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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
//! Definition of a lexer for the WebAssembly text format.
//!
//! This module provides a [`Lexer`][] type which is an iterate over the raw
//! tokens of a WebAssembly text file. A [`Lexer`][] accounts for every single
//! byte in a WebAssembly text field, returning tokens even for comments and
//! whitespace. Typically you'll ignore comments and whitespace, however.
//!
//! If you'd like to iterate over the tokens in a file you can do so via:
//!
//! ```
//! # fn foo() -> Result<(), wast::Error> {
//! use wast::lexer::Lexer;
//!
//! let wat = "(module (func $foo))";
//! for token in Lexer::new(wat).iter(0) {
//! println!("{:?}", token?);
//! }
//! # Ok(())
//! # }
//! ```
//!
//! Note that you'll typically not use this module but will rather use
//! [`ParseBuffer`](crate::parser::ParseBuffer) instead.
//!
//! [`Lexer`]: crate::lexer::Lexer
use crate::token::Span;
use crate::Error;
use std::borrow::Cow;
use std::char;
use std::fmt;
use std::slice;
use std::str;
use std::str::Utf8Error;
/// A structure used to lex the s-expression syntax of WAT files.
///
/// This structure is used to generate [`Token`] items, which should account for
/// every single byte of the input as we iterate over it. A [`LexError`] is
/// returned for any non-lexable text.
#[derive(Clone)]
pub struct Lexer<'a> {
input: &'a str,
allow_confusing_unicode: bool,
}
/// A single token parsed from a `Lexer`.
#[derive(Copy, Clone, Debug, PartialEq)]
pub struct Token {
/// The kind of token this represents, such as whether it's whitespace, a
/// keyword, etc.
pub kind: TokenKind,
/// The byte offset within the original source for where this token came
/// from.
pub offset: usize,
/// The byte length of this token as it resides in the original source.
//
// NB: this is `u32` to enable packing `Token` into two pointers of size.
// This does limit a single token to being at most 4G large, but that seems
// probably ok.
pub len: u32,
}
#[test]
fn token_is_not_too_big() {
assert!(std::mem::size_of::<Token>() <= std::mem::size_of::<u64>() * 2);
}
/// Classification of what was parsed from the input stream.
///
/// This enumeration contains all kinds of fragments, including comments and
/// whitespace.
#[derive(Copy, Clone, Debug, PartialEq)]
pub enum TokenKind {
/// A line comment, preceded with `;;`
LineComment,
/// A block comment, surrounded by `(;` and `;)`. Note that these can be
/// nested.
BlockComment,
/// A fragment of source that represents whitespace.
Whitespace,
/// A left-parenthesis, including the source text for where it comes from.
LParen,
/// A right-parenthesis, including the source text for where it comes from.
RParen,
/// A string literal, which is actually a list of bytes.
String,
/// An identifier (like `$foo`).
///
/// All identifiers start with `$` and the payload here is the original
/// source text.
Id,
/// A keyword, or something that starts with an alphabetic character.
///
/// The payload here is the original source text.
Keyword,
/// An annotation (like `@foo`).
///
/// All annotations start with `@` and the payload will be the name of the
/// annotation.
Annotation,
/// A reserved series of `idchar` symbols. Unknown what this is meant to be
/// used for, you'll probably generate an error about an unexpected token.
Reserved,
/// An integer.
Integer(IntegerKind),
/// A float.
Float(FloatKind),
}
/// Description of the parsed integer from the source.
#[derive(Copy, Clone, Debug, PartialEq)]
pub struct IntegerKind {
sign: Option<SignToken>,
has_underscores: bool,
hex: bool,
}
/// Description of a parsed float from the source.
#[allow(missing_docs)]
#[derive(Copy, Clone, Debug, PartialEq)]
pub enum FloatKind {
#[doc(hidden)]
Inf { negative: bool },
#[doc(hidden)]
Nan { negative: bool },
#[doc(hidden)]
NanVal {
negative: bool,
has_underscores: bool,
},
#[doc(hidden)]
Normal { has_underscores: bool, hex: bool },
}
enum ReservedKind {
/// "..."
String,
/// anything that's just a sequence of `idchars!()`
Idchars,
/// $"..."
IdString,
/// @"..."
AnnotationString,
/// everything else (a conglomeration of strings, idchars, etc)
Reserved,
}
/// Errors that can be generated while lexing.
///
/// All lexing errors have line/colum/position information as well as a
/// `LexError` indicating what kind of error happened while lexing.
#[derive(Debug, Clone, PartialEq, Eq)]
#[non_exhaustive]
pub enum LexError {
/// A dangling block comment was found with an unbalanced `(;` which was
/// never terminated in the file.
DanglingBlockComment,
/// An unexpected character was encountered when generally parsing and
/// looking for something else.
Unexpected(char),
/// An invalid `char` in a string literal was found.
InvalidStringElement(char),
/// An invalid string escape letter was found (the thing after the `\` in
/// string literals)
InvalidStringEscape(char),
/// An invalid hexadecimal digit was found.
InvalidHexDigit(char),
/// An invalid base-10 digit was found.
InvalidDigit(char),
/// Parsing expected `wanted` but ended up finding `found` instead where the
/// two characters aren't the same.
Expected {
/// The character that was expected to be found
wanted: char,
/// The character that was actually found
found: char,
},
/// We needed to parse more but EOF (or end of the string) was encountered.
UnexpectedEof,
/// A number failed to parse because it was too big to fit within the target
/// type.
NumberTooBig,
/// An invalid unicode value was found in a `\u{...}` escape in a string,
/// only valid unicode scalars can be escaped that way.
InvalidUnicodeValue(u32),
/// A lone underscore was found when parsing a number, since underscores
/// should always be preceded and succeeded with a digit of some form.
