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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503
extern crate regex; extern crate serde_regex; use crate::grammar::Grammar; use crate::util::{Expression, Flags}; use regex::{Captures, Error, Regex}; use std::cmp::{Eq, Ord, Ordering, PartialEq, PartialOrd}; use std::collections::HashMap; /// This is functionally equivalent to a [`Regex`]: you can use it repeatedly to /// search a string. It cannot itself be used directly to split strings, but /// its regular expression is public and may be so used. It improves on regular /// expressions in that the [`Match`] object it returns is the root node in a /// parse tree, so its matches preserve parse structure. /// /// [`Regex`]: ../regex/struct.Regex.html /// [`Match`]: ../pidgin/struct.Match.html #[derive(Debug, Serialize, Deserialize)] pub struct Matcher { /// The `Regex` used for parsing. #[serde(with = "serde_regex")] pub rx: Regex, root: String, translation: HashMap<String, String>, parentage: HashMap<String, Vec<String>>, } impl Matcher { /// Returns `Some(Match)` if the grammar can parse the string. Note that /// unless the grammar is string-bounded, this only means it can parse /// the string at some point. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let m = grammar!{ /// /// (?bB) /// /// S -> r(r"\A") <subject> <VP> r(r"\.\z") /// /// subject => [["Amy", "Bob", "Carter", "Dianne"]] /// VP -> <verb_intransitive> | <verb_transitive> <object> /// verb_intransitive => [["naps", "doodles", "exercises", "meditates"]] /// verb_transitive => [["eats", "programs", "sees", "throws"]] /// object => (?w) [["a sandwich", "eggs", "the sunset"]] /// /// }.matcher()?; /// /// let parse_tree = m.parse("Amy programs the sunset.").unwrap(); /// /// println!("{}", parse_tree); /// /// // S (0, 24): "Amy programs the sunset." /// // subject (0, 3): "Amy" /// // VP (4, 23): "programs the sunset" /// // verb_transitive (4, 12): "programs" /// // object (13, 23): "the sunset" /// # Ok(()) } /// ``` pub fn parse<'t>(&self, s: &'t str) -> Option<Match<'t>> { match self.rx.captures(s) { Some(captures) => { let c = captures.get(0).unwrap(); let mut m = Match { rule: self.root.clone(), text: s, start: c.start(), end: c.end(), children: None, }; self.build_tree(&mut m, &self.root, s, &captures); let mut m = self.simplify_tree(m); self.sort_tree(&mut m); Some(m) } None => None, } } // put the children in match order fn sort_tree<'t>(&self, m: &mut Match) { if let Some(ref mut children) = m.children { children.sort(); for c in children { self.sort_tree(c); } } } fn simplify_tree<'t>(&self, m: Match<'t>) -> Match<'t> { if m.children.is_some() { let mut clone = m.clone(); let children: Vec<Match> = m .children .unwrap() .into_iter() .map(|c| self.simplify_tree(c)) .collect(); if children.len() == 1 && m.rule == children[0].rule { children[0].clone() } else { clone.children = Some(children); clone } } else { m } } /// Returns whether the grammar can parse the string. This is a cheaper /// operation than parsing. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let m = grammar!{ /// /// (?bB) /// /// S -> r(r"\A") <subject> <VP> r(r"\.\z") /// /// subject => [["Amy", "Bob", "Carter", "Dianne"]] /// VP -> <verb_intransitive> | <verb_transitive> <object> /// verb_intransitive => [["naps", "doodles", "exercises", "meditates"]] /// verb_transitive => [["eats", "programs", "sees", "throws"]] /// object => (?w) [["a sandwich", "eggs", "the sunset"]] /// /// }.matcher()?; /// /// assert!