List view
Support all of the models from https://github.com/aws/aws-cryptographic-material-providers-library and https://github.com/aws/aws-database-encryption-sdk-dynamodb, with all features and polish (e.g. desired improvements to naming, code layout, ergonomics, documents, etc)
No due date•9/25 issues closedSupport all of the models from https://github.com/aws/aws-cryptographic-material-providers-library and https://github.com/aws/aws-database-encryption-sdk-dynamodb, at least well enough to run successfully, even if production-ready features like `@documentation` or improved error handling are not yet supported.
No due date•10/10 issues closed("beneration" == "manually write out the code generation target code") See attached issues for all test models that are in scope. Models we can explicitly skip: * AggregateReferences - only tests Dafny code generation, not intended for runtime testing. * CodegenPatches - only tests the patching functionality, which we're already heavily using for Rust. :) * Dependencies - creating and using DependenciesLite instead * Under SimpleTypes: BigDecimal/BigInteger/SimpleByte/SimpleFloat/SimpleShort - not yet used in downstream projects, no runtime testing for existing languages either. * aws-sdks - Not worth the effort of beneration, but this is assuming we add SDKLite and benerate that first.
No due date•28/32 issues closedAt least nice-to-have if not blocking tasks before flipping the repo public. Constraints that make this smaller: * We'll only be publishing the Smithy build plugin (`codegen/smithy-dafny-codegen`) and not the CLI, which we hope to deprecate anyway. * We'll only be claiming support for Java initially. * We'll be very clear about what what shapes and features are not yet supported (and #185 is particularly critical for that) * We'll be publishing a 0.x version, just like all other `smithy-<lang>` projects.
No due date•9/23 issues closedIssues that are at least nice-to-have before we officially support other customers using this repo. #151 is the primary new functionality we need, but there are other clean-up tasks around generalizing to arbitrary service models that will drastically improve the UX.
No due date•5/5 issues closed