1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252
//! An ANSI escape sequence parser module. //! //! **This module is not available with default features. You have to enable `parser` feature //! if you'd like to use it.** //! //! # Parser //! //! The ANSI escape sequence parser parses input data in two steps: //! //! * transforms input data into valid characters, generic csi & escape sequences, throws away invalid data, //! * give them meaning, throws away sequences without known meaning. //! //! ## First step //! //! State machine implementation for the first step is inspired by the //! [A parser for DEC’s ANSI-compatible video terminals](https://vt100.net/emu/dec_ansi_parser) article //! and the [vte](https://crates.io/crates/vte) crate. The goal of this step is to transform an input //! data into characters, generic csi & escape sequences, validate them and throw away malformed input. //! //! An example of valid csi sequence: `b"\x1B[20;10R"`. Output of the first step will be: //! //! * valid csi sequence //! * with two parameters `[20, 10]` //! * and the final character `R`. //! //! ## Second step //! //! An input of this step is output of the first one. We know that the final character `R` represents //! cursor position and two parameters should be provided. They were provided, we can give it a //! meaning and return `Sequence::CursorPosition(10, 20)`. //! //! All sequences without known meaning are discarded. //! //! ## Implementation //! //! Both steps are considered as an implementation detail and there's no plan to make them //! publicly available. //! //! The `parser` module provides the [`Parser`](struct.Parser.html) structure you can feed with //! the [`advance`](struct.Parser.html#method.advance) method. It also implements the standard //! library `Iterator<Item = Sequence>` trait which allows you to consume valid sequences with //! known meaning via the `next()` method. Check the [`Sequence`](enum.Sequence.html) enum to learn //! what this module can parse. use std::collections::VecDeque; use engine::{Engine, Provide}; pub use types::{KeyCode, KeyModifiers, Mouse, MouseButton, Sequence}; mod engine; mod parsers; pub(crate) mod types; /// An ANSI escape sequence parser. /// /// `Parser` implements the `Iterator<Item = Sequence>` trait, thus you can use the /// `next()` method to consume all valid sequences with known meaning. /// /// # Examples /// /// Parse cursor position: /// /// ``` /// use anes::parser::{Parser, Sequence}; /// /// let mut parser = Parser::default(); /// parser.advance(b"\x1B[20;10R", false); /// /// assert_eq!(Some(Sequence::CursorPosition(10, 20)), parser.next()); /// assert!(parser.next().is_none()); /// ``` /// /// Parse keyboard event: /// /// ``` /// use anes::parser::{KeyCode, KeyModifiers, Parser, Sequence}; /// /// let mut parser = Parser::default(); /// parser.advance("𐌼a".as_bytes(), false); /// /// assert_eq!(Some(Sequence::Key(KeyCode::Char('𐌼'), KeyModifiers::empty())), parser.next()); /// assert_eq!(Some(Sequence::Key(KeyCode::Char('a'), KeyModifiers::empty())), parser.next()); /// assert!(parser.next().is_none()); /// ``` #[derive(Default)] pub struct Parser { engine: Engine, provider: SequenceProvider, } impl Parser { /// Advances parser state machine with additional input data. /// /// # Arguments /// /// * `buffer` - input data (stdin in raw mode, etc.) /// * `more` - more input data available right now /// /// It's crucial to provide correct `more` value in order to receive `KeyCode::Esc` events /// as soon as possible. /// /// # Examples /// /// Esc key: /// /// ``` /// use anes::parser::{KeyCode, KeyModifiers, Parser, Sequence}; /// /// let mut parser = Parser::default(); /// // User pressed Esc key & nothing else which means that there's no additional input available /// // aka no possible escape sequence = `KeyCode::Esc` dispatched. /// parser.advance(&[0x1b], false); /// /// assert_eq!(Some(Sequence::Key(KeyCode::Esc, KeyModifiers::empty())), parser.next()); /// assert!(parser.next().is_none()); /// ``` /// /// Possible escape sequence: /// /// ``` /// use anes::parser::{KeyCode, KeyModifiers, Parser, Sequence}; /// /// let mut parser = Parser::default(); /// // User pressed F1 = b"\x1BOP" /// /// // Every escape sequence starts with Esc (0x1b). There's more input available /// // aka possible escape sequence = `KeyCode::Esc` isn't dispatched even when the parser /// // doesn't know rest of the sequence. /// parser.advance(&[0x1b], true); /// assert!(parser.next().is_none()); /// /// // Advance parser with rest of the sequence /// parser.advance(&[b'O', b'P'], false); /// assert_eq!(Some(Sequence::Key(KeyCode::F(1), KeyModifiers::empty())), parser.next()); /// assert!(parser.next().is_none()); /// ``` pub fn advance(&mut self, buffer: &[u8], more: bool) { let len = buffer.len(); for (idx, byte) in buffer.iter().enumerate() { self.engine .advance(&mut self.provider, *byte, idx < len - 1 || more); } } } impl Iterator for Parser { type Item = Sequence; fn next(&mut self) -> Option<Self::Item> { self.provider.next() } } #[derive(Default)] struct SequenceProvider { esc_o: bool, seqs: VecDeque<Sequence>, } impl Iterator for SequenceProvider { type Item = Sequence; fn next(&mut self) -> Option<Self::Item> { self.seqs.pop_front() } } impl Provide for SequenceProvider { fn provide_char(&mut self, ch: char) { // eprintln!("dispatch_char: {}", ch); if let Some(seq) = parsers::parse_char(ch, self.esc_o) { self.seqs.push_back(seq); } self.esc_o = false; } fn provide_esc_sequence(&mut self, ch: char) { if ch == 'O' { // Exception // // Esc O - dispatched as an escape sequence followed by single character (P-S) representing // F1-F4 keys. We store Esc O flag only which is then used in the dispatch_char method. self.esc_o = true; } else { self.esc_o = false; if let Some(seq) = parsers::parse_esc_sequence(ch) { self.seqs.push_back(seq); } } } fn provide_csi_sequence(&mut self, parameters: &[u64], ignored_count: usize, ch: char) { if let Some(seq) = parsers::parse_csi_sequence(parameters, ignored_count, ch) { self.seqs.push_back(seq); } self.esc_o = false; } } #[cfg(test)] mod tests { use super::Parser; #[test] fn dispatch_char() { let mut parser = Parser::default(); parser.advance(&[b'a'], false); assert!(parser.next().is_some()); } #[test] fn dispatch_esc_sequence() { let mut parser = Parser::default(); parser.advance(&[b'\x1B'], true); assert!(parser.next().is_none()); parser.advance(&[b'a'], false); assert!(parser.next().is_some()); } #[test] fn does_not_dispatch_esc_sequence_with_upper_case_o() { let mut parser = Parser::default(); parser.advance(&[b'\x1B'], true); assert!(parser.next().is_none()); parser.advance(&[b'O'], true); assert!(parser.next().is_none()); } #[test] fn dispatch_esc_with_upper_case_o_followed_by_char_as_single_sequence() { let mut parser = Parser::default(); parser.advance(&[b'\x1B'], true); assert!(parser.next().is_none()); parser.advance(&[b'O'], true); assert!(parser.next().is_none()); parser.advance(&[b'P'], false); assert!(parser.next().is_some()); assert!(parser.next().is_none()); } #[test] fn dispatch_csi_sequence() { let mut parser = Parser::default(); parser.advance(&[b'\x1B'], true); assert!(parser.next().is_none()); parser.advance(&[b'['], true); assert!(parser.next().is_none()); parser.advance(&[b'D'], false); assert!(parser.next().is_some()); } }