Tyler Gregg
Tyler Gregg
All implementations have some equivalent of a "write all values" API that simply re-writes the data parsed by the given reader using the given writer's configuration. Useful examples are: *...
Because path extractor APIs are more user-friendly (and often more performant) than the alternatives, we should add cookbook examples that show how to use these APIs for the implementations that...
The symbols section needs to have crisp definitions of its important terms. To start, we can come to a consensus about the terms defined in the [developer guide](http://amzn.github.io/ion-docs/guides/symbols-guide.html) and add...
The symbols section of the spec should include some additional guidance (i.e. *may* statements) around how implementations can choose handle preservation of symbol context. For example, implementations may wish to...
* [ ] Enable the `LazyReader` to read from streams (in addition to byte slices) * [ ] Expose byte slice accessors for stream elements' encodings (e.g. `fn annotations(&self) ->...
https://github.com/amazon-ion/ion-rust/pull/485 introduces the ability to iteratively parse top-level values from a stream. Previously, only `@[u8]` was supported as input, forcing `inspect` and `to` to memory-map the stream (e.g., https://github.com/amazon-ion/ion-cli/blob/master/src/bin/ion/commands/beta/inspect.rs#L182). The...
Allow users to analyze streams for information about: * Number of symbols * Number of symbol tables (and appends vs fresh tables) * Minimum/maximum/average size of top-level values (this helps...
* Replace current CCs with the project maintainers. * Add more info to the build to make errors easier to diagnose.
When the user (or, more likely, a system reader via writeValues) writes a symbol table or IVM, the writer should flush any previously buffered data to the output. This is...