diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-17 12:02:58 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-17 12:02:58 +0000 |
commit | 698f8c2f01ea549d77d7dc3338a12e04c11057b9 (patch) | |
tree | 173a775858bd501c378080a10dca74132f05bc50 /vendor/pest/src/iterators | |
parent | Initial commit. (diff) | |
download | rustc-698f8c2f01ea549d77d7dc3338a12e04c11057b9.tar.xz rustc-698f8c2f01ea549d77d7dc3338a12e04c11057b9.zip |
Adding upstream version 1.64.0+dfsg1.upstream/1.64.0+dfsg1
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'vendor/pest/src/iterators')
-rw-r--r-- | vendor/pest/src/iterators/flat_pairs.rs | 177 | ||||
-rw-r--r-- | vendor/pest/src/iterators/mod.rs | 22 | ||||
-rw-r--r-- | vendor/pest/src/iterators/pair.rs | 389 | ||||
-rw-r--r-- | vendor/pest/src/iterators/pairs.rs | 419 | ||||
-rw-r--r-- | vendor/pest/src/iterators/queueable_token.rs | 27 | ||||
-rw-r--r-- | vendor/pest/src/iterators/tokens.rs | 144 |
6 files changed, 1178 insertions, 0 deletions
diff --git a/vendor/pest/src/iterators/flat_pairs.rs b/vendor/pest/src/iterators/flat_pairs.rs new file mode 100644 index 000000000..bb5bbb185 --- /dev/null +++ b/vendor/pest/src/iterators/flat_pairs.rs @@ -0,0 +1,177 @@ +// pest. The Elegant Parser +// Copyright (c) 2018 Dragoș Tiselice +// +// Licensed under the Apache License, Version 2.0 +// <LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0> or the MIT +// license <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. All files in the project carrying such notice may not be copied, +// modified, or distributed except according to those terms. + +use std::fmt; +use std::rc::Rc; + +use super::pair::{self, Pair}; +use super::queueable_token::QueueableToken; +use super::tokens::{self, Tokens}; +use RuleType; + +/// An iterator over [`Pair`]s. It is created by [`Pairs::flatten`]. +/// +/// [`Pair`]: struct.Pair.html +/// [`Pairs::flatten`]: struct.Pairs.html#method.flatten +pub struct FlatPairs<'i, R> { + /// # Safety + /// + /// All `QueueableToken`s' `input_pos` must be valid character boundary indices into `input`. + queue: Rc<Vec<QueueableToken<R>>>, + input: &'i str, + start: usize, + end: usize, +} + +/// # Safety +/// +/// All `QueueableToken`s' `input_pos` must be valid character boundary indices into `input`. +pub unsafe fn new<R: RuleType>( + queue: Rc<Vec<QueueableToken<R>>>, + input: &str, + start: usize, + end: usize, +) -> FlatPairs<R> { + FlatPairs { + queue, + input, + start, + end, + } +} + +impl<'i, R: RuleType> FlatPairs<'i, R> { + /// Returns the `Tokens` for these pairs. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a + /// } + /// + /// let input = ""; + /// let pairs = pest::state(input, |state| { + /// // generating Token pair with Rule::a ... + /// # state.rule(Rule::a, |s| Ok(s)) + /// }).unwrap(); + /// let tokens: Vec<_> = pairs.flatten().tokens().collect(); + /// + /// assert_eq!(tokens.len(), 2); + /// ``` + #[inline] + pub fn tokens(self) -> Tokens<'i, R> { + tokens::new(self.queue, self.input, self.start, self.end) + } + + fn next_start(&mut self) { + self.start += 1; + + while self.start < self.end && !self.is_start(self.start) { + self.start += 1; + } + } + + fn next_start_from_end(&mut self) { + self.end -= 1; + + while self.end >= self.start && !self.is_start(self.end) { + self.end -= 1; + } + } + + fn is_start(&self, index: usize) -> bool { + match self.queue[index] { + QueueableToken::Start { .. } => true, + QueueableToken::End { .. } => false, + } + } +} + +impl<'i, R: RuleType> Iterator for FlatPairs<'i, R> { + type Item = Pair<'i, R>; + + fn next(&mut self) -> Option<Self::Item> { + if