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
|
//===----------------------------------------------------------------------===//
//
// This source file is part of the Swift.org open source project
//
// Copyright (c) 2014 - 2023 Apple Inc. and the Swift project authors
// Licensed under Apache License v2.0 with Runtime Library Exception
//
// See https://swift.org/LICENSE.txt for license information
// See https://swift.org/CONTRIBUTORS.txt for the list of Swift project authors
//
//===----------------------------------------------------------------------===//
#if swift(>=6)
@_spi(RawSyntax) internal import SwiftSyntax
#else
@_spi(RawSyntax) import SwiftSyntax
#endif
// MARK: - Check multiline string literal indentation
fileprivate class StringLiteralExpressionIndentationChecker {
// MARK: Entry
init(expectedIndentation: SyntaxText, arena: SyntaxArena) {
self.expectedIndentation = expectedIndentation
self.arena = arena
}
func checkIndentation(of expressionSegment: RawExpressionSegmentSyntax) -> RawExpressionSegmentSyntax? {
if let rewrittenSegment = self.visit(node: RawSyntax(expressionSegment)) {
return rewrittenSegment.as(RawExpressionSegmentSyntax.self)
} else {
return nil
}
}
// MARK: Implementation
private let expectedIndentation: SyntaxText
private let arena: SyntaxArena
private func visit(node: RawSyntax) -> RawSyntax? {
if node.isToken {
return visitTokenNode(token: node.as(RawTokenSyntax.self)!)
} else {
return visitLayoutNode(node: node)
}
}
private func visitTokenNode(token: RawTokenSyntax) -> RawSyntax? {
if !token.leadingTriviaPieces.contains(where: { $0.isNewline }) {
// Only checking tokens on a newline
return nil
}
let hasSufficientIndentation = token.tokenView.leadingTrivia { leadingTrivia -> Bool in
let indentationStartIndex =
leadingTrivia
.lastIndex(where: { $0 == UInt8(ascii: "\n") || $0 == UInt8(ascii: "\r") })?
.advanced(by: 1)
?? leadingTrivia.startIndex
return SyntaxText(rebasing: leadingTrivia[indentationStartIndex...]).hasPrefix(expectedIndentation)
}
if hasSufficientIndentation {
return nil
}
if let tokenDiagnostic = token.tokenView.tokenDiagnostic, tokenDiagnostic.severity == .error {
// Token already has a lexer error, ignore the indentation error until that
// error is fixed
return nil
}
let tokenWithDiagnostic = token.tokenView.withTokenDiagnostic(
tokenDiagnostic: TokenDiagnostic(.insufficientIndentationInMultilineStringLiteral, byteOffset: 0),
arena: arena
)
return RawSyntax(tokenWithDiagnostic)
}
private func visitLayoutNode(node: RawSyntax) -> RawSyntax? {
let layoutView = node.layoutView!
var hasRewrittenChild = false
var rewrittenChildren: [RawSyntax?] = []
for child in layoutView.children {
if let child, let rewrittenChild = visit(node: child) {
hasRewrittenChild = true
rewrittenChildren.append(rewrittenChild)
} else {
rewrittenChildren.append(child)
}
}
precondition(rewrittenChildren.count == layoutView.children.count)
if hasRewrittenChild {
return layoutView.replacingLayout(with: rewrittenChildren, arena: arena)
} else {
return nil
}
}
}
// MARK: - Post-process multi-line string literals
fileprivate extension SyntaxText {
/// If the text ends with any newline character, return that character,
/// otherwise `nil`.
var newlineSuffix: SyntaxText? {
if hasSuffix("\r\n") {
return "\r\n"
} else if hasSuffix("\n") {
return "\n"
} else if hasSuffix("\r") {
return "\r"
} else {
return nil
}
}
/// If the text is a single newline character, return the trivia piece that
/// represents it, otherwise `nil`.
var triviaPieceIfNewline: RawTriviaPiece? {
if self == "\r\n" {
return .carriageReturnLineFeeds(1)
} else if self == "\n" {
return .newlines(1)
} else if self == "\r" {
return .carriageReturns(1)
} else {
return nil
}
}
}
extension Parser {
/// If `text` only consists of indentation whitespace (space and tab), return
/// the trivia pieces that represent `text`, otherwise return `nil`.
