blob: b61a8180e4c5b02f095b20c7505765232b0cb908 [file] [log] [blame]
// Copyright (c) 2012-2018 Ugorji Nwoke. All rights reserved.
// Use of this source code is governed by a MIT license found in the LICENSE file.
Package codec provides a
High Performance, Feature-Rich Idiomatic Go 1.4+ codec/encoding library
for binc, msgpack, cbor, json.
Supported Serialization formats are:
- msgpack:
- binc:
- cbor:
- json:
- simple:
To install:
go get
This package will carefully use 'unsafe' for performance reasons in specific places.
You can build without unsafe use by passing the safe or appengine tag
i.e. 'go install -tags=safe ...'. Note that unsafe is only supported for the last 3
go sdk versions e.g. current go release is go 1.9, so we support unsafe use only from
go 1.7+ . This is because supporting unsafe requires knowledge of implementation details.
For detailed usage information, read the primer at .
The idiomatic Go support is as seen in other encoding packages in
the standard library (ie json, xml, gob, etc).
Rich Feature Set includes:
- Simple but extremely powerful and feature-rich API
- Support for go1.4 and above, while selectively using newer APIs for later releases
- Excellent code coverage ( > 90% )
- Very High Performance.
Our extensive benchmarks show us outperforming Gob, Json, Bson, etc by 2-4X.
- Careful selected use of 'unsafe' for targeted performance gains.
100% mode exists where 'unsafe' is not used at all.
- Lock-free (sans mutex) concurrency for scaling to 100's of cores
- Coerce types where appropriate
e.g. decode an int in the stream into a float, decode numbers from formatted strings, etc
- Corner Cases:
Overflows, nil maps/slices, nil values in streams are handled correctly
- Standard field renaming via tags
- Support for omitting empty fields during an encoding
- Encoding from any value and decoding into pointer to any value
(struct, slice, map, primitives, pointers, interface{}, etc)
- Extensions to support efficient encoding/decoding of any named types
- Support encoding.(Binary|Text)(M|Unm)arshaler interfaces
- Support IsZero() bool to determine if a value is a zero value.
Analogous to time.Time.IsZero() bool.
- Decoding without a schema (into a interface{}).
Includes Options to configure what specific map or slice type to use
when decoding an encoded list or map into a nil interface{}
- Mapping a non-interface type to an interface, so we can decode appropriately
into any interface type with a correctly configured non-interface value.
- Encode a struct as an array, and decode struct from an array in the data stream
- Option to encode struct keys as numbers (instead of strings)
(to support structured streams with fields encoded as numeric codes)
- Comprehensive support for anonymous fields
- Fast (no-reflection) encoding/decoding of common maps and slices
- Code-generation for faster performance.
- Support binary (e.g. messagepack, cbor) and text (e.g. json) formats
- Support indefinite-length formats to enable true streaming
(for formats which support it e.g. json, cbor)
- Support canonical encoding, where a value is ALWAYS encoded as same sequence of bytes.
This mostly applies to maps, where iteration order is non-deterministic.
- NIL in data stream decoded as zero value
- Never silently skip data when decoding.
User decides whether to return an error or silently skip data when keys or indexes
in the data stream do not map to fields in the struct.
- Detect and error when encoding a cyclic reference (instead of stack overflow shutdown)
- Encode/Decode from/to chan types (for iterative streaming support)
- Drop-in replacement for encoding/json. `json:` key in struct tag supported.
- Provides a RPC Server and Client Codec for net/rpc communication protocol.
- Handle unique idiosyncrasies of codecs e.g.
- For messagepack, configure how ambiguities in handling raw bytes are resolved
- For messagepack, provide rpc server/client codec to support
msgpack-rpc protocol defined at:
Extension Support
Users can register a function to handle the encoding or decoding of
their custom types.
There are no restrictions on what the custom type can be. Some examples:
type BisSet []int
type BitSet64 uint64
type UUID string
type MyStructWithUnexportedFields struct { a int; b bool; c []int; }
type GifImage struct { ... }
As an illustration, MyStructWithUnexportedFields would normally be
encoded as an empty map because it has no exported fields, while UUID
would be encoded as a string. However, with extension support, you can
encode any of these however you like.
Custom Encoding and Decoding
This package maintains symmetry in the encoding and decoding halfs.
We determine how to encode or decode by walking this decision tree
- is type a codec.Selfer?
- is there an extension registered for the type?
- is format binary, and is type a encoding.BinaryMarshaler and BinaryUnmarshaler?
- is format specifically json, and is type a encoding/json.Marshaler and Unmarshaler?
- is format text-based, and type an encoding.TextMarshaler?
- else we use a pair of functions based on the "kind" of the type e.g. map, slice, int64, etc
This symmetry is important to reduce chances of issues happening because the
encoding and decoding sides are out of sync e.g. decoded via very specific
encoding.TextUnmarshaler but encoded via kind-specific generalized mode.
