schala/schala-lang/language/src/lib.rs

#![feature(trace_macros)]
#![feature(custom_attribute)]
//#![feature(unrestricted_attribute_tokens)]
#![feature(slice_patterns, box_patterns, box_syntax)]

//! `schala-lang` is where the Schala programming language is actually implemented. 
//! It defines the `Schala` type, which contains the state for a Schala REPL, and implements
//! `ProgrammingLanguageInterface` and the chain of compiler passes for it.

extern crate itertools;
#[macro_use]
extern crate lazy_static;
#[macro_use]
extern crate maplit;
extern crate schala_repl;
#[macro_use]
extern crate schala_lang_codegen;
extern crate ena;

use std::cell::RefCell;
use std::rc::Rc;

use itertools::Itertools;
use schala_repl::{ProgrammingLanguageInterface, ComputationRequest, ComputationResponse, LangMetaRequest, LangMetaResponse, GlobalOutputStats};

macro_rules! bx {
  ($e:expr) => { Box::new($e) }
}

#[macro_use]
mod util;
#[macro_use]
mod typechecking;

mod tokenizing;
mod ast;
mod parsing;
mod symbol_table;
mod builtin;
mod reduced_ast;
mod eval;

/// All bits of state necessary to parse and execute a Schala program are stored in this struct.
/// `state` represents the execution state for the AST-walking interpreter, the other fields
/// should be self-explanatory.
pub struct Schala {
  source_reference: SourceReference,
  state: eval::State<'static>,
  symbol_table: Rc<RefCell<symbol_table::SymbolTable>>,
  type_context: typechecking::TypeContext<'static>,
  active_parser: Option<parsing::Parser>,
}

impl Schala {
  fn handle_docs(&self, source: String) -> LangMetaResponse {
    LangMetaResponse::Docs {
      doc_string: format!("<<Schala-lang documentation not yet implemented>>")
    }
  }
}

impl Schala {
  /// Creates a new Schala environment *without* any prelude.
  fn new_blank_env() -> Schala {
    let symbols = Rc::new(RefCell::new(symbol_table::SymbolTable::new()));
    Schala {
      source_reference: SourceReference::new(),
      symbol_table: symbols.clone(),
      state: eval::State::new(symbols),
      type_context: typechecking::TypeContext::new(),
      active_parser: None,
    }
  }

  /// Creates a new Schala environment with the standard prelude, which is defined as ordinary
  /// Schala code in the file `prelude.schala`
  pub fn new() -> Schala {
    let prelude = include_str!("prelude.schala");
    let mut s = Schala::new_blank_env();

    let request = ComputationRequest { source: prelude, debug_requests: vec![] };
    s.run_computation(request);
    s
  }
}

fn tokenizing(input: &str, _handle: &mut Schala, comp: Option<&mut PassDebugArtifact>) -> Result<Vec<tokenizing::Token>, String> {
  let tokens = tokenizing::tokenize(input);
  comp.map(|comp| {
    let token_string = tokens.iter().map(|t| t.to_string_with_metadata()).join(", ");
    comp.add_artifact(token_string);
  });

  let errors: Vec<String> = tokens.iter().filter_map(|t| t.get_error()).collect();
  if errors.len() == 0 {
    Ok(tokens)
  } else {
    Err(format!("{:?}", errors))
  }
}

fn parsing(input: Vec<tokenizing::Token>, handle: &mut Schala, comp: Option<&mut PassDebugArtifact>) -> Result<ast::AST, String> {
  use crate::parsing::Parser;

  let mut parser = match handle.active_parser.take() {
    None => Parser::new(input),
    Some(parser) => parser
  };

  let ast = parser.parse();
  let trace = parser.format_parse_trace();

  comp.map(|comp| {
    /*
    //TODO need to control which of these debug stages get added
    let opt = comp.cur_debug_options.get(0).map(|s| s.clone());
    match opt {
      None => comp.add_artifact(TraceArtifact::new("ast", format!("{:?}", ast))),
      Some(ref s) if s == "compact" => comp.add_artifact(TraceArtifact::new("ast", format!("{:?}", ast))),
      Some(ref s) if s == "expanded" => comp.add_artifact(TraceArtifact::new("ast", format!("{:#?}", ast))),
      Some(ref s) if s == "trace" => comp.add_artifact(TraceArtifact::new_parse_trace(trace)),
      Some(ref x) => println!("Bad parsing debug option: {}", x),
    };
    */
  });
  ast.map_err(|err| format_parse_error(err, handle))
}

fn format_parse_error(error: parsing::ParseError, handle: &mut Schala) -> String {
  let line_num = error.token.line_num;
  let ch = error.token.char_num;
  let line_from_program = handle.source_reference.get_line(line_num);
  let location_pointer = format!("{}^", " ".repeat(ch));

