Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support left recursion syntax #533

Draft
wants to merge 9 commits into
base: master
Choose a base branch
from
Draft

Support left recursion syntax #533

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
40c9900
add ParserState Clone traits
Abdillah Aug 21, 2021
dd9aa6a
add recursive processor
Abdillah Aug 21, 2021
b3a7532
add description docs and remove debugging code
Abdillah Aug 21, 2021
6aabb1c
fix leftrecursive test symbol should not be optional
Abdillah Aug 26, 2021
b2fb874
change leftrecursive mark tests to contains a whole rule
Abdillah Aug 26, 2021
3aca137
change leftrecursive mark impl to contains a whole rule
Abdillah Aug 26, 2021
3ea837a
add recursive modifier syntax on rule
Abdillah Aug 28, 2021
1800161
add recursive modifier code generation
Abdillah Aug 28, 2021
6390c30
add loop guard when recursion not 'recurse'
Abdillah Aug 28, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
43 changes: 43 additions & 0 deletions generator/src/generator.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -240,6 +240,16 @@ fn generate_rule(rule: OptimizedRule) -> TokenStream {
generate_expr(rule.expr)
};

let expr = if rule.rec {
quote! {
state.recursive(Rule::#name, |state| {
#expr
})
}
} else {
expr
};

match rule.ty {
RuleType::Normal => quote! {
#[inline]
Expand Down Expand Up @@ -636,6 +646,7 @@ mod tests {
let rules = vec![OptimizedRule {
name: "f".to_owned(),
ty: RuleType::Normal,
rec: false,
expr: OptimizedExpr::Ident("g".to_owned()),
}];

Expand Down Expand Up @@ -926,13 +937,45 @@ mod tests {
);
}

#[test]
fn generate_from_recursive_rule() {
let rule = OptimizedRule {
name: "f".to_owned(),
ty: RuleType::Normal,
rec: true,
expr: OptimizedExpr::Choice(
Box::new(OptimizedExpr::Ident("f".to_owned())),
Box::new(OptimizedExpr::Ident("g".to_owned())),
),
};

assert_eq!(
generate_rule(rule).to_string(),
quote! {
#[inline]
#[allow (non_snake_case , unused_variables)]
pub fn f (state: Box<::pest::ParserState<Rule>>) -> ::pest::ParseResult<Box<::pest::ParserState<Rule>>> {
state.rule(Rule::f, |state| {
state.recursive(Rule::f, |state| {
self::f(state).or_else(|state| {
self::g(state)
})
})
})
}
}
.to_string()
);
}

#[test]
fn generate_complete() {
let name = Ident::new("MyParser", Span::call_site());
let generics = Generics::default();
let rules = vec![OptimizedRule {
name: "a".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: OptimizedExpr::Str("b".to_owned()),
}];
let defaults = vec!["ANY"];
Expand Down
1 change: 1 addition & 0 deletions generator/src/lib.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,6 +64,7 @@ pub fn derive_parser(input: TokenStream, include_grammar: bool) -> TokenStream {
Rule::grammar_rule => "rule".to_owned(),
Rule::_push => "PUSH".to_owned(),
Rule::assignment_operator => "`=`".to_owned(),
Rule::recursive_modifier => "`*`".to_owned(),
Rule::silent_modifier => "`_`".to_owned(),
Rule::atomic_modifier => "`@`".to_owned(),
Rule::compound_atomic_modifier => "`$`".to_owned(),
Expand Down
1 change: 1 addition & 0 deletions meta/src/ast.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@
pub struct Rule {
pub name: String,
pub ty: RuleType,
pub rec: bool,
pub expr: Expr,
}

Expand Down
3 changes: 2 additions & 1 deletion meta/src/grammar.pest
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@
grammar_rules = _{ SOI ~ grammar_rule+ ~ EOI }

grammar_rule = {
identifier ~ assignment_operator ~ modifier? ~
identifier ~ assignment_operator ~ recursive_modifier? ~ modifier? ~
opening_brace ~ expression ~ closing_brace
}

Expand All @@ -29,6 +29,7 @@ modifier = _{
non_atomic_modifier
}

recursive_modifier = { "*" }
silent_modifier = { "_" }
atomic_modifier = { "@" }
compound_atomic_modifier = { "$" }
Expand Down
3 changes: 2 additions & 1 deletion meta/src/optimizer/concatenator.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,10 +10,11 @@
use ast::*;

pub fn concatenate(rule: Rule) -> Rule {
let Rule { name, ty, expr } = rule;
let Rule { name, ty, rec, expr } = rule;
Rule {
name,
ty,
rec,
expr: expr.map_bottom_up(|expr| {
if ty == RuleType::Atomic {
// TODO: Use box syntax when it gets stabilized.
Expand Down
3 changes: 2 additions & 1 deletion meta/src/optimizer/factorizer.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,10 +10,11 @@
use ast::*;

pub fn factor(rule: Rule) -> Rule {
let Rule { name, ty, expr } = rule;
let Rule { name, ty, rec, expr } = rule;
Rule {
name,
ty,
rec,
expr: expr.map_top_down(|expr| {
// TODO: Use box syntax when it gets stabilized.
match expr {
Expand Down
3 changes: 2 additions & 1 deletion meta/src/optimizer/lister.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,10 +10,11 @@
use ast::*;

