example:


Since Bond supports multiple serialization protocols, application endpoints either have to agree on a particular protocol, or include protocol metadata in the payload. Marshaling APIs provide the standard way to do the latter by automatically adding a payload header with the protocol identifier and version.

The Marshal and Unmarshal APIs are similar to Serializer and Deserializer, except that when calling Unmarshal the application simply provides an input stream with payload data, rather than an instance of a particular protocol reader:

final Record src = new Record(); = "foo";

final ByteArrayOutputStream output = new ByteArrayOutputStream();
final CompactBinaryWriter writer = new CompactBinaryWriter(output, 1);

Marshal.marshal(src, writer);

final ByteArrayInputStream input = new ByteArrayInputStream(output.toByteArray());
final Record dst = Unmarshal.unmarshal(input, Record.BOND_TYPE);

See also the following example:


Java type erasure makes it necessary to express the parameters of generic Bond types separately from normal Java declarations. This information is carried in BondType instances, which are required for all serialization and deserialization calls. Concrete Bond-generated types provide their BondType instances in the static member BOND_TYPE, but generic Bond-generated types require their parameters to be explicitly passed to their constructors and then expose their BondType instances via the instance method .getBondType().

See the following example:

Schema evolution

Bond does not use explicit versioning to deal with changes to schemas (and the resulting types) over time. Instead, Bond supports certain schema evolution operations which allow the producer and consumer of Bond types to evolve without lockstep coordination.

The following changes to a schema will never break compatibility across the wire:

The following changes to a type are generally safe but require some consideration about how the change is rolled out:

These following changes will break wire compatibility and are not recommended:

Some best practices and other considerations to keep in mind:

Default values

Fields of a Bond defined struct always have a default value, either explicitly specified in the .bond file, or the implicit default.

The implicit default is

There is no implicit default for enum fields: they must have an explicit default value in the .bond file.

Explicit default values (other than nothing) may not be specified for nullable or container fields. Struct and bonded fields may not have an explicit default value. They always use their implicit default values.

The default values of fields matter because this is what an application will see after deserialization for any optional field that wasn’t present in the payload (e.g. when the payload was created from an older version of the schema).

Additionally, some protocols can omit optional non-struct fields set to their default values, reducing payload size.

Default value of nothing

Sometimes it is necessary to distinguish between any of the possible values of a field and absence of a value. To support such scenarios Bond allows non-struct fields’ default values to be explicitly set to nothing 1:

struct AboutNothing
    0: uint16 n = nothing;
    1: string name = nothing;
    2: list<float> floats = nothing;

Setting a field’s default to nothing doesn’t affect the schema type of the field, but does change the generated Java type. Types that would be references in Java - structs, collections, bondeds, and anything nullable - become Something<T>, while primitives become one of several specializations (e.g., SomethingInteger). A field with a default of nothing that is not present in a payload will result in a field of the appropriate Something type with a null value.

The fact that setting the default value of a field to nothing doesn’t affect the field’s schema type has an important consequence: the default value of the field doesn’t have a serialized representation. What this means in practice depends on whether the field is optional or required. Optional fields set to nothing are usually omitted during serialization 2, just like for any other default values. Required fields, by definition, can never be omitted. Since nothing has no serialized representation, an attempt to serialize an object with required fields set to nothing will result in a runtime exception. If a null value needs to be represented in the serialized form, then a default of nothing is the wrong choice and a nullable type should be used instead.

Nullable types

For any type in the Bond meta-schema, nullable<T> defines a nullable type. A nullable type can store all the same values as its base type plus one additional value: null.

struct Nullables
    0: nullable<bool>         b; // can be true, false, or null
    1: list<nullable<string>> l; // can be a (possibly empty) list or null

The default value for a field of a nullable type is always implicitly set to null. Explicit default values for nullable fields are not supported.

Java reference types are always nullable, so a nullable<T> where T maps to a Java reference type will not change the output of code generation. Attemping to serialize a field whose Bond type is not nullable that has a null value will result in a runtime exception. Where T maps to a Java value type, such as Bond’s int32 and Java’s int, code generation will produce a field of the appropriate boxed reference type (in this case, Integer).

Since a nullable type must represent the additional value of null, its serialized representation necessarily incurs some overhead compared to the base type. Often it is more efficient to avoid using a nullable type and instead to designate one of the normal values to handle the special case that otherwise would be represented by null. For example empty is usually a good choice for string and container types and 0 for arithmetic types. Another option that may sometimes be appropriate is setting the default value of a non-struct field to nothing. Struct fields can have neither an explicit default value nor be set to nothing, so nullable needs to be used if null semantics are needed for these fields.

The canonical scenario where a nullable type is the right choice is recursive structures. For example here’s how Bond TypeDef struct is defined:

struct TypeDef
    // Type identifier
    0: BondDataType id = BT_STRUCT;

    // Index of struct definition in SchemaDef.structs when id == BT_STRUCT
    1: uint16 struct_def = 0;

    // Type definition for:
    //  list elements (id == BT_LIST),
    //  set elements (id == BT_SET),
    //  or mapped value (id == BT_MAP)
    2: nullable<TypeDef> element;

    // Type definition for map key when id == BT_MAP
    3: nullable<TypeDef> key;

    // True if the type is bonded<T>; used only when id == BT_STRUCT
    4: bool bonded_type;

The TypeDef struct is used to represent the type of a field in a Bond schema. If the type is a container such as a list or map, the type definition becomes recursive. For example, a list type definition contains the type of the list element which of course itself can be a container of elements of some other type, and so on, until the recursion is terminated with a null value for the element and key fields.