private func parseIndentationTrivia(text: SyntaxText) -> [RawTriviaPiece]? {
let trivia = TriviaParser.parseTrivia(text, position: .leading)
if trivia.allSatisfy({ $0.isIndentationWhitespace }) {
return trivia
} else {
return nil
}
}
private func reclassifyTrivia(
in token: RawTokenSyntax,
leading reclassifyLeading: SyntaxText = "",
trailing reclassifyTrailing: SyntaxText = "",
tokenDiagnostic: TokenDiagnostic? = nil
) -> RawTokenSyntax {
precondition(SyntaxText(rebasing: token.tokenText.prefix(reclassifyLeading.count)) == reclassifyLeading)
precondition(SyntaxText(rebasing: token.tokenText.suffix(reclassifyTrailing.count)) == reclassifyTrailing)
return RawTokenSyntax(
kind: token.tokenKind,
text: SyntaxText(rebasing: token.tokenText.dropFirst(reclassifyLeading.count).dropLast(reclassifyTrailing.count)),
leadingTriviaPieces: token.leadingTriviaPieces + TriviaParser.parseTrivia(reclassifyLeading, position: .leading),
trailingTriviaPieces: TriviaParser.parseTrivia(reclassifyTrailing, position: .trailing)
+ token.trailingTriviaPieces,
presence: token.presence,
tokenDiagnostic: token.tokenView.tokenDiagnostic ?? tokenDiagnostic,
arena: self.arena
)
}
/// Re-classify the newline of the last line as trivia since the newline is
/// not part of the represented string. If the last line has its newline
/// escaped by a trailing `\`, mark that string segment as unexpected and
/// generate a missing segment that doesn't have a trailing `\`.
///
/// Returns `true` if the closing quote is on its own line and `false` otherwise.
private func reclassifyNewlineOfLastSegmentAsTrivia(
rawStringDelimitersToken: RawTokenSyntax?,
openQuoteHasTrailingNewline: Bool,
middleSegments: inout [RawStringLiteralSegmentListSyntax.Element]
) -> Bool {
guard let segment = middleSegments.last else {
return openQuoteHasTrailingNewline
}
switch segment {
case .stringSegment(let lastMiddleSegment):
if !lastMiddleSegment.content.trailingTriviaPieces.isEmpty {
precondition(
lastMiddleSegment.content.trailingTriviaPieces.contains(where: { $0.isBackslash }),
"The lexer should only add trailing trivia to a string segment if the newline is escaped by a backslash"
)
if rawStringDelimitersToken != nil {
// ... except in raw string literals where the C++ parser accepts the
// last line to be escaped. To match the C++ parser's behavior, we also
// need to allow escaped newline in raw string literals.
return true
}
let unexpectedBeforeContent = lastMiddleSegment.content
let content = RawTokenSyntax(
missing: .stringSegment,
text: lastMiddleSegment.content.tokenText,
leadingTriviaPieces: lastMiddleSegment.content.leadingTriviaPieces,
trailingTriviaPieces: lastMiddleSegment.content.trailingTriviaPieces.filter({ $0.isNewline }),
arena: self.arena
)
middleSegments[middleSegments.count - 1] = .stringSegment(
RawStringSegmentSyntax(
RawUnexpectedNodesSyntax(
combining: lastMiddleSegment.unexpectedBeforeContent,
unexpectedBeforeContent,
arena: self.arena
),
content: content,
lastMiddleSegment.unexpectedAfterContent,
arena: self.arena
)
)
return true
} else if let newlineSuffix = lastMiddleSegment.content.tokenText.newlineSuffix {
// The newline at the end of the last line in the string literal is not part of the represented string.