self.start >= self.end { + return None; + } + + let pair = unsafe { pair::new(Rc::clone(&self.queue), self.input, self.start) }; + + self.next_start(); + + Some(pair) + } +} + +impl<'i, R: RuleType> DoubleEndedIterator for FlatPairs<'i, R> { + fn next_back(&mut self) -> Option<Self::Item> { + if self.end <= self.start { + return None; + } + + self.next_start_from_end(); + + let pair = unsafe { pair::new(Rc::clone(&self.queue), self.input, self.end) }; + + Some(pair) + } +} + +impl<'i, R: RuleType> fmt::Debug for FlatPairs<'i, R> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("FlatPairs") + .field("pairs", &self.clone().collect::<Vec<_>>()) + .finish() + } +} + +impl<'i, R: Clone> Clone for FlatPairs<'i, R> { + fn clone(&self) -> FlatPairs<'i, R> { + FlatPairs { + queue: Rc::clone(&self.queue), + input: self.input, + start: self.start, + end: self.end, + } + } +} + +#[cfg(test)] +mod tests { + use super::super::super::macros::tests::*; + use super::super::super::Parser; + + #[test] + fn iter_for_flat_pairs() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + + assert_eq!( + pairs.flatten().map(|p| p.as_rule()).collect::<Vec<Rule>>(), + vec![Rule::a, Rule::b, Rule::c] + ); + } + + #[test] + fn double_ended_iter_for_flat_pairs() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + assert_eq!( + pairs + .flatten() + .rev() + .map(|p| p.as_rule()) + .collect::<Vec<Rule>>(), + vec![Rule::c, Rule::b, Rule::a] + ); + } +} diff --git a/vendor/pest/src/iterators/mod.rs b/vendor/pest/src/iterators/mod.rs new file mode 100644 index 000000000..1a7896371 --- /dev/null +++ b/vendor/pest/src/iterators/mod.rs @@ -0,0 +1,22 @@ +// pest. The Elegant Parser +// Copyright (c) 2018 Dragoș Tiselice +// +// Licensed under the Apache License, Version 2.0 +// <LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0> or the MIT +// license <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. All files in the project carrying such notice may not be copied, +// modified, or distributed except according to those terms. + +//! Types and iterators for parser output. + +mod flat_pairs; +mod pair; +pub(crate) mod pairs; +mod queueable_token; +mod tokens; + +pub use self::flat_pairs::FlatPairs; +pub use self::pair::Pair; +pub use self::pairs::Pairs; +pub(crate) use self::queueable_token::QueueableToken; +pub use self::tokens::Tokens; diff --git a/vendor/pest/src/iterators/pair.rs b/vendor/pest/src/iterators/pair.rs new file mode 100644 index 000000000..88844a353 --- /dev/null +++ b/vendor/pest/src/iterators/pair.rs @@ -0,0 +1,389 @@ +// pest. The Elegant Parser +// Copyright (c) 2018 Dragoș Tiselice +// +// Licensed under the Apache License, Version 2.0 +// <LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0> or the MIT +// license <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. All files in the project carrying such notice may not be copied, +// modified, or distributed except according to those terms. + +use std::fmt; +use std::hash::{Hash, Hasher}; +use std::ptr; +use std::rc::Rc; +use std::str; + +#[cfg(feature = "pretty-print")] +use serde::ser::SerializeStruct; + +use super::pairs::{self, Pairs}; +use super::queueable_token::QueueableToken; +use super::tokens::{self, Tokens}; +use span::{self, Span}; +use RuleType; + +/// A matching pair of [`Token`]s and everything between them. +/// +/// A matching `Token` pair is formed by a `Token::Start` and a subsequent `Token::End` with the +/// same `Rule`, with the condition that all `Token`s between them can form such pairs as well. +/// This is similar to the [brace matching problem](https://en.wikipedia.org/wiki/Brace_matching) in +/// editors. +/// +/// [`Token`]: ../enum.Token.html +#[derive(Clone)] +pub struct Pair<'i, R> { + /// # Safety + /// + /// All `QueueableToken`s' `input_pos` must be valid character boundary indices into `input`. + queue: Rc<Vec<QueueableToken<R>>>, + input: &'i str, + /// Token index into `queue`. + start: usize, +} + +/// # Safety +/// +/// All `QueueableToken`s' `input_pos` must be valid character boundary indices into `input`. +pub unsafe fn new<R: RuleType>( + queue: Rc<Vec<QueueableToken<R>>>, + input: &str, + start: usize, +) -> Pair<R> { + Pair { + queue, + input, + start, + } +} + +impl<'i, R: RuleType> Pair<'i, R> { + /// Returns the `Rule` of the `Pair`. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a + /// } + /// + /// let input = ""; + /// let pair = pest::state(input, |state| { + /// // generating Token pair with Rule::a ... + /// # state.rule(Rule::a, |s| Ok(s)) + /// }).unwrap().next().unwrap(); + /// + /// assert_eq!(pair.as_rule(), Rule::a); + /// ``` + #[inline] + pub fn as_rule(&self) -> R { + match self.queue[self.pair()] { + QueueableToken::End { rule, .. } => rule, + _ => unreachable!(), + } + } + + /// Captures a slice from the `&str` defined by the token `Pair`. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// ab + /// } + /// + /// let input = "ab"; + /// let pair = pest::state(input, |state| { + /// // generating Token pair with Rule::ab ... + /// # state.rule(Rule::ab, |s| s.match_string("ab")) + /// }).unwrap().next().unwrap(); + /// + /// assert_eq!(pair.as_str(), "ab"); + /// ``` + #[inline] + pub fn as_str(&self) -> &'i str { + let start = self.pos(self.start); + let end = self.pos(self.pair()); + + // Generated positions always come from Positions and are UTF-8 borders. + &self.input[start..end] + } + + /// Returns the `Span` defined by the `Pair`, consuming it. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// ab + /// } + /// + /// let input = "ab"; + /// let pair = pest::state(input, |state| { + /// // generating Token pair with Rule::ab ... + /// # state.rule(Rule::ab, |s| s.match_string("ab")) + /// }).unwrap().next().unwrap(); + /// + /// assert_eq!(pair.into_span().as_str(), "ab"); + /// ``` + #[inline] + #[deprecated(since = "2.0.0", note = "Please use `as_span` instead")] + pub fn into_span(self) -> Span<'i> { + self.as_span() + } + + /// Returns the `Span` defined by the `Pair`, **without** consuming it. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// ab + /// } + /// + /// let input = "ab"; + /// let pair = pest::state(input, |state| { + /// // generating Token pair with Rule::ab ... + /// # state.rule(Rule::ab, |s| s.match_string("ab")) + /// }).unwrap().next().unwrap(); + /// + /// assert_eq!(pair.as_span().as_str(), "ab"); + /// ``` + #[inline] + pub fn as_span(&self) -> Span<'i> { + let start = self.pos(self.start); + let end = self.pos(self.pair()); + + // Generated positions always come from Positions and are UTF-8 borders. + unsafe { span::Span::new_unchecked(self.input, start, end) } + } + + /// Returns the inner `Pairs` between the `Pair`, consuming it. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a + /// } + /// + /// let input = ""; + /// let pair = pest::state(input, |state| { + /// // generating Token pair with Rule::a ... + /// # state.rule(Rule::a, |s| Ok(s)) + /// }).unwrap().next().unwrap(); + /// + /// assert!(pair.into_inner().next().is_none()); + /// ``` + #[inline] + pub fn into_inner(self) -> Pairs<'i, R> { + let pair = self.pair(); + + pairs::new(self.queue, self.input, self.start + 1, pair) + } + + /// Returns the `Tokens` for the `Pair`. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a + /// } + /// + /// let input = ""; + /// let pair = pest::state(input, |state| { + /// // generating Token pair with Rule::a ... + /// # state.rule(Rule::a, |s| Ok(s)) + /// }).unwrap().next().unwrap(); + /// let tokens: Vec<_> = pair.tokens().collect(); + /// + /// assert_eq!(tokens.len(), 2); + /// ``` + #[inline] + pub fn tokens(self) -> Tokens<'i, R> { + let end = self.pair(); + + tokens::new(self.queue, self.input, self.start, end + 1) + } + + /// Generates a string that stores the lexical information of `self` in + /// a pretty-printed JSON format. + #[cfg(feature = "pretty-print")] + pub fn to_json(&self) -> String { + ::serde_json::to_string_pretty(self).expect("Failed to pretty-print Pair to json.") + } + + fn pair(&self) -> usize { + match self.queue[self.start] { + QueueableToken::Start { + end_token_index, .. + } => end_token_index, + _ => unreachable!(), + } + } + + fn pos(&self, index: usize) -> usize { + match self.queue[index] { + QueueableToken::Start { input_pos, .. } | QueueableToken::End { input_pos, .. } => { + input_pos + } + } + } +} + +impl<'i, R: RuleType> Pairs<'i, R> { + /// Create a new `Pairs` iterator containing just the single `Pair`. + pub fn single(pair: Pair<'i, R>) -> Self { + let end = pair.pair(); + pairs::new(pair.queue, pair.input, pair.start, end) + } +} + +impl<'i, R: RuleType> fmt::Debug for Pair<'i, R> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_struct("Pair") + .field("rule", &self.as_rule()) + .field("span", &self.as_span()) + .field("inner", &self.clone().into_inner().collect::<Vec<_>>()) + .finish() + } +} + +impl<'i, R: RuleType> fmt::Display for Pair<'i, R> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + let rule = self.as_rule(); + let start = self.pos(self.start); + let end = self.pos(self.pair()); + let mut pairs = self.clone().into_inner().peekable(); + + if pairs.peek().is_none() { + write!(f, "{:?}({}, {})", rule, start, end) + } else { + write!( + f, + "{:?}({}, {}, [{}])", + rule, + start, + end, + pairs + .map(|pair| format!("{}", pair)) + .collect::<Vec<_>>() + .join(", ") + ) + } + } +} + +impl<'i, R: PartialEq> PartialEq for Pair<'i, R> { + fn eq(&self, other: &Pair<'i, R>) -> bool { + Rc::ptr_eq(&self.queue, &other.queue) + && ptr::eq(self.input, other.input) + && self.start == other.start + } +} + +impl<'i, R: Eq> Eq for Pair<'i, R> {} + +impl<'i, R: Hash> Hash for Pair<'i, R> { + fn hash<H: Hasher>(&self, state: &mut H) { + (&*self.queue as *const Vec<QueueableToken<R>>).hash(state); + (self.input as *const str).hash(state); + self.start.hash(state); + } +} + +#[cfg(feature = "pretty-print")] +impl<'i, R: RuleType> ::serde::Serialize for Pair<'i, R> { + fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error> + where + S: ::serde::Serializer, + { + let start = self.pos(self.start); + let end = self.pos(self.pair()); + let rule = format!("{:?}", self.as_rule()); + let inner = self.clone().into_inner(); + + let mut ser = serializer.serialize_struct("Pairs", 3)?; + ser.serialize_field("pos", &(start, end))?; + ser.serialize_field("rule", &rule)?; + + if inner.peek().is_none() { + ser.serialize_field("inner", &self.as_str())?; + } else { + ser.serialize_field("inner", &inner)?; + } + + ser.end() + } +} + +#[cfg(test)] +mod tests { + use macros::tests::*; + use parser::Parser; + + #[test] + #[cfg(feature = "pretty-print")] + fn test_pretty_print() { + let pair = AbcParser::parse(Rule::a, "abcde").unwrap().next().unwrap(); + + let expected = r#"{ + "pos": [ + 0, + 3 + ], + "rule": "a", + "inner": { + "pos": [ + 1, + 2 + ], + "pairs": [ + { + "pos": [ + 1, + 2 + ], + "rule": "b", + "inner": "b" + } + ] + } +}"#; + + assert_eq!