LoneUnderscore,
/// A "confusing" unicode character is present in a comment or a string
/// literal, such as a character that changes the direction text is
/// typically displayed in editors. This could cause the human-read
/// version to behave differently than the compiler-visible version, so
/// these are simply rejected for now.
ConfusingUnicode(char),
/// An invalid utf-8 sequence was found in a quoted identifier, such as
/// `$"\ff"`.
InvalidUtf8Id(Utf8Error),
/// An empty identifier was found, or a lone `$`.
EmptyId,
/// An empty identifier was found, or a lone `@`.
EmptyAnnotation,
}
/// A sign token for an integer.
#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub enum SignToken {
/// Plus sign: "+",
Plus,
/// Minus sign: "-",
Minus,
}
/// A fully parsed integer from a source string with a payload ready to parse
/// into an integral type.
#[derive(Debug, PartialEq)]
pub struct Integer<'a> {
sign: Option<SignToken>,
val: Cow<'a, str>,
hex: bool,
}
/// Possible parsed float values
#[derive(Debug, PartialEq, Eq)]
pub enum Float<'a> {
/// A float `NaN` representation
Nan {
/// The specific bits to encode for this float, optionally
val: Option<Cow<'a, str>>,
/// Whether or not this is a negative `NaN` or not.
negative: bool,
},
/// An float infinite representation,
Inf {
#[allow(missing_docs)]
negative: bool,
},
/// A parsed and separated floating point value
Val {
/// Whether or not the `integral` and `decimal` are specified in hex
hex: bool,
/// The float parts before the `.`
integral: Cow<'a, str>,
/// The float parts after the `.`
decimal: Option<Cow<'a, str>>,
/// The exponent to multiple this `integral.decimal` portion of the
/// float by. If `hex` is true this is `2^exponent` and otherwise it's
/// `10^exponent`
exponent: Option<Cow<'a, str>>,
},
}
// https://webassembly.github.io/spec/core/text/values.html#text-idchar
macro_rules! idchars {
() => {
b'0'..=b'9'
| b'A'..=b'Z'
| b'a'..=b'z'
| b'!'
| b'#'
| b'$'
| b'%'
| b'&'
| b'\''
| b'*'
| b'+'
| b'-'
| b'.'
| b'/'
| b':'
| b'<'
| b'='
| b'>'
| b'?'
| b'@'
| b'\\'
| b'^'
| b'_'
| b'`'
| b'|'
| b'~'
}
}
impl<'a> Lexer<'a> {
/// Creates a new lexer which will lex the `input` source string.
pub fn new(input: &str) -> Lexer<'_> {
Lexer {
input,
allow_confusing_unicode: false,
}
}
/// Returns the original source input that we're lexing.
pub fn input(&self) -> &'a str {
self.input
}
/// Configures whether "confusing" unicode characters are allowed while
/// lexing.
///
/// If allowed then no error will happen if these characters are found, but
/// otherwise if disallowed a lex error will be produced when these
/// characters are found. Confusing characters are denied by default.
///
/// For now "confusing characters" are primarily related to the "trojan
/// source" problem where it refers to characters which cause humans to read
/// text differently than this lexer, such as characters that alter the
/// left-to-right display of the source code.
pub fn allow_confusing_unicode(&mut self, allow: bool) -> &mut Self {
self.allow_confusing_unicode = allow;
self
}
/// Lexes the next at the byte position `pos` in the input.
///
/// Returns `Some` if a token is found or `None` if we're at EOF.
///
/// The `pos` argument will be updated to point to the next token on a
/// successful parse.
///
/// # Errors
///
/// Returns an error if the input is malformed.
pub fn parse(&self, pos: &mut usize) -> Result<Option<Token>, Error> {
let offset = *pos;
Ok(match self.parse_kind(pos)? {
Some(kind) => Some(Token {
kind,
offset,
len: (*pos - offset).try_into().unwrap(),
}),
None => None,
})
}
fn parse_kind(&self, pos: &mut usize) -> Result<Option<TokenKind>, Error> {
let start = *pos;
// This `match` generally parses the grammar specified at
//
// https://webassembly.github.io/spec/core/text/lexical.html#text-token
let remaining = &self.input.as_bytes()[start..];
let byte = match remaining.first() {
Some(b) => b,
None => return Ok(None),
};
match byte {
// Open-parens check the next character to see if this is the start
// of a block comment, otherwise it's just a bland left-paren
// token.
b'(' => match remaining.get(1) {
Some(b';') => {
let mut level = 1;
// Note that we're doing a byte-level search here for the
// close-delimiter of `;)`. The actual source text is utf-8
// encode in `remaining` but due to how utf-8 works we
// can safely search for an ASCII byte since it'll never
// otherwise appear in the middle of a codepoint and if we
// find it then it's guaranteed to be the right byte.
//
// Mainly we're avoiding the overhead of decoding utf-8
// characters into a Rust `char` since it's otherwise
// unnecessary work.
let mut iter = remaining[2..].iter();
while let Some(ch) = iter.next() {
match ch {
b'(' => {
if let Some(b';') = iter.as_slice().first() {
level += 1;
iter.next();
}
}
b';' => {
if let Some(b')') = iter.as_slice().first() {
level -= 1;
iter.next();
if level == 0 {
let len = remaining.len() - iter.as_slice().len();
let comment = &self.input[start..][..len];
*pos += len;
self.check_confusing_comment(*pos, comment)?;
return Ok(Some(TokenKind::BlockComment));
}
}
}
_ => {}
}
}
Err(self.error(start, LexError::DanglingBlockComment))
}
_ => {
*pos += 1;
Ok(Some(TokenKind::LParen))
}
},
b')' => {
*pos += 1;
Ok(Some(TokenKind::RParen))
}
// https://webassembly.github.io/spec/core/text/lexical.html#white-space
b' ' | b'\n' | b'\r' | b'\t' => {
self.skip_ws(pos);
Ok(Some(TokenKind::Whitespace))
}
c @ (idchars!() | b'"') => {
let (kind, src) = self.parse_reserved(pos)?;
match kind {
// If the reserved token was simply a single string then
// that is converted to a standalone string token
ReservedKind::String => return Ok(Some(TokenKind::String)),
// If only idchars were consumed then this could be a
// specific kind of standalone token we're interested in.