(m.is_match("Bob doodles.")); /// # Ok(()) } /// ``` pub fn is_match(&self, text: &str) -> bool { self.rx.is_match(text) } // recursively build a parse tree showing the groups matched fn build_tree<'t>(&self, m: &mut Match<'t>, parent: &str, text: &'t str, captures: &Captures) { if let Some(children) = self.parentage.get(parent) { for c in children { if let Some(n) = captures.name(c) { let name = self.translation.get(c).unwrap(); let mut child = Match { text, rule: name.clone(), start: n.start(), end: n.end(), children: None, }; self.build_tree(&mut child, &c, text, captures); m.children.get_or_insert_with(|| Vec::new()).push(child); } } } } pub(crate) fn new(g: &Grammar) -> Result<Matcher, Error> { let mut idx = 0; let mut translation = HashMap::new(); let mut parentage = HashMap::new(); let root = g.name.clone().unwrap_or(String::from("0")); let mut g = g.clone(); g.name = Some(root.clone()); let mut g = Expression::Grammar(g.clone(), false); Matcher::walk(&mut idx, &root, &mut g, &mut translation, &mut parentage); if let Expression::Grammar(g, _) = g { match Regex::new(&g.to_s(&Flags::defaults(), false, false)) { Ok(rx) => Ok(Matcher { translation, parentage, rx, root, }), Err(e) => Err(e), } } else { unreachable!() } } fn walk( idx: &mut usize, parent: &str, e: &mut Expression, translation: &mut HashMap<String, String>, parentage: &mut HashMap<String, Vec<String>>, ) { match e { Expression::Sequence(v, _) => { for e in v { Matcher::walk(idx, parent, e, translation, parentage); } } Expression::Repetition(e, _, _) => { Matcher::walk(idx, parent, e, translation, parentage) } Expression::Alternation(v, _) => { for e in v { Matcher::walk(idx, parent, e, translation, parentage); } } Expression::Grammar(g, _) => { if let Some(n) = g.name.clone() { let new_name = format!("m{}", idx); *idx += 1; translation.insert(new_name.clone(), n); parentage .entry(parent.to_string()) .or_insert_with(|| Vec::new()) .push(new_name.clone()); for e in &mut g.sequence { Matcher::walk(idx, &new_name, &mut *e, translation, parentage); } g.name = Some(new_name.clone()); } else { for ref mut e in &mut g.sequence { Matcher::walk(idx, parent, e, translation, parentage); } } } _ => (), } } } /// This is a node in a parse tree. It is functionally similar to [`regex::Match`], /// in fact providing much the same API, but unlike a `regex::Match` a `pidgin::Match` /// always corresponds to some rule, it knows what rule it corresponds to, /// and it records any sub-matches involved in its parsing. /// /// The lifetime parameter `'t` represents the lifetime of the `&str` matched /// against. /// /// [`regex::Match`]: ../regex/struct.Match.html #[derive(Debug, Clone)] pub struct Match<'t> { rule: String, text: &'t str, start: usize, end: usize, children: Option<Vec<Match<'t>>>, } impl<'t> Match<'t> { /// Returns the matched text. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let m = grammar!{ /// foo => ("bar") /// }.matcher()?.parse(" bar ").unwrap(); /// /// assert_eq!("bar", m.as_str()); /// # Ok(()) } /// ``` pub fn as_str(&self) -> &'t str { &self.text[self.start..self.end] } /// Returns the starting offset of the match. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let m = grammar!{ /// foo => ("bar") /// }.matcher()?.parse(" bar ").unwrap(); /// /// assert_eq!(3, m.start()); /// # Ok(()) } /// ``` pub fn start(&self) -> usize { self.start } /// Returns the ending offset of the match. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let m = grammar!{ /// foo => ("bar") /// }.matcher()?.parse(" bar ").unwrap(); /// /// assert_eq!(6, m.end()); /// # Ok(()) } /// ``` pub fn end(&self) -> usize { self.end } /// Returns the grammar rule matched. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let m = grammar!{ /// foo => ("bar") /// }.matcher()?.parse(" bar ").unwrap(); /// /// assert_eq!("foo", m.rule()); /// # Ok(()) } /// ``` pub fn rule(&self) -> &str { &self.rule } /// Returns the sub-matches of this match, if any. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let m = grammar!