Interactive parser development example writing.

Author: Gohla

src/4_example/a_1_Cargo.toml

@@ -0,0 +1,14 @@
+[package]
+name = "pie"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+pie_graph = "0.0.1"
+
+[dev-dependencies]
+dev_shared = { path = "../dev_shared" }
+assert_matches = "1"
+pest = "2"
+pest_meta = "2"
+pest_vm = "2"

src/4_example/a_2_main.rs

@@ -0,0 +1,3 @@
+fn main() {
+
+}

src/4_example/a_3_main_parse_mod.rs

@@ -0,0 +1,5 @@
+pub mod parse;
+
+fn main() {
+
+}

src/4_example/a_4_grammar.rs

@@ -0,0 +1,38 @@
+use std::collections::HashSet;
+use std::fmt::Write;
+
+/// Parse programs with a compiled pest grammar.
+#[derive(Clone, Eq, PartialEq, Debug)]
+pub struct CompiledGrammar {
+  rules: Vec<pest_meta::optimizer::OptimizedRule>,
+  rule_names: HashSet<String>,
+}
+
+impl CompiledGrammar {
+  /// Compile the pest grammar from `grammar_text`, using `path` to annotate errors. Returns a [`Self`] instance.
+  ///
+  /// # Errors
+  ///
+  /// Returns `Err(error_string)` when compiling the grammar fails.
+  pub fn new(grammar_text: &str, path: Option<&str>) -> Result<Self, String> {
+    match pest_meta::parse_and_optimize(grammar_text) {
+      Ok((builtin_rules, rules)) => {
+        let mut rule_names = HashSet::with_capacity(builtin_rules.len() + rules.len());
+        rule_names.extend(builtin_rules.iter().map(|s| s.to_string()));
+        rule_names.extend(rules.iter().map(|s| s.name.clone()));
+        Ok(Self { rules, rule_names })
+      },
+      Err(errors) => {
+        let mut error_string = String::new();
+        for mut error in errors {
+          if let Some(path) = path.as_ref() {
+            error = error.with_path(path);
+          }
+          error = error.renamed_rules(pest_meta::parser::rename_meta_rule);
+          let _ = writeln!(error_string, "{}", error); // Ignore error: writing to String cannot fail.
+        }
+        Err(error_string)
+      }
+    }
+  }
+}

src/4_example/a_5_parse.rs

@@ -0,0 +1,64 @@
+use std::collections::HashSet;
+use std::fmt::Write;
+
+/// Parse programs with a compiled pest grammar.
+#[derive(Clone, Eq, PartialEq, Debug)]
+pub struct CompiledGrammar {
+  rules: Vec<pest_meta::optimizer::OptimizedRule>,
+  rule_names: HashSet<String>,
+}
+
+impl CompiledGrammar {
+  /// Compile the pest grammar from `grammar_text`, using `path` to annotate errors. Returns a [`Self`] instance.
+  ///
+  /// # Errors
+  ///
+  /// Returns `Err(error_string)` when compiling the grammar fails.
+  pub fn new(grammar_text: &str, path: Option<&str>) -> Result<Self, String> {
+    match pest_meta::parse_and_optimize(grammar_text) {
+      Ok((builtin_rules, rules)) => {
+        let mut rule_names = HashSet::with_capacity(builtin_rules.len() + rules.len());
+        rule_names.extend(builtin_rules.iter().map(|s| s.to_string()));
+        rule_names.extend(rules.iter().map(|s| s.name.clone()));
+        Ok(Self { rules, rule_names })
+      },
+      Err(errors) => {
+        let mut error_string = String::new();
+        for mut error in errors {
+          if let Some(path) = path.as_ref() {
+            error = error.with_path(path);
+          }
+          error = error.renamed_rules(pest_meta::parser::rename_meta_rule);
+          let _ = writeln!(error_string, "{}", error); // Ignore error: writing to String cannot fail.
+        }
+        Err(error_string)
+      }
+    }
+  }
+
+  /// Parse `program_text` with rule `rule_name` using this compiled grammar, using `path` to annotate errors. Returns
+  /// parsed pairs formatted as a string.
+  ///
+  /// # Errors
+  ///
+  /// Returns `Err(error_string)` when parsing fails.
+  pub fn parse(&self, program_text: &str, rule_name: &str, path: Option<&str>) -> Result<String, String> {
+    if !self.rule_names.contains(rule_name) {
+      let message = format!("rule '{}' was not found", rule_name);
+      return Err(message);
+    }
+    // Note: can't store `Vm` in `CompiledGrammar` because `Vm` is not `Clone` nor `Eq`.
+    let vm = pest_vm::Vm::new(self.rules.clone());
+    match vm.parse(rule_name, program_text) {
+      Ok(pairs) => Ok(format!("{}", pairs)),
+      Err(mut error) => {
+        if let Some(path) = path {
+          error = error.with_path(path);
+        }
+        error = error.renamed_rules(|r| r.to_string());
+        let error_string = format!("{}", error);
+        Err(error_string)
+      }
+    }
+  }
+}