ReservedKind::Idchars => {
// https://webassembly.github.io/spec/core/text/values.html#integers
if let Some(ret) = self.classify_number(src) {
return Ok(Some(ret));
// https://webassembly.github.io/spec/core/text/values.html#text-id
} else if *c == b'$' {
return Ok(Some(TokenKind::Id));
// part of the WebAssembly/annotations proposal
// (no online url yet)
} else if *c == b'@' {
return Ok(Some(TokenKind::Annotation));
// https://webassembly.github.io/spec/core/text/lexical.html#text-keyword
} else if b'a' <= *c && *c <= b'z' {
return Ok(Some(TokenKind::Keyword));
}
}
ReservedKind::IdString => return Ok(Some(TokenKind::Id)),
ReservedKind::AnnotationString => return Ok(Some(TokenKind::Annotation)),
// ... otherwise this was a conglomeration of idchars,
// strings, or just idchars that don't match a prior rule,
// meaning this falls through to the fallback `Reserved`
// token.
ReservedKind::Reserved => {}
}
Ok(Some(TokenKind::Reserved))
}
// This could be a line comment, otherwise `;` is a reserved token.
// The second byte is checked to see if it's a `;;` line comment
//
// Note that this character being considered as part of a
// `reserved` token is part of the annotations proposal.
b';' => match remaining.get(1) {
Some(b';') => {
let remaining = &self.input[*pos..];
let byte_pos = memchr::memchr2(b'\n', b'\r', remaining.as_bytes())
.unwrap_or(remaining.len());
*pos += byte_pos;
let comment = &remaining[..byte_pos];
self.check_confusing_comment(*pos, comment)?;
Ok(Some(TokenKind::LineComment))
}
_ => {
*pos += 1;
Ok(Some(TokenKind::Reserved))
}
},
// Other known reserved tokens other than `;`
//
// Note that these characters being considered as part of a
// `reserved` token is part of the annotations proposal.
b',' | b'[' | b']' | b'{' | b'}' => {
*pos += 1;
Ok(Some(TokenKind::Reserved))
}
_ => {
let ch = self.input[start..].chars().next().unwrap();
Err(self.error(*pos, LexError::Unexpected(ch)))
}
}
}
fn skip_ws(&self, pos: &mut usize) {
// This table is a byte lookup table to determine whether a byte is a
// whitespace byte. There are only 4 whitespace bytes for the `*.wat`
// format right now which are ' ', '\t', '\r', and '\n'. These 4 bytes
// have a '1' in the table below.
//
// Due to how utf-8 works (our input is guaranteed to be utf-8) it is
// known that if these bytes are found they're guaranteed to be the
// whitespace byte, so they can be safely skipped and we don't have to
// do full utf-8 decoding. This means that the goal of this function is
// to find the first non-whitespace byte in `remaining`.
//
// For now this lookup table seems to be the fastest, but projects like
// https://github.com/lemire/despacer show other simd algorithms which
// can possibly accelerate this even more. Note that `*.wat` files often
// have a lot of whitespace so this function is typically quite hot when
// parsing inputs.
#[rustfmt::skip]
const WS: [u8; 256] = [
// \t \n \r
/* 0x00 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1, 0, 0,
/* 0x10 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
// ' '
/* 0x20 */ 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0x30 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0x40 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0x50 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0x60 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0x70 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0x80 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0x90 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0xa0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0xb0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0xc0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0xd0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0xe0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
/* 0xf0 */ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
];
let remaining = &self.input[*pos..];
let non_ws_pos = remaining
.as_bytes()
.iter()
.position(|b| WS[*b as usize] != 1)
.unwrap_or(remaining.len());
*pos += non_ws_pos;
}
/// Splits off a "reserved" token which is then further processed later on
/// to figure out which kind of token it is `depending on `ReservedKind`.
///
/// For more information on this method see the clarification at
/// <https://github.com/WebAssembly/spec/pull/1499> but the general gist is
/// that this is parsing the grammar:
///
/// ```text
/// reserved := (idchar | string)+
/// ```
///
/// which means that it is eating any number of adjacent string/idchar
/// tokens (e.g. `a"b"c`) and returning the classification of what was
/// eaten. The classification assists in determining what the actual token
/// here eaten looks like.
fn parse_reserved(&self, pos: &mut usize) -> Result<(ReservedKind, &'a str), Error> {
let mut idchars = 0u32;
let mut strings = 0u32;
let start = *pos;
while let Some(byte) = self.input.as_bytes().get(*pos) {
match byte {
// Normal `idchars` production which appends to the reserved
// token that's being produced.
idchars!() => {
idchars += 1;
*pos += 1;
}
// https://webassembly.github.io/spec/core/text/values.html#text-string
b'"' => {
strings += 1;
*pos += 1;
let mut it = self.input[*pos..].chars();
let result = Lexer::parse_str(&mut it, self.allow_confusing_unicode);
*pos = self.input.len() - it.as_str().len();
match result {
Ok(_) => {}
Err(e) => {
let err_pos = match &e {
LexError::UnexpectedEof => self.input.len(),
_ => self.input[..*pos].char_indices().next_back().unwrap().0,
};
return Err(self.error(err_pos, e));
}
}
}
// Nothing else is considered part of a reserved token
_ => break,
}
}
let ret = &self.input[start..*pos];
Ok(match (idchars, strings) {
(0, 0) => unreachable!(),
(0, 1) => (ReservedKind::String, ret),
(_, 0) => (ReservedKind::Idchars, ret),
// Pattern match `@"..."` and `$"..."` for string-based
// identifiers and annotations.