  let line_num_digits = format!("{}", line_num).chars().count();
  let space_padding = " ".repeat(line_num_digits);

  format!(r#"
{error_msg}
{space_padding} |
{line_num} | {}
{space_padding} | {}
"#, line_from_program, location_pointer, error_msg=error.msg, space_padding=space_padding, line_num=line_num)
}

fn symbol_table(input: ast::AST, handle: &mut Schala, comp: Option<&mut PassDebugArtifact>) -> Result<ast::AST, String> {
  let add = handle.symbol_table.borrow_mut().add_top_level_symbols(&input);
  match add {
    Ok(()) => {
      let debug = handle.symbol_table.borrow().debug_symbol_table();
      comp.map(|comp| comp.add_artifact(debug));
      Ok(input)
    },
    Err(msg) => Err(msg)
  }
}

fn typechecking(input: ast::AST, handle: &mut Schala, comp: Option<&mut PassDebugArtifact>) -> Result<ast::AST, String> {
  let result = handle.type_context.typecheck(&input);

  comp.map(|comp| {
    comp.add_artifact(match result {
      Ok(ty) => ty.to_string(),
      Err(err) => format!("Type error: {}", err.msg)
    });
  });

  Ok(input)
}

fn ast_reducing(input: ast::AST, handle: &mut Schala, comp: Option<&mut PassDebugArtifact>) -> Result<reduced_ast::ReducedAST, String> {
  let ref symbol_table = handle.symbol_table.borrow();
  let output = input.reduce(symbol_table);
  comp.map(|comp| comp.add_artifact(format!("{:?}", output)));
  Ok(output)
}

fn eval(input: reduced_ast::ReducedAST, handle: &mut Schala, comp: Option<&mut PassDebugArtifact>) -> Result<String, String> {
  comp.map(|comp| comp.add_artifact(handle.state.debug_print()));
  let evaluation_outputs = handle.state.evaluate(input, true);
  let text_output: Result<Vec<String>, String> = evaluation_outputs
    .into_iter()
    .collect();

  let eval_output: Result<String, String> = text_output
    .map(|v| { v.into_iter().intersperse(format!("\n")).collect() });
  eval_output
}

/// Represents lines of source code
struct SourceReference {
  lines: Option<Vec<String>>
}

impl SourceReference {
  fn new() -> SourceReference {
    SourceReference { lines: None }
  }

  fn load_new_source(&mut self, source: &str) {
    //TODO this is a lot of heap allocations - maybe there's a way to make it more efficient?
    self.lines = Some(source.lines().map(|s| s.to_string()).collect()); }

  fn get_line(&self, line: usize) -> String {
    self.lines.as_ref().and_then(|x| x.get(line).map(|s| s.to_string())).unwrap_or(format!("NO LINE FOUND"))
  }
}

struct PassDebugArtifact {
  artifact: Option<String>
}
impl PassDebugArtifact {
  fn add_artifact(&mut self, artifact: String) {
    self.artifact = Some(artifact)
  }
}

impl ProgrammingLanguageInterface for Schala {
  fn get_language_name(&self) -> String { format!("Schala") }
  fn get_source_file_suffix(&self) -> String { format!("schala") }

  fn run_computation(&mut self, request: ComputationRequest) -> ComputationResponse {
    let ComputationRequest { source, debug_requests } = request;

    let mut token_debug_artifact = None;
    let mut parsing_debug_artifact = None;
    let mut symbol_debug_artifact = None;
    let mut typechecking_debug_artifact = None;
    let mut reducing_debug_artifact = None;
    let mut eval_debug_artifact = None;

    self.source_reference.load_new_source(source);
    let main_output: Result<String, String> = tokenizing(source, self, token_debug_artifact)
      .and_then(|tokens| parsing(tokens, self, parsing_debug_artifact))
      .and_then(|ast| symbol_table(ast, self, symbol_debug_artifact))
      .and_then(|ast| typechecking(ast, self, typechecking_debug_artifact))
      .and_then(|ast| ast_reducing(ast, self, reducing_debug_artifact))
      .and_then(|reduced_ast| eval(reduced_ast, self, eval_debug_artifact));

    ComputationResponse {
      main_output,
      global_output_stats: GlobalOutputStats::default(),
      debug_responses: vec![],
    }
  }

  fn request_meta(&mut self, request: LangMetaRequest) -> LangMetaResponse {
    match request {
      LangMetaRequest::StageNames => LangMetaResponse::StageNames(
        vec!["tokenizing".into(), "parsing".into(), "typechecking".into(),
        "ast reduction".into(), "ast-walking evaluation".into()]
      ),
      LangMetaRequest::Docs { source } => self.handle_docs(source),
      LangMetaRequest::Custom { .. } => LangMetaResponse::Custom { kind: format!("not-implemented"), value: format!("") }
    }
  }
}