// Mark it as trivia.
let content = self.reclassifyTrivia(in: lastMiddleSegment.content, trailing: newlineSuffix)
middleSegments[middleSegments.count - 1] = .stringSegment(
RawStringSegmentSyntax(
lastMiddleSegment.unexpectedBeforeContent,
content: content,
lastMiddleSegment.unexpectedAfterContent,
arena: self.arena
)
)
return true
} else {
return false
}
case .expressionSegment:
return false
#if RESILIENT_LIBRARIES
@unknown default:
fatalError()
#endif
}
}
/// Re-classifying the indentation of the middle lines as trivia, such that
/// the string segment token only contains whitespace that is part of the
/// actual string. If a line is under-indented a
/// `.insufficientIndentationInMultilineStringLiteral` lexer error will be
/// attached to the string segment token.
private func postProcessIndentationAndEscapedNewlineOfMiddleSegments(
rawStringDelimitersToken: RawTokenSyntax?,
middleSegments: inout [RawStringLiteralSegmentListSyntax.Element],
isFirstSegmentOnNewLine: Bool,
indentation: SyntaxText
) {
let expressionIndentationChecker = StringLiteralExpressionIndentationChecker(
expectedIndentation: indentation,
arena: self.arena
)
var isSegmentOnNewLine = isFirstSegmentOnNewLine
for (index, segment) in middleSegments.enumerated() {
switch segment {
case .stringSegment(var segment):
// We are not considering leading trivia for indentation computation.
// If these assertions are violated, we can probably lift them but we
// would need to check that they produce the expected results.
precondition(segment.content.leadingTriviaByteLength == 0)
// Re-classify indentation as leading trivia
if isSegmentOnNewLine {
if segment.content.tokenText.hasPrefix(indentation) {
segment = RawStringSegmentSyntax(
segment.unexpectedBeforeContent,
content: self.reclassifyTrivia(in: segment.content, leading: indentation),
segment.unexpectedAfterContent,
arena: self.arena
)
} else if (segment.content.tokenText == "" || segment.content.tokenText.triviaPieceIfNewline != nil)
&& segment.content.trailingTriviaPieces.allSatisfy({ $0.isNewline })
{
// Empty lines don't need to be indented and there's no indentation we need to strip away.
} else {
let actualIndentation = SyntaxText(
rebasing: segment.content.tokenText.prefix(while: { $0 == UInt8(ascii: " ") || $0 == UInt8(ascii: "\t") })
)
let tokenDiagnostic = TokenDiagnostic(.insufficientIndentationInMultilineStringLiteral, byteOffset: 0)
let content = self.reclassifyTrivia(
in: segment.content,
leading: actualIndentation,
tokenDiagnostic: tokenDiagnostic
)
segment = RawStringSegmentSyntax(
segment.unexpectedBeforeContent,
content: content,
segment.unexpectedAfterContent,
arena: self.arena
)
}
}
isSegmentOnNewLine =
segment.content.tokenText.newlineSuffix != nil
|| (segment.content.trailingTriviaPieces.last?.isNewline ?? false)
middleSegments[index] = .stringSegment(segment)
case .expressionSegment(let segment):
isSegmentOnNewLine = segment.rightParen.trailingTriviaPieces.contains(where: { $0.isNewline })
if let rewrittenSegment = expressionIndentationChecker.checkIndentation(of: segment) {
middleSegments[index] = .expressionSegment(rewrittenSegment)
}
#if RESILIENT_LIBRARIES
@unknown default:
fatalError()
#endif
}
}
}
/// After parsing a multi-line string literal from tokens produced by the lexer,
/// post-process the string literal by performing the following steps
/// - Ensuring that the open quote is followed by a newline (i.e. the opening
/// quote shouldn't have any any text on the same line). If this is not the
/// case, mark the existing opening quote as unexpected and synthesize
/// a missing quote that is followed by a newline
/// - Re-classifying the indentation of the middle lines as trivia, such that
/// the string segment token only contains whitespace that is part of the
/// actual string. If a line is under-indented a
/// `.insufficientIndentationInMultilineStringLiteral` lexer error will be
/// attached to the string segment token.