(1, 1) if ret.starts_with("$") => (ReservedKind::IdString, ret),
(1, 1) if ret.starts_with("@") => (ReservedKind::AnnotationString, ret),
_ => (ReservedKind::Reserved, ret),
})
}
fn classify_number(&self, src: &str) -> Option<TokenKind> {
let (sign, num) = if let Some(stripped) = src.strip_prefix('+') {
(Some(SignToken::Plus), stripped)
} else if let Some(stripped) = src.strip_prefix('-') {
(Some(SignToken::Minus), stripped)
} else {
(None, src)
};
let negative = sign == Some(SignToken::Minus);
// Handle `inf` and `nan` which are special numbers here
if num == "inf" {
return Some(TokenKind::Float(FloatKind::Inf { negative }));
} else if num == "nan" {
return Some(TokenKind::Float(FloatKind::Nan { negative }));
} else if let Some(stripped) = num.strip_prefix("nan:0x") {
let mut it = stripped.as_bytes().iter();
let has_underscores = skip_underscores(&mut it, |x| char::from(x).is_ascii_hexdigit())?;
if it.next().is_some() {
return None;
}
return Some(TokenKind::Float(FloatKind::NanVal {
negative,
has_underscores,
}));
}
// Figure out if we're a hex number or not
let test_valid: fn(u8) -> bool;
let (mut it, hex) = if let Some(stripped) = num.strip_prefix("0x") {
test_valid = |x: u8| char::from(x).is_ascii_hexdigit();
(stripped.as_bytes().iter(), true)
} else {
test_valid = |x: u8| char::from(x).is_ascii_digit();
(num.as_bytes().iter(), false)
};
// Evaluate the first part, moving out all underscores
let mut has_underscores = skip_underscores(&mut it, test_valid)?;
match it.clone().next() {
// If we're followed by something this may be a float so keep going.
Some(_) => {}
// Otherwise this is a valid integer literal!
None => {
return Some(TokenKind::Integer(IntegerKind {
has_underscores,
sign,
hex,
}))
}
}
// A number can optionally be after the decimal so only actually try to
// parse one if it's there.
if it.clone().next() == Some(&b'.') {
it.next();
match it.clone().next() {
Some(c) if test_valid(*c) => {
if skip_underscores(&mut it, test_valid)? {
has_underscores = true;
}
}
Some(_) | None => {}
}
};
// Figure out if there's an exponential part here to make a float, and
// if so parse it but defer its actual calculation until later.
match (hex, it.next()) {
(true, Some(b'p')) | (true, Some(b'P')) | (false, Some(b'e')) | (false, Some(b'E')) => {
match it.clone().next() {
Some(b'-') => {
it.next();
}
Some(b'+') => {
it.next();
}
_ => {}
}
if skip_underscores(&mut it, |x| char::from(x).is_ascii_digit())? {
has_underscores = true;
}
}
(_, None) => {}
_ => return None,
}
// We should have eaten everything by now, if not then this is surely
// not a float or integer literal.
if it.next().is_some() {
return None;
}
return Some(TokenKind::Float(FloatKind::Normal {
has_underscores,
hex,
}));
fn skip_underscores<'a>(
it: &mut slice::Iter<'_, u8>,
good: fn(u8) -> bool,
) -> Option<bool> {
let mut last_underscore = false;
let mut has_underscores = false;
let first = *it.next()?;
if !good(first) {
return None;
}
while let Some(c) = it.clone().next() {
if *c == b'_' && !last_underscore {
has_underscores = true;
it.next();
last_underscore = true;
continue;
}
if !good(*c) {
break;
}
last_underscore = false;
it.next();
}
if last_underscore {
return None;
}
Some(has_underscores)
}
}
/// Verifies that `comment`, which is about to be returned, has a "confusing
/// unicode character" in it and should instead be transformed into an
/// error.
fn check_confusing_comment(&self, end: usize, comment: &str) -> Result<(), Error> {
if self.allow_confusing_unicode {
return Ok(());
}
// In an effort to avoid utf-8 decoding the entire `comment` the search
// here is a bit more optimized. This checks for the `0xe2` byte because
// in the utf-8 encoding that's the leading encoding byte for all
// "confusing characters". Each instance of 0xe2 is checked to see if it
// starts a confusing character, and if so that's returned.
//
// Also note that 0xe2 will never be found in the middle of a codepoint,
// it's always the start of a codepoint. This means that if our special
// characters show up they're guaranteed to start with 0xe2 bytes.
let bytes = comment.as_bytes();
for pos in memchr::Memchr::new(0xe2, bytes) {
if let Some(c) = comment[pos..].chars().next() {
if is_confusing_unicode(c) {
// Note that `self.cur()` accounts for already having
// parsed `comment`, so we move backwards to where
// `comment` started and then add the index within
// `comment`.
