Skip to main content
Version: 1.2.0

Schema Evolution

Schema evolution lets different versions of your service exchange messages safely — a v2 writer can produce a message a v1 reader still understands, and vice versa.

Compatible Mode​

Compatible mode is the default. It writes extra field metadata so readers can skip unknown fields and tolerate missing ones. Keep it for independent deployments, rolling upgrades, and xlang services. For payloads whose reader and writer schemas never differ, see When to Use Each Mode.

Compatible readers also tolerate selected scalar field type changes when the conversion is lossless. A matched field can read between boolean, string, numeric scalars, and Decimal when the converted value has the same logical value. For example, "true", "false", "1", and "0" can be read as booleans, exact finite ASCII numeric strings can be read as numeric fields that can hold them, numbers and decimals can be read as canonical strings, and numeric widening or narrowing succeeds only when no precision or range is lost. Invalid strings and lossy conversions fail during deserialization. Nullable fields still compose with these conversions, but reference-tracked scalar type changes are incompatible.

Default Compatible Mode​

const fory = new Fory();

Use this when:

  • services deploy schema changes independently
  • older readers may see newer payloads
  • newer readers may see older payloads from before a field was added

Example​

Writer schema:

const writerType = Type.struct(
  { typeId: 1001 },
  {
    name: Type.string(),
    age: Type.int32(),
  },
);

Reader schema with fewer fields:

const readerType = Type.struct(
  { typeId: 1001 },
  {
    name: Type.string(),
  },
);

With compatible mode, the reader ignores fields it does not know about, and fills unknown fields with default values.

When to Use Each Mode​

RequirementSame-schema opt-outCompatible mode
Every reader and writer uses the same schemaworksworks
Independent deploymentsunsaferecommended
Best size and speed for same-schema datayesno
Rolling upgradesunsaferecommended

Set compatible: false for xlang payloads only after verifying that every language uses the same schema, or when native types are generated from Fory schema IDL.

Same-Schema Per-Struct Opt-Out​

You can disable evolution metadata for a specific struct even inside a compatible: true instance:

const fixedType = Type.struct(
  { typeId: 1002, evolving: false },
  {
    name: Type.string(),
  },
);

evolving: false can be faster and smaller for that struct. Use it only when every reader and writer always uses the same struct schema. If one side writes with evolving: false and the other reads expecting compatible metadata, deserialization will fail.

Xlang Requirement​

Compatible mode only protects you from schema differences in the fields of a type. You still need the same type identity (same numeric ID or same typeName) on every side. See Xlang Serialization.