1 //===--- FormatTokenLexer.h - Format C++ code ----------------*- C++ ----*-===//
2 //
3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4 // See https://llvm.org/LICENSE.txt for license information.
5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6 //
7 //===----------------------------------------------------------------------===//
8 ///
9 /// \file
10 /// This file contains FormatTokenLexer, which tokenizes a source file
11 /// into a token stream suitable for ClangFormat.
12 ///
13 //===----------------------------------------------------------------------===//
14 
15 #ifndef LLVM_CLANG_LIB_FORMAT_FORMATTOKENLEXER_H
16 #define LLVM_CLANG_LIB_FORMAT_FORMATTOKENLEXER_H
17 
18 #include "Encoding.h"
19 #include "FormatToken.h"
20 #include "clang/Basic/SourceLocation.h"
21 #include "clang/Basic/SourceManager.h"
22 #include "clang/Format/Format.h"
23 #include "llvm/ADT/MapVector.h"
24 #include "llvm/Support/Regex.h"
25 
26 #include <stack>
27 
28 namespace clang {
29 namespace format {
30 
31 enum LexerState {
32   NORMAL,
33   TEMPLATE_STRING,
34   TOKEN_STASHED,
35 };
36 
37 class FormatTokenLexer {
38 public:
39   FormatTokenLexer(const SourceManager &SourceMgr, FileID ID, unsigned Column,
40                    const FormatStyle &Style, encoding::Encoding Encoding);
41 
42   ArrayRef<FormatToken *> lex();
43 
44   const AdditionalKeywords &getKeywords() { return Keywords; }
45 
46 private:
47   void tryMergePreviousTokens();
48 
49   bool tryMergeLessLess();
50   bool tryMergeNSStringLiteral();
51   bool tryMergeJSPrivateIdentifier();
52   bool tryMergeCSharpVerbatimStringLiteral();
53   bool tryMergeCSharpKeywordVariables();
54   bool tryMergeCSharpNullConditionals();
55   bool tryMergeCSharpDoubleQuestion();
56   bool tryTransformCSharpForEach();
57 
58   bool tryMergeTokens(ArrayRef<tok::TokenKind> Kinds, TokenType NewType);
59 
60   // Returns \c true if \p Tok can only be followed by an operand in JavaScript.
61   bool precedesOperand(FormatToken *Tok);
62 
63   bool canPrecedeRegexLiteral(FormatToken *Prev);
64 
65   // Tries to parse a JavaScript Regex literal starting at the current token,
66   // if that begins with a slash and is in a location where JavaScript allows
67   // regex literals. Changes the current token to a regex literal and updates
68   // its text if successful.
69   void tryParseJSRegexLiteral();
70 
71   // Handles JavaScript template strings.
72   //
73   // JavaScript template strings use backticks ('`') as delimiters, and allow
74   // embedding expressions nested in ${expr-here}. Template strings can be
75   // nested recursively, i.e. expressions can contain template strings in turn.
76   //
77   // The code below parses starting from a backtick, up to a closing backtick or
78   // an opening ${. It also maintains a stack of lexing contexts to handle
79   // nested template parts by balancing curly braces.
80   void handleTemplateStrings();
81 
82   void tryParsePythonComment();
83 
84   bool tryMerge_TMacro();
85 
86   bool tryMergeConflictMarkers();
87 
88   FormatToken *getStashedToken();
89 
90   FormatToken *getNextToken();
91 
92   FormatToken *FormatTok;
93   bool IsFirstToken;
94   std::stack<LexerState> StateStack;
95   unsigned Column;
96   unsigned TrailingWhitespace;
97   std::unique_ptr<Lexer> Lex;
98   const SourceManager &SourceMgr;
99   FileID ID;
100   const FormatStyle &Style;
101   IdentifierTable IdentTable;
102   AdditionalKeywords Keywords;
103   encoding::Encoding Encoding;
104   llvm::SpecificBumpPtrAllocator<FormatToken> Allocator;
105   // Index (in 'Tokens') of the last token that starts a new line.
106   unsigned FirstInLineIndex;
107   SmallVector<FormatToken *, 16> Tokens;
108 
109   llvm::SmallMapVector<IdentifierInfo *, TokenType, 8> Macros;
110 
111   bool FormattingDisabled;
112 
113   llvm::Regex MacroBlockBeginRegex;
114   llvm::Regex MacroBlockEndRegex;
115 
116   void readRawToken(FormatToken &Tok);
117 
118   void resetLexer(unsigned Offset);
119 };
120 
121 } // namespace format
122 } // namespace clang
123 
124 #endif
125