let pos = end - comment.len() + pos;
return Err(self.error(pos, LexError::ConfusingUnicode(c)));
}
}
}
Ok(())
}
fn parse_str(
it: &mut str::Chars<'a>,
allow_confusing_unicode: bool,
) -> Result<Cow<'a, [u8]>, LexError> {
enum State {
Start,
String(Vec<u8>),
}
let orig = it.as_str();
let mut state = State::Start;
loop {
match it.next().ok_or(LexError::UnexpectedEof)? {
'"' => break,
'\\' => {
match state {
State::String(_) => {}
State::Start => {
let pos = orig.len() - it.as_str().len() - 1;
state = State::String(orig[..pos].as_bytes().to_vec());
}
}
let buf = match &mut state {
State::String(b) => b,
State::Start => unreachable!(),
};
match it.next().ok_or(LexError::UnexpectedEof)? {
'"' => buf.push(b'"'),
'\'' => buf.push(b'\''),
't' => buf.push(b'\t'),
'n' => buf.push(b'\n'),
'r' => buf.push(b'\r'),
'\\' => buf.push(b'\\'),
'u' => {
Lexer::must_eat_char(it, '{')?;
let n = Lexer::hexnum(it)?;
let c = char::from_u32(n).ok_or(LexError::InvalidUnicodeValue(n))?;
buf.extend(c.encode_utf8(&mut [0; 4]).as_bytes());
Lexer::must_eat_char(it, '}')?;
}
c1 if c1.is_ascii_hexdigit() => {
let c2 = Lexer::hexdigit(it)?;
buf.push(to_hex(c1) * 16 + c2);
}
c => return Err(LexError::InvalidStringEscape(c)),
}
}
c if (c as u32) < 0x20 || c as u32 == 0x7f => {
return Err(LexError::InvalidStringElement(c))
}
c if !allow_confusing_unicode && is_confusing_unicode(c) => {
return Err(LexError::ConfusingUnicode(c))
}
c => match &mut state {
State::Start => {}
State::String(v) => {
v.extend(c.encode_utf8(&mut [0; 4]).as_bytes());
}
},
}
}
match state {
State::Start => Ok(orig[..orig.len() - it.as_str().len() - 1].as_bytes().into()),
State::String(s) => Ok(s.into()),
}
}
/// Parses an id-or-string-based name from `it`.
///
/// Note that `it` should already have been lexed and this is just
/// extracting the value. If the token lexed was `@a` then this should point
/// to `a`.
///
/// This will automatically detect quoted syntax such as `@"..."` and the
/// byte string will be parsed and validated as utf-8.
///
/// # Errors
///
/// Returns an error if a quoted byte string is found and contains invalid
/// utf-8.
fn parse_name(it: &mut str::Chars<'a>) -> Result<Cow<'a, str>, LexError> {
if it.clone().next() == Some('"') {
it.next();
match Lexer::parse_str(it, true)? {
Cow::Borrowed(bytes) => match std::str::from_utf8(bytes) {
Ok(s) => Ok(Cow::Borrowed(s)),
Err(e) => Err(LexError::InvalidUtf8Id(e)),
},
Cow::Owned(bytes) => match String::from_utf8(bytes) {
Ok(s) => Ok(Cow::Owned(s)),
Err(e) => Err(LexError::InvalidUtf8Id(e.utf8_error())),
},
}
} else {
Ok(Cow::Borrowed(it.as_str()))
}
}
fn hexnum(it: &mut str::Chars<'_>) -> Result<u32, LexError> {
let n = Lexer::hexdigit(it)?;
let mut last_underscore = false;
let mut n = n as u32;
while let Some(c) = it.clone().next() {
if c == '_' {
it.next();
last_underscore = true;
continue;
}
if !c.is_ascii_hexdigit() {
break;
}
last_underscore = false;
it.next();
n = n
.checked_mul(16)
.and_then(|n| n.checked_add(to_hex(c) as u32))
.ok_or(LexError::NumberTooBig)?;
}
if last_underscore {
return Err(LexError::LoneUnderscore);
}
Ok(n)
}
/// Reads a hexidecimal digit from the input stream, returning where it's
/// defined and the hex value. Returns an error on EOF or an invalid hex
/// digit.
fn hexdigit(it: &mut str::Chars<'_>) -> Result<u8, LexError> {
let ch = Lexer::must_char(it)?;
if ch.is_ascii_hexdigit() {
Ok(to_hex(ch))
} else {
Err(LexError::InvalidHexDigit(ch))
}
}
/// Reads the next character from the input string and where it's located,
/// returning an error if the input stream is empty.
fn must_char(it: &mut str::Chars<'_>) -> Result<char, LexError> {
it.next().ok_or(LexError::UnexpectedEof)
}
/// Expects that a specific character must be read next
fn must_eat_char(it: &mut str::Chars<'_>, wanted: char) -> Result<(), LexError> {
let found = Lexer::must_char(it)?;
if wanted == found {
Ok(())
} else {
Err(LexError::Expected { wanted, found })
}
}
/// Creates an error at `pos` with the specified `kind`
fn error(&self, pos: usize, kind: LexError) -> Error {
Error::lex(Span { offset: pos }, self.input, kind)
}
/// Returns an iterator over all tokens in the original source string
/// starting at the `pos` specified.
pub fn iter(&self, mut pos: usize) -> impl Iterator<Item = Result<Token, Error>> + '_ {
std::iter::from_fn(move || self.parse(&mut pos).transpose())
}
/// Returns whether an annotation is present at `pos`. If it is present then
/// `Ok(Some(token))` is returned corresponding to the token, otherwise
/// `Ok(None)` is returned. If the next token cannot be parsed then an error
/// is returned.
pub fn annotation(&self, mut pos: usize) -> Result<Option<Token>, Error> {
let bytes = self.input.as_bytes();
// Quickly reject anything that for sure isn't an annotation since this
// method is used every time an lparen is parsed.
if bytes.get(pos) != Some(&b'@') {
return Ok(None);
}
match self.parse(&mut pos)? {
Some(token) => match token.kind {
TokenKind::Annotation => Ok(Some(token)),
_ => Ok(None),
},
None => Ok(None),
}
}
}
impl Token {
/// Returns the original source text for this token.
pub fn src<'a>(&self, s: &'a str) -> &'a str {
&s[self.offset..][..self.len.try_into().unwrap()]
}
/// Returns the identifier, without the leading `$` symbol, that this token
/// represents.
///
/// Note that this method returns the contents of the identifier. With a
/// string-based identifier this means that escapes have been resolved to
/// their string-based equivalent.
