Paul Colomiets

> The spec/graphql-js also implements a validation rule (`UniqueInputFieldNamesRule`) to strictly enforce the existence of only one field. > > Currently, this creates an ambiguity issue, since the consumer of...

I'm not sure yet. I wonder what others think on this. /cc @graphql-rust/graphql-parser-maintainers, @graphql-rust/graphql-client-maintainers

The idea was that position is good trade-off of memory vs specificity of the position. I.e. this is not enough for doing in-place modification of the text but should be...

This is true for inputs and outputs. But I'm still not sure it is true for literals.

Ah, yes, it's because travis stopped working with github at some point. Should setup a github actions instead. But this doesn't mean there are no tests or that we don't...

I don't think that exporting `*` is a good idea here. Those parsers are meant to be private API, that can be restructured at any time. Also we don't expose...

I'm not sure what you refer specifically. I think there are two possible formats: `Debug` which is mostly generated by Rust, so you can probably refer to the Rust documentation...

Well, I usually use normal functions and matches. Since every node is unique. And having a method per node type or similar usually not very convenient in Rust (because either...

It's just that "Display" trait of the AST outputs the query. This is convenient if you want to extract a piece of query or modify query and convert back to...

> However, it is possible to print information provided in panic payload, at least if you have a custom panic fmt. But im not sure how panics are reported in...