(expected, pair.to_json()); + } + + #[test] + fn pair_into_inner() { + let pair = AbcParser::parse(Rule::a, "abcde").unwrap().next().unwrap(); // the tokens a(b()) + + let pairs = pair.into_inner(); // the tokens b() + + assert_eq!(2, pairs.tokens().count()); + } +} diff --git a/vendor/pest/src/iterators/pairs.rs b/vendor/pest/src/iterators/pairs.rs new file mode 100644 index 000000000..abae123e1 --- /dev/null +++ b/vendor/pest/src/iterators/pairs.rs @@ -0,0 +1,419 @@ +// pest. The Elegant Parser +// Copyright (c) 2018 Dragoș Tiselice +// +// Licensed under the Apache License, Version 2.0 +// <LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0> or the MIT +// license <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. All files in the project carrying such notice may not be copied, +// modified, or distributed except according to those terms. + +use std::fmt; +use std::hash::{Hash, Hasher}; +use std::ptr; +use std::rc::Rc; +use std::str; + +#[cfg(feature = "pretty-print")] +use serde::ser::SerializeStruct; + +use super::flat_pairs::{self, FlatPairs}; +use super::pair::{self, Pair}; +use super::queueable_token::QueueableToken; +use super::tokens::{self, Tokens}; +use RuleType; + +/// An iterator over [`Pair`]s. It is created by [`pest::state`] and [`Pair::into_inner`]. +/// +/// [`Pair`]: struct.Pair.html +/// [`pest::state`]: ../fn.state.html +/// [`Pair::into_inner`]: struct.Pair.html#method.into_inner +#[derive(Clone)] +pub struct Pairs<'i, R> { + queue: Rc<Vec<QueueableToken<R>>>, + input: &'i str, + start: usize, + end: usize, +} + +pub fn new<R: RuleType>( + queue: Rc<Vec<QueueableToken<R>>>, + input: &str, + start: usize, + end: usize, +) -> Pairs<R> { + Pairs { + queue, + input, + start, + end, + } +} + +impl<'i, R: RuleType> Pairs<'i, R> { + /// Captures a slice from the `&str` defined by the starting position of the first token `Pair` + /// and the ending position of the last token `Pair` of the `Pairs`. This also captures + /// the input between those two token `Pair`s. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a, + /// b + /// } + /// + /// let input = "a b"; + /// let pairs = pest::state(input, |state| { + /// // generating Token pairs with Rule::a and Rule::b ... + /// # state.rule(Rule::a, |s| s.match_string("a")).and_then(|s| s.skip(1)) + /// # .and_then(|s| s.rule(Rule::b, |s| s.match_string("b"))) + /// }).unwrap(); + /// + /// assert_eq!(pairs.as_str(), "a b"); + /// ``` + #[inline] + pub fn as_str(&self) -> &'i str { + if self.start < self.end { + let start = self.pos(self.start); + let end = self.pos(self.end - 1); + // Generated positions always come from Positions and are UTF-8 borders. + &self.input[start..end] + } else { + "" + } + } + + /// Captures inner token `Pair`s and concatenates resulting `&str`s. This does not capture + /// the input between token `Pair`s. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a, + /// b + /// } + /// + /// let input = "a b"; + /// let pairs = pest::state(input, |state| { + /// // generating Token pairs with Rule::a and Rule::b ... + /// # state.rule(Rule::a, |s| s.match_string("a")).and_then(|s| s.skip(1)) + /// # .and_then(|s| s.rule(Rule::b, |s| s.match_string("b"))) + /// }).unwrap(); + /// + /// assert_eq!(pairs.concat(), "ab"); + /// ``` + #[inline] + pub fn concat(&self) -> String { + self.clone() + .fold(String::new(), |string, pair| string + pair.as_str()) + } + + /// Flattens the `Pairs`. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a, + /// b + /// } + /// + /// let input = ""; + /// let pairs = pest::state(input, |state| { + /// // generating nested Token pair with Rule::b inside Rule::a + /// # state.rule(Rule::a, |state| { + /// # state.rule(Rule::b, |s| Ok(s)) + /// # }) + /// }).unwrap(); + /// let tokens: Vec<_> = pairs.flatten().tokens().collect(); + /// + /// assert_eq!(tokens.len(), 4); + /// ``` + #[inline] + pub fn flatten(self) -> FlatPairs<'i, R> { + unsafe { flat_pairs::new(self.queue, self.input, self.start, self.end) } + } + + /// Returns the `Tokens` for the `Pairs`. + /// + /// # Examples + /// + /// ``` + /// # use std::rc::Rc; + /// # use pest; + /// # #[allow(non_camel_case_types)] + /// # #[derive(Clone, Copy, Debug, Eq, Hash, Ord, PartialEq, PartialOrd)] + /// enum Rule { + /// a + /// } + /// + /// let input = ""; + /// let pairs = pest::state(input, |state| { + /// // generating Token pair with Rule::a ... + /// # state.rule(Rule::a, |s| Ok(s)) + /// }).unwrap(); + /// let tokens: Vec<_> = pairs.tokens().collect(); + /// + /// assert_eq!(tokens.len(), 2); + /// ``` + #[inline] + pub fn tokens(self) -> Tokens<'i, R> { + tokens::new(self.queue, self.input, self.start, self.end) + } + + /// Peek at the first inner `Pair` without changing the position of this iterator. + #[inline] + pub fn peek(&self) -> Option<Pair<'i, R>> { + if self.start < self.end { + Some(unsafe { pair::new(Rc::clone(&self.queue), self.input, self.start) }) + } else { + None + } + } + + /// Generates a string that stores the lexical information of `self` in + /// a pretty-printed JSON format. + #[cfg(feature = "pretty-print")] + pub fn to_json(&self) -> String { + ::serde_json::to_string_pretty(self).expect("Failed to pretty-print Pairs to json.") + } + + fn pair(&self) -> usize { + match self.queue[self.start] { + QueueableToken::Start { + end_token_index, .. + } => end_token_index, + _ => unreachable!(), + } + } + + fn pair_from_end(&self) -> usize { + match self.queue[self.end - 1] { + QueueableToken::End { + start_token_index, .. + } => start_token_index, + _ => unreachable!(), + } + } + + fn pos(&self, index: usize) -> usize { + match self.queue[index] { + QueueableToken::Start { input_pos, .. } | QueueableToken::End { input_pos, .. } => { + input_pos + } + } + } +} + +impl<'i, R: RuleType> Iterator for Pairs<'i, R> { + type Item = Pair<'i, R>; + + fn next(&mut self) -> Option<Self::Item> { + let pair = self.peek()?; + self.start = self.pair() + 1; + Some(pair) + } +} + +impl<'i, R: RuleType> DoubleEndedIterator for Pairs<'i, R> { + fn next_back(&mut self) -> Option<Self::Item> { + if self.end <= self.start { + return None; + } + + self.end = self.pair_from_end(); + + let pair = unsafe { pair::new(Rc::clone(&self.queue), self.input, self.end) }; + + Some(pair) + } +} + +impl<'i, R: RuleType> fmt::Debug for Pairs<'i, R> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_list().entries(self.clone()).finish() + } +} + +impl<'i, R: RuleType> fmt::Display for Pairs<'i, R> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + write!( + f, + "[{}]", + self.clone() + .map(|pair| format!("{}", pair)) + .collect::<Vec<_>>() + .join(", ") + ) + } +} + +impl<'i, R: PartialEq> PartialEq for Pairs<'i, R> { + fn eq(&self, other: &Pairs<'i, R>) -> bool { + Rc::ptr_eq(&self.queue, &other.queue) + && ptr::eq(self.input, other.input) + && self.start == other.start + && self.end == other.end + } +} + +impl<'i, R: Eq> Eq for Pairs<'i, R> {} + +impl<'i, R: Hash> Hash for Pairs<'i, R> { + fn hash<H: Hasher>(&self, state: &mut H) { + (&*self.queue as *const Vec<QueueableToken<R>>).hash(state); + (self.input as *const str).hash(state); + self.start.hash(state); + self.end.hash(state); + } +} + +#[cfg(feature = "pretty-print")] +impl<'i, R: RuleType> ::serde::Serialize for Pairs<'i, R> { + fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error> + where + S: ::serde::Serializer, + { + let start = self.pos(self.start); + let end = self.pos(self.end - 1); + let pairs = self.clone().collect::<Vec<_>>(); + + let mut ser = serializer.serialize_struct("Pairs", 2)?; + ser.serialize_field("pos", &(start, end))?; + ser.serialize_field("pairs", &pairs)?; + ser.end() + } +} + +#[cfg(test)] +mod tests { + use super::super::super::macros::tests::*; + use super::super::super::Parser; + + #[test] + #[cfg(feature = "pretty-print")] + fn test_pretty_print() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + + let expected = r#"{ + "pos": [ + 0, + 5 + ], + "pairs": [ + { + "pos": [ + 0, + 3 + ], + "rule": "a", + "inner": { + "pos": [ + 1, + 2 + ], + "pairs": [ + { + "pos": [ + 1, + 2 + ], + "rule": "b", + "inner": "b" + } + ] + } + }, + { + "pos": [ + 4, + 5 + ], + "rule": "c", + "inner": "e" + } + ] +}"#; + + assert_eq!(expected, pairs.to_json()); + } + + #[test] + fn as_str() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + + assert_eq!(pairs.as_str(), "abcde"); + } + + #[test] + fn as_str_empty() { + let mut pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + + assert_eq!(pairs.nth(1).unwrap().into_inner().as_str(), ""); + } + + #[test] + fn concat() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + + assert_eq!(pairs.concat(), "abce"); + } + + #[test] + fn pairs_debug() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + + #[rustfmt::skip] + assert_eq!( + format!("{:?}", pairs), + "[\ + Pair { rule: a, span: Span { str: \"abc\", start: 0, end: 3 }, inner: [\ + Pair { rule: b, span: Span { str: \"b\", start: 1, end: 2 }, inner: [] }\ + ] }, \ + Pair { rule: c, span: Span { str: \"e\", start: 4, end: 5 }, inner: [] }\ + ]" + .to_owned() + ); + } + + #[test] + fn pairs_display() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + + assert_eq!( + format!("{}", pairs), + "[a(0, 3, [b(1, 2)]), c(4, 5)]".to_owned() + ); + } + + #[test] + fn iter_for_pairs() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + assert_eq!( + pairs.map(|p| p.as_rule()).collect::<Vec<Rule>>(), + vec![Rule::a, Rule::c] + ); + } + + #[test] + fn double_ended_iter_for_pairs() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + assert_eq!( + pairs.rev().map(|p| p.as_rule()).collect::<Vec<Rule>>(), + vec![Rule::c, Rule::a] + ); + } +} diff --git a/vendor/pest/src/iterators/queueable_token.rs b/vendor/pest/src/iterators/queueable_token.rs new file mode 100644 index 000000000..7d56749bb --- /dev/null +++ b/vendor/pest/src/iterators/queueable_token.rs @@ -0,0 +1,27 @@ +// pest. The Elegant Parser +// Copyright (c) 2018 Dragoș Tiselice +// +// Licensed under the Apache License, Version 2.0 +// <LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0> or the MIT +// license <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. All files in the project carrying such notice may not be copied, +// modified, or distributed except according to those terms. + +// This structure serves to improve performance over Token objects in two ways: +// +// * it is smaller than a Token, leading to both less memory use when stored in the queue but also +// increased speed when pushing to the queue +// * it finds its pair in O(1) time instead of O(N), since pair positions are known at parse time +// and can easily be stored instead of recomputed +#[derive(Debug)] +pub enum QueueableToken<R> { + Start { + end_token_index: usize, + input_pos: usize, + }, + End { + start_token_index: usize, + rule: R, + input_pos: usize, + }, +} diff --git a/vendor/pest/src/iterators/tokens.rs b/vendor/pest/src/iterators/tokens.rs new file mode 100644 index 000000000..59b75c520 --- /dev/null +++ b/vendor/pest/src/iterators/tokens.rs @@ -0,0 +1,144 @@ +// pest. The Elegant Parser +// Copyright (c) 2018 Dragoș Tiselice +// +// Licensed under the Apache License, Version 2.0 +// <LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0> or the MIT +// license <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your +// option. All files in the project carrying such notice may not be copied, +// modified, or distributed except according to those terms. + +use std::fmt; +use std::rc::Rc; +use std::str; + +use super::queueable_token::QueueableToken; +use position; +use token::Token; +use RuleType; + +/// An iterator over [`Token`]s. It is created by [`Pair::tokens`] and [`Pairs::tokens`]. +/// +/// [`Token`]: ../enum.Token.html +/// [`Pair::tokens`]: struct.Pair.html#method.tokens +/// [`Pairs::tokens`]: struct.Pairs.html#method.tokens +#[derive(Clone)] +pub struct Tokens<'i, R> { + /// # Safety: + /// + /// All `QueueableToken`s' `input_pos` must be valid character boundary indices into `input`. + queue: Rc<Vec<QueueableToken<R>>>, + input: &'i str, + start: usize, + end: usize, +} + +// TODO(safety): QueueableTokens must be valid indices into input. +pub fn new<R: RuleType>( + queue: Rc<Vec<QueueableToken<R>>>, + input: &str, + start: usize, + end: usize, +) -> Tokens<R> { + if cfg!(debug_assertions) { + for tok in queue.iter() { + match *tok { + QueueableToken::Start { input_pos, .. } | QueueableToken::End { input_pos, .. } => { + assert!( + input.get(input_pos..).is_some(), + "💥 UNSAFE `Tokens` CREATED 💥" + ) + } + } + } + } + + Tokens { + queue, + input, + start, + end, + } +} + +impl<'i, R: RuleType> Tokens<'i, R> { + fn create_token(&self, index: usize) -> Token<'i, R> { + match self.queue[index] { + QueueableToken::Start { + end_token_index, + input_pos, + } => { + let rule = match self.queue[end_token_index] { + QueueableToken::End { rule, .. } => rule, + _ => unreachable!(), + }; + + Token::Start { + rule, + // QueueableTokens are safely created. + pos: unsafe { position::Position::new_unchecked(self.input, input_pos) }, + } + } + QueueableToken::End { + rule, input_pos, .. + } => { + Token::End { + rule, + // QueueableTokens are safely created. + pos: unsafe { position::Position::new_unchecked(self.input, input_pos) }, + } + } + } + } +} + +impl<'i, R: RuleType> Iterator for Tokens<'i, R> { + type Item = Token<'i, R>; + + fn next(&mut self) -> Option<Self::Item> { + if self.start >= self.end { + return None; + } + + let token = self.create_token(self.start); + + self.start += 1; + + Some(token) + } +} + +impl<'i, R: RuleType> DoubleEndedIterator for Tokens<'i, R> { + fn next_back(&mut self) -> Option<Self::Item> { + if self.end <= self.start { + return None; + } + + let token = self.create_token(self.end - 1); + + self.end -= 1; + + Some(token) + } +} + +impl<'i, R: RuleType> fmt::Debug for Tokens<'i, R> { + fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result { + f.debug_list().entries(self.clone()).finish() + } +} + +#[cfg(test)] +mod tests { + use super::super::super::macros::tests::*; + use super::super::super::Parser; + use super::Token; + + #[test] + fn double_ended_iter_for_tokens() { + let pairs = AbcParser::parse(Rule::a, "abcde").unwrap(); + let mut tokens = pairs.clone().tokens().collect::<Vec<Token<Rule>>>(); + tokens.reverse(); + let reverse_tokens = pairs.tokens().rev().collect::<Vec<Token<Rule>>>(); + assert_eq!(tokens, reverse_tokens); + } +} |