///
/// Should only be used with `TokenKind::Id`.
///
/// # Errors
///
/// Returns an error if this is a string-based identifier (e.g. `$"..."`)
/// which is invalid utf-8.
pub fn id<'a>(&self, s: &'a str) -> Result<Cow<'a, str>, Error> {
let mut ch = self.src(s).chars();
let dollar = ch.next();
debug_assert_eq!(dollar, Some('$'));
let id = Lexer::parse_name(&mut ch).map_err(|e| self.error(s, e))?;
if id.is_empty() {
return Err(self.error(s, LexError::EmptyId));
}
Ok(id)
}
/// Returns the annotation, without the leading `@` symbol, that this token
/// represents.
///
/// Note that this method returns the contents of the identifier. With a
/// string-based identifier this means that escapes have been resolved to
/// their string-based equivalent.
///
/// Should only be used with `TokenKind::Annotation`.
///
/// # Errors
///
/// Returns an error if this is a string-based identifier (e.g. `$"..."`)
/// which is invalid utf-8.
pub fn annotation<'a>(&self, s: &'a str) -> Result<Cow<'a, str>, Error> {
let mut ch = self.src(s).chars();
let at = ch.next();
debug_assert_eq!(at, Some('@'));
let id = Lexer::parse_name(&mut ch).map_err(|e| self.error(s, e))?;
if id.is_empty() {
return Err(self.error(s, LexError::EmptyAnnotation));
}
Ok(id)
}
/// Returns the keyword this token represents.
///
/// Should only be used with [`TokenKind::Keyword`].
pub fn keyword<'a>(&self, s: &'a str) -> &'a str {
self.src(s)
}
/// Returns the reserved string this token represents.
///
/// Should only be used with [`TokenKind::Reserved`].
pub fn reserved<'a>(&self, s: &'a str) -> &'a str {
self.src(s)
}
/// Returns the parsed string that this token represents.
///
/// This returns either a raw byte slice into the source if that's possible
/// or an owned representation to handle escaped characters and such.
///
/// Should only be used with [`TokenKind::String`].
pub fn string<'a>(&self, s: &'a str) -> Cow<'a, [u8]> {
let mut ch = self.src(s).chars();
ch.next().unwrap();
Lexer::parse_str(&mut ch, true).unwrap()
}
/// Returns the decomposed float token that this represents.
///
/// This will slice up the float token into its component parts and return a
/// description of the float token in the source.
///
/// Should only be used with [`TokenKind::Float`].
pub fn float<'a>(&self, s: &'a str, kind: FloatKind) -> Float<'a> {
match kind {
FloatKind::Inf { negative } => Float::Inf { negative },
FloatKind::Nan { negative } => Float::Nan {
val: None,
negative,
},
FloatKind::NanVal {
negative,
has_underscores,
} => {
let src = self.src(s);
let src = if src.starts_with("n") { src } else { &src[1..] };
let mut val = Cow::Borrowed(src.strip_prefix("nan:0x").unwrap());
if has_underscores {
*val.to_mut() = val.replace("_", "");
}
Float::Nan {
val: Some(val),
negative,
}
}
FloatKind::Normal {
has_underscores,
hex,
} => {
let src = self.src(s);
let (integral, decimal, exponent) = match src.find('.') {
Some(i) => {
let integral = &src[..i];
let rest = &src[i + 1..];
let exponent = if hex {
rest.find('p').or_else(|| rest.find('P'))
} else {
rest.find('e').or_else(|| rest.find('E'))
};
match exponent {
Some(i) => (integral, Some(&rest[..i]), Some(&rest[i + 1..])),
None => (integral, Some(rest), None),
}
}
None => {
let exponent = if hex {
src.find('p').or_else(|| src.find('P'))
} else {
src.find('e').or_else(|| src.find('E'))
};
match exponent {
Some(i) => (&src[..i], None, Some(&src[i + 1..])),
None => (src, None, None),
}
}
};
let mut integral = Cow::Borrowed(integral.strip_prefix('+').unwrap_or(integral));
let mut decimal = decimal.and_then(|s| {
if s.is_empty() {
None
} else {
Some(Cow::Borrowed(s))
}
});
let mut exponent =
exponent.map(|s| Cow::Borrowed(s.strip_prefix('+').unwrap_or(s)));
if has_underscores {
*integral.to_mut() = integral.replace("_", "");
if let Some(decimal) = &mut decimal {
*decimal.to_mut() = decimal.replace("_", "");
}
if let Some(exponent) = &mut exponent {
*exponent.to_mut() = exponent.replace("_", "");
}
}
if hex {
*integral.to_mut() = integral.replace("0x", "");
}
Float::Val {
hex,
integral,
decimal,
exponent,
}
}
}
}
/// Returns the decomposed integer token that this represents.
///
/// This will slice up the integer token into its component parts and
/// return a description of the integer token in the source.
///
/// Should only be used with [`TokenKind::Integer`].
pub fn integer<'a>(&self, s: &'a str, kind: IntegerKind) -> Integer<'a> {
let src = self.src(s);
let val = match kind.sign {
Some(SignToken::Plus) => src.strip_prefix('+').unwrap(),
Some(SignToken::Minus) => src,
None => src,
};
let mut val = Cow::Borrowed(val);
if kind.has_underscores {
*val.to_mut() = val.replace("_", "");
}
if kind.hex {
*val.to_mut() = val.replace("0x", "");
}
Integer {
sign: kind.sign,
hex: kind.hex,
val,
}
}
fn error(&self, src: &str, err: LexError) -> Error {
Error::lex(
Span {
offset: self.offset,
},
src,
err,
)
}
}
impl<'a> Integer<'a> {
/// Returns the sign token for this integer.
pub fn sign(&self) -> Option<SignToken> {
self.sign
}
/// Returns the value string that can be parsed for this integer, as well
/// as the base that it should be parsed in
pub fn val(&self) -> (&str, u32) {
(&self.val, if self.hex { 16 } else { 10 })
}
}
fn to_hex(c: char) -> u8 {
match c {
'a'..='f' => c as u8 - b'a' + 10,
'A'..='F' => c as u8 - b'A' + 10,
_ => c as u8 - b'0',
}
}
impl fmt::Display for LexError {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
use LexError::*;
match self {
DanglingBlockComment => f.write_str("unterminated block comment")?,
Unexpected(c) => write!(f, "unexpected character '{}'", escape_char(*c))?,
InvalidStringElement(c) => {
write!(f, "invalid character in string '{}'", escape_char(*c))?
