Index: lib/Format/WhitespaceManager.h =================================================================== --- lib/Format/WhitespaceManager.h +++ lib/Format/WhitespaceManager.h @@ -106,11 +106,11 @@ /// \p StartOfTokenColumn and \p InPPDirective will be used to lay out /// trailing comments and escaped newlines. Change(bool CreateReplacement, SourceRange OriginalWhitespaceRange, - unsigned IndentLevel, int Spaces, unsigned StartOfTokenColumn, - unsigned NewlinesBefore, StringRef PreviousLinePostfix, - StringRef CurrentLinePrefix, tok::TokenKind Kind, - bool ContinuesPPDirective, bool IsStartOfDeclName, - bool IsInsideToken); + unsigned IndentLevel, unsigned NestingLevel, int Spaces, + unsigned StartOfTokenColumn, unsigned NewlinesBefore, + StringRef PreviousLinePostfix, StringRef CurrentLinePrefix, + tok::TokenKind Kind, TokenType Type, bool ContinuesPPDirective, + bool IsStartOfDeclName, bool IsInsideToken); bool CreateReplacement; // Changes might be in the middle of a token, so we cannot just keep the @@ -125,13 +125,19 @@ // FIXME: Currently this is not set correctly for breaks inside comments, as // the \c BreakableToken is still doing its own alignment. tok::TokenKind Kind; + // Same comment applies as for Kind + TokenType Type; bool ContinuesPPDirective; bool IsStartOfDeclName; - // The number of nested blocks the token is in. This is used to add tabs + // A combination of nesting level and indent level. + // The nesting level of this token, i.e. the number of surrounding (), + // [], {} or <>. Note that the value stored here is slightly different + // to the NestingLevel passed in. See propagateIndentLevels() for details. + // The second element in the pair, the indent level, is used to add tabs // only for the indentation, and not for alignment, when // UseTab = US_ForIndentation. - unsigned IndentLevel; + std::pair NestingAndIndentLevel; // The number of spaces in front of the token or broken part of the token. // This will be adapted when aligning tokens. @@ -170,6 +176,10 @@ /// \c EscapedNewlineColumn for the first tokens or token parts in a line. void calculateLineBreakInformation(); + /// \brief Propagate IndentLevel from the first token on a line, to all of + /// the tokens on that line. + void propagateIndentLevels(); + /// \brief Align consecutive assignments over all \c Changes. void alignConsecutiveAssignments(); @@ -202,6 +212,10 @@ void appendIndentText(std::string &Text, unsigned IndentLevel, unsigned Spaces, unsigned WhitespaceStartColumn); + /// \brief Determine whether this token is considered as the start of a + /// declaration name. + static bool IsStartOfDeclName(const FormatToken &Tok); + SmallVector Changes; const SourceManager &SourceMgr; tooling::Replacements Replaces; Index: lib/Format/WhitespaceManager.cpp =================================================================== --- lib/Format/WhitespaceManager.cpp +++ lib/Format/WhitespaceManager.cpp @@ -27,19 +27,21 @@ WhitespaceManager::Change::Change( bool CreateReplacement, SourceRange OriginalWhitespaceRange, - unsigned IndentLevel, int Spaces, unsigned StartOfTokenColumn, - unsigned NewlinesBefore, StringRef PreviousLinePostfix, - StringRef CurrentLinePrefix, tok::TokenKind Kind, bool ContinuesPPDirective, + unsigned IndentLevel, unsigned NestingLevel, int Spaces, + unsigned StartOfTokenColumn, unsigned NewlinesBefore, + StringRef PreviousLinePostfix, StringRef CurrentLinePrefix, + tok::TokenKind Kind, TokenType Type, bool ContinuesPPDirective, bool IsStartOfDeclName, bool