Consequently, if a type only defines one-half of the symmetry
(e.g. it implements UnmarshalJSON() but not MarshalJSON() ),
then that type doesn't satisfy the check and we will continue walking down the
decision tree.
RPC Client and Server Codecs are implemented, so the codecs can be used
with the standard net/rpc package.
The Handle is SAFE for concurrent READ, but NOT SAFE for concurrent modification.
The Encoder and Decoder are NOT safe for concurrent use.
Consequently, the usage model is basically:
- Create and initialize the Handle before any use.
Once created, DO NOT modify it.
- Multiple Encoders or Decoders can now use the Handle concurrently.
They only read information off the Handle (never write).
- However, each Encoder or Decoder MUST not be used concurrently
- To re-use an Encoder/Decoder, call Reset(...) on it first.
This allows you use state maintained on the Encoder/Decoder.
Sample usage model:
// create and configure Handle
var (
bh codec.BincHandle
mh codec.MsgpackHandle
ch codec.CborHandle
mh.MapType = reflect.TypeOf(map[string]interface{}(nil))
// configure extensions
// e.g. for msgpack, define functions and enable Time support for tag 1
// mh.SetExt(reflect.TypeOf(time.Time{}), 1, myExt)
// create and use decoder/encoder
var (
r io.Reader
w io.Writer
b []byte
h = &bh // or mh to use msgpack
dec = codec.NewDecoder(r, h)
dec = codec.NewDecoderBytes(b, h)
err = dec.Decode(&v)
enc = codec.NewEncoder(w, h)
enc = codec.NewEncoderBytes(&b, h)
err = enc.Encode(v)
//RPC Server
go func() {
for {
conn, err := listener.Accept()
rpcCodec := codec.GoRpc.ServerCodec(conn, h)
//OR rpcCodec := codec.MsgpackSpecRpc.ServerCodec(conn, h)
//RPC Communication (client side)
conn, err = net.Dial("tcp", "localhost:5555")
rpcCodec := codec.GoRpc.ClientCodec(conn, h)
//OR rpcCodec := codec.MsgpackSpecRpc.ClientCodec(conn, h)
client := rpc.NewClientWithCodec(rpcCodec)
Running Tests
To run tests, use the following:
go test
To run the full suite of tests, use the following:
go test -tags alltests -run Suite
You can run the tag 'safe' to run tests or build in safe mode. e.g.
go test -tags safe -run Json
go test -tags "alltests safe" -run Suite
Running Benchmarks
Please see .
Struct fields matching the following are ignored during encoding and decoding
- struct tag value set to -
- func, complex numbers, unsafe pointers
- unexported and not embedded
- unexported and embedded and not struct kind
- unexported and embedded pointers (from go1.10)
Every other field in a struct will be encoded/decoded.
Embedded fields are encoded as if they exist in the top-level struct,
with some caveats. See Encode documentation.
package codec
// TODO:
// - For Go 1.11, when mid-stack inlining is enabled,
// we should use committed functions for writeXXX and readXXX calls.
// This involves uncommenting the methods for decReaderSwitch and encWriterSwitch
// and using those (decReaderSwitch and encWriterSwitch) in all handles
// instead of encWriter and decReader.
// The benefit is that, for the (En|De)coder over []byte, the encWriter/decReader
// will be inlined, giving a performance bump for that typical case.
// However, it will only be inlined if mid-stack inlining is enabled,
// as we call panic to raise errors, and panic currently prevents inlining.
// - To make Handle comparable, make extHandle in BasicHandle a non-embedded pointer,
// and use overlay methods on *BasicHandle to call through to extHandle after initializing
// the "xh *extHandle" to point to a real slice.
// - Look through and fix padding for each type, to eliminate false sharing
// - critical shared objects that are read many times
// TypeInfos
// - pooled objects:
// decNaked, decNakedContainers, codecFner, typeInfoLoadArray,
// - small objects allocated independently, that we read/use much across threads:
// codecFn, typeInfo
// - Objects allocated independently and used a lot
// Decoder, Encoder,
// xxxHandle, xxxEncDriver, xxxDecDriver (xxx = json, msgpack, cbor, binc, simple)
// - In all above, arrange values modified together to be close to each other.
// For all of these, either ensure that they occupy full cache lines,
// or ensure that the things just past the cache line boundary are hardly read/written
// e.g. JsonHandle.RawBytesExt - which is copied into json(En|De)cDriver at init
// Occupying full cache lines means they occupy 8*N words (where N is an integer).
// Check this out by running: ./ -z
// - look at those tagged ****, meaning they are not occupying full cache lines
// - look at those tagged <<<<, meaning they are larger than 32 words (something to watch)
// - Run "golint -min_confidence 0.81"