}
InvalidStringEscape(c) => write!(f, "invalid string escape '{}'", escape_char(*c))?,
InvalidHexDigit(c) => write!(f, "invalid hex digit '{}'", escape_char(*c))?,
InvalidDigit(c) => write!(f, "invalid decimal digit '{}'", escape_char(*c))?,
Expected { wanted, found } => write!(
f,
"expected '{}' but found '{}'",
escape_char(*wanted),
escape_char(*found)
)?,
UnexpectedEof => write!(f, "unexpected end-of-file")?,
NumberTooBig => f.write_str("number is too big to parse")?,
InvalidUnicodeValue(c) => write!(f, "invalid unicode scalar value 0x{:x}", c)?,
LoneUnderscore => write!(f, "bare underscore in numeric literal")?,
ConfusingUnicode(c) => write!(f, "likely-confusing unicode character found {:?}", c)?,
InvalidUtf8Id(_) => write!(f, "malformed UTF-8 encoding of string-based id")?,
EmptyId => write!(f, "empty identifier")?,
EmptyAnnotation => write!(f, "empty annotation id")?,
}
Ok(())
}
}
fn escape_char(c: char) -> String {
match c {
'\t' => String::from("\\t"),
'\r' => String::from("\\r"),
'\n' => String::from("\\n"),
'\\' => String::from("\\\\"),
'\'' => String::from("\\\'"),
'\"' => String::from("\""),
'\x20'..='\x7e' => String::from(c),
_ => c.escape_unicode().to_string(),
}
}
/// This is an attempt to protect agains the "trojan source" [1] problem where
/// unicode characters can cause editors to render source code differently
/// for humans than the compiler itself sees.
///
/// To mitigate this issue, and because it's relatively rare in practice,
/// this simply rejects characters of that form.
///
/// [1]: https://www.trojansource.codes/
fn is_confusing_unicode(ch: char) -> bool {
matches!(
ch,
'\u{202a}'
| '\u{202b}'
| '\u{202d}'
| '\u{202e}'
| '\u{2066}'
| '\u{2067}'
| '\u{2068}'
| '\u{206c}'
| '\u{2069}'
)
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn ws_smoke() {
fn get_whitespace(input: &str) -> &str {
let token = get_token(input);
match token.kind {
TokenKind::Whitespace => token.src(input),
other => panic!("unexpected {:?}", other),
}
}
assert_eq!(get_whitespace(" "), " ");
assert_eq!(get_whitespace(" "), " ");
assert_eq!(get_whitespace(" \n "), " \n ");
assert_eq!(get_whitespace(" x"), " ");
assert_eq!(get_whitespace(" ;"), " ");
}
#[test]
fn line_comment_smoke() {
fn get_line_comment(input: &str) -> &str {
let token = get_token(input);
match token.kind {
TokenKind::LineComment => token.src(input),
other => panic!("unexpected {:?}", other),
}
}
assert_eq!(get_line_comment(";;"), ";;");
assert_eq!(get_line_comment(";; xyz"), ";; xyz");
assert_eq!(get_line_comment(";; xyz\nabc"), ";; xyz");
assert_eq!(get_line_comment(";;\nabc"), ";;");
assert_eq!(get_line_comment(";; \nabc"), ";; ");
assert_eq!(get_line_comment(";; \rabc"), ";; ");
assert_eq!(get_line_comment(";; \r\nabc"), ";; ");
}
#[test]
fn block_comment_smoke() {
fn get_block_comment(input: &str) -> &str {
let token = get_token(input);
match token.kind {
TokenKind::BlockComment => token.src(input),
other => panic!("unexpected {:?}", other),
}
}
assert_eq!(get_block_comment("(;;)"), "(;;)");
assert_eq!(get_block_comment("(; ;)"), "(; ;)");
assert_eq!(get_block_comment("(; (;;) ;)"), "(; (;;) ;)");
}
fn get_token(input: &str) -> Token {
Lexer::new(input)
.parse(&mut 0)
.expect("no first token")
.expect("no token")
}
#[test]
fn lparen() {
assert_eq!(get_token("((").kind, TokenKind::LParen);
}
#[test]
fn rparen() {
assert_eq!(get_token(")(").kind, TokenKind::RParen);
}
#[test]
fn strings() {
fn get_string(input: &str) -> Vec<u8> {
let token = get_token(input);
match token.kind {
TokenKind::String => token.string(input).to_vec(),
other => panic!("not keyword {:?}", other),
}
}
assert_eq!(&*get_string("\"\""), b"");
assert_eq!(&*get_string("\"a\""), b"a");
assert_eq!(&*get_string("\"a b c d\""), b"a b c d");
assert_eq!(&*get_string("\"\\\"\""), b"\"");
assert_eq!(&*get_string("\"\\'\""), b"'");
assert_eq!(&*get_string("\"\\n\""), b"\n");
assert_eq!(&*get_string("\"\\t\""), b"\t");
assert_eq!(&*get_string("\"\\r\""), b"\r");
assert_eq!(&*get_string("\"\\\\\""), b"\\");
assert_eq!(&*get_string("\"\\01\""), &[1]);
assert_eq!(&*get_string("\"\\u{1}\""), &[1]);
assert_eq!(
&*get_string("\"\\u{0f3}\""),
'\u{0f3}'.encode_utf8(&mut [0; 4]).as_bytes()
);
assert_eq!(
&*get_string("\"\\u{0_f_3}\""),
