Start work on generated code for encoding/decoding proto messages. #268
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
blackgnezdo
approved these changes
Dec 7, 2018
| @@ -113,7 +113,7 @@ generateModule modName imports syntaxType modifyImport definitions importedEnv s | |||
| mainImports = map (modifyImport . importSimple) | |||
| [ "Control.DeepSeq", "Lens.Labels.Prism" ] | |||
| sharedImports = map (modifyImport . importSimple) | |||
| [ "Prelude", "Data.Int", "Data.Word" | |||
| [ "Prelude", "Data.Int", "Data.Monoid", "Data.Word" | |||
There was a problem hiding this comment.
We should probably decide on our supported GHC versions some time soon. If you can think of a way to poll proto-lens users for which version of GHC they care about, maybe we can formulate a support policy of some sort.
There was a problem hiding this comment.
Currently the "ground truth" for what's supported is whatever's covered by our CI:
https://github.com/google/proto-lens/blob/master/.travis.yml#L20
It goes back to lts-7 which corresponds to ghc-8.0.1 and cabal-1.24.2.0.
I'll fix the build breakages before merging.
|
I presume you'll take care of builds before merging. |
This stubs out new methods in the Message class for decoding and encoding.
By generating the implementations rather than going through the reified
FieldDescriptor, we should ultimately get much faster code.
Currently, the implementations are trivial (e.g., `buildMessage = const ""`) and
also named `unfinished{Parse,Build}Message`.
The code for the old, "reflected" encoding/decoding (i.e., using
`FieldDescriptor`) has been moved to a new module namespace
`Data.ProtoLens.Encoding.Reflected`. It is still used by
`Data.ProtoLens.Encoding` by default. A Cabal flag `generated-encoding` lets
`Data.ProtoLens.Encoding use the new implementation.
My intention is to finish this implementation and delete the old code before
making a new major release of proto-lens. In the meantime, this setup should
make it easier to implement the new API as a series of distinct PRs.
I changed CI to build the tests and benchmarks with the new API, but not to
actually run them (since they don't pass).
judah
force-pushed
the
generate-encoding
branch
from
caf88be
to
73c616b
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This stubs out new methods in the Message class for decoding and encoding.
By generating the implementations rather than going through the reified
FieldDescriptor, we should ultimately get much faster code.
Currently, the implementations are trivial (e.g.,
buildMessage = const "") andalso named
unfinished{Parse,Build}Message.The code for the old, "reflected" encoding/decoding (i.e., using
FieldDescriptor) has been moved to a new module namespaceData.ProtoLens.Encoding.Reflected. It is still used byData.ProtoLens.Encodingby default. A Cabal flaggenerated-encodinglets`Data.ProtoLens.Encoding use the new implementation.
My intention is to finish this implementation and delete the old code before
making a new major release of proto-lens. In the meantime, this setup should
make it easier to implement the new API as a series of distinct PRs.
I changed CI to build the tests and benchmarks with the new API, but not to
actually run them (since they don't pass).
This change is