IsInsideToken) : CreateReplacement(CreateReplacement), OriginalWhitespaceRange(OriginalWhitespaceRange), StartOfTokenColumn(StartOfTokenColumn), NewlinesBefore(NewlinesBefore), PreviousLinePostfix(PreviousLinePostfix), - CurrentLinePrefix(CurrentLinePrefix), Kind(Kind), + CurrentLinePrefix(CurrentLinePrefix), Kind(Kind), Type(Type), ContinuesPPDirective(ContinuesPPDirective), - IsStartOfDeclName(IsStartOfDeclName), IndentLevel(IndentLevel), - Spaces(Spaces), IsInsideToken(IsInsideToken), IsTrailingComment(false), - TokenLength(0), PreviousEndOfTokenColumn(0), EscapedNewlineColumn(0), + IsStartOfDeclName(IsStartOfDeclName), + NestingAndIndentLevel(NestingLevel, IndentLevel), Spaces(Spaces), + IsInsideToken(IsInsideToken), IsTrailingComment(false), TokenLength(0), + PreviousEndOfTokenColumn(0), EscapedNewlineColumn(0), StartOfBlockComment(nullptr), IndentationOffset(0) {} void WhitespaceManager::reset() { @@ -56,10 +58,9 @@ Tok.Decision = (Newlines > 0) ? FD_Break : FD_Continue; Changes.push_back( Change(/*CreateReplacement=*/true, Tok.WhitespaceRange, IndentLevel, - Spaces, StartOfTokenColumn, Newlines, "", "", Tok.Tok.getKind(), - InPPDirective && !Tok.IsFirst, - Tok.is(TT_StartOfName) || Tok.is(TT_FunctionDeclarationName), - /*IsInsideToken=*/false)); + Tok.NestingLevel, Spaces, StartOfTokenColumn, Newlines, "", "", + Tok.Tok.getKind(), Tok.Type, InPPDirective && !Tok.IsFirst, + IsStartOfDeclName(Tok), /*IsInsideToken=*/false)); } void WhitespaceManager::addUntouchableToken(const FormatToken &Tok, @@ -68,10 +69,10 @@ return; Changes.push_back(Change( /*CreateReplacement=*/false, Tok.WhitespaceRange, /*IndentLevel=*/0, + Tok.NestingLevel, /*Spaces=*/0, Tok.OriginalColumn, Tok.NewlinesBefore, "", "", - Tok.Tok.getKind(), InPPDirective && !Tok.IsFirst, - Tok.is(TT_StartOfName) || Tok.is(TT_FunctionDeclarationName), - /*IsInsideToken=*/false)); + Tok.Tok.getKind(), Tok.Type, InPPDirective && !Tok.IsFirst, + IsStartOfDeclName(Tok), /*IsInsideToken=*/false)); } void WhitespaceManager::replaceWhitespaceInToken( @@ -81,13 +82,13 @@ if (Tok.Finalized) return; SourceLocation Start = Tok.getStartOfNonWhitespace().getLocWithOffset(Offset); - Changes.push_back(Change( - true, SourceRange(Start, Start.getLocWithOffset(ReplaceChars)), - IndentLevel, Spaces, std::max(0, Spaces), Newlines, PreviousPostfix, - CurrentPrefix, Tok.is(TT_LineComment) ? tok::comment : tok::unknown, - InPPDirective && !Tok.IsFirst, - Tok.is(TT_StartOfName) || Tok.is(TT_FunctionDeclarationName), - /*IsInsideToken=*/Newlines == 0)); + Changes.push_back( + Change(true, SourceRange(Start, Start.getLocWithOffset(ReplaceChars)), + IndentLevel, Tok.NestingLevel, Spaces, std::max(0, Spaces), + Newlines, PreviousPostfix, CurrentPrefix, + Tok.is(TT_LineComment) ? tok::comment : tok::unknown, Tok.Type, + InPPDirective && !Tok.IsFirst, IsStartOfDeclName(Tok), + /*IsInsideToken=*/Newlines == 0)); } const tooling::Replacements &WhitespaceManager::generateReplacements() { @@ -95,6 +96,7 @@ return Replaces; std::sort(Changes.begin(), Changes.end(), Change::IsBeforeInFile(SourceMgr)); + propagateIndentLevels(); calculateLineBreakInformation(); alignConsecutiveDeclarations(); alignConsecutiveAssignments(); @@ -160,6 +162,41 @@ } } +struct TokenTypeAndLevel { + TokenType Type; + std::pair NestingAndIndentLevel; +}; + +// Upon entry, IndentLevel is only set on the first token of the line. +// Here we propagate IndentLevel to every token on the line. +// In addition, we tweak the NestingLevel slightly, so that it works naturally +// with the way we recursively process scope depths when +// performing alignment. +void WhitespaceManager::propagateIndentLevels() { + unsigned IndentLevel = + Changes.size() > 0 ? Changes[0].NestingAndIndentLevel.second : 0; + for (auto &Change : Changes) { + if (Change.NewlinesBefore != 0) + IndentLevel = Change.NestingAndIndentLevel.second; + Change.NestingAndIndentLevel.second = IndentLevel; + + // NestingLevel is raised