pub fn list(rule: Rule) -> Rule {
let Rule { name, ty, expr } = rule;
let Rule { name, ty, rec, expr } = rule;
Rule {
name,
ty,
rec,
expr: expr.map_bottom_up(|expr| {
// TODO: Use box syntax when it gets stabilized.
match expr {
Expand Down
26 changes: 26 additions & 0 deletions meta/src/optimizer/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -76,6 +76,7 @@ fn rule_to_optimized_rule(rule: Rule) -> OptimizedRule {
OptimizedRule {
name: rule.name,
ty: rule.ty,
rec: rule.rec,
expr: to_optimized(rule.expr),
}
}
Expand All @@ -91,6 +92,7 @@ fn to_hash_map(rules: &[OptimizedRule]) -> HashMap<String, OptimizedExpr> {
pub struct OptimizedRule {
pub name: String,
pub ty: RuleType,
pub rec: bool,
pub expr: OptimizedExpr,
}

Expand Down Expand Up @@ -285,6 +287,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Normal,
rec: false,
expr: box_tree!(Choice(
Choice(
Choice(Str(String::from("a")), Str(String::from("b"))),
Expand All @@ -299,6 +302,7 @@ mod tests {
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Normal,
rec: false,
expr: box_tree!(Choice(
Str(String::from("a")),
Choice(
Expand All @@ -319,6 +323,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: box_tree!(Rep(Seq(
NegPred(Choice(Str(String::from("a")), Str(String::from("b")))),
Ident(String::from("ANY"))
Expand All @@ -328,6 +333,7 @@ mod tests {
let skipped = vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: OptimizedExpr::Skip(vec![String::from("a"), String::from("b")]),
}];

Expand All @@ -341,6 +347,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: box_tree!(Seq(
Seq(Str(String::from("a")), Str(String::from("b"))),
Seq(Str(String::from("c")), Str(String::from("d")))
Expand All @@ -350,6 +357,7 @@ mod tests {
let concatenated = vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: OptimizedExpr::Str(String::from("abcd")),
}];

Expand All @@ -361,13 +369,15 @@ mod tests {
let rules = vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: Expr::RepExact(Box::new(Expr::Ident(String::from("a"))), 3),
}];
let unrolled = {
use optimizer::OptimizedExpr::*;
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: box_tree!(Seq(
Ident(String::from("a")),
Seq(Ident(String::from("a")), Ident(String::from("a")))
Expand All @@ -383,13 +393,15 @@ mod tests {
let rules = vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: Expr::RepMax(Box::new(Expr::Str("a".to_owned())), 3),
}];
let unrolled = {
use optimizer::OptimizedExpr::*;
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: box_tree!(Seq(
Opt(Str(String::from("a"))),
Seq(Opt(Str(String::from("a"))), Opt(Str(String::from("a"))))
Expand All @@ -405,13 +417,15 @@ mod tests {
let rules = vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: Expr::RepMin(Box::new(Expr::Str("a".to_owned())), 2),
}];
let unrolled = {
use optimizer::OptimizedExpr::*;
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: box_tree!(Seq(
Str(String::from("a")),
Seq(Str(String::from("a")), Rep(Str(String::from("a"))))
Expand All @@ -427,13 +441,15 @@ mod tests {
let rules = vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: Expr::RepMinMax(Box::new(Expr::Str("a".to_owned())), 2, 3),
}];
let unrolled = {
use optimizer::OptimizedExpr::*;
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
/* TODO possible room for improvement here:
* if the sequences were rolled out in the opposite
* order, we could further optimize the strings
Expand All @@ -458,6 +474,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: box_tree!(Seq(
Seq(Insens(String::from("a")), Insens(String::from("b"))),
Seq(Insens(String::from("c")), Insens(String::from("d")))
Expand All @@ -467,6 +484,7 @@ mod tests {
let concatenated = vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Atomic,
rec: false,
expr: OptimizedExpr::Insens(String::from("abcd")),
}];

Expand All @@ -480,6 +498,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Choice(
Seq(
Ident(String::from("a")),
Expand All @@ -497,6 +516,7 @@ mod tests {
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Seq(
Ident(String::from("a")),
Seq(
Expand All @@ -517,6 +537,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Choice(
Seq(Ident(String::from("a")), Ident(String::from("b"))),
Ident(String::from("a"))
Expand All @@ -528,6 +549,7 @@ mod tests {
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Seq(Ident(String::from("a")), Opt(Ident(String::from("b"))))),
}]
};
Expand All @@ -542,6 +564,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Choice(
Ident(String::from("a")),
Seq(Ident(String::from("a")), Ident(String::from("b")))
Expand All @@ -553,6 +576,7 @@ mod tests {
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Ident(String::from("a"))),
}]
};
Expand All @@ -567,6 +591,7 @@ mod tests {
vec![Rule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Seq(
Rep(Seq(Ident(String::from("a")), Ident(String::from("b")))),
Ident(String::from("a"))
Expand All @@ -578,6 +603,7 @@ mod tests {
vec![OptimizedRule {
name: "rule".to_owned(),
ty: RuleType::Silent,
rec: false,
expr: box_tree!(Seq(
Ident(String::from("a")),
Rep(Seq(Ident(String::from("b")), Ident(String::from("a"))))
Expand Down
Loading