src/4_example/a_6_test.rs

@@ -0,0 +1,32 @@
+
+#[cfg(test)]
+mod tests {
+  use super::*;
+
+  #[test]
+  fn test_compile_parse() -> Result<(), String> {
+    // Grammar compilation failure.
+    let result = CompiledGrammar::new("asd = { fgh } qwe = { rty }", None);
+    assert!(result.is_err());
+    println!("{}", result.unwrap_err());
+
+    // Grammar that parses numbers.
+    let compiled_grammar = CompiledGrammar::new("num = { ASCII_DIGIT+ }", None)?;
+    println!("{:?}", compiled_grammar);
+
+    // Parse failure
+    let result = compiled_grammar.parse("a", "num", None);
+    assert!(result.is_err());
+    println!("{}", result.unwrap_err());
+    // Parse failure due to non-existent rule.
+    let result = compiled_grammar.parse("1", "asd", None);
+    assert!(result.is_err());
+    println!("{}", result.unwrap_err());
+    // Parse success
+    let result = compiled_grammar.parse("1", "num", None);
+    assert!(result.is_ok());
+    println!("{}", result.unwrap());
+
+    Ok(())
+  }
+}

src/4_example/b_1_main_task_mod.rs

@@ -0,0 +1,6 @@
+pub mod parse;
+pub mod task;
+
+fn main() {
+
+}

src/4_example/b_2_tasks_outputs.rs

@@ -0,0 +1,41 @@
+use std::io::Read;
+use std::path::{Path, PathBuf};
+
+use pie::{Context, Task};
+
+use crate::parse::CompiledGrammar;
+
+/// Tasks for compiling a grammar and parsing files with it.
+#[derive(Clone, Eq, PartialEq, Hash, Debug)]
+pub enum Tasks {
+  CompileGrammar { grammar_file_path: PathBuf },
+  Parse { compiled_grammar_task: Box<Tasks>, program_file_path: PathBuf, rule_name: String }
+}
+
+impl Tasks {
+  /// Create a [`Self::CompileGrammar`] task that compiles the grammar in file `grammar_file_path`.
+  pub fn compile_grammar(grammar_file_path: impl Into<PathBuf>) -> Self {
+    Self::CompileGrammar { grammar_file_path: grammar_file_path.into() }
+  }
+
+  /// Create a [`Self::Parse`] task that uses the compiled grammar returned by requiring `compiled_grammar_task` to
+  /// parse the program in file `program_file_path`, starting parsing with `rule_name`.
+  pub fn parse(
+    compiled_grammar_task: &Tasks,
+    program_file_path: impl Into<PathBuf>,
+    rule_name: impl Into<String>
+  ) -> Self {
+    Self::Parse {
+      compiled_grammar_task: Box::new(compiled_grammar_task.clone()),
+      program_file_path: program_file_path.into(),
+      rule_name: rule_name.into()
+    }
+  }
+}
+
+/// Outputs for [`Tasks`].
+#[derive(Clone, Eq, PartialEq, Debug)]
+pub enum Outputs {
+  CompiledGrammar(CompiledGrammar),
+  Parsed(Option<String>)
+}

src/4_example/b_3_require_file.rs

@@ -0,0 +1,11 @@
+
+fn require_file_to_string<C: Context<Tasks>>(context: &mut C, path: impl AsRef<Path>) -> Result<String, String> {
+  let path = path.as_ref();
+  let mut file = context.require_file(path)
+    .map_err(|e| format!("Opening file '{}' for reading failed: {}", path.display(), e))?
+    .ok_or_else(|| format!("File '{}' does not exist", path.display()))?;
+  let mut text = String::new();
+  file.read_to_string(&mut text)
+    .map_err(|e| format!("Reading file '{}' failed: {}", path.display(), e))?;
+  Ok(text)
+}