/// - Re-classify the newline of the last line as trivia since the newline is
/// not part of the represented string. If the last line has its newline
/// escaped by a trailing `\`, mark that string segment as unexpected and
/// generate a missing segment that doesn't have a trailing `\`.
private func postProcessMultilineStringLiteral(
rawStringDelimitersToken: RawTokenSyntax?,
openQuote: RawTokenSyntax,
segments allSegments: [RawStringLiteralSegmentListSyntax.Element],
closeQuote: RawTokenSyntax
) -> (
unexpectedBeforeOpeningQuote: [RawTokenSyntax],
openingQuote: RawTokenSyntax,
segments: [RawStringLiteralSegmentListSyntax.Element],
unexpectedBeforeClosingQuote: [RawTokenSyntax],
closingQuote: RawTokenSyntax
) {
// -------------------------------------------------------------------------
// Precondition
precondition(
allSegments.allSatisfy {
if case .stringSegment(let segment) = $0 {
return segment.unexpectedBeforeContent == nil
&& segment.unexpectedAfterContent == nil
&& segment.content.leadingTriviaByteLength == 0
} else {
return true
}
},
"String segment produced by the lexer should not have unexpected text or trivia because we would drop it during post-processing"
)
// -------------------------------------------------------------------------
// Variables
var middleSegments = allSegments
// In a well-formed string literal, the last segment only consists of whitespace and contains the closing quote's indentation
let lastSegment = middleSegments.popLast()?.as(RawStringSegmentSyntax.self)
let indentation: SyntaxText
let indentationTrivia: [RawTriviaPiece]
var unexpectedBeforeOpenQuote: [RawTokenSyntax] = []
var openQuote = openQuote
var unexpectedBeforeCloseQuote: [RawTokenSyntax] = []
var closeQuote = closeQuote
// -------------------------------------------------------------------------
// Check that the close quote is on new line
let openQuoteHasTrailingNewline = openQuote.trailingTriviaPieces.last?.isNewline ?? false
let closeDelimiterOnNewLine = reclassifyNewlineOfLastSegmentAsTrivia(
rawStringDelimitersToken: rawStringDelimitersToken,
openQuoteHasTrailingNewline: openQuoteHasTrailingNewline,
middleSegments: &middleSegments
)
if !closeDelimiterOnNewLine || closeQuote.leadingTriviaByteLength != 0 {
unexpectedBeforeCloseQuote = [closeQuote]
closeQuote = RawTokenSyntax(missing: closeQuote.tokenKind, leadingTriviaPieces: [.newlines(1)], arena: self.arena)
// The closing delimiter doesn't start on a new line and thus it doesn't
// make sense to try and extract indentation from it.
return (unexpectedBeforeOpenQuote, openQuote, allSegments, unexpectedBeforeCloseQuote, closeQuote)
}
// -------------------------------------------------------------------------
// Parse indentation of the closing quote
if let lastSegment,
let parsedTrivia = parseIndentationTrivia(text: lastSegment.content.tokenText)
{
indentationTrivia = parsedTrivia
indentation = lastSegment.content.tokenText
closeQuote = RawTokenSyntax(
kind: closeQuote.tokenKind,
text: closeQuote.tokenText,
leadingTriviaPieces: parsedTrivia + closeQuote.leadingTriviaPieces,
trailingTriviaPieces: closeQuote.trailingTriviaPieces,
presence: closeQuote.presence,
tokenDiagnostic: closeQuote.tokenView.tokenDiagnostic,
arena: self.arena
)
} else {
if let lastSegment = lastSegment {
indentationTrivia = TriviaParser.parseTrivia(lastSegment.content.tokenText, position: .leading).prefix(while: {
$0.isIndentationWhitespace
})
let indentationByteLength = indentationTrivia.reduce(0, { $0 + $1.byteLength })
indentation = SyntaxText(rebasing: lastSegment.content.tokenText[0..<indentationByteLength])
middleSegments.append(.stringSegment(lastSegment))
} else {
indentationTrivia = []
indentation = ""
}
unexpectedBeforeCloseQuote = [closeQuote]
closeQuote = RawTokenSyntax(
missing: closeQuote.tokenKind,
leadingTriviaPieces: [.newlines(1)] + indentationTrivia,
arena: self.arena
)
}
// -------------------------------------------------------------------------
// Check that the open quote is followed by newline
// Condition for the loop below that indicates whether the segment we are
// iterating over is on a new line.
