1.2.12

Release 1.2.12

This is a production release of go-codec.

Changes include:

• expose (En|De)coder.HandleName()
• cbor: fix indefinite stream conformance
• codecgen: use base32 to generate field names
• return io.ErrUnexpectedEOF (not io.EOF) when EOF is reached while needing more data

Please try it out and share your experiences.

1.2.11

Release 1.2.11

This is a production release of go-codec.

Changes include:

• codecgen: fix decoding missing fields of struct from array, due to double counter increment

Please try it out and share your experiences.

1.2.10

Release 1.2.10

This is a production release of go-codec.

Changes include:

• codecgen: fix error decoding struct from array, due to misplaced counter increment
• fix error encoding interface{} map key in Canonical mode

Please try it out and share your experiences.

1.2.9

This is a production release of go-codec.

Changes include:

• faster growslice (as zero'ing memory only done if needed)
• make decoding via io.Reader more robust
• read and write optimizations
• clean up code across the board
• benchmark decoding for different engines against same set of encoded bytes
• improve inline'ability of code across runtime and codecgen'erated codebases

Please try it out and share your experiences.

1.2.8

This is a production release of go-codec.

Changes include:

• treat an array with all zero values as empty, and omit if omitempty=true
• module support: ensure module dependencies are auto-generated by go tool
• support go 1.20: needed fix for growslice in unsafe mode
• add support for go 1.13 error wrapping
• codecgen: optimize canonical encoding, handling extensions, named types and ordered keys
• codecgen: handle canonical encoding of map and struct well
• codecgen: skip type aliases
• codecgen: fix spurious errors in generated code calling codec.Selfer encode methods

Please try it out and share your experiences.

1.2.7

This is a production release of go-codec.

Changes include:

• support registering extension functions for time.Time
• fix bugs with encoding/decoding very-nested slices in specific cases

Please try it out and share your experiences.

1.2.6

This is a production release of go-codec.

Changes include:

• support EncodeOptions.NoAddressableReadonly option
• streamline/reduce size of helper types
• optimize side(En|De)code and (En|De)codeExt calls
• decode a number from any number in the stream i.e.
  you can decode a float/integer/unsigned integer from any number in stream
• clean up tests so they work completely regardless if we recover from panics
• improve inlining of common functions

Please try it out and share your experiences.

1.2.5

This is a production release of go-codec.

This is a very important release, building upon the optimization introduced in v1.2.4.

Changes include:

• support gollvm in addition to gccgo in default high-performance mode
• support codec.safe and codec.notfastpath as preferred build tags (replacing safe and notfastpath)
• support Canonical mode with MissingFielder
• robust handling of codecgen caveats wrt Canonical|CheckCircularRef flags and MissingFielder implementations
• robust handling of transient values, where we track if a value has internal pointers and handle appropriately
• reduce use of global values: all in-use values are scoped to an Encoder or Decoder
• implement stateManager for Encoder/Decoder, supporting capture, reset and restore functions.
• side encoding or decoding can now reuse an Encoder/Decoder, leveraging state management to
  capture its state, do something else, then restore its state and continue
• support running all tests in parallel, so we can shake out any concurrency issues
• Fix issue in rpc where writer was not being flushed, causing a hang where reader was
  waiting on an encoded value
• Support shallow copies of a Handle, to support parallel execution where we need to
  temporarily change some Handler fields

This is the best release yet. Please try it out and share your experiences.

1.2.4

Release 1.2.4

This is a production release of go-codec.

This is a very important release, which focused on optimization across the board to
ensure that go-codec is the most performant library for popular encoding formats.

Optimizations across the board include:

• optimization by eliding allocation where not necessary e.g.
  • when encoding non-addressable value with ptr receiver
  • map iteration/access
  • decoding floats
  • creating slices or sub-slices (removing temporary slice allocated on heap)
  • decoding map keys if both key and value kinds are scalars
  • decoding map values if value kind is a scalar
• better reuse of pool'ed []byte(s)
• leveraging mapXXX_fastXXX calls where possible
• better inlining (where applicable) e.g. arrayStart, mapStart, etc
• support complex numbers (where we treat complex numbers as floats with imaginary part = 0)
• optimize encoding and decoding arrays, including using fast-paths for slices as applicable
• use runtime's growslice algorithm when growing arrays for better performance
• optimize nextValueBytes and swallow so they do not allocate when applicable e.g.
  do not allocate/copy bytes if swallow is called or you were decoding from a []byte (just return a sub-slice)
• optimize isEmptyStruct to just compare if value == zero'es (instead of field by field),
  simulating comparing to zero value.

To guarantee better fairness in benchmarking, we did the following:

• use common types in benchmarks and common methods to ensure fairness in benchmarks
• run benchmarks with ZeroCopy=true and (MapValue|Interface|SliceElement)Reset = true,
  aligning with how other libraries operate by default

With these changes, we have reflection-based mode performing within 25% of codecgen,
with similar allocation numbers as codecgen. For example, encoding typically has just 1
allocation without codecgen (just as in codecgen).

This is the best release yet. Please try it out and share your experiences.

1.2.3

This is a production release of go-codec.

This optimizes decoding for zero-copy mode when decoding from a string or []byte.

With this mode, we see 50% reduction in allocation in both codecgen and normal execution,
and this gives performance better than other libraries (easyjson, json-iterator, etc)
while providing much more features and supported formats (json, cbor, msgpack, simple, etc).