src/4_example/b_4_task.rs

@@ -0,0 +1,24 @@
+
+impl Task for Tasks {
+  type Output = Result<Outputs, String>;
+
+  fn execute<C: Context<Self>>(&self, context: &mut C) -> Self::Output {
+    match self {
+      Tasks::CompileGrammar { grammar_file_path } => {
+        let grammar_text = require_file_to_string(context, grammar_file_path)?;
+        let compiled_grammar = CompiledGrammar::new(&grammar_text, Some(grammar_file_path.to_string_lossy().as_ref()))?;
+        Ok(Outputs::CompiledGrammar(compiled_grammar))
+      }
+      Tasks::Parse { compiled_grammar_task, program_file_path, rule_name } => {
+        let Ok(Outputs::CompiledGrammar(compiled_grammar)) = context.require_task(compiled_grammar_task.as_ref()) else {
+          // Return `None` if compiling grammar failed. Don't propagate the error, otherwise the error would be
+          // duplicated for all `Parse` tasks.
+          return Ok(Outputs::Parsed(None));
+        };
+        let program_text = require_file_to_string(context, program_file_path)?;
+        let output = compiled_grammar.parse(&program_text, rule_name, Some(program_file_path.to_string_lossy().as_ref()))?;
+        Ok(Outputs::Parsed(Some(output)))
+      }
+    }
+  }
+}

src/4_example/c_1_Cargo.toml

@@ -0,0 +1,15 @@
+[package]
+name = "pie"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+pie_graph = "0.0.1"
+
+[dev-dependencies]
+dev_shared = { path = "../dev_shared" }
+assert_matches = "1"
+pest = "2"
+pest_meta = "2"
+pest_vm = "2"
+clap = { version = "4", features = ["derive"] }

src/4_example/c_2_cli.rs

@@ -0,0 +1,20 @@
+use std::path::PathBuf;
+
+use clap::Parser;
+
+pub mod parse;
+pub mod task;
+
+#[derive(Parser)]
+pub struct Args {
+  /// Path to the pest grammar file.
+  grammar_file_path: PathBuf,
+  /// Rule name (from the pest grammar file) used to parse program files.
+  rule_name: String,
+  /// Paths to program files to parse with the pest grammar.
+  program_file_paths: Vec<PathBuf>
+}
+
+fn main() {
+  let args = Args::parse();
+}

src/4_example/c_3_compile_parse.rs

@@ -0,0 +1,51 @@
+use std::fmt::Write;
+use std::path::PathBuf;
+
+use clap::Parser;
+use pie::Pie;
+use pie::tracker::writing::WritingTracker;
+
+use crate::task::{Outputs, Tasks};
+
+pub mod parse;
+pub mod task;
+
+#[derive(Parser)]
+pub struct Args {
+  /// Path to the pest grammar file.
+  grammar_file_path: PathBuf,
+  /// Rule name (from the pest grammar file) used to parse program files.
+  rule_name: String,
+  /// Paths to program files to parse with the pest grammar.
+  program_file_paths: Vec<PathBuf>
+}
+
+fn main() {
+  let args = Args::parse();
+  compile_grammar_and_parse(args);
+}
+
+fn compile_grammar_and_parse(args: Args) {
+  let mut pie = Pie::with_tracker(WritingTracker::with_stderr());
+
+  let mut session = pie.new_session();
+  let mut errors = String::new();
+
+  let compile_grammar_task = Tasks::compile_grammar(&args.grammar_file_path);
+  if let Err(error) = session.require(&compile_grammar_task) {
+    let _ = writeln!(errors, "{}", error); // Ignore error: writing to String cannot fail.
+  }
+
+  for path in args.program_file_paths {
+    let task = Tasks::parse(&compile_grammar_task, &path, &args.rule_name);
+    match session.require(&task) {
+      Err(error) => { let _ = writeln!(errors, "{}", error); }
+      Ok(Outputs::Parsed(Some(output))) => println!("Parsing '{}' succeeded: {}", path.display(), output),
+      _ => {}
+    }
+  }
+
+  if !errors.is_empty() {
+    println!("Errors:\n{}", errors);
+  }
+}

src/4_example/c_4_grammar.pest

@@ -0,0 +1,5 @@
+num = { ASCII_DIGIT+ }
+
+main = { SOI ~ num ~ EOI }
+
+WHITESPACE = _{ " " | "\t" | "\n" | "\r" }

src/4_example/c_4_test_1.txt

@@ -0,0 +1 @@
+42

src/4_example/c_4_test_2.txt

@@ -0,0 +1 @@
+foo