if !openQuoteHasTrailingNewline {
unexpectedBeforeOpenQuote = [openQuote]
openQuote = RawTokenSyntax(
missing: openQuote.tokenKind,
trailingTriviaPieces: [.newlines(1)] + indentationTrivia,
arena: self.arena
)
}
// -------------------------------------------------------------------------
// Check indentation of segments and escaped newlines at end of segment
postProcessIndentationAndEscapedNewlineOfMiddleSegments(
rawStringDelimitersToken: rawStringDelimitersToken,
middleSegments: &middleSegments,
isFirstSegmentOnNewLine: openQuoteHasTrailingNewline,
indentation: indentation
)
// -------------------------------------------------------------------------
// Done
return (
unexpectedBeforeOpenQuote,
openQuote,
middleSegments,
unexpectedBeforeCloseQuote,
closeQuote
)
}
}
// MARK: - Parse string literals
extension Parser {
/// Consumes a raw string or extended regex delimiter that has the same
/// number of `#` as `openDelimiter`.
mutating func parsePoundDelimiter(
_ kind: RawTokenKind,
matching openDelimiter: RawTokenSyntax?
) -> (unexpectedBeforeCheckedDelimiter: RawUnexpectedNodesSyntax?, checkedDelimiter: RawTokenSyntax?) {
// Check for leadingTriviaText == "" so we don't consume the leading raw
// string delimiter of an upcoming string literal, e.g. in
// ```
// "normal literal"
// #"raw literal"#
// ```
let delimiter: RawTokenSyntax?
if self.at(TokenSpec(kind)) && self.currentToken.leadingTriviaText == "" {
delimiter = self.consumeAnyToken()
} else {
delimiter = nil
}
switch (openDelimiter, delimiter) {
case (nil, nil):
return (nil, nil)
case (let open?, nil):
return (nil, missingToken(kind, text: open.tokenText))
case (nil, .some):
return (RawUnexpectedNodesSyntax([delimiter], arena: self.arena), nil)
case (let open?, let close?):
if open.tokenText == close.tokenText {
return (nil, close)
} else {
return (RawUnexpectedNodesSyntax([delimiter], arena: self.arena), missingToken(kind, text: open.tokenText))
}
}
}
/// Parse a string literal expression.
mutating func parseStringLiteral() -> RawStringLiteralExprSyntax {
/// Parse opening raw string delimiter if exist.
let openingPounds = self.consume(if: .rawStringPoundDelimiter)
/// Try to parse @ in order to recover from Objective-C style literals
let unexpectedAtSign = self.consume(if: .atSign)
/// Parse open quote.
var (unexpectedBeforeOpeningQuote, openQuote) = self.expect(
.stringQuote,
.multilineStringQuote,
default: .stringQuote
)
unexpectedBeforeOpeningQuote = RawUnexpectedNodesSyntax(
combining: unexpectedAtSign,
unexpectedBeforeOpeningQuote,
arena: self.arena
)
var openQuoteKind: RawTokenKind = openQuote.tokenKind
if openQuote.isMissing, let singleQuote = self.consume(if: .singleQuote) {
unexpectedBeforeOpeningQuote = RawUnexpectedNodesSyntax(
combining: unexpectedBeforeOpeningQuote,
singleQuote,
arena: self.arena
)
openQuoteKind = .singleQuote
}
/// Parse segments.
var segments: [RawStringLiteralSegmentListSyntax.Element] = []
var loopProgress = LoopProgressCondition()
while self.hasProgressed(&loopProgress) {
// If we encounter a token with leading trivia, we're no longer in the
// string literal.
guard currentToken.leadingTriviaText.isEmpty else { break }
if let stringSegment = self.consume(if: .stringSegment, TokenSpec(.identifier, remapping: .stringSegment)) {
segments.append(.stringSegment(RawStringSegmentSyntax(content: stringSegment, arena: self.arena)))
} else if let backslash = self.consume(if: .backslash) {
let (unexpectedBeforeDelimiter, delimiter) = self.parsePoundDelimiter(
.rawStringPoundDelimiter,
matching: openingPounds
)
let leftParen = self.expectWithoutRecoveryOrLeadingTrivia(.leftParen)
let expressions = RawLabeledExprListSyntax(
elements: self.parseArgumentListElements(pattern: .none),
arena: self.arena
)
// For recovery, eat anything up to the next token that either starts a new string segment or terminates the string.
