From 698f8c2f01ea549d77d7dc3338a12e04c11057b9 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Wed, 17 Apr 2024 14:02:58 +0200 Subject: Adding upstream version 1.64.0+dfsg1. Signed-off-by: Daniel Baumann --- vendor/pest/src/iterators/pairs.rs | 419 +++++++++++++++++++++++++++++++++++++ 1 file changed, 419 insertions(+) create mode 100644 vendor/pest/src/iterators/pairs.rs (limited to 'vendor/pest/src/iterators/pairs.rs') 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 +// or the MIT +// license , 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>>, + input: &'i str, + start: usize, + end: usize, +} + +pub fn new( + queue: Rc>>, + input: &str, + start: usize, + end: usize, +) -> Pairs { + 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> { + 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 { + 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 { + 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::>() + .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(&self, state: &mut H) { + (&*self.queue as *const Vec>).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(&self, serializer: S) -> Result + where + S: ::serde::Serializer, + { + let start = self.pos(self.start); + let end = self.pos(self.end - 1); + let pairs = self.clone().collect::>(); + + 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::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::c, Rule::a] + ); + } +} -- cgit v1.2.3