on the opening paren/square, and remains raised + // until AFTER the closing paren/square. This means that with a statement + // like: + // + // for (int i = 0; ...) + // int x = 1;. + // + // The "int x = 1" line is going to have the same NestingLevel as + // the tokens inside the parentheses of the "for" statement. + // In order to break this continuity, we force the closing paren/square to + // reduce it's nesting level. + // If we don't do this, then "x = 1" ends up getting aligned to "i = 1" + if (Change.Kind == tok::r_paren || Change.Kind == tok::r_square) + Change.NestingAndIndentLevel.first--; + } +} + // Align a single sequence of tokens, see AlignTokens below. template static void @@ -167,21 +204,54 @@ SmallVector &Changes) { bool FoundMatchOnLine = false; int Shift = 0; + + // ScopeStack keeps track of the current scope depth. + // We only run the "Matches" function on tokens from the outer-most scope. + // However, we do need to adjust some special tokens within the entire + // Start..End range, regardless of their scope. This special rule applies + // only to function parameters which are split across multiple lines, and + // who's function names have been shifted for declaration alignment's sake. + // See "split function parameter alignment" in FormatTest.cpp for an example. + SmallVector ScopeStack; + for (unsigned i = Start; i != End; ++i) { - if (Changes[i].NewlinesBefore > 0) { - FoundMatchOnLine = false; + if (ScopeStack.size() != 0 && + Changes[i].NestingAndIndentLevel < + ScopeStack.back().NestingAndIndentLevel) + ScopeStack.pop_back(); + + if (i != Start) { + if (Changes[i].NestingAndIndentLevel > + Changes[i - 1].NestingAndIndentLevel) { + ScopeStack.push_back( + {Changes[i - 1].Type, Changes[i].NestingAndIndentLevel}); + if (i > Start + 1 && Changes[i - 2].Type == TT_FunctionDeclarationName) + ScopeStack.back().Type = Changes[i - 2].Type; + } + } + + bool InsideNestedScope = ScopeStack.size() != 0; + + if (Changes[i].NewlinesBefore > 0 && !InsideNestedScope) { Shift = 0; + FoundMatchOnLine = false; } // If this is the first matching token to be aligned, remember by how many // spaces it has to be shifted, so the rest of the changes on the line are // shifted by the same amount - if (!FoundMatchOnLine && Matches(Changes[i])) { + if (!FoundMatchOnLine && !InsideNestedScope && Matches(Changes[i])) { FoundMatchOnLine = true; Shift = Column - Changes[i].StartOfTokenColumn; Changes[i].Spaces += Shift; } + // This is for function parameters that are split across multiple lines, + // as mentioned in the ScopeStack comment. + if (InsideNestedScope && Changes[i].NewlinesBefore > 0 && + ScopeStack.back().Type == TT_FunctionDeclarationName) + Changes[i].Spaces += Shift; + assert(Shift >= 0); Changes[i].StartOfTokenColumn += Shift; if (i + 1 != Changes.size()) @@ -189,15 +259,37 @@ } } -// Walk through all of the changes and find sequences of matching tokens to -// align. To do so, keep track of the lines and whether or not a matching token -// was found on a line. If a matching token is found, extend the current -// sequence. If the current line cannot be part of a sequence, e.g. because -// there is an empty line before it or it contains only non-matching tokens, -// finalize the previous sequence. +// Walk through a subset of the changes, starting at StartAt, and find +// sequences of matching tokens to align. To do so, keep track of the lines and +// whether or not a matching token was found on a line. If a matching token is +// found, extend the current sequence. If the current line cannot be part of a +// sequence, e.g. because there is an empty line before it or it contains only +// non-matching tokens, finalize the previous sequence. +// The value returned is the token on which we stopped, either because we +// exhausted all items inside Changes, or because we hit a scope level higher +// than our initial scope. +// This function is recursive. Each invocation processes only the scope level +// equal to the initial level, which is the level of Changes[StartAt]. +// If we encounter a scope level greater than the initial level, then we call +// ourselves recursively, thereby avoiding the pollution of the current state +// with the alignment requirements of the nested sub-level. This recursive +// behavior is necessary for aligning function prototypes that have one or more +// arguments. +// If this function encounters a scope level less than the initial level, +// it exits. +// There is a non-obvious subtlety in the recursive behavior: Even though we +// defer processing of nested levels to recursive invocations of this +// function, when it comes time to align a sequence of tokens, we run the +// alignment on the entire sequence, including the nested levels. +// When doing so, most of the nested tokens are skipped, because their +// alignment was already handled by the recursive invocations of this function. +// However, the special exception is that we do NOT skip function parameters +// that are split across multiple lines. See the test case in FormatTest.cpp +// that mentions "split function parameter alignment" for an example of this. template -static void AlignTokens(const FormatStyle &Style, F &&Matches, - SmallVector &Changes) { +static unsigned AlignTokens(const FormatStyle &Style, F &&Matches, + SmallVector &Changes, + unsigned StartAt) { unsigned MinColumn = 0; unsigned MaxColumn = UINT_MAX; @@ -205,14 +297,11 @@ unsigned StartOfSequence = 0; unsigned EndOfSequence = 0; - // Keep track of the nesting level of matching tokens, i.e. the number of - // surrounding (), [], or {}. We will only align a sequence of matching - // token that share the same scope depth. - // - // FIXME: This could use FormatToken::NestingLevel information, but there is - // an outstanding issue wrt the brace scopes. - unsigned NestingLevelOfLastMatch = 0; - unsigned NestingLevel = 0; + // Measure the scope level (i.e. depth of (), [], {}) of the first token, and + // abort when we hit any token in a higher scope than the starting one. + auto NestingAndIndentLevel = StartAt < Changes.size() + ? Changes[StartAt].NestingAndIndentLevel + : std::pair(0, 0); // Keep track of the number of commas before the matching tokens, we will only // align a sequence of matching tokens if they are preceded by the same number @@ -240,7 +329,11 @@ EndOfSequence = 0; }; - for (unsigned i = 0, e = Changes.size(); i != e; ++i) { + unsigned i = StartAt; + for (unsigned e = Changes.size(); + i != e && Changes[i].NestingAndIndentLevel >= NestingAndIndentLevel; + ++i) { + if (Changes[i].NewlinesBefore != 0) { CommasBeforeMatch = 0; EndOfSequence = i; @@ -254,17 +347,11 @@ if (Changes[i].Kind == tok::comma) { ++CommasBeforeMatch; - } else if (Changes[i].Kind == tok::r_brace || - Changes[i].Kind == tok::r_paren || - Changes[i].Kind == tok::r_square) { - --NestingLevel; - } else if (Changes[i].Kind == tok::l_brace || - Changes[i].Kind == tok::l_paren || - Changes[i].Kind == tok::l_square) { - // We want sequences to skip over child scopes if possible, but not the - // other way around. - NestingLevelOfLastMatch = std::min(NestingLevelOfLastMatch, NestingLevel); - ++NestingLevel; + } else if (Changes[i].NestingAndIndentLevel != NestingAndIndentLevel) { + // Call AlignTokens recursively, skipping over this scope block. + unsigned StoppedAt = AlignTokens(Style, Matches, Changes, i); + i = StoppedAt - 1; + continue; } if (!Matches(Changes[i])) @@ -273,12 +360,10 @@ // If there is more than one matching token per line, or if the number of // preceding commas, or the scope depth, do not match anymore, end the // sequence. - if (FoundMatchOnLine || CommasBeforeMatch != CommasBeforeLastMatch || - NestingLevel != NestingLevelOfLastMatch) + if (FoundMatchOnLine || CommasBeforeMatch != CommasBeforeLastMatch) AlignCurrentSequence(); CommasBeforeLastMatch = CommasBeforeMatch; - NestingLevelOfLastMatch = NestingLevel; FoundMatchOnLine = true; if (StartOfSequence == 0) @@ -301,27 +386,27 @@ MaxColumn = std::min(MaxColumn, ChangeMaxColumn); } - EndOfSequence = Changes.size(); + unsigned StoppedAt = i; + EndOfSequence = i; AlignCurrentSequence(); + return StoppedAt; } void WhitespaceManager::alignConsecutiveAssignments() { if (!Style.AlignConsecutiveAssignments) return; - AlignTokens(Style, - [&](const Change &C) { - // Do not align on equal signs that are first on a line. - if (C.NewlinesBefore > 0) - return false; + AlignTokens(Style, [&](const Change &C) { + // Do not align on equal signs that are first on a line. + if (C.NewlinesBefore > 0) + return false; - // Do not align on equal signs that are last on a line. - if (&C != &Changes.back() && (&C + 1)->NewlinesBefore > 0) - return false; + // Do not align on equal signs that are last on a line. + if (&C != &Changes.back() && (&C + 1)->NewlinesBefore > 0) + return false; - return C.Kind == tok::equal; - }, - Changes); + return C.Kind == tok::equal; + }, Changes, 0); } void WhitespaceManager::alignConsecutiveDeclarations() { @@ -334,9 +419,8 @@ // const char* const* v1; // float const* v2; // SomeVeryLongType const& v3; - AlignTokens(Style, [](Change const &C) { return C.IsStartOfDeclName; }, - Changes); + Changes, 0); } void WhitespaceManager::alignTrailingComments() { @@ -372,8 +456,7 @@ unsigned CommentColumn = SourceMgr.getSpellingColumnNumber( Changes[i].OriginalWhitespaceRange.getEnd()); for (unsigned j = i + 1; j != e; ++j) { - if (Changes[j].Kind == tok::comment || - Changes[j].Kind == tok::unknown) + if (Changes[j].Kind == tok::comment || Changes[j].Kind == tok::unknown) // Skip over comments and unknown tokens. "unknown tokens are used for // the continuation of multiline comments. continue; @@ -486,7 +569,8 @@ C.PreviousEndOfTokenColumn, C.EscapedNewlineColumn); else appendNewlineText(ReplacementText, C.NewlinesBefore); - appendIndentText(ReplacementText, C.IndentLevel, std::max(0, C.Spaces), + appendIndentText(ReplacementText, C.NestingAndIndentLevel.second, + std::max(0, C.Spaces), C.StartOfTokenColumn - std::max(0, C.Spaces)); ReplacementText.append(C.CurrentLinePrefix); storeReplacement(C.OriginalWhitespaceRange, ReplacementText); @@ -494,8 +578,7 @@ } } -void WhitespaceManager::storeReplacement(SourceRange Range, - StringRef Text) { +void WhitespaceManager::storeReplacement(SourceRange Range, StringRef Text) { unsigned WhitespaceLength = SourceMgr.getFileOffset(Range.getEnd()) - SourceMgr.getFileOffset(Range.getBegin()); // Don't create a replacement, if it does not change anything. @@ -574,5 +657,10 @@ } } +bool WhitespaceManager::IsStartOfDeclName(const FormatToken &Tok) { + return Tok.is(TT_StartOfName) || Tok.is(TT_FunctionDeclarationName) || + Tok.is(tok::kw_operator); +} + } // namespace format } // namespace clang Index: unittests/Format/FormatTest.cpp =================================================================== --- unittests/Format/FormatTest.cpp +++ unittests/Format/FormatTest.cpp @@ -9250,12 +9250,11 @@ "};", Alignment); - // FIXME: Should align all three assignments verifyFormat( "int i = 1;\n" "SomeType a = SomeFunction(looooooooooooooooooooooongParameterA,\n" " loooooooooooooooooooooongParameterB);\n" - "int j = 2;", + "int j = 2;", Alignment); verifyFormat("template