// This allows us to skip over extraneous identifiers etc. in an unterminated string interpolation.
var unexpectedBeforeRightParen: [RawTokenSyntax] = []
var unexpectedProgress = LoopProgressCondition()
while !self.at(.rightParen, .stringSegment, .backslash) && !self.at(TokenSpec(openQuoteKind), .endOfFile)
&& self.hasProgressed(&unexpectedProgress)
{
unexpectedBeforeRightParen.append(self.consumeAnyToken())
}
// Consume the right paren if present, ensuring that it's on the same
// line if this is a single-line literal. Leading trivia is fine as
// we allow e.g "\(foo )".
let rightParen: Token
if self.at(.rightParen) && self.atStartOfLine && openQuote.tokenKind != .multilineStringQuote {
rightParen = missingToken(.rightParen)
} else {
rightParen = self.expectWithoutRecovery(.rightParen)
}
if case .inStringInterpolation = self.currentToken.cursor.currentState {
// The parser has more knowledge that we have reached the end of the
// string interpolation now, even if we haven't seen the closing ')'.
// For example, consider the following code
// "\(abc "
// Since the lexer doesn't know anything about the expression structure,
// it assumes that the `"` starts a new string literal. But since we
// know in the parser that an identifier cannot be followed by a string
// literal without a connecting binary operator and can thus consider
// it as the surrounding string literal end, which thus also terminates
// the string interpolation.
self.lexemes.perform(stateTransition: .pop, currentToken: &self.currentToken)
}
segments.append(
.expressionSegment(
RawExpressionSegmentSyntax(
backslash: backslash,
unexpectedBeforeDelimiter,
pounds: delimiter,
leftParen: leftParen,
expressions: expressions,
RawUnexpectedNodesSyntax(unexpectedBeforeRightParen, arena: self.arena),
rightParen: rightParen,
arena: self.arena
)
)
)
} else {
break
}
}
/// Parse close quote.
let unexpectedBeforeClosingQuote: RawUnexpectedNodesSyntax?
let closingQuote: RawTokenSyntax
if openQuoteKind == .singleQuote {
let singleQuote = self.expectWithoutRecoveryOrLeadingTrivia(.singleQuote)
unexpectedBeforeClosingQuote = RawUnexpectedNodesSyntax([singleQuote], arena: self.arena)
closingQuote = missingToken(.stringQuote)
} else {
unexpectedBeforeClosingQuote = nil
closingQuote = self.expectWithoutRecoveryOrLeadingTrivia(TokenSpec(openQuote.tokenKind))
}
let (unexpectedBeforeClosingPounds, closingPounds) = self.parsePoundDelimiter(
.rawStringPoundDelimiter,
matching: openingPounds
)
if openQuote.tokenKind == .multilineStringQuote, !openQuote.isMissing, !closingQuote.isMissing {
let postProcessed = postProcessMultilineStringLiteral(
rawStringDelimitersToken: openingPounds,
openQuote: openQuote,
segments: segments,
closeQuote: closingQuote
)
return RawStringLiteralExprSyntax(
openingPounds: openingPounds,
RawUnexpectedNodesSyntax(
combining: unexpectedBeforeOpeningQuote,
postProcessed.unexpectedBeforeOpeningQuote,
arena: self.arena
),
openingQuote: postProcessed.openingQuote,
segments: RawStringLiteralSegmentListSyntax(elements: postProcessed.segments, arena: self.arena),
RawUnexpectedNodesSyntax(
combining: postProcessed.unexpectedBeforeClosingQuote,
unexpectedBeforeClosingQuote,
arena: self.arena
),
closingQuote: postProcessed.closingQuote,
unexpectedBeforeClosingPounds,
closingPounds: closingPounds,
arena: self.arena
)
} else {
return RawStringLiteralExprSyntax(
openingPounds: openingPounds,
unexpectedBeforeOpeningQuote,
openingQuote: openQuote,
segments: RawStringLiteralSegmentListSyntax(elements: segments, arena: self.arena),
unexpectedBeforeClosingQuote,
closingQuote: closingQuote,
unexpectedBeforeClosingPounds,
closingPounds: closingPounds,
arena: self.arena
)
}
}
mutating func parseSimpleString() -> RawSimpleStringLiteralExprSyntax {
let openDelimiter = self.consume(if: .rawStringPoundDelimiter)
let (unexpectedBeforeOpenQuote, openQuote) = self.expect(
anyIn: SimpleStringLiteralExprSyntax.OpeningQuoteOptions.self,
default: .stringQuote
)
/// Parse segments.