'\u{0f3}'.encode_utf8(&mut [0; 4]).as_bytes()
);
for i in 0..=255i32 {
let s = format!("\"\\{:02x}\"", i);
assert_eq!(&*get_string(&s), &[i as u8]);
}
}
#[test]
fn id() {
fn get_id(input: &str) -> String {
let token = get_token(input);
match token.kind {
TokenKind::Id => token.id(input).unwrap().to_string(),
other => panic!("not id {:?}", other),
}
}
assert_eq!(get_id("$x"), "x");
assert_eq!(get_id("$xyz"), "xyz");
assert_eq!(get_id("$x_z"), "x_z");
assert_eq!(get_id("$0^"), "0^");
assert_eq!(get_id("$0^;;"), "0^");
assert_eq!(get_id("$0^ ;;"), "0^");
assert_eq!(get_id("$\"x\" ;;"), "x");
}
#[test]
fn annotation() {
fn get_annotation(input: &str) -> String {
let token = get_token(input);
match token.kind {
TokenKind::Annotation => token.annotation(input).unwrap().to_string(),
other => panic!("not annotation {:?}", other),
}
}
assert_eq!(get_annotation("@foo"), "foo");
assert_eq!(get_annotation("@foo "), "foo");
assert_eq!(get_annotation("@f "), "f");
assert_eq!(get_annotation("@\"x\" "), "x");
assert_eq!(get_annotation("@0 "), "0");
}
#[test]
fn keyword() {
fn get_keyword(input: &str) -> &str {
let token = get_token(input);
match token.kind {
TokenKind::Keyword => token.keyword(input),
other => panic!("not keyword {:?}", other),
}
}
assert_eq!(get_keyword("x"), "x");
assert_eq!(get_keyword("xyz"), "xyz");
assert_eq!(get_keyword("x_z"), "x_z");
assert_eq!(get_keyword("x_z "), "x_z");
assert_eq!(get_keyword("x_z "), "x_z");
}
#[test]
fn reserved() {
fn get_reserved(input: &str) -> &str {
let token = get_token(input);
match token.kind {
TokenKind::Reserved => token.reserved(input),
other => panic!("not reserved {:?}", other),
}
}
assert_eq!(get_reserved("^_x "), "^_x");
}
#[test]
fn integer() {
fn get_integer(input: &str) -> String {
let token = get_token(input);
match token.kind {
TokenKind::Integer(i) => token.integer(input, i).val.to_string(),
other => panic!("not integer {:?}", other),
}
}
assert_eq!(get_integer("1"), "1");
assert_eq!(get_integer("0"), "0");
assert_eq!(get_integer("-1"), "-1");
assert_eq!(get_integer("+1"), "1");
assert_eq!(get_integer("+1_000"), "1000");
assert_eq!(get_integer("+1_0_0_0"), "1000");
assert_eq!(get_integer("+0x10"), "10");
assert_eq!(get_integer("-0x10"), "-10");
assert_eq!(get_integer("0x10"), "10");
}
#[test]
fn float() {
fn get_float(input: &str) -> Float<'_> {
let token = get_token(input);
match token.kind {
TokenKind::Float(f) => token.float(input, f),
other => panic!("not float {:?}", other),
}
}
assert_eq!(
get_float("nan"),
Float::Nan {
val: None,
negative: false
},
);
assert_eq!(
get_float("-nan"),
Float::Nan {
val: None,
negative: true,
},
);
assert_eq!(
get_float("+nan"),
Float::Nan {
val: None,
negative: false,
},
);
assert_eq!(
get_float("+nan:0x1"),
Float::Nan {
val: Some("1".into()),
negative: false,
},
);
assert_eq!(
get_float("nan:0x7f_ffff"),
Float::Nan {
val: Some("7fffff".into()),
negative: false,
},
);
assert_eq!(get_float("inf"), Float::Inf { negative: false });
assert_eq!(get_float("-inf"), Float::Inf { negative: true });
assert_eq!(get_float("+inf"), Float::Inf { negative: false });
assert_eq!(
get_float("1.2"),
Float::Val {
integral: "1".into(),
decimal: Some("2".into()),
exponent: None,
hex: false,
},
);
assert_eq!(
get_float("1.2e3"),
Float::Val {
integral: "1".into(),
decimal: Some("2".into()),
exponent: Some("3".into()),
hex: false,
},
);
assert_eq!(
get_float("-1_2.1_1E+0_1"),
Float::Val {
integral: "-12".into(),
decimal: Some("11".into()),
exponent: Some("01".into()),
hex: false,
},
);
assert_eq!(
get_float("+1_2.1_1E-0_1"),
Float::Val {
integral: "12".into(),
decimal: Some("11".into()),
exponent: Some("-01".into()),
hex: false,
},
);
assert_eq!(
get_float("0x1_2.3_4p5_6"),
Float::Val {
integral: "12".into(),
decimal: Some("34".into()),
exponent: Some("56".into()),
hex: true,
},
);
assert_eq!(
get_float("+0x1_2.3_4P-5_6"),
Float::Val {
integral: "12".into(),
decimal: Some("34".into()),
exponent: Some("-56".into()),
hex: true,
},
);
assert_eq!(
get_float("1."),
Float::Val {
integral: "1".into(),
decimal: None,
exponent: None,
hex: false,
},
);
assert_eq!(
get_float("0x1p-24"),
Float::Val {
integral: "1".into(),
decimal: None,
exponent: Some("-24".into()),
hex: true,
},
);
}
}