{ /// TOP -> <foo> <bar> <baz> /// foo => (1) /// bar => (2) /// baz => (3) /// }.matcher()?.parse(" 1 2 3 ").unwrap(); /// /// let children = m.children().unwrap(); /// /// assert_eq!(3, children.len()); /// assert_eq!("1", children[0].as_str()); /// assert_eq!("foo", children[0].rule()); /// assert_eq!("2", children[1].as_str()); /// assert_eq!("bar", children[1].rule()); /// assert_eq!("3", children[2].as_str()); /// assert_eq!("baz", children[2].rule()); /// # Ok(()) } /// ``` pub fn children(&self) -> Option<&[Match<'t>]> { match self.children.as_ref() { Some(v) => Some(v), None => None, } } /// Returns the first `Match` defined by the given rule under this parse /// node searching recursively, depth-first, left-to-right. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let matcher = grammar!{ /// TOP -> <foo> <bar> /// foo -> (1) <baz> /// bar -> (2) <baz> /// baz -> [["prawn", "shrimp", "crevette"]] /// }.matcher()?; /// /// let p = matcher.parse("1 crevette 2 shrimp").unwrap(); /// let baz = p.name("baz").unwrap(); /// /// assert_eq!("crevette", baz.as_str()); /// # Ok(()) } /// ``` pub fn name(&self, name: &str) -> Option<&Match> { if self.rule == name { Some(self) } else { if self.children.is_some() { for m in self.children.as_ref().unwrap() { if let Some(n) = m.name(name) { return Some(n); } } } None } } /// Returns all `Match`es matching the given rule in the parse tree under /// this node. Matches are ordered as found by a depth-first left-to-right /// search of the parse tree. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let matcher = grammar!{ /// TOP -> <foo> <bar> /// foo -> (1) <baz> /// bar -> (2) <baz> /// baz -> [["prawn", "shrimp", "crevette"]] /// }.matcher()?; /// /// let p = matcher.parse("1 crevette 2 shrimp").unwrap(); /// let names = p.all_names("baz"); /// /// assert_eq!("crevette", names[0].as_str()); /// assert_eq!("shrimp", names[1].as_str()); /// # Ok(()) } /// ``` pub fn all_names(&self, name: &str) -> Vec<&Match> { let mut v = Vec::new(); self.collect(name, &mut v); v } /// Returns whether the given rule matched for any node in the parse tree. /// /// # Examples /// /// ```rust /// # #[macro_use] extern crate pidgin; /// # use std::error::Error; /// # fn demo() -> Result<(), Box<Error>> { /// let g = grammar!{ /// TOP => <animal> | <thing> /// animal => [["cat", "dog", "camel"]] /// thing => [["carpet", "crate", "cartoon"]] /// }; /// /// let m = g.matcher()?; /// /// assert!(m.parse("cat").unwrap().has("animal")); /// # Ok(())} /// ``` pub fn has(&self, name: &str) -> bool { self.rule == name || self.children.is_some() && self .children .as_ref() .unwrap() .iter() .any(|ref m| m.has(name)) } fn collect(&'t self, name: &str, names: &mut Vec<&'t Match<'t>>) { if self.rule == name { names.push(self); } if self.children.is_some() { for m in self.children.as_ref().unwrap() { m.collect(name, names); } } } // display the parse tree with indentation fn _fmt(&self, depth: usize, f: &mut std::fmt::Formatter) -> std::fmt::Result { let own = format!( "{}{} ({}, {}): {}\n", " ".repeat(depth), self.rule, self.start, self.end, format!("{:?}", self.as_str()) ); if let Some(ref children) = &self.children { let mut r = write!(f, "{}", own); for c in children { r = c._fmt(depth + 1, f); } r } else { write!(f, "{}", own) } } } impl<'t> std::fmt::Display for Match<'t> { fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result { self._fmt(0, f) } } impl<'t> PartialEq for Match<'t> { fn eq(&self, other: &Match<'t>) -> bool { self.cmp(other) == Ordering::Equal && self.rule == other.rule } } impl<'t> Eq for Match<'t> {} impl<'t> PartialOrd for Match<'t> { fn partial_cmp(&self, other: &Match<'t>) -> Option<Ordering> { Some(self.cmp(other)) } } impl<'t> Ord for Match<'t> { fn cmp(&self, other: &Match<'t>) -> Ordering { let o = self.start.cmp(&other.start); match o { Ordering::Equal => self.end.cmp(&other.end), _ => o, } } }