var segments: [RawStringSegmentSyntax] = []
var loopProgress = LoopProgressCondition()
while hasProgressed(&loopProgress) {
// If we encounter a token with leading trivia, we're no longer in the
// string literal.
guard currentToken.leadingTriviaText.isEmpty else { break }
if let stringSegment = self.consume(if: .stringSegment, TokenSpec(.identifier, remapping: .stringSegment)) {
var unexpectedAfterContent: RawUnexpectedNodesSyntax?
if let (backslash, leftParen) = self.consume(if: .backslash, followedBy: .leftParen) {
var unexpectedTokens: [RawSyntax] = [RawSyntax(backslash), RawSyntax(leftParen)]
let (unexpectedBeforeRightParen, rightParen) = self.expect(TokenSpec(.rightParen, allowAtStartOfLine: false))
unexpectedTokens += unexpectedBeforeRightParen?.elements ?? []
unexpectedTokens.append(RawSyntax(rightParen))
unexpectedAfterContent = RawUnexpectedNodesSyntax(
unexpectedTokens,
arena: self.arena
)
}
segments.append(RawStringSegmentSyntax(content: stringSegment, unexpectedAfterContent, arena: self.arena))
} else {
break
}
}
let (unexpectedBetweenSegmentAndCloseQuote, closeQuote) = self.expect(
anyIn: SimpleStringLiteralExprSyntax.ClosingQuoteOptions.self,
default: openQuote.closeTokenKind
)
let closeDelimiter = self.consume(if: .rawStringPoundDelimiter)
if openQuote.tokenKind == .multilineStringQuote, !openQuote.isMissing, !closeQuote.isMissing {
let postProcessed = postProcessMultilineStringLiteral(
rawStringDelimitersToken: openDelimiter,
openQuote: openQuote,
segments: segments.compactMap { RawStringLiteralSegmentListSyntax.Element.stringSegment($0) },
closeQuote: closeQuote
)
return RawSimpleStringLiteralExprSyntax(
RawUnexpectedNodesSyntax(
combining: openDelimiter,
unexpectedBeforeOpenQuote,
postProcessed.unexpectedBeforeOpeningQuote,
arena: self.arena
),
openingQuote: postProcessed.openingQuote,
segments: RawSimpleStringLiteralSegmentListSyntax(
// `RawSimpleStringLiteralSegmentListSyntax` only accepts `RawStringSegmentSyntax`.
// So we can safely cast.
elements: postProcessed.segments.map { $0.cast(RawStringSegmentSyntax.self) },
arena: self.arena
),
RawUnexpectedNodesSyntax(
combining: unexpectedBetweenSegmentAndCloseQuote,
postProcessed.unexpectedBeforeClosingQuote,
arena: self.arena
),
closingQuote: postProcessed.closingQuote,
RawUnexpectedNodesSyntax(
[closeDelimiter],
arena: self.arena
),
arena: self.arena
)
} else {
return RawSimpleStringLiteralExprSyntax(
RawUnexpectedNodesSyntax(combining: unexpectedBeforeOpenQuote, openDelimiter, arena: self.arena),
openingQuote: openQuote,
segments: RawSimpleStringLiteralSegmentListSyntax(elements: segments, arena: self.arena),
unexpectedBetweenSegmentAndCloseQuote,
closingQuote: closeQuote,
RawUnexpectedNodesSyntax([closeDelimiter], arena: self.arena),
arena: self.arena
)
}
}
}
// MARK: - Utilities
fileprivate extension RawTokenSyntax {
var closeTokenKind: SimpleStringLiteralExprSyntax.ClosingQuoteOptions {
switch self {
case .multilineStringQuote:
return .multilineStringQuote
case .stringQuote:
return .stringQuote
default:
return .stringQuote
}
}
}
fileprivate extension SyntaxText {
private func hasSuffix(_ other: String) -> Bool {
var other = other
return other.withSyntaxText { self.hasSuffix($0) }
}
}
fileprivate extension RawTriviaPiece {
var isBackslash: Bool {
switch self {
case .backslashes: return true
default: return false
}
}
}
|