Releases: rowtype-yoga/purescript-protobuf
v4.6.0
What's Changed
- spago bundle instructions by @jamesdbrock in #44
- spago polyrepo reorganization by @jamesdbrock in #47
- protoc plugin code generation speed improvement for large input by @jamesdbrock in #50
Full Changelog: v4.5.0...v4.6.0
v4.5.0
What's Changed
spago@nextbuild system. Reorganized packages. by @jamesdbrock in #40
Full Changelog: v4.4.0...v4.5.0
v4.4.0
Stack-safety bugs fixed for encoding. putRepeated and putPacked will be stack-safe.
- Add
MonadRecconstraint toputfunctions. - Replace all internal use of
traverse_withfoldRecM.
Full Changelog: v4.3.0...v4.4.0
v4.3.0
- Development: purs v0.15.7, node-streams-aff v5.0.0, assorted improvements.
Full Changelog: v4.2.0...v4.3.0
v4.2.0
- Add doc comments from the
.protomessages and enums to the generated.purscode. - Improved dev environment for purescript-language-server and ide-purescript
withPURS_IDE_SOURCES. - Use
spago2nix_nativeBuildInputs.
Full Changelog: v4.1.0...v4.2.0
v4.1.0
Full Changelog: v4.0.0...v4.1.0
- Nix flake. With one-step spago2nix.
- Upgrade dependencies.
- Delete
mkUint8Array. Upstreamed this to purescript-arraybuffer. - Delete
manyArray,parseMaybe,parseEither,parseExceptT. Upstreamed these to purescript-parsing. - Refactor
manyLength. Get rid of the FFI, and a slight speedup in the benchmarks. - Delete
Runtime.positionZero.
v2.0.0
Upgraded to PureScript 0.14.3.
For byte fields we’re now using the DataBuff type
from arraybuffer-builder v2.1.0 which allows us to avoid array copies,
and so may speed up encoding and decoding in some cases.
Created a Protobuf.Prelude module to eliminate many import warnings
in the generated code.
Breaking changes
The Bytes field type was formerly a wrapper for ArrayBuffer, and now
it is a wrapper for DataBuff, which may be either ArrayBuffer or DataView.
This changes the API and will require some changes in dependent code.
Bugfixes
Fixed a bug which sometimes incorrectly errored with “index out of bounds” when
decoding packed repeated fields of double, float, fixed32, or sfixed32.
Fixed a bug which would cause a compilation failure in generated code for
enums with negative values.