cynic-parser tracking issue

[obmarg]

Things left to do:

For v0.1.0:

• Parse scalar definitions
• Parse directive definitions
• Parse extensions
• Tests
  • Import them
  • Double import tests on actual schemas Add more tests of actual schemas to cynic-parser #829
  • Extension tests
  • Fix the kitchen sink tests
  • Make sure there's nothing missing
• Optimise ID types
  • Make them use non-zero integers Update cynic-parser to use NonZeroU32 for IDs #826
  • Probably use something smaller than usize - do not need that many indices
• Make the API make sense
  • Parsing functions should return errors not panic Add error handling to cynic-parser #831
  • Add optional error reports w/ ariadne Add error report feature to cynic-parser #833
• Stop using lalrpop in build.rs Remove direct lalrpop dependency from cynic-parser #823
• Disable pretty by default Disable printing by default in cynic-parser #832
• Integrate with cynic-codegen Update cynic-codegen to use cynic-parser #824
• Finish the printing code
• Look into & benchmark the range idea I had Update cynic-parser with IdRange #827
• Re-org
  • At the least the reader.rs can be split up
  • Writer can probably also get its own file
• Write a readme etc.
• Setup release process
• Strings should live behind a reader as well I think, maybe? I'm flip flopping on this somewhat

Later:

• Benchmark
  • Write benchmarks against graphql-parser, async-graphql-parser, apollo-parser
  • Ideally figure out how to benchmark against go, js & other language parsers
  • Publish benchmarks somehow
• Escape strings in output code fix: escape printed strings in cynic-parser #914
• Validate the schema roots
• Generate Debug impls for the reader types
• Fix the formatting in printing code
  • fix: a lot of bugs in cynic-parser pretty printing #916
  • Spaces between fields w/ docstrings fix: blank lines between things with docstrings #954
  • Spaces between arguments w/ docstrings fix: blank lines between things with docstrings #954
  • Fix union formatting
  • Get parity w/ prettier on the github schema test: parity with prettier on github schema #955
• Parse executable documents
• Value vs ConstValue (particularly for directives)
• Update cynic-introspection printing code to use the parser?
• Iterate on error reports (and test more)
• Make sure documentation is complete
  • finish a readme
  • at the very least write some lib.rs docs that give an overview of the layout of the crate
• Special case empty schemas (can have a special error for this instead of whatever the parser says by default)
• put spans on relevant bits of the AST
• Clean up the rest of the TODOs in the code
• Handle escape characters, and string literals generally (iirc I keep the quotes, probably not the best idea) string escaping in cynic-parser #868
• Pick a structure for the type_system/executable modules and make sure they both use it.
• Fix the ignored executable parser tests fix: a query formatting issue in parser #947
• Use IdRange for Value?
• Figure out how to avoid making every keyword a token of its own
• Maybe try an inlined small string optimisation