Bond protocols are pluggable, allowing application to choose the most appropriate encoding format. Bond supports three kinds of protocols:

Compact Binary

A binary, tagged protocol using variable integer encoding and compact field header. A good choice, along with Fast Binary, for RPC scenarios.

Implemented in CompactBinaryReader and CompactBinaryWriter classes. Version 2 of Compact Binary adds length prefix for structs. This enables deserialization of bonded<T> and skipping of unknown fields in constant time.

See also Compact Binary encoding reference.

Fast Binary

A binary, tagged protocol similar to Compact Binary but optimized for deserialization speed rather than payload compactness.

Implemented in FastBinaryReader andFastBinaryWriter classes.

See also Fast Binary encoding reference.

Simple Binary

A binary, untagged protocol which is a good choice for storage scenarios as it offers potential for big saving on payload size. Because Simple Binary is an untagged protocol, it requires that the payload schema is available during deserialization. In typical storage scenario application would store a runtime schema and pass it to any deserialization calls made later.

Implemented in SimpleBinaryReader and SimpleBinaryWriter classes.

Version 2 of Simple Protocol uses variable integer encoding for string and container lengths, resulting in more compact payloads.

See example: examples/java/core/untagged_protocols.

Runtime schema

Some generic applications may need to work with Bond schemas unknown at compile-time. In order to address such scenarios, Bond defines a type SchemaDef to represent schemas at runtime. Applications can obtain an instance of SchemaDef for a particular type through its BondType:

// from a concrete class:
final SchemaDef schema = Record.BOND_TYPE.buildSchemaDef();

// from an instance of a concrete or generic type:
final SchemaDef schema = myInstance.getBondType().buildSchemaDef();

The SchemaDef object is always self contained, including the runtime schema definitions for all nested types (if any). SchemaDef is a Bond type, defined in, and as such can be de/serialized like any other Bond type:

serializer.serialize(Record.BOND_TYPE.buildSchemaDef(), writer);

A serialized representation of SchemaDef can be also obtained directly from a schema definition IDL file using the bond compiler.

See also the following example:

Understanding bonded<T>

The generic type bonded<T> is a simple yet powerful abstraction which is a fundamental part of Bond APIs and enables such usage scenarios as lazy deserialization, pass-through, and polymorphism.

In Java, bonded<T> maps to the Bonded<T> abstract class, which supports four operations: serialize, deserialize, convert, and cast. Bond provides several implementations that represent delayed serialization backed by an instance or a stream. Bonded<T> exposes several static methods that allow the creation of bonded objects and streams.

Lazy deserialization

Because bonded<T> can store (or, more accurately, refer to) data representing a serialized object, it can be used to delay deserialization of some parts of a payload:

struct Example
    0: Always always;
    1: bonded<Sometimes> sometimes;

The schema defined above contains two nested fields. When an object of type Example is deserialized, the field always will be fully instantiated and deserialized, but the field sometimes, which is declared as bonded<Sometimes>, will be merely initialized with a reference to its serialized representation. Applications can then deserialize the object only when needed:

final Example ex = deserializer.deserialize(reader);

// Deserialize sometimes only when needed
if (needSometimes) {
    final Sometimes sometimes = ex.sometimes.deserialize();


The type parameter T in a Bonded<T> is invariant. The instance method .cast(BondType) can be used to upcast and .convert(BondType) can be used to downcast. In both cases, the data backing the Bonded is unmodified.

Warning: instantiating a Bonded<Base> from an instance of Derived and then serializing it will serialize only the Base fields.

Codegen parameters

--namespace: Allows mapping Bond namespaces into Java packages. If you have a Bond file containing a namespace examples declaration and want your classes generated into org.bondlib.examples, you can invoke gbc like this:

gbc java --namespace="examples=org.bondlib.examples"

Multiple aliases may be given to a single --namespace option by separating them with semicolons.

Platform limitations

Bond for Java currently targets JDK 1.6.

Build instructions

Bond for Java is currently only available in source form. You will need to clone the Bond repository, install the appropriate dependencies for your platform, and build and install the Bond compiler. Follow the instructions in the top-level to do all of this.

Java has two additional requirements:

With all of this done, you’re ready to build the Bond library and the Bond gradle plugin (optional, but strongly recommended). To build the plugin and install it to your local maven repository:

cd java/gradle-plugin; gradle build install

To build the library and install it to your local maven repository:

cd java/core; gradle build install

To consume either component from your local maven repository, see the build.gradle in any of our Java example projects.

Build tooling

We provide a plugin for the gradle build tool. Once you’ve completed all of the build and installation steps in the build instructions section, the gradle plugin will recursively discover and compile all Bond files in src/main/bond and src/test/bond and add the generated code as production and test sources, respectively. You can specify Bond files outside those directories and pass options to gbc by adding an explicit compileBond or compileTestBond container to your build file. You can see an example of this in the build.gradle of the Bond Core library itself.


Bond compiler reference

C++ User’s Manual

C# User’s Manual

Python User’s Manual

  1. In Bond there is no concept of a default value for structs and thus a default of nothing can’t be set for fields of struct types or bonded<T>.

  2. Some protocols might not support omitting optional fields (e.g. Simple Protocol). In such cases an attempt to serialize an object with field(s) set to nothing will result in a runtime exception.