aida_2012 VS tonic

Some years ago I wrote a JSON parser, see for example https://github.com/joakim-strandberg/aida_2012/blob/master/src/tests/aida-json_dom_parser_tests.adb. I've been thinking of refactoring but never gotten around to do it. One of the ideas for writing the code was minimizing external dependencies. The JSON parser is written in SPARK and at the time 2018 when the code was written SPARK did not allow heap allocations which means all the data-structures used are statically allocated. I can imagine onox json-ada has a nicer API.

For getting UTF-8 support one can also use Dmitry Kazakovs Simple components (http://www.dmitry-kazakov.de/ada/components). Based on it have made a UTF-8 version which is Ada/Spark based (https://github.com/joakim-strandberg/aida_2012). Another one I have been toying with is cross-compiler: https://github.com/joakim-strandberg/portable_ada . To print to standard out using UTF-8 the easiest way is probably enabling "-gnatW8" in order for normal String types be interpreted as UTF-8 strings. Using the UTF-8 support in the links I would recommend making one's own Ada bindings to the OS. It should be easy to do on Mac OS X and Linux where UTF-8 is enabled by default but more of an issue on Windows where the UTF-8 encoded strings need to be converted to UTF-16 before being handed over to the OS.

Use tonic-build directly from Rust.

} // Omit the remaining code and refer to the example in Tonic: https://github.com/hyperium/tonic/blob/master/examples/src/multiplex/server.rs ```

If I’m not misunderstanding, you could both decode the gRPC protobuf AND write to delta lake in Rust. Tonic, Delta-rs.

Original post: https://github.com/hyperium/tonic/issues/1405. Cross-post here in case the problem is not specific to tonic.

Tonic

Just to double check Is this the correct repo?

Wasm isn't involved much actually. Basically, the frontend asks the backend to perform a gRPC call on its behalf using Tauri events. They are like named channels on which you can send any serde-compatible value. But the backend is a normal Rust program so there are no constraints there. I use prost-reflect to encode/decode Protobuf messages according to Protobuf descriptors loaded at runtime and make the actual requests using tonic from the tokio ecosystem. prost-reflect is necessary because, normally, tonic expects the Protobuf descriptor to be known at compile time so it can make some code generation behind the scenes.

You could have a look at gRPC i.e. https://github.com/hyperium/tonic

Looking at this example code: https://github.com/hyperium/tonic/blob/master/examples/src/tower/server.rs

The remaining gap is remote actors, since you still need some kind of serialization between them, and take your pick of